METHOD AND APPARATUS FOR TRANSMITTING AND RECEIVING POLAR CODE

Abstract

A 5th generation (5G) or 6th generation (6G) communication system for supporting a higher data rate than a beyond 4th generation (4G) communication system, such as long-term evolution (LTE) is provided. An electronic device performs interleaving on codeword bits, according to a polar code sequence corresponding to the codeword bits, allocates the interleaved codeword bits respectively to blocks, and transmits at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocation, on at least one resource block corresponding to each of the blocks.

Description

JOINT RESEARCH AGREEMENT

The disclosure was made by or on behalf of the below listed parties to a joint research agreement. The joint research agreement was in effect on or before the date the disclosure was made and the disclosure was made as a result of activities undertaken within the scope of the joint research agreement. The parties to the joint research agreement are 1) Samsung Electronics Co., Ltd. and 2) Research & Business Foundation Sungkyunkwan University.

BACKGROUND

1. Field

The disclosure relates to a technology for transmission and reception of a polar code. More particularly, the disclosure relates to a method and apparatus for performing interleaving and deinterleaving for transmission and reception of a polar code.

2. Description of the Related Art

Looking back on processes of development over generations of wireless communication, technologies for human-targeted services, such as voice, multimedia, and data, have been mainly developed. It is expected that connected devices, which have been explosively increasing after the commercialization of 5th generation (5G) communication systems, will be connected to communication networks. Examples of an object connected to a network may include a vehicle, a robot, a drone, a home appliance, a display, smart sensors installed in various infrastructures, a construction machine, and factory equipment. It is expected that mobile devices may evolve into various form factors, such as augmented reality glasses, an augmented reality headset, and a hologram device. In the 6th generation (6G) era, efforts are being made to develop enhanced 6G communication systems so as to provide various services by connecting hundreds of billions of devices and objects to each other. In this regard, 6G communication systems are referred to as beyond 5G communication systems.

The 6G communication systems, which are expected to be realized around 2030, have a maximum data rate of tera (i.e., 1,000 gigabytes) bps and a wireless latency time of 100 microseconds (μsec). In other words, compared to the 5G communication systems, a data rate is increased 50 times and a wireless latency time is reduced to 1/10 in the 6G communication systems.

To achieve such a high data rate and an ultra low latency time, the 6G communication systems are considered to be realized in terahertz (THz) bands (e.g., from 95 gigahertz (GHz) to 3 THz). In the THz bands, it is expected that the importance of a technology for guaranteeing a signal arrival distance, i.e., coverage, will increase due to a severer path loss and atmosphere absorption phenomenon compared to millimeter wave (mmWave) bands introduced in 5G. As a main technology for guaranteeing the coverage, a multiple antenna transmission technology, such as new waveform, beamforming, massive multiple-input and multiple-output (MIMO), full dimensional (FD) MIMO, array antenna, or large scale antenna, which is superior in terms of coverage compared to a radio frequency (RF) device, antenna, and orthogonal frequency division multiplexing (OFDM), needs to be developed. In addition, to improve the coverage of THz band signals, new technologies, such as a metamaterial-based lens and antenna, a high-dimensional space multiplexing technology using orbital angular momentum (OAM), and a reconfigurable intelligent surface (RIS), are being discussed.

Also, to enhance frequency efficiency and improve system networks, in the 6G communication systems, a full duplex technology in which an uplink and a downlink simultaneously use a same frequency resource at a same time, a network technology integrally using a satellite and high-altitude platform stations (HAPSs), a network structure innovative technology supporting a mobile base station and enabling optimization and automation of a network operation, a dynamic spectrum sharing technology through spectrum usage prediction-based collision avoidance, an artificial intelligence (AI)-based communication technology that uses AI from a design stage and realizes system optimization by including an end-to-end AI support function, and a next-generation distributed computing technology that realizes a complex service exceeding the limit of terminal computing capability by using ultra-high performance communication and computing resources (mobile edge computing (MEC), cloud, and the like) are being developed. Moreover, attempts are continued to further strengthen connectivity between devices, further optimize a network, accelerate softwarization of a network entity, and increase openness of wireless communication, through a design of a new protocol to be used in the 6G communication systems, mechanism development for implementing a hardware-based security environment and safely using data, and technology development related to privacy maintaining method.

According to studies on and development of such 6G communication systems, it is expected that a new level of the next hyper-connected experience may become possible through hyper-connectivity of the 6G communication systems including not only a connection between things, but also a connection between a person and a thing. In particular, it is expected that services, such as truly immersive extended reality (XR), high-fidelity mobile hologram, and digital replica, may be provided through the 6G communication systems. In addition, services, such as remote surgery, industrial automation, and emergency response, through security and reliability enhancement may be provided through the 6G communication systems and applied to various fields, such as industry, medical care, vehicles, and home appliances.

The above information is presented as background information only to assist with an understanding of the disclosure. No determination has been made, and no assertion is made, as to whether any of the above might be applicable as prior art with regard to the disclosure.

SUMMARY

Aspects of the disclosure are to address at least the above-mentioned problems and/or disadvantages and to provide at least the advantages described below. Accordingly, an aspect of the disclosure is to provide an interleaving method, a resource allocation method, and a code construction method, which may increase a decoding performance in a block fading channel while maintaining a decoding performance in an additive white Gaussian noise channel, with respect to a bit-interleaved coded modulation (BICM) system to which a general polar code is applied.

Additional aspects will be set forth in part in the description which follows and, in part, will be apparent from the description, or may be learned by practice of the presented embodiments.

In accordance with an aspect of the disclosure, a method, performed by a first electronic device, of transmitting a polar code is provided. The method includes performing interleaving on codeword bits, according to a polar code sequence corresponding to the codeword bits, allocating the interleaved codeword bits respectively to blocks, and transmitting at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocating, on at least one resource block corresponding to each of the blocks.

In accordance with another aspect of the disclosure, a method, performed by a second electronic device, of receiving a polar code is provided. The method includes obtaining codeword bits by performing demodulation on a signal received from a first electronic device on each of at least one resource block, performing deinterleaving on the codeword bits by using a polar code sequence corresponding to the codeword bits, and performing decoding on a sequence including the deinterleaved codeword bits.

In accordance with another aspect of the disclosure, a first electronic device for transmitting a polar code is provided. The first electronic device includes a transceiver, and at least one processor connected to the transceiver and configured to perform interleaving on codeword bits, according to a polar code sequence corresponding to the codeword bits, allocate the interleaved codeword bits respectively to blocks, and transmit at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocation, on at least one resource block corresponding to each of the blocks.

In accordance with another aspect of the disclosure, a second electronic device for receiving a polar code is provided. The second electronic device includes a transceiver, and at least one processor connected to the transceiver and configured to obtain codeword bits by performing demodulation on a signal received from a first electronic device on each of at least one resource block, perform deinterleaving on the codeword bits by using a polar code sequence corresponding to the codeword bits, and perform decoding on a sequence including the deinterleaved codeword bits.

According to an embodiment of the disclosure, a computer-readable recording medium may have recorded thereon a program for executing, on a computer, the method according to at least one embodiment of the disclosure.

Other aspects, advantages, and salient features of the disclosure will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses various embodiments of the disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

The above and other aspects, features, and advantages of certain embodiments of the disclosure will be more apparent from the following description taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a diagram for describing channel combining and channel splitting of a polar code according to an embodiment of the disclosure;

FIG. 2 is a diagram for describing a sequence Q₁₀₂₃ and a sequence for N=512 , which are adopted in the 5th generation-new radio (5G-NR) standard according to an embodiment of the disclosure;

FIG. 3 is a diagram for describing puncturing and shortening of a polar code according to an embodiment of the disclosure;

FIG. 4 is a diagram for describing a 4-bit punctured polar code when N=8 according to an embodiment of the disclosure;

FIG. 5 is a diagram for describing a 4-bit shortened polar code when N=8 according to an embodiment of the disclosure;

FIG. 6 is a diagram of an example of decoding a polar code in which a length is N=8 on a bipartite graph according to an embodiment of the disclosure;

FIG. 7 is a diagram for describing a transmission system model using a polar code for 5G-NR uplink according to an embodiment of the disclosure;

FIG. 8 is a diagram for describing a method of performing interleaving in units of subblocks according to an embodiment of the disclosure;

FIG. 9 is a diagram for describing a bit-interleaved coded modulation (BICM) transmission/reception system model to which a polar code is applied according to an embodiment of the disclosure;

FIG. 10A is a diagram for describing a method by which logical resource blocks, to which an interleaving method is applied, are allocated to physical resource blocks according to frequency hopping according to an embodiment of the disclosure;

FIG. 10B is a diagram for describing allocating of logical resource blocks, to which an interleaving method is applied, to physical resource blocks according to an embodiment of the disclosure;

FIG. 11 is a diagram for describing a case where, when certain codeword bit patterns are simultaneously faded, reliability of a certain source bit is greatly reduced due to a structural feature of a polar code, and thus decoding becomes difficult according to an embodiment of the disclosure;

FIG. 12 is a diagram for describing a polar-coded BICM system transmitter and a BICM system transmitter during multi-block fading channel transmission according to an embodiment of the disclosure;

FIG. 13 is a diagram for describing a polar-coded BICM system receiver and a BICM system receiver during multi-block fading channel transmission according to an embodiment of the disclosure;

FIG. 14 is a diagram for describing a method by which a signal transmitted from an electronic device avoids a structural weakness and secures a high diversity by effectively using a non-homogeneous channel according to an embodiment of the disclosure;

FIG. 15 is a diagram for describing a process of updating a fading variable when updating a check node according to an embodiment of the disclosure;

FIG. 16 is a diagram for describing a process of updating a fading stochastic variable when updating a variable node according to an embodiment of the disclosure;

FIG. 17 is a diagram for describing updating of a fading stochastic variables included in source bits according to an embodiment of the disclosure;

FIG. 18 is a diagram for describing a change in a diversity for each bit, according to fading block allocation through suitable interleaving of a codeword bit according to an embodiment of the disclosure;

FIG. 19 is a diagram for analyzing a diversity according to an interleaving method and comparing a correlation between the diversity and a simulation performance, based on a Boolean approximation-based diversity analysis according to an embodiment of the disclosure;

FIG. 20 is a diagram for describing an interleaving method and effects, which guarantee a diversity, according to an embodiment of the disclosure;

FIG. 21 is a diagram for describing a block allocation method of modulated codeword bits, according to an embodiment of the disclosure;

FIG. 22 is a diagram for describing a block allocation method of modulated codeword bits, according to an embodiment of the disclosure;

FIG. 23 is a diagram for describing bit indexes arranged in an order of a code sequence C_N and diversity values corresponding thereto according to an embodiment of the disclosure;

FIG. 24 is a diagram showing reliabilities of bits for each of an interleaver and a triangular interleaver, in a descending order, according to an embodiment of the disclosure according to an embodiment of the disclosure;

FIG. 25 is a graph comparing required signal-to-noise ratio (SNR) [dB] performances achieving a block error rate (BLER) 0.1% of a code modified to have symmetry with an asymmetric 5G-NR code in an additive white Gaussian noise channel (AWGNC) and a block fading channel according to an embodiment of the disclosure;

FIG. 26 is a diagram for comparing the maximum numbers K_max of information bits guaranteeing diversity-2, according to an interleaving type and a code length according to an embodiment of the disclosure;

FIG. 27 is a diagram for describing a method of allocating a diversity, based on auxiliary sequences of a polar code mother code to which rate-matching is not applied and a polar code in which 6 bits are punctured according to an embodiment of the disclosure;

FIG. 28 is a diagram for describing a method of allocating a diversity according to auxiliary sequences of a polar code mother code without rate-matching and a polar code in which 2 bits are shortened according to an embodiment of the disclosure;

FIG. 29 is a diagram for describing a method of allocating modulated codeword bits to resource blocks by separately defining a sequence and an auxiliary sequence, and a method of allocating the modulated codeword bits by using a unified sequence, according to an embodiment of the disclosure;

FIG. 30 is a diagram for describing a diversity calculating method for a punctured polar code according to an embodiment of the disclosure;

FIG. 31 is a diagram for describing a diversity updating method for a shortened polar code according to an embodiment of the disclosure;

FIG. 32 is a graph showing a required SNR [dB] achieving BLER=0.1% according to the number of information bits, according to an interleaver type, under successive cancellation (SC) decoding according to an embodiment of the disclosure;

FIG. 33 is a graph for describing a performance of an interleaving method in an AWGNC, according to an embodiment of the disclosure;

FIG. 34 is a graph for describing a performance of an interleaving method when rate-matching is considered in an AWGNC according to an embodiment of the disclosure;

FIG. 35 is a graph for describing a performance of an interleaving method according to various decoding methods according to an embodiment of the disclosure;

FIG. 36 is a diagram showing diversities included in source bits in an order of a code sequence C_N, according to an interleaver type and the number of punctured bits according to an embodiment of the disclosure;

FIG. 37 is a graph for describing a decoding performance according to an interleaver type according to an embodiment of the disclosure;

FIG. 38 is a diagram showing diversities included in source bits in an order of a code sequence C_N, according to an interleaver type and the number of shortened bits according to an embodiment of the disclosure;

FIG. 39 is a graph for analyzing a shortening performance of a polar code according to an embodiment of the disclosure;

FIG. 40 is a block diagram of an electronic device for transmitting a signal to which an interleaving method is applied according to an embodiment of the disclosure; and

FIG. 41 is a block diagram of an electronic device for receiving a signal to which an interleaving method is applied according to an embodiment of the disclosure.

Throughout the drawings, it should be noted that like reference numbers are used to depict the same or similar elements, features, and structures.

DETAILED DESCRIPTION

The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of various embodiments of the disclosure as defined by the claims and their equivalents. It includes various specific details to assist in that understanding but these are to be regarded as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the various embodiments described herein can be made without departing from the scope and spirit of the disclosure. In addition, descriptions of well-known functions and constructions may be omitted for clarity and conciseness.

The terms and words used in the following description and claims are not limited to the bibliographical meanings, but, are merely used by the inventor to enable a clear and consistent understanding of the disclosure. Accordingly, it should be apparent to those skilled in the art that the following description of various embodiments of the disclosure is provided for illustration purpose only and not for the purpose of limiting the disclosure as defined by the appended claims and their equivalents.

It is to be understood that the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a component surface” includes reference to one or more of such surfaces.

Before describing the Detailed Description, certain terms and phrases used throughout the specification will be defined. The term “couple” and the derivatives thereof refer to any direct or indirect communication between two or more elements, regardless of whether the elements physically contact each other. Not only the terms “transmit”, “receive”, and “communicate”, but also the derivatives thereof include both direct communication and indirect communication. The terms “comprise” and “include”, as well as the derivatives thereof, denote unlimited inclusion. The term “or” is inclusive and denote “and/or”. Not only the phase “associated with”, but also the derivatives thereof denote “include”, “included in”, “interconnected to”, “contain”, “contained in”, “connect or connected to”, “couple or coupled to”, “able to communicate with”, “cooperate with”, “interleave”, “juxtaposed with”, “approach”, “related to”, “have”, “characterized by”, or “relevant to”.

Also, in the disclosure, it will be understood that when elements are “connected” or “coupled” to each other, the elements may be directly connected or coupled to each other, but may alternatively be connected or coupled to each other with an intervening element therebetween, unless specified otherwise.

As the disclosure allows for various changes and numerous examples, particular embodiments of the disclosure will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the embodiments of the disclosure to particular modes of practice, and it will be understood that all changes, equivalents, and substitutes that do not depart from the spirit and technical scope of the disclosure are encompassed in the embodiments of the disclosure.

In the description of embodiments of the disclosure, certain detailed explanations of related art are omitted when it is deemed that they may unnecessarily obscure the essence of the disclosure. Also, numbers (for example, a first, a second, and the like) used in the description of the specification are merely identifier codes for distinguishing one element from another.

In the disclosure, regarding an element represented as a “-er (or)”, “unit”, or a “module”, two or more elements may be combined into one element or one element may be divided into two or more elements according to subdivided functions. In addition, each element described hereinafter may additionally perform some or all of functions performed by another element, in addition to main functions of itself, and some of the main functions of each element may be performed entirely by another component.

It will be understood that blocks in flowcharts or combinations of the flowcharts may be performed by computer program instructions. Because these computer program instructions may be loaded into a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus, the instructions, which are executed via the processor of the computer or other programmable data processing apparatus generate means for implementing the functions specified in the flowchart block(s). The computer program instructions may be stored in a computer-executable or computer-readable memory capable of directing a computer or another programmable data processing apparatus to implement a function in a particular manner, and thus the instructions stored in the computer-executable or computer-readable memory may also be capable of producing manufacturing items containing instruction units for performing the functions described in the flowchart block(s). The computer program instructions may also be loaded into a computer or another programmable data processing apparatus, and thus, instructions for operating the computer or the other programmable data processing apparatus by generating a computer-executed process when a series of operations are performed in the computer or the other programmable data processing apparatus may provide operations for performing the functions described in the flowchart block(s).

In addition, each block may represent a portion of a module, segment, or code that includes one or more executable instructions for executing specified logical function(s). It should also be noted that in some alternative implementations, functions mentioned in blocks may occur out of order. For example, two blocks illustrated successively may actually be executed substantially concurrently, or the blocks may sometimes be performed in a reverse order according to the corresponding function.

Channel coding is a technique for transmitting a message with high reliability by using redundancy bits when transmitting data, and to this end, the use of an error correction code is important. Claude Shannon has defined a maximum information transmission amount for reliable communication as channel capacity, and since then, studies have been conducted on numerous error correction codes to approach channel capacity. Starting with classic codes, such as Hamming, Bose-Chaudhuri-Hocquenghem (BCH), and Reed-Solomon (RS) codes, various modern codes, such as turbo codes, low-density parity-check (LDPC) codes, and polar codes, have been newly proposed, wherein the LDPC codes are referred to as representative capacity-approaching codes.

Thereamong, the polar codes are error correction codes first proposed by Arikan in 2009, and are first codes that have been theoretically demonstrated that when a code length increases in various types of binary-input discrete memoryless channels (BI-DMC), channel capacity is asymptotically achieved (capacity-achieving).

In particular, the polar codes secure excellent performances by simultaneously using cyclic redundancy check (CRC) codes and parity-check (PC) codes through the 3rd generation partnership project (3GPP) 5th generation (5G) standardization process, and have been adopted as error correction codes for enhanced mobile broadband (eMBB) control channels, based thereon. In addition, the polar codes are considered as channel code candidates for beyond 5G (B5G) and 6th generation (6G) communication, based on strong decoding performances in short lengths.

Following 5G communication, B5G and 6G communication require deep studies on code constructions or coding and decoding techniques, which may guarantee excellent performances for more practical channel models. The 6G communication has high requirements, such as ultra-high bandwidth, ultra-high reliability, and ultra-low latency, and has low latency conditions and diversity of channels. Here, to secure diversity, an encoded signal may be transmitted through a non-homogeneous or non-ergodic channels and resources. For example, in the 5G communication, scenarios of frequency hopping or resource allocation type (3GPP, new radio (NR); Multiplexing and channel coding, TS 38.212, v 15.2.0, September 2018, hereinafter, 3GPP-1), which allocate a modulated codeword to two or more different resource blocks, are considered. When high diversity is securable by effectively using a given non-homogeneous channel, a further excellent decoding performance may be provided by supplementing weaknesses of a polar-coded bit-interleaved coded modulation (BICM) construction of

On the other hand, in conventional codes and 5G-NR codes showing excellent performances in additive white Gaussian noise channels (AWGNCs) or flat fading channels, when certain bit patterns are simultaneously faded in nonergodic block fading channels or multi-block fading channels, decoding performances may greatly decrease due to structural features of polar codes. Because a conventional BICM system includes a bit interleaver, codeword bits are evenly spread over fading blocks, and thus it is possible to obtain a certain degree of diversity. However, it is unable to provide a construction that consistently guarantees diversity for various code parameters during decoding.

The disclosure proposes reinforced polarized BICM construction and transmission technologies, which improve block error rate (BLER) performances in a nonergodic block fading channel transmission environment while maintaining an AWGNC performance under successive cancellation (SC) decoding, through a BICM construction that aligns high diversity to information bits of pre-constructed polar codes.

FIG. 1 is a diagram for describing channel combining and channel splitting of a polar code according to an embodiment of the disclosure.

Polar codes are error correction codes for theoretically achieving channel capacity, and obtain channels highly suitable for communication and channels unsuitable for communication due to low reliability, through a two-stage process of channel combining and channel splitting.

Referring to FIG. 1, codeword bits x_i(i∈_N) are generated through multiplication of N source bits u_i(i∈_N={0,1, . . . , N−1}) and a generator matrix

$G_N=G_2^{\otimes n}(G_2=\left(\begin{array}{cc}1& 0\\ 1& 1\end{array}\right),$

(·)^⊗ with a Kronecker product. A relationship between a codeword vector x and a source vector u may be represented as x=uG_N. Then, each codeword bit may go through a modulation process and then pass through an independent and identically distributed (i.i.d.) channel W, and thus a reception signal y_i(i∈_N) may be generated. Here, one large combined channel W_N using a bit vector of a u=(u₀, . . . , u_N−1) domain as an input and a reception signal y as an output may be obtained, and such a process is referred to as channel combining. Channel splitting refers to a process of obtaining virtual channels (split channels or bit channels) W_n⁽ⁱ⁾(y₀^N−1, u₀ⁱ⁻¹|u_i) (i∈_N={0,1, . . . , N−1}) experienced by each source bit, from the combined channel W_N. A polar code uses a one-bit sequential decoding method, and decoding may be performed with the help of all reception signals y₀^N−1=(y₀. . . . , y_N−1) and values (û₀ⁱ⁻¹) of previously decoded source bits, when determining a value of an ith source bit. In other words, a value of a source bit is determined according to a log-likelihood ratio (LLR) code of a probability that y₀^N−1 and u₀ⁱ⁻¹ may be obtained when u_i=1, and a probability that y₀^N−1 and u₀ⁱ⁻¹ may be obtained when u_i=0. u_i=0 is determined when the LLR has a value greater than 0, and u_i=1 is determined when the LLR has a value less than 0. As a result, N split channels generated by performing channel combining and channel splitting on i.i.d. N channel have different channel qualities, and capacities thereof are polarized and code lengths are increased, and thus a proportion of split channels converged to a mutual information amount I(W_N⁽ⁱ⁾)=1 becomes equal to I(W) . On the other hand, a proportion of split channels converted to I(W_N⁽ⁱ⁾)=0 is 1−I(W). Methods, such as density evolution, Gaussian approximation, polarization weight, may be used to approximate and evaluate qualities of N split channels. In addition, in the 5G-NR standard, a single sequence Q_Nmax for a maximum mother code length N_max(=1024) is determined through an experiential method, based on a partial order method, and a code sequence for a smaller mother code length N(<N_max) is determined as an ordered set or sequence Q_N=[i<N, i∈Q_Nmax]. In the disclosure, effectiveness of a technology of the disclosure is evaluated considering both a 5G-NR code sequence and a code sequence obtained through a polarization weight-based construction.

First, a polarization weight (PW) method described in “G. He et al., β-expansion: a theoretical framework for fast and recursive construction of polar codes, in Proc. Institute of Electrical and Electronics Engineers Global Communications Conference (IEEE GLOBECOM), Singapore, December 2017” will be described in detail. Polarization weight is a method of expressing a quality of each split channel in a weighted sum, and a quality of a bit channel W_N⁽ⁱ⁾, in which a decimal index for a code length N(=2ⁿ) is i, may be represented as Equation 1.

P_N⁽ⁱ⁾=Σ_i=0ⁿ⁻¹b_iβⁱ   Equation 1

Here, b_j(j∈_N={0,1, . . . , N−1}) indicates a bit value in a binary expression of an index (i∈_N={0,1, . . . , N−1}), and may be expressed as i=(b_n−1b_n−2 . . . b₁b₀)₂. Here, b₀ is a least significant bit (LSB) and b_n−1 is a most significant bit (MSB). Also, β denotes a fitting parameter, and β=2^1/4≈1.1892 may be finally determined by adjusting a value of β such that an AWGNC performance is remarkable. For example, when N=8, qualities of all eight split channels are P₈⁽⁰⁾=0.00, P₈⁽¹⁾=1.00, P₈⁽²⁾=1.19, P₈⁽³⁾=2.19, P₈⁽⁴⁾=1.41, P₈⁽⁵⁾=2.41, P₈⁽⁶⁾=2.60, P₈⁽⁷⁾=3.60, and reliability is high when a value thereof is high. In a polarization weight evaluation method, the qualities of the split channels may be calculated as a simple weighted sum regardless of a channel type and a code parameter (N, K). In other words, unlike a construction based on density evolution or Gaussian approximation method, which separately constructs and uses an optimum code according to each channel code parameter, the polarization weight evaluation method may support several code rates through a single sequence.

The 5G-NR code sequence is similar to a PW-based code sequence in that the 5G-NR code sequence is constructed to support all code rates through a single sequence and is based on a partial order, but there are several differences. First, the 5G-NR code sequence stores and uses only a code sequence Q₁₀₂₃ for a maximum mother code length N_max(=1024). For example, when a sequence for a polar code, in which a code length is N=512, is to be used, an index less than 512 is selected from a code sequence Q₁₀₂₃ for N=1024, and a sequence is configured by using selected indexes (sub-sequence; Q_N=[i∈Q₁₀₂₃|i<N]). Also, a sequence Q₀¹⁰²³ adopted in the standard has been determined such that a performance is remarkable through an empirical construction.

FIG. 2 is a diagram for describing a sequence Q₁₀₂₃ and a sequence for N=512, which are adopted in the 5G-NR standard according to an embodiment of the disclosure.

Referring to FIG. 2, as indicated in shades, indexes less than 512 may be selected from the code sequence Q₁₀₂₃, and a sequence may be configured with the selected indexes.

However, an initial construction code length is limited to N=2ⁿ(n=1,2, . . . ) because a 2*2 polarization kernel matrix is used while constructing a code of conventional polar code. Various code lengths or code rates need to be supported for a polar code to be actually applied to an application, and thus a required code length needs to be achieved through various rate-matching technologies, such as puncturing, shortening, and repetition.

The puncturing is a type of a code modification method, and is a method of excluding J codeword bits in a mother code x from transmission. A receiving end including a decoder does not contain any stochastic information regarding punctured bits, and thus an LLR of the punctured bits needs to be input as 0 during decoding. When a certain codeword bit is punctured, symmetric capacity of some split channels is decreased to 0, and at this time, source bits in a u domain corresponding to the capacity-0 split channels are referred to as incapable bits. The incapable bits are selected as frozen bits in a code construction, and the frozen bit generally has a value of 0. _p⊂_N={0,1, . . . ,N−1} is referred to as an index set (or a puncturing pattern) of punctured bits in an x domain, and _p⊂_N is referred to as an index set (or an incapable pattern) of incapable bits corresponding thereto in the u domain. In other words, when bits corresponding to _p are punctured, bits corresponding to _p become incapable, and thus become bits unavailable for data transmission.

FIG. 3 is a diagram for describing puncturing and shortening of a polar code according to an embodiment of the disclosure.

Referring to FIG. 3, when some output bits of an encoder are punctured, input bits of the encoder corresponding thereto become incapable bits, and when some input bits of the encoder are shortened, output bits of the encoder corresponding thereto become fixed.

FIG. 4 is a diagram for describing a 4-bit punctured polar code when N=8 according to an embodiment of the disclosure.

Referring to FIG. 4, when codeword bits having an index of _p={0,1,2,4} from among output bits of an encoder are punctured, input bits _p={0,1,2,4} of the encoder may become incapable bits.

Shortening uses features in the form of a lower triangular matrix included in a generator matrix G_N=G₂^⊗n of a polar code. In detail, when J bits in a u domain to be shortened during encoding are selected and values thereof are fixed to 0, values of J bits may also become 0 in an x domain. Codeword bits fixed as such are not transmitted, but a receiving end accurately contains information about the codeword bits, and thus an absolute value of LLR, i.e., reliability, is set to a high value, for example, a maximum value settable in hardware. _s⊂_N={0,1, . . . ,N−1} is an index set of bits to be shortened in the u domain, and _s⊂_N is an index set of bits fixed in the x domain. In other words, when bits corresponding to _s are shortened, bits corresponding to _s are fixed.

FIG. 5 is a diagram for describing a 4-bit shortened polar code when N=8 according to an embodiment of the disclosure.

Referring to FIG. 5, when source bits having an index of _s={3,5,6,7}, from among input bits of an encoder, are shortened, bits corresponding to _s={3,5,6,7} are fixed.

An SC decoding method may be used as a decoding method of the polar code. An SC decoding may be realized as belief-propagation (BP) considering SC decoding in a bipartite graph corresponding to a generator matrix G_N=G₂^⊗n. All SC decodings may be completed as each node performs 2-input 1-output operation one time. An SC decoder sequentially determines bit values using a rule according to Equation 2, for an estimated source bit vector û₀^N−1=(û₀, û₁, . . . û_N−1).

$\begin{array}{cc}{\hat{u}}_i=\{\begin{array}{c}0,\mathrm{if}i\in A^c\\ h_i\left(y_0^{N-1},{\hat{u}}_0^{i-1}\right).\mathrm{otherwise}\end{array}& \mathrm{Equation}2\end{array}$

In Equation 2, h_i(y₀^N−1, û₀^N−1) is

$h_i\left(y_0^{N-1},{\hat{u}}_0^{i-1}\right)=\{\begin{array}{c}0,\mathrm{if}\ln \frac{w_N^{\left(i\right)}\left(y_0^{N-1},{\hat{u}}_0^{i-1}❘u_i=0\right)}{w_N^{\left(i\right)}\left(y_0^{N-1},{\hat{u}}_0^{i-1}❘u_i=1\right)}\\ 1,\mathrm{otherwise}\end{array}\ge 0,$

and the above method may be realized based on LLR-BP.

FIG. 6 is a diagram of an example of decoding a polar code in which a length is N=8 on a bipartite graph according to an embodiment of the disclosure.

The bipartite graph includes a variable node (circle), a check node (square), and edges. The variable node is a single bit whereas the check node indicates a linear constraint condition, such as a binary sum of all adjacent variable node values being 0. When _N denotes a bipartite graph corresponding to a polar code having a length of N=2ⁿ, the graph may be divided into (n+1) stages in which indexes are from 0 (adjacent to a u-domain) to n (adjacent to a x-domain). Also, v_i(t) and c_j(t) respectively indicate ith variable node and jth check node at a stage stage-t. The graph _N includes a set ^(t)={v_i^(t)|i∈_N} of variable nodes and a set ^(l)={c_j^(l)|j∈_N} of check node for t=[0:n] , and edges connecting such nodes. ⁽⁰⁾ and ⁽ⁿ⁾ respectively correspond to an input vector u and an output vector x of an encoder.

For t∈[0:n−1], the variable nodes connected to ^(t) and ^(t+1) are connected to the check nodes in ^(t). On the bipartite graph, two types of information of LLR α_i^(t)∈ and a hard-decision bit value β_i^(t)∈{0,1} are exchanged for every v_i^(t) during an SC decoding process. The LLR values are propagated from right to left on the graph, and the hard-decision bit values are propagated from left to right, based on currently estimated encoding input bits. When l_i is indicated as intrinsic LLR (log p(y|x=1) log p (y|x=0)) for a codeword bit x_i, l_i=0 when x_i is a puncturing bit as described above, and l_i=∞ (or a large value available for implementation, for example, a maximum value l_max) when x_i is a bit fixed by shortening. Intrinsic LLRs are set to α_i⁽ⁿ⁾=l_i in a stage-n for all i∈_N, and an LLR message is calculated in a direction of t=0 from a stage t=n−1. In SC decoding realized based on LLR (A. B. Stimming, M. B. Parizi, and A. Burg, “LLR-based successive cancellation list (SCL) decoding of polar codes,” IEEE Trans. Signal Process., vol. 63, no. 19, pp. 5165-5179, October 2015, hereinafter, stimming 15), the LLR values are updated by two types of functions in each stage, and LLR update according to Equation 3 is performed when i_t=0 for i₂=(i_n−1 . . . i₁i₀).

α_i^{(t) =f(α}_i^(t+1), α_i+ht^(t+1))≈sgn(α_i^(t+1))sgn(α_i+ht^(t+1))·min (α_i^(t+1), α_i+ht^(t+1))   Equation 3

Here, to α_i^(t)=0 when even one of a α_i^(t+1) and α_i+2t^(t+1) is 0. On the other hand, when i_t=1, LLR update according to Equation 4 is performed.

α_i^{(t) =g(α}_i+ht^(t+1), α_i^(t+1), β_i=2t^(t))≈(1−2β_i=2t^(t))α_i=2t^(t)+β_i^(t)   Equation 4

Here, to α_i^(t)=0 when even one of a α_i^(t+1) and α_i+2t^(t+1) is 0.

For an integer i∈_N(={0,1, . . . , N−1}), considering binary expansion Σ_t=0ⁿ⁻¹i_t2^t(i_t∈{0,1}), (i_n−1i_n−2 . . . i₀) will be referred to as a binary expression of i, and will be interchangeably used with i. In other words, 7=(0111). In particular, when a binary sequence expression is to be emphasized, i₂=(i_n−1i_n−2 . . . i₀). d_H(i) denotes a Hamming weight of the binary expression (i_n−1i_n−2 . . . i₀), and denotes the number of symbols, which is not 0. In other words, 7₂=(0111) and d_H(7)=3. Also, l the one's complement of i, and values of all i_t may be changed 0→1 and 1→0. 7=(0111)=(1000)=8.

Hereinafter, important concepts for describing a definition and operations of a polar code will be described. First, a partial order relation in an integer set, called binary domination, will be defined.

Definition 1. Any two integers i, j∈_N and binary expressions thereof i₂=(i_n−1i_n−2 . . . i₀), j₂=(j_n−1j_n−2 . . . j₀) are considered. When i_t≤j_t is satisfied for all t∈[0:n−1], it is phrased that “i is dominated by j” and indicated as ij. For example, 5₂=(0101) and 13₂=(1101)₂ for ₁₆, and thus 513. On the other hand, 7₂=(0111), 8₂=(1000) and thus 7 and 8 are not in a comparison relation. In other words, a domination relation is unable to be determined for 7 and 8.

Definition 2. A dominated integer set _i of i∈_N is defined as _i≙{k∈_N|ki}.

For example, when i=6 for ₈, _6={(000), (010), (100), (110)}={0,2,4,6}.

Definition 3. A dominating integer set _i of i∈_N is defined as _i≙{k∈_N|ki}.

For example, when i=6(6₂=(110)) for ₈, ₆={(110), (111)}={6,7}.

Definition 4. (Property of Set) When _j∈ for integer sets ∈_N and ∀j∈, it is said that a set has a binary domination (BD) property ( complies with binary domination).

Definition 5. (BD Sequence) There is a sequence C₀^N−1=(c₀, c₁, . . . , c_N−) having a length of N. The sequence C₀^N−1 has a BD property when c_ic_j for different integers i,j ∈_Nor domination comparison is not possible therebetween. Alternatively, C₀^N−1will be referred to as a BD sequence.

Proposition 6. A subsequence C₀ⁱ(i∈_N) of one BD sequence C₀^N−1 is assumed. Then, a set _i={c₀, c₁, . . . , c_N−} including elements of C₀ⁱ satisfies a BD property.

Binary domination also plays an important role in determining puncturing and shortening patterns of a polar code. In detail, when an incapable bit pattern for a puncturing pattern _p satisfying the binary domination is _p, a set _p has a BD property of Definition 4 and |_p|=|_p| is satisfied. Similarly, a shortening pattern satisfying the binary domination satisfies |_p|=|_S|.

A technique proposed in the disclosure is based on a result of following lemma.

Lemma 7. For a valid incapable bit pattern _p, identical puncturing bit pattern _p(=_p) generates an incapable bit pattern _p.

According to Lemma 7, when any incapable bit pattern _p is valid, a puncturing pattern including a same index set may make Up incapable. A same incapable bit pattern is generated even when a reverse puncturing bit pattern, in which the one's complement is applied to an index, is considered. (M. Jang et al., “Rate-matching of polar codes based on binary domination,” IEEE Trans. on Commun. vol. 67, issue: 10, Oct. 2019).

FIG. 6 is a diagram showing an identical puncturing bit pattern and a reverse puncturing bit pattern, which comply with binary domination.

Referring to FIG. 6, it is verified that, for the incapable bit pattern _p, a same incapable bit pattern _p satisfying _p32_p makes Up incapable.

The related art corresponding to embodiments of the disclosure is a BICM model to which a polar code is applied. A BICM system based on 5G-NR polar code may be considered. Hereinafter, functions of blocks in a 5G-NR transmission/reception system and notations will be first described.

FIG. 7 is a diagram for describing a transmission system model using a polar code for 5G-NR uplink according to an embodiment of the disclosure.

Referring to FIG. 7, a, c, u, d, y, x′, f is a vector corresponding to input/output of each function, and K, K′, N, E is a size of the vector.

A BICM model using a 5G-NR polar code may be used as the transmission system model according to FIG. 7. The polar code is used to encode uplink control information (UCI) through a physical uplink control channel (PUCCH) and a physical uplink shared channel (PUSCH). In downlink, the polar code is used to encode payloads of a physical downlink control channel (PDCCH) and a physical broadcast channel (PBCH) through a physical downlink control channel.

In a 5G application, the number of information bits is fixed to A, and transmission of a codeword having a length E for achieving a data-rate R=A/E required by a higher communication layer is considered. To satisfy such requirements, rate-matching may be performed by applying methods, such as puncturing, shortening, and repetition, to a polar code having an initially constructed length N=2ⁿ.

Referring to FIG. 7, there is a binary message vector a that has a length of A′. An encoder may {circle around (1)} generate an input vector c having a length of K=A′+L by concatenating an external code, such as a cyclic redundancy check (CR) code or PC code, to the vector a.

Then, through the above-described code construction method (regarding the standard, based on an order of 5G-NR code sequence Q_N), {circle around (2)} bits of the vector a may be mapped to K split channels having high reliability from among N split channels. Here, the remaining split channels are pre-determined, i.e., frozen, to vectors known to a transceiver (for example, zero vectors). The vector generated through the mapping is u and a dimension of u is N . Then, {circle around (3)} polar encoding is performed through

$G_N=G_2^{\otimes n}\left(G_2=\left[\begin{array}{cc}1& 0\\ 1& 1\end{array}\right]\right),$

multiplication with a generator matrix and a vector x generated as a result of performing the polar encoding is x=uG_N (Here, the multiplication is a calculation on a Galois field (GF(2).). After the two stages of encoding (external encoding +polar encoding), {circle around (4)} a subblock interleaver splits the codeword vector into 32 blocks having a same size and permutates each block according to a simple rule to generate a vector y. The subblock interleaver performs such a process to facilitate puncturing/shortening to be performed in a next stage, and here, a subblock interleaving sequence is also referred to as a rate-matching sequence. In detail, N encoded bits are interleaved prior to the rate-matching, and during the rate-matching, puncturing is performed from the front or shortening is performed from the back according to an interleaved sequence to adjust a length. An interleaver J splits the N encoded bits stored in x into 32 subblocks having a length of B=N/32, and the interleaving may be performed in units of subblocks.

FIG. 8 is a diagram for describing a method of performing interleaving in units of subblocks according to an embodiment of the disclosure.

Referring to FIG. 8, 32 indexes are block indexes including several bits. A bit y_j after interleaving from an encoded bit x_i(i∈_N) may be represented as Equation 5 below.

i=J(j)=B·P(└j/B┘)+q   Equation 5

Here, q=j mod B and j=0,1, . . . , N−1. Also, └·┘ is a descending operator that is a largest integer equal to or less than an input.

Then, {circle around (5)} a vector x′ having a length of E(=N — J) may be provided from a vector y having a length of N, through a rate-matching technology, such as puncturing/shortening/repetition. In a 5G-NR system, the puncturing is applied to E≤N and code rate (R) R≤7/16 (i.e., a low-to-moderate code rate) areas, and first U=N−E bits of interleaved codeword bits are not transmitted (x′_i=y_i+U and i=0, . . . . E−1). On the other hand, the shortening is applied to E N and R>7/16 areas, and last U=N — E bits of the interleaved codeword bits are not transmitted. In other words, x;=y 1 (i=0, . . . , E−1) . Lastly, although not described in the disclosure, the repetition is applied when E >N, and first U=N−E codeword bits are used for retransmission (x′_i=y_{i mod N}) Also, {circle around (6)} a vector f is acquired through a channel interleaver, and {circle around (7)} a vector g after modulation is transmitted to a modulator and {circle around (7)} passes through a channel. The channel interleaver used in the 5G-NR standard is a triangular interleaver. The triangular interleaver is defined by a single parameter of a width (or height) T of an isosceles right triangle. In detail, the length T of one side of the isosceles right angle is determined to be a minimum integer value satisfying a condition of

$\frac{T\left(T+1\right)}{2}\ge E.$

Indexes of codeword bits having a length of E are written on the triangular interleaver in a horizontal direction, and are read in a vertical direction when the write is completed. For example, when N=E=8 , T=4 is determined. For example, (0,1,2,3,4,5,6,7) are sequentially written in a row-major manner and read in a column-major manner to obtain an interleaved sequence of (0,4,7,1,5,2,6,3). In general, the channel interleaver is used to distribute, throughout a codeword, a fading effect on bits configuring a symbol during high-order modulation, and it has been experimentally verified that AWGNC has a high performance gain when the interleaver is not used. Structural vulnerability of a code may be improved through the interleaver.

6G communication have high requirements, such as ultra-high bandwidth, ultra-high reliability, and ultra-low latency. Here, due to low-latency conditions and diversity of channels, an encoded signal may be transmitted through a non-homogeneous channel or resource to secure diversity. Also, such transmission may be approximated to nonergodic channel or multi-block fading channel transmission. In general, a bit that experienced a weak channel may be adjusted to evenly spread throughout a codeword, through channel interleaving according to a BICM system, for a codeword bit before transmission. This is to prevent a performance from deteriorating by a combination of symbols that have a low channel coefficient due to a weak point of a finite length code. In particular, the BICM system may achieve a highest effect from convolutional codes where consecutive errors are fatal, and the BICM system has been also applied to other codes, such as turbo codes.

Regarding the triangular interleaver adopted as the channel interleaver of the conventional 5G-NR standard, weak points of codes have been avoided in overall through pseudo-random interleaving, and the performance has been improved by easing a performance deterioration effect in the AWGNC and fast fading channel. However, the triangular interleaver is empirically constructed in the block fading channel, and is not a construction that theoretically guarantees the diversity.

FIG. 9 is a diagram for describing a conventional BICM transmission/reception system model to which a polar code is applied according to an embodiment of the disclosure.

Referring to FIG. 9, a code length is adjusted through a rate-matching technology, such as puncturing or shortening, after polar encoding, and an AWGNC or fading channel is experienced after BICM interleaving/modulation.

Meanwhile, in 5G-NR, resource allocation similar to block fading channel transmission is defined to secure frequency diversity. For example, a scenario in which a modulated codeword is transmitted to two frequency blocks separated from each other by a half of a system bandwidth in one slot, as in intra-slot frequency hopping, or a modulated codeword is transmitted through several split frequency blocks as in resource allocation type [3GPP-1] is defined. When M denotes the number of resource blocks to which a single codeword is to be allocated, intra-slot frequency hopping may be applied when M=2, and the resource allocation type may be applied not only when M=2 but also when M>2 , and the modulated codeword is transmitted through M different resource blocks. In the disclosure, codeword bit allocation when M=2 is mainly described, and a method of securing full-diversity when M>2 is also described. Here, when both M=2 and M>2, one codeword may be allocated to several resource blocks in a same quantity (uniform allocation) or in different quantities (non-uniform allocation).

FIG. 10A is a diagram for describing a method by which logical resource blocks, to which an interleaving method is applied, are allocated to physical resource blocks according to frequency hopping according to an embodiment of the disclosure.

Referring to FIG. 10A, resources in two logical resource blocks may be transmitted by being allocated to two different physical resource blocks, and this may be an example of intra-slot frequency hopping resource allocation. The interleaving method according to an embodiment of the disclosure may be applied when logical resource blocks are allocated to physical resource blocks on one or more slots. However, this is only an example and a resource unit on a time axis is not limited to one slot. For example, a resource block may be allocated in units of mini-slots.

FIG. 10B is a diagram for describing allocating of logical resource blocks, to which an interleaving method is applied, to physical resource blocks according to an embodiment of the disclosure.

Referring to FIG. 10B, bits in four logical resource blocks may be allocated to four physical resource blocks. Embodiments of the disclosure consider modulated codeword transmission through several split frequency blocks, as in the scenarios of FIGS. 9, 10A, and 10B. It may be assumed that the above scenarios are similar to experiencing a fading channel in block units because resources in a block have homogeneous channel characteristics and have a very low correlation with a channel of a physically separated block. Accordingly, it is important to develop a modulation encoding technology considering a block fading channel and study an enhanced BICM technology, which are basic assumptions of channels of the disclosure.

FIG. 11 is a diagram for describing a case where, when certain codeword bit patterns are simultaneously faded, reliability of a certain source bit is greatly reduced due to a structural feature of a polar code, and thus decoding becomes difficult according to an embodiment of the disclosure.

Referring to FIG. 11, N=8 and codewords may be allocated to two blocks. When first, third, and fourth codeword bits are faded due to fading, reliability of a fourth information bit may be very low due to a structural feature.

A triangular interleaver used as a BICM interleaver of the 5G-NR standard is a type of pseudo-random interleaver, and has a decoding performance almost similar to a random interleaver through an effect of averaging a fading effect experienced by a symbol. Clearly, the triangular interleaver is still considered as an interleaver supporting a highly excellent performance in high-order modulation and fading, there may be an area where performance deterioration occurs due to structural incompleteness of the triangular interleaver in some code parameters. For example, when certain bit patterns are simultaneously faded as in FIG. 11, reliability of a certain information bit may greatly decrease due to a structural construction characteristic of a polar code, which may be deemed as low diversity. The decoding performance may greatly decrease due to the low diversity. According to embodiments of the disclosure, all information bits have full-diversity under SC decoding, through a premeditated BICM system construction. Here, when all bits included in an information set have full-diversity, a code also has full-diversity. Ultimately, an AWGNC performance is to be maintained while improving a fading performance through a diversity effect, by using an improved BICM construction.

FIG. 12 is a diagram for describing a conventional polar-coded BICM system transmitter and a BICM system transmitter during multi-block fading channel transmission according to an embodiment of the disclosure.

A BICM interleaver in a conventional polar-coded BICM model has been mostly pseudo-randomly and heuristically constructed to evenly distribute various values of a fading channel to a codeword. In particular, a triangular interleaver adopted as a BICM system for a polar code exhibits a satisfactory BICM performance according to incoherence with a polar code structure. However, such a heuristic construction is unable to provide a guaranteed diversity in a block fading channel having a small number of fading diversities. On the other hand, an interleaver proposed in an embodiment of the disclosure is an interleaver constructed premeditatedly to secure a diversity in which decoding of information bits is high during SC decoding, by effectively using a non-homogeneous channel.

Referring to FIG. 12, an electronic device according to an embodiment of the disclosure may classify a rate-matched codeword x′ into a plurality of groups considering diversity when a modulated codeword is divided and allocated to two or more multi-blocks, perform modulation on the classified codeword of each group, and allocate the same to a fading channel. To consider the diversity, the electronic device may perform interleaving on a codeword x or x′ by using a polar code sequence based on reliability (e.g., polarization weight) of source bits corresponding to indexes of codeword bits.

Here, a same number of bits or different numbers of bits may be allocated to bit groups corresponding to different channel blocks or different resource groups.

FIG. 13 is a diagram for describing a conventional polar-coded BICM system receiver and a BICM system receiver during multi-block fading channel transmission according to an embodiment of the disclosure.

Referring to FIG. 13, reception signals that passed through a channel are combined immediately before decoding, after demodulation is performed on each reception signal.

FIG. 14 is a diagram for describing a method by which a signal transmitted from an electronic device avoids a structural weakness and secures a high diversity by effectively using a non-homogeneous channel according to an embodiment of the disclosure.

Referring to FIG. 14, eight codeword bits may be allocated to two blocks. When premeditated interleaving is performed as described above with reference to FIG. 12, a non-homogeneous channel may be effectively used and a structural weakness may be premeditatedly avoided, thereby securing a high diversity, as shown in FIG. 14.

There is a need for strategies for evaluating diversities of bits provided through interleaving and allocating a diversity to each bit. First, embodiments of the disclosure propose a method of calculating a diversity for each information bit by using a diversity updating method through Boolean variable approximation of a fading event.

Embodiments of the disclosure propose an interleaver construction method considering a diversity. The conventional 5G-NR polar code has been mainly constructed based on an AWGNC performance. Embodiments of the disclosure propose a method of aligning high/full diversity to information bits through premeditated fading block allocation or interleaving of codeword bits, without changing a conventional code (e.g., a 5G-NR code or a PW-based code) (i.e., a code sequence or information set). A diversity alignment method according to an embodiment of the disclosure effectively operates even for a punctured or shortened polar code and provides high/full diversity in a wider region compared to the related art, and thus provides a high performance gain in a certain code parameter.

Boolean Approximation-Based Diversity Analysis in Block Fading Channel.

As described above, a codeword (x′) that is punctured/shortened after encoding may be divided and allocated to different non-homogeneous resource blocks to be transmitted, and at this time, code diversity may be secured from diversity of a channel. It is assumed that symbols/bits in one resource all experience a same fading coefficient h_i, and it is assumed that h_is are independent from each other. In a case of resource allocation considering diversity, such as actual frequency hopping or resource allocation type allocation, resources in a block have similar channel characteristics and channels of blocks allocated by being physically separated from each other have a low correlation, and thus a similar channel situation assumed above is provided. For a theoretical analysis of a proposed system, an ideal block fading channel is assumed, and thus a resource allocation technique for securing diversity of an actual system may be satisfactorily approximated as described above.

Here, as described above, the number of blocks of a block fading channel for a codeword is M, and M has a value of 2 or greater. When M=2, the block fading channel will be referred to as a 2 fading-block channel, and when M=4, the block fading channel will be referred to as a 4 fading-block channel. In the disclosure, an interleaver construction method when M=2 (e.g., an intra-slot frequency hopping or resource allocation type scenario) will be described first, and an interleaver construction method expanding to a scenario where M≥3 will be described.

A source vector u=(u₀, u₁, . . . , u_N−1) is multiplied by a generator matrix G_N to generate a codeword vector x=(x₀, x₁, . . . , x_N−1), and codeword bits in x may be rate-matched and then allocated to M different resource blocks. Index sets of bits allocated to blocks A, B when M=2 are referred to as _A, _B, and the numbers of elements included in the sets are n_A, n_B. When n_A=n_B uniform allocation is performed and when n_A≠n_B, non-uniform allocation is performed. When M=4, index sets of bits allocated to blocks A, B, C, and D are _A, _B, _C, _D,the numbers of elements included in the sets are n_A, n_B, n_C, n_D. When n_A=n_B=n_C=n_D, uniform allocation is performed, and non-uniform allocation is performed otherwise. A similar definition may be provided when M≠2,4.

FIG. 14 illustrates example of uniform allocation and an encoding structure of polar code where N=4 and M=2. Four source bits are encoded to generate four codeword bits, and at this time, x₁ and x₂ are allocated to the block A having a fading coefficient of h_A, and x₃ and x₄ are allocated to the block B having a fading coefficient of h_B. Here, fading experienced by codewords allocated to the blocks A and B may be represented as Boolean stochastic variables A and B. Irrelevant to the above drawing, when one codeword is allocated to M different resource blocks, fading events corresponding to the blocks may be respectively represented as A₁, A₂, . . . , A_M (When there are four blocks, A, B, C, D for convenience of expression).

In embodiments of the disclosure, a fading coefficient h_i having a real value is approximated to a Boolean variable. When I₁(or I₂)=0, h_i has a sufficiently small value and thus a fading situation is indicated, and when I₁(or I₂)=1, h has a sufficiently large value and thus a normal transmission/reception situation is indicated. In other words, each fading event is to be approximated to on-off, and fading manners of messages during Genie-aided SC decoding process are to be analyzed. A fading stochastic variable corresponding to a check/variable node is updated for each decoding stage, such that fading stochastic variables finally included in source bits are simply obtained. Such dichotomous approximation is unable to express a case where a fading coefficient value is moderate, but enables a fading operation of an SC decoding process to be effectively expressed and analyzed.

Diversity update corresponding to check/variable node processing of SC decoding will now be described. As described above, a polar code may obtain an LLR of bits in a u-domain by updating a check node and variable node, from an LLR of bits of an x-domain through LLR-based Genie-aided SC decoding (Stimming15). Here, a check node arithmetic operation f(l₁, l₂) and a result thereof for two input LLRs l₁, l₂ may be represented as Equation 6.

l′≙f(l₁, l₂)=sgn(l₁)sgn(l₂) min (|l₁|, |l₂|)   Equation 6

In other words, only minimum values for two inputs are obtained as outputs during the check node arithmetic operation. Here, when Boolean approximation proposed in the disclosure is applied to l₁ and l₂ having real values, l′ is l′≠0 only when both l₁ and l₂ are not 0. Because the check node arithmetic operation outputs a minimum value of an absolute value, when fading trends of two inputs are represented as Boolean variables l₁, l₂, fading of an output message may be represented as l₁, l₂ that is Boolean multiplication.

FIG. 15 is a diagram for describing a process of updating a fading variable when updating a check node according to an embodiment of the disclosure.

Referring to FIG. 15, a process of updating a fading stochastic variable during a first stage check node update of _A={1,2}, _B={3,4} when N=4 is illustrated. When a codeword bit x₁ mapped to a fading block A and x₃ mapped to B, or a codeword bit x₂ mapped to A and x₄ mapped to B are combined at a check node and updated, AB is obtained.

On the other hand, for a variable node arithmetic operation g(l₁, l₂, u_i), an LLR may be obtained by using two input LLRs l₁, l₂ and a previously determined source bit value u_i, as in Equation 7.

l″=g(l₁, l₂, β_i)=(−1)^βil_l+l₂  Equation 7

In other words, addition of two LLRs are performed in the variable node. An output message is highly unlikely to be faded, excluding a case where two message inputs are both faded. When a fading operation of an input message is represented as a Boolean stochastic variable l₁, l₂, a fading operation of an output may be represented as l₁+l₂ that is an OR arithmetic operation.

In summary, a Boolean function-based fading approximation technique first represents a fading event of each fading block as a single Boolean variable, and evaluates a fading characteristic of an LLR message as a Boolean function in each decoding stage while tracking LLR-based Genie-aided SC decoding process. During this process, two arithmetic operations are performed, wherein multiplication of two Boolean variables is performed in a check node corresponding to downgrade, and an OR arithmetic operation of two Boolean variables is performed in a variable node corresponding to upgrade. Accordingly, a Boolean function corresponding to bits of a u domain may be calculated, and a diversity corresponding thereto may be calculated.

FIG. 16 is a diagram for describing a process of updating a fading stochastic variable when updating a variable node according to an embodiment of the disclosure.

Referring to FIG. 16, it will be described how fading stochastic variables included in codeword bits are evaluated through a check node/variable node arithmetic operation, and how diversities of source bits are calculated. Referring to FIG. 16, N=4 and _A={1,2}, _b={3,4}.

For u₁, total two check node arithmetic operations are performed. The check node arithmetic operation is performed with respect to a fading stochastic variable AB included in bits allocated to x₁ and x₃ in a first stage, and the check node arithmetic operation is performed again with respect to AB obtained through a check node arithmetic operation of x₂ and x₄ in a second stage, and thus a fading stochastic variable of u₁ is AB. Unlike u₁, u₂ experiences one variable node arithmetic operation after one check node arithmetic operation. When represented as a diversity, (AB)+(AB)=AB. Similarly, update processes of u₃ and u₄ may be respectively represented as (A+B) (A+B)=A+B and (A+B)+(A+B)=A+B in equations.

FIG. 17 is a diagram for describing updating of a fading stochastic variables included in source bits according to an embodiment of the disclosure.

Referring to FIG. 17, N=8 and _A={1,2,3,4}, _B={5,6,7,8}.

In this case, a fading (Boolean) function of u₁·u₄ is represented as AB and a fading function of u₅˜u₈ is represented as A+B. Here, source bits may be A (or B), AB, or A+B depending on an allocation pattern. In the disclosure, for convenience of description, a concept of survival count n s (when M=2, n_s∈{1,2,2²−1}, and when M=4, n_s∈{1,2, . . . , 2⁴−1}) corresponding to a state of a fading stochastic variable included in the source bits is introduced and a diversity is to be directly or indirectly evaluated based on the survival count. AB survives only in a case of M=2, A≠0, B≠0 , and thus n_s=1, and for A (or B), n_s=2, and for A+B , n_s=3. In the example of FIG. 17, for u₁˜u₄, n_s=1 , and for u₅˜u₈, n_s=3. When M=2, n_s=3 corresponds to full diversity 2.

Evaluation of diversity for each source bit is based on Genie-aided SC decoding and assumes that previous bits have been completely decoded. When SC decoding is actually performed, it is unable to achieve an evaluated diversity. This is because, due to characteristics of the SC decoding, when previous information bits are wrongly determined, it is unable to guarantee reliability or diversity of following information bits, since when a diversity of previous bits is low, the following information bits (may be subject to the previous bits. However, when a source bit is a fixed bit (frozen bit), determining (decoding) of the source bit does not affect decoding of a following source bit. Accordingly, to secure full-diversity of a code, all bits corresponding to an information set need to have full-diversity. A polar code interleaver construction guaranteeing a diversity is provided by maximizing that satisfies the full-diversity.

Meanwhile, considering a situation where a parameter of a code is provided, only a diversity of pre-determined information bits may be considered and a decoding performance may be affected by the lowest diversity among the diversity of each of pre-determined information bits In the disclosure, when the information bits in the information set all have the full-diversity, it is defined that a diversity of the code also has the full-diversity. In particular, the code has the full-diversity when n_s of all information bits is 3. In addition, it is defined that a diversity-1 is provided when the diversity of the code has a stochastic variable corresponding to A (or B) , and a weak diversity-1 is provided when the diversity of the code has a stochastic variable corresponding to AB. For example, there may be a situation where data is transmitted by selecting four source bits from among eight source bits, i.e., a code rate is 0.5. In this case, an information set is _AWGN={4,6,7,8} for a code constructed based on an AWGNC performance, but minimum n_s of bits in the set is 1 (a fading stochastic variable of the code is AB), and thus there may be a bit having a weak diversity-1 in the information bits for an AWGN code (a code constructed considering AWGNC: PW-based code or 5G-NR code). On the other hand, when a code is constructed considering only a diversity, an information set is provided according to _div={5,6,7,8}, and because bits in the set are all n_s=3 , the code has a diversity-2.

According to embodiments of the disclosure, to provide a high code diversity without changing a conventional code, fading block allocation is applied through suitable BICM interleaving, thereby assigning a high diversity to information bits of a pre-constructed code.

FIG. 18 is a diagram for describing a change in a diversity for each bit, according to fading block allocation through suitable interleaving of a codeword bit according to an embodiment of the disclosure.

Referring to FIG. 18, in both a case where allocation is performed according to _A={1,2,3,4}, _B={5,6,7,8} as in a left diagram of FIG. 18 and a case where allocation is performed according to X _A={1,2,3,5}, _A={4,6,7,8} as in a right diagram of FIG. 18, four source bits have a fading approximation function A+B. However, as shown in FIG. 18, a location of an information bit having a code diversity-2 varies. Information bits in an AWGN code _AWGN={4,6,7,8} have a weak code diversity-1 according to minimum n_s=1 in the left diagram, but have a code diversity-2 according to minimum n_s=3 in the right diagram. In other words, a high diversity may be allocated to pre-determined information bits via codeword bit distribution through suitable interleaving, and this will be referred to as diversity alignment in the disclosure.

FIG. 19 is a diagram for analyzing a diversity according to a conventional interleaving method and comparing a correlation between the diversity and a simulation performance, based on a Boolean approximation-based diversity analysis according to an embodiment of the disclosure.

Referring to FIG. 19, n_s included in each bit is indicated in a descending order from a bit having high reliability when N=64, and 2nd to 4th columns indicate n_s for each bit respectively when there is no interleaver (No), a block interleaver (Block) is applied, and a triangular interleaver (Tri) is applied. Also, a right diagram in FIG. 19 indicates a required signal-to-noise ratio (SNR) [dB] achieving BLER=10⁻³ according to the number of information bits. In other words, the smaller the value of the required SNR, the better the performance.

Also, FIG. 19 illustrates n_s (in an order of a code sequence C_N) for each source bit when there is no interleaver (NO), when a block interleaver is applied (BLK), and when a triangular interleaver is applied (TRI), in a 2-fading block channel, and a required SNR (dB) achieving BLER=10⁻³, which correspond thereto.

Based on experiment results, when the block interleaver is used and K=3→4, a bit where n_s=1 is included in an information set , and thus a diversity of a code is decreased from 2 to a weak diversity-1, thereby greatly increasing a required SNR. In addition, when there is no interleaver, a decoding performance is greatly reduced when a bit in which an index is 31 is included in (K=7→8). In the above example, the diversity is decreased to diversity-1 (or weak diversity-1) when K=10 when there is no interleaver, K=4 when the block interleaver is applied, and K=27 when the triangular interleaver is applied. In other words, the triangular interleaver relatively maintains diversity-2 even when K is high, and thus has an excellent performance. In summary, the Boolean approximation-based diversity analysis newly considered in embodiments of the disclosure has a very close relation with an actual simulation performance, and accordingly, an excellent performance of a conventional triangular interleaver may also be described.

Referring to FIG. 19, referring to n_s for each bit included in the triangular interleaver, the triangular interleaver includes high n_s in general, but does not include an optimum diversity arrangement. Here, the optimum diversity arrangement denotes an arrangement in which diversity-2 is provided as information bits of a code where R=0.5 all include n_s=3. In this regard, embodiments of the disclosure newly propose an interleaving method having an excellent fading performance by supplementing weaknesses of the triangular interleaver and approximating to the optimum diversity arrangement of a 5G-NR code. First, interleaver optimization through repetitive diversity evaluation may be performed. However, an interleaver optimized as such may have a limited code rate and limited length flexibility, and thus a more basic diversity alignment technique will be proposed.

Interleaving Method for Transmitting Diversity of Polar Code.

1) Interleaving Method for Fading Block Distribution of Modulated Codeword Bit When M=2.

Through an interleaving method for fading block distribution of a codeword bit, according to an embodiment of the disclosure, a code diversity-2 may be satisfied when R≤0.5 by transmitting a pre-constructed polar code of a full N=2ⁿ length. Before the introduction, a theoretical setoff regarding the number of source bits having full-diversity when M=2 will be first proposed.

Lemma. When M=2 , the maximum number of source bits having diversity-2 is N/2.

Even when interleaving is not performed, the number of bits having full-diversity is always N/2, and thus as long as the degree of freedom of code construction is provided, it is very easy for a code to achieve full-diversity. However, because a satisfactorily constructed code is pre-determined in AWGNC, an interleaver (or a bit allocation pattern) needs to be well determined such that existing information source bits all have full-diversity. Accordingly, the disclosure proposes a new greedy algorithm that uses, as an output, an optimum bit allocation pattern guaranteeing full-diversity up to a code rate of 0.5. An algorithm when M=2 will be proposed first, and an optimum bit allocation pattern search algorithm of M>2 will be proposed next.

First, when a polarization weight or a polar code sequence C_N defined according to a 5G-NR method is provided, the key is to create full-diversity one by one from an information bit having highest reliability. First, all codeword bits are allocated to block X₁. Here, according to a Boolean approximation-based diversity analysis tool, all source bits have diversity-1. After the initial setting process above, one bit is randomly selected at a time to be moved to a block X₂, and it is verified whether the bit has full-diversity in an order of C_N. Whenever each bit is moved, it is verified whether full-diversity is achieved by using the Boolean diversity analysis tool proposed above. The above process is shown in Algorithm 1.

Algorithm 1 Optimization of bit allocation for M = 2.
Input: Fading blocks X1, and X2;
Output: x1, X2;
1: Allocate all xi's to block X1: XX1 = [N], X2 = 0.
2: while The number of information bits with diversity 2 ≤ N/2 do
3:  Find bit index i such that the next reliable information
bit has diversity 2 if xi; is moved to block X2.
4:  if found then
5:   Move the i from X1 to X2.
6:  else
7:   Stop optimization.
8:  end if
9: end while

For example, an optimum bit allocation pattern search example when a code length is N=8 and C₈=(7,6,5,3,4,2,1,0) will be described. First, all codeword bits are allocated to the block X₁ (_X1=₈, _X2=∅). Then, after selecting a random bit (index: i) in _X1 and moving the same to the block X 2 , it is verified whether a source bit u₇ has diversity-2. When u₇ has diversity-2, updates of _X1=₈\{i}, _X2={i} are performed, and then another bit is moved to the block X 2 to verify whether a next highly reliable u₆ has diversity-2. Algorithm 1 operates until all N/2 information source bits have full-diversity or until no particular source bit has full-diversity in any movement. In other words, Algorithm 1 ends when K=N/2 is reached or when a diversity-interleaver is not found.

Table 1 shows a diversity-interleaver search success probability of Algorithm 1. Code lengths were N∈{64,128, . . . , 2048} and a polarization weight-based code construction sequence was considered. According to Algorithm 1, an optimum interleaver search was successful with a very high probability in a short-medium code length, and the search was successful with a probability of 20% even in a very long code length of N=2048. As a result, for a well-constructed code provided in AWGNC, many interleavers having full-diversity are found through Algorithm 1.

TABLE 1
Diversity-Interleaver Search Success Probability N∈
{64, 128, . . . , 2048} PW Code Assumption
N	64	128	256	512	1024	2048
Rate	0.998	0.983	0.933	0.795	0.525	0.190

In Table 2, maximum code rates guaranteeing full-diversity for various interleaver types are compared. Unlike identity (when interleaver is not applied), a square block interleaver, and a triangular interleaver, diversity-2 is always guaranteed at a code rate 0.5.

TABLE 2
Comparison of Full-Diversity Guaranteed Maximum Code Rate
according to Code Length and Interleaver Type (M = 2)
	N	64	128	256	512	1024
Identity	Rmax	0.125	0.094	0.066	0.051	0.040
Block	Rmax	0.063	0.195	0.020	0.158	0.006
Triangular	Rmax	0.406	0.336	0.426	0.336	0.406
Proposed	Rmax	0.500	0.500	0.500	0.500	0.500

Among many types of optimized interleavers, it was verified that a code sequence-based interleaver that conforms to BD while satisfying symmetry is one of the interleavers. In other words, a method of using a code sequence C_N as an interleaver sequence after Algorithm 1 is also proposed.

As described above, n_A and n_B(n_A+n_B=N) may be different from each other. This concept may be expanded and referred to as uniform allocation when bits allocated to all blocks are the same when M>2. Here,

$X_A=C_N\left[0:\frac{N}{2}-1\right],X_B=C_N\left[\frac{N}{2}:N-1\right]$

are determined for _p, _p satisfying binary domination and an ordered set (code ordered set) _N of indexes provided in an order of a code sequence C_N of split channels. This process may be achieved by performing interleaving on a codeword vector in an order of C_N, dividing the interleaved codeword vector forward and backward, and allocating the same to two blocks. Here, following results may be obtained based on Lemma 7.

Theorem 8. When a code sequence C_N is symmetric for a polar code defined by the code sequence C_N in which a mother code length is N=2ⁿ, u-domain bits corresponding to information bits of

$C_N\left[0:\frac{N}{2}-1\right]$

have diversity-2 when a codeword vector x is allocated to two fading blocks based on

$X_A=C_N\left[0:\frac{N}{2}-1\right],X_B=C_N\left[\frac{N}{2}:N-1\right].$

In other words, codeword interleaving through _N and a same block allocation method may satisfy full-diversity at a code rate R<0.5. As will be described below, even when a codeword sequence is not symmetric, a full-diversity performance approximating to a code rate 0.5 may be realized when the codeword sequence has high symmetry.

FIG. 20 is a diagram for describing an interleaving method and effects, which guarantee a diversity, according to an embodiment of the disclosure.

Referring to FIG. 20, a diversity included in source bits when same numbers (n_A=n_B=4) of codeword bits are allocated to two different resource blocks, based on a code sequence (i.e., interleaver sequence I_N=code sequence _N), is illustrated. When a code length is 8, ₈=[7 6 5 3 4 2 1 0], and at this time, index sets of bits allocated to blocks A and B, based on uniform allocation, are respectively A=set(C₈[0:3])={7,6,5,3}, _B=set(C₈[4:7])={4,2,1,0}. Here, set(·) is a function converting an ordered set or sequence into a set. In this case, bits in a u-domain having a same index as a bit index allocated to the block A in an x-domain may include n_s=3, and bits in the u-domain having a same index as a bit index allocated to the block B includes n_s=. However, when the code sequence _N is not symmetric, the number of source bits of the u-domain having a full-diversity may be reduced. This will be described in detail below.

2) Interleaving Method for Fading Block Distribution After Modulation of Codeword Bit When M>2.

Algorithm 2 indicates a method of searching for an optimum bit allocation pattern when M>2. Similar to Algorithm 1, all codeword bits are allocated to the block X₁ for initial setting. Then, (M−1)-tuple codeword bits (x_i−1, x_i−2, . . . , x_i−M−1, ) are respectively moved to blocks X₂, . . . X_M−1, and then it is verified whether a first source bit of C_N has diversity-M. When successful, updates of _X1=_N\{i₁. . . , i_M−1}, _X2={i₁}, . . . , _XM={i_M−1} are performed, and then the above processes are repeated. Algorithm 2 is repeated when K=[N/M] is reached or when full-diversity is not achieved even by moving any tuple.

Algorithm 2 Optimization of bit allocation for M > 2.
Input: Fading blocks X1, X2, . . . , XM;
Output: X1, X2, . . . , XM;
1: Allocate all xi's to block X1: X1 = [N], X2 = . . . = XM = 0.
2: while The number of information bits with diversity 2 ≤ N/M] do
3:  Find (i1, . . . , iM-1) such that the next reliable information bit has
diversity M if (xi1, . . . , xiM-1) is moved to block (X2, . . . , XM).
4:  if found then
5:   Move i1, i2, . . . , iM-1 from X1 to
6: X2, X3, . . . , XM, respectively.
7:  else
8:   Stop optimization.
9:  end if
10: end while

In Tables 3 and 4, maximum code rates R_max guaranteeing full-diversity of interleavers found through Algorithm 2 and conventional interleavers are compared respectively when M=3, 4. For an identity interleaver, R_max is gradually decreased when a code length is increased. For a triangular interleaver, R_max is high compared to the identity interleaver and a square block interleaver, but a proposed interleaver has full-diversity up to a much higher code rate area.

TABLE 3
Comparison of Full-Diversity Guaranteed Maximum Code Rates
according to Code Length and Interleaver Type (M = 3)
	N	32	64	128	256	512
Identity	Rmax	0.156	0.109	0.086	0.063	0.049
Block	Rmax	0.156	0.047	0.086	0.016	0.092
Triangular	Rmax	0.219	0.234	0.250	0.215	0.156
Proposed	Rmax	0.313	0.328	0.313	0.316	0.320

TABLE 4
Comparison of Full-Diversity Guaranteed Maximum Code Rates
according to Code Length and Interleaver Type (M = 4)
	N	32	64	128	256	512
Identity	Rmax	0.125	0.078	0.055	0.043	0.031
Block	Rmax	0.125	0.031	0.047	0.012	0.031
Triangular	Rmax	0.156	0.109	0.156	0.102	0.129
Proposed	Rmax	0.250	0.250	0.250	0.246	0.245

FIG. 21 is a diagram for describing a block allocation method of modulated codeword bits, according to an embodiment of the disclosure.

Referring to FIG. 21, M=3. As in M=2, when M>2, it is important to secure a diversity of 2 in as many source bits as possible. When the numbers of bits allocated to blocks A, B, and C are respectively indicated as n_A, n_B, n_C, and puncturing/shortening is performed by J bits, values of n_A, n_B, n_C may be determined to be very close to (N−J)/3 and may be non-homogeneously allocated to n_A≠n_B≠n_C depending on a situation. When indexes corresponding to a bit set _B where n_s=3 when M=2 are arbitrarily allocated to the blocks B and C, a value of diversity for each bit under genie-aided SC decoding may be guaranteed to be at least 2. However, such a case does not guarantee a full-diversity.

FIG. 22 is a diagram for describing a block allocation method of modulated codeword bits, according to an embodiment of the disclosure. Referring to FIG. 22, modulated codeword bits may be allocated to four different resource blocks. In this case, as described above, allocation of

$X_A=C_N\left[0:\frac{N}{2}-1\right],X_B=C_N\left[\frac{N}{2}:N-1\right]$

satisfying n_A=n_B when M=2 satisfies code diversity-2 up to R=0.5 , and thus even when the modulated codeword bits are allocated to four blocks as in FIG. 22, a minimum value-2 may be guaranteed as a diversity for each bit regardless of a pattern allocated to C and D in _B. According to situations, same numbers of bits, i.e., n_A=n_B=n_C=n_D, may be allocated to fading blocks, and when n_A, n_B, n_C, n_D have different values, the bits may be non-homogeneously allocated.

A range of code rates satisfying diversity 2 for each bit may vary depending on symmetry of a code sequence. Hereinafter, a PW sequence in which a code sequence is symmetric and a 5G-NR code sequence that is asymmetric are considered, and code rates at which the PW sequence and the 5G-NR code sequence guarantee diversity 2 (full-diversity when M=2) will be identified. Also, a diversity performance is expanded while guaranteeing flexibility and good AWCNC performance through symmetry modification of the 5G-NR code sequence.

Before detailed descriptions, additional notations will be briefly introduced. For a set ⊂_N={0,1, . . . , N−1}, a complementary set of is indicated as a symbol ′, and ′=_N\. Also, a reverse set of the set (or the one's complement set) is defined as ={N−i|i∈}. When cardinality of the set is ||=a, sizes of the complementary set and the reverse set are respectively |′|=N−a, ||a. For example, when ={0,1,2} for ₈, ′ and are respectively ′={3,4,5,6,7} and ={5,6,7}.

Also, for a listed integer set or sequence (ordereu JCL) , when

$𝓍=𝓏\left[0:\frac{N}{2}-1\right],𝓎=𝓏\left[\frac{N}{2}:N-1\right]\mathrm{is}𝓍\bigcap \widetilde{𝓎}=𝓍\mathrm{or}\widetilde{𝓍}\bigcap 𝓎=𝓎,$

is symmetric (In this case, when is symmetric while satisfying BD, is BD-symmetric). For example, when ₁={7,6,5,3,4,2,1,0}, and ₁={7,6,5,3}, ₁={4,2,1,0} and ∩₁=₁ is satisfied, and thus a symmetric sequence and a BD property are satisfied. On the other hand, when ₂={7,6,5,2,4,3,1,0}, BD is not satisfied, and ₂={7,6,5,2}, ₂={4,3,1,0}, and ∩₂={0,1}≠₂, and thus symmetry is also not satisfied. As a result, for a code constructed to satisfy a BD property and symmetry, a diversity-2 is guaranteed at a code rate R≤0.5. In the disclosure, a proposed interleaving and block allocation technique has a BD property, but even for an asymmetric conventional 5G-NR code, a high code diversity is securable for a general code parameter compared to conventional interleavers. Descriptions will be developed based on a polarization-weight (PW) sequence as an example of a symmetric code sequence having a BD property and a 5G-NR code sequence as an example of asymmetric code having a BD property.

1) Symmetric Code: Polarization-Weight (PW)-Based Code Sequence.

As described in the background of a polar code, a polarization weight is a method of approximating and evaluating qualities of split channels to a weighted sum, and symmetry is satisfied as in _A∩_B=_B for a sequence _N=[_A|_B] obtained during a PW-based code construction.

For example, when a PW-based code construction method is assumed when N=64, C₆₄=[63 62 61 59 55 47 60 31 58 57 54 53 46 51 45 30 43 29 56 39 27 52 23 50 44 49 15 42 28 41 38 26 37 25 22 35 21 48 14 19 13 40 11 36 24 7 34 20 33 18 12 17 10 9 6 5 32 3 16 8 4 21 0]. Accordingly, _A=[26 38 41 28 42 15 49 44 50 23 52 27 39 56 29 43 30 45 51 46 53 54 57 58 31 60 47 55 59 61 62 63] and _B=[0 1 2 4 8 16 3 32 5 6 9 10 17 12 18 33 20 34 7 24 36 11 40 13 19 14 48 21 35 22 25 37], and for all i, (i)+(N−1−i)=N−1 is satisfied, and thus a PW-based code sequence satisfies a BD property and symmetry (_A∩_B=_B).

In other words, when a single codeword is allocated to two channels having different fading coefficients under SC decoding, a diversity-2 is guaranteed when R≤0.5 in a case where a code satisfying a BD property and symmetry is _A=

${𝒞}_N\left[0:\frac{N}{2}-1\right]{𝒳}_B={𝒞}_N\left[\frac{N}{2}:N-1\right].$

2) Asymmetric Code:5G-NR Code Sequence.

Unlike the PW-based code sequence, the 5G-NR code sequence stores a single code sequence (reliability ascending order code sequence) Q₁₀₂₄ for N_max=1024, and determines C_N={i<N₁|i∈Q₁₀₂₄]} when a code sequence for N=N₁<N_max is used. In other words, a new sub-code sequence is configured by selecting an index of i<N₁. Here, Q₁₀₂₄ is a code determined such that an AWGNC performance is excellent through an empirical method.

FIG. 23 is a diagram for describing bit indexes arranged in an order of a code sequence C_N and diversity values corresponding thereto according to an embodiment of the disclosure.

Referring to FIG. 23, when N=64 , a 5G-NR code sequence is ₀⁶³C₆₄=[63 62 61 59 55 47 31 60 58 57 54 53 46 51 45 30 43 29 39 27 56 52 15 50 44 49 42 28 41 38 22 25 37 26 35 21 14 48 13 19 40 11 7 36 24 34 20 33 18 10 17 6 9 32 16 8 4 2 1 0]. Accordingly, C₆₄[0:31]=[22 38 41 28 42 49 44 50 15 52 23 56 27 39 29 43 30 45 51 46 53 54 57 58 60 31 47 55 59 61 62 63] and C₆₄[0:31]=[0 1 2 4 8 16 32 3 5 9 6 17 10 18 12 33 20 34 24 36 7 11 40 19 13 48 14 21 35 26 37 25]. Here, _B∩_B^c={41,22}≠∅ and thus a 5G-NR code for N=64 is an asymmetric code.

Due to a bit set {41,22}, {37,26} inducing asymmetry, bits in which n_s=2 (n_s≠3) are generated compared to a symmetric sequence.

Referring to FIG. 23, indexes (source bits) of the 5G-NR code are arranged in an order of a code sequence C_N when N=64 , and n_s of each corresponding source bit is illustrated. Here, a first row of a first box corresponds to

$C_A\left(=C_N\left[0:\frac{N}{2}-1\right]\right),$

and a second row corresponds to

$C_B\left(=C_N\left[0:\frac{N}{2}-1\right]\right),$

Bit pairs violating symmetry in C_A, C_B are respectively {41,22} and {37,26}.

First, source bits corresponding to _a\{41,22} have n_s=3, bits corresponding to _B\{37,26} have n_s=1, and bits of {41,22,37,26} inducing asymmetry have n_s=2. In other words, a location of a first bit generating asymmetry is a point where a diversity is decreased to a diversity-1 (or a weak diversity-1), and thus a maximum code rate for maintaining a code diversity-2 varies depending on the location of the first bit. In embodiments of the disclosure, it is reviewed whether 1) an excellent decoding performance is exhibited compared to a conventional triangular interleaver even when a BD asymmetric code is used, and 2) a fading channel performance and an AWGNC performance are securable even when a code or code sequence is reconstructed to remove asymmetry.

First, an electronic device according to an embodiment of the disclosure 1) identifies whether an excellent diversity and decoding performance is exhibited compared to a conventional triangular interleaver even when an asymmetric code sequence is used. In other words, the electronic device identifies symmetry of a code when the code, such as N=64,128,256, is selected from a 5G-NR code sequence C₁₀₂₄ and used, and analyzes how much the code affects a diversity.

FIG. 24 is a diagram showing reliabilities of bits for each of an interleaver and a triangular interleaver, in a descending order, according to an embodiment of the disclosure.

Referring to FIG. 24, n_s of bits when an interleaver and triangular interleaver according to an embodiment of the disclosure are each applied to N=64,128,256 are illustrated in an order of a code sequence C_N.

n_s of bits arranged in the order of the code sequence C_N are compared when a 5G-NR code (BD and asymmetric) is used when N=64,128,256. As described above, the 5G-NR code is a sub-sequence of a pre-constructed sequence C₁₀₂₃ having a length of 1024, and may be symmetric or asymmetric depending on a length. In conclusion, N=128 guarantees a diversity-2 when R=0.5, but for N=64,256, there are bits where a diversity-2 is damaged when R<0.5.

First, when a code length N=64, an asymmetric code sequence is obtained as a result. Here, a diversity-2 is guaranteed up to K=29 considering a codeword bit

$\left(R\le \frac{29}{64}\right)$

allocation method (or an interleaving method) of the disclosure, and diversity-2 is provided in a wider area than a triangular interleaver that is the related art guaranteeing diversity-2 only up to K=23. When N=128, a 5G-NR code sequence is symmetric as a result, and thus a diversity-2 is guaranteed up to K=64, i.e., R≤0.5, as in a case where a symmetric sequence is used. On the other hand, when the triangular interleaver is applied, a diversity is decreased to a diversity-2 when K=39, and thus a performance gain is secured when K=39−64. Lastly, when N=256, a bit having n_s=2 as an asymmetric code sequence first appears when K=127. A bit causing asymmetry in the 5G-NR code generally appears very close to R=0.5. A diversity-2 maximum guarantee K value of the conventional triangular interleaver is only K=105, whereas a combination of interleaver 5G-NR code sequences of the disclosure guarantees a diversity-2 up to K=126, and thus there is only a difference of 2 from an achievable maximum value 128. In this regard, asymmetry is not great even when the 5G-NR code sequence is used, and thus an AWGNC performance may be maintained while obtaining a performance gain of approaching the optimum in a fading channel.

Next, 2) a diversity in a block fading channel may be optimized by reconstructing a code for summarization when a provided code sequence is asymmetric. When an asymmetric code sequence is used, a code is not changed, and thus an AWGNC is maintained, whereas according to a method of reconstructing a code, an excellent performance may be guaranteed as a diversity-2 is maintained up to R=0.5 in a block fading channel, but an AWGNC performance may deteriorate because the code is changed. However, the number of bits causing asymmetry in the 5G-NR code is only 1 to 2, and thus the bits barely affects an actual simulation performance. Accordingly, a case where the 5G-NR code is used and a case where a symmetric PW code is used both maintain an AWGNC performance and at the same time, secure a great performance gain compared to the related art in a fading situation.

FIG. 25 is a graph comparing required SNR [dB] performances achieving a BLER 0.1% of a code modified to have symmetry with an asymmetric 5G-NR code in an AWGNC and a block fading channel according to an embodiment of the disclosure.

Referring to FIG. 25, N=64 and biphase shift keying (BPSK) are assumed. Also, a 5G-NR code may be modified to be symmetric by substituting bits generating asymmetry in _A and _B with each other.

The 5G-NR code has a sequence of [63 62 61 59 55 47 31 60 58 57 54 53 46 51 45 30 43 29 39 27 56 52 15 50 44 49 42 28 41 38 22 25 37 26 35 21 14 48 13 19 11 7 36 24 34 20 33 18 10 17 6 9 5 32 16 8 4 2 1 0] and the modified code has a sequence of [63 62 61 59 55 47 31 60 58 57 54 53 46 51 45 30 43 29 39 27 56 52 15 50 44 49 42 28 41 38 37 25 22 26 35 21 14 48 13 19 40 11 7 36 24 34 20 33 18 10 17 6 9 32 16 8 4 2 1 0] by substituting {41,22} and {37,26}, which induce asymmetry, with each other.

Through the graphs of FIG. 25, required SNRs [dB] of a BD-asymmetric code and a symmetrically modified code, which achieve BLER=0.1%, in a BPSK-AWGNC and block fading channel are identified.

A left graph of FIG. 25 indicates a BPSK-AWGNC performance, and it is identified that performances achieving BLER=0.1% before and after the modification are almost the same. Also, it is identified, through a right graph of FIG. that an area where an AWGNC decoding performance is improved by removing asymmetry from a BPSK-block fading channel.

It is identified, through the graphs of FIG. 25, that when an interleaving method according to an embodiment of the disclosure is used, a diversity-2 is provided in a wider area than the related art even when the asymmetric 5G-NR code is used, and a fading performance is improved while maintaining an AWGNC performance even when a code is modified to remove asymmetry.

FIG. 26 is a diagram for comparing the maximum numbers K_max of information bits guaranteeing diversity-2, according to an interleaving type and a code length according to an embodiment of the disclosure.

When there is no interleaver or a block interleaver (square block interleaver) is considered, a diversity-2 is not fully guaranteed even at a low code rate, such as R=0.2, and in particular, in a case of the block interleaver, when a code length is N=(2^b)²(b=1, . . . ), a diversity is greatly decreased in association with a structural feature. On the other hand, a conventional triangular interleaver guarantees a diversity-2 even for a relatively high code rate. However, an interleaver according to an embodiment of the disclosure has a code diversity-2 up to a higher code rate, and may have a great performance gain by a higher diversity than the triangular interleaver in a code rate area of R=0.33˜0.50.

• • Interleaving and codeword bit allocation method of polar code considering rate-matching.

A code length of a polar code is limited to N=2ⁿ(n=1,2, . . . ) according to a construction characteristic, and to apply the limited code length to actual practical 5G/B5G/6G communication, various code lengths and code rates need to be flexibly supported. For example, an effective code length E required to transmit 50-bit data at a ½ code rate is

$E=\frac{K}{R}=50/\left(0.5\right)=100.$

In tnis case, 36 bits may be repeatedly transmitted from N=2⁶=64 bits or 28 bits may be punctured or shortened from N=2⁷=128 bits. In the standard, the puncturing is applied to R≤7/16 that is a relatively low-to-moderate code rate area, and the shortening is applied to higher code rate transmission.

TABLE 5
Comparison of K areas to which puncturing and shortening are applied,
according to code length and number of punctured/shortened bits
	Punctured
N	bit	0	4	8	12	16	20	24	28	32	36	40	44	48	52	56
32	M	32	28	24	20	16
	Puncturing	14	12	10	8	7
	(Max. K)
	Shortening	15~	13~	11~	9~	8~
	(Min. K)
64	M	64	60	56	52	48	44	40	36	32
	Puncturing	28	26	24	22	21	19	17	15	14
	(Max. K)
	Shortening		27~	25~	23~	22~	20~	18~	16~	15~
	(Min. K)	29~
128	M	128	124	120	116	112	108	104	100	96	92	88	84	80	76	72
	Puncturing	56	54	52	50	49	47	45	43	42	40	38	36	35	33	31
	(Max. K)
	Shortening	57~	55~	53~	51~	50~	48~	46~	44~	43~	41~	39~	37~	36~	34~	32~
	(Min. K)

Table 5 shows, in detail, the K areas to which techniques are applied according to the number of punctured/shortened bits. For example, when N=64, an area to which the puncturing is applied when the number of punctured/shortened bits is 16 is K≤21, and the shortening is applied when K>22 that is a relatively high code rate. The puncturing is used in a relatively low code rate area, and thus a sufficient diversity improvement effect may be exhibited through the disclosure in an area where n_s=1 or 2 due to structural vulnerability of the related art. On the other hand, the shortening is considered when R≥7/16. Accordingly, a full-diversity is not obtained in an R>1/2 area, and thus performance improvement may be expectea in an

$\frac{7}{16}\le R\le \frac{1}{2}$

area. In embodiments of the disclosure, efficient diversity transmission using a polar code is enabled through a construction of an interleaver comprehensively considering even a code rate.

- Interleaver and codeword allocation operating method for guaranteeing diversity of punctured/shortened polar code.

As described above, in a case where there is no rate-matching, when codeword bits having indexes corresponding to the first half and codeword bits having indexes corresponding to remaining half are respectively allocated to blocks A and B, regarding a code sequence C_N satisfying BD property and symmetry, through an interleaving method according to an embodiment of the disclosure, a diversity-2 may be guaranteed when R=0.5. Even when the code sequence C_N is asymmetric, the diversity-2 may be guaranteed to a very high area compared to the related art. However, when code modification, such as puncturing/shortening, is applied, natural application and effect of a technique of the disclosure are limited due to characteristics of puncturing/shortening techniques of 5G-NR. Accordingly, herein, an interleaving and fading block bit allocation technique for a modified code is to be proposed. Uniform allocation is mainly considered, but quasi-uniform allocation in which the number of bits for each block is almost the same does not exceed an intention and application range of the technique of the disclosure. A code length of a punctured polar code is adjusted to (N−|_p|), and in a case of uniform allocation when the same number of modulated codeword bits are allocated to different fading blocks, the number of bits allocated to each block is (N−|_p|)/2. Similarly, a code length of a shortened polar code is also adjusted to (N−|_s|), and in a case of uniform allocation, the number of bits allocated to each block is (N−|_s|)/2. For example, a polar code (C₈=[7 6 5 3 4 2 1 0]) in which a code length is 8 is allocated to blocks A and B by 4 bits each (_A={7,6,5,3}, _B={4,2,1,0}). However, when puncturing of 2 bits is applied, 3 bits may be allocated to each of the blocks A and B. The puncturing is generally performed from a codeword bit corresponding to a source bit having a low reliability, and when _p={0,1}, 2 bits from among the 4 bits previously allocated to the block B are punctured A:4→4, B:4→2), and thus 1 bit needs to be moved from the block A to the block B and reallocated. When a first bit index reallocated to the block B is r₀, sets ′_A, ′_B of bit indexes allocated to the blocks A and B are ′_A={3,5,6,7}\r₀ and ′_B={2,4}∪r₀.

The above scenario may be represented as a formula below.

When an information bit index set for a punctured bit number J (a puncturing index set (puncturing pattern): _p, J=|_p|) and code rate R=0.5 is , _A=, _B=^c, and _A=_B=0.5N when J=0. On the other hand, when J≠0, _A≠, _B≠^c, |_A|=|_B|=0.5(N−J) . Also, when 2 bits are additionally punctured to J=2i(i≥0)→2 (i+1), ′_B=(_B\_p)∪r_i and ′_A=(_A\_p)\r_i, and when J is increased, r_i is an (i+1)th bit index moved from _B to _A. Here, r_i may be determined such that diversities of information bits are maximized, and a high code diversity may be secured by greedily determining r_i in an order for various J values.

The above method may be identically applied to a code shortening situation. When a polar code (C₈=[7 6 5 3 4 2 1 0]) in which a code length is 8 is 2-bit shortened (_s={6,7}), 3 bits may be allocated to each of the blocks A and B. The shortening is generally performed from a codeword bit corresponding to a source bit having a high reliability, and when _s={6,7}, 2 bits from among the 4 bits previously allocated to the block A are shortened (A:4→2, B:4→4) (unlike the puncturing), and thus 1 bit needs to be moved from the block B to the block A. When a first bit index moved to the block A is s o , sets of bit indexes moved to the block A are ′_B={0,1,2,4}\s₀ and ′_B={3,5}∪s₀.

When J≠0 for an information bit index set regarding a shortened bit number J and a code rate R=0.5, _B≠^c, _A≠ and |_A|=|_B|=0.5 (N−J). Also, when 2 bits are additionally shortened to J=2i(i≥0)→2(i+1) , ′_B=(_B\_s)\s_i and ′_A=(_B\_s)\s_i, and when J is increased, s_i is a bit index moved from _B to _A. Like r_i, s_i may also be determined such that diversities of information bits are improved, and s_i values optimized for the determined J may be determined, thereby securing a high diversity.

The bits may be reallocated to secure a high diversity in various puncturing/shortening situations with the r_i values and s_i values determined through the above processes, and in this regard, embodiments of the disclosure introduce “auxiliary sequences

$T_N=\left[r_0{r}_1\cdots r_{\frac{N}{4}-1}{s}_0{s}_1\cdots s_{\frac{N}{4}-1}\right]$

and len (T_N)=N/2 (len(·) is a length of)”. Here,

$r_i\left(i\in \left[0:\frac{N}{4}-1\right]\right)$

is an (i+1)th bit index moving a punctured polar code from the block B to the block A as described above, and s_i

$\left(i\in \left[0:\frac{N}{4}-1\right]\right)$

is an (i+1)th bit index moving the shortened polar code from the block A to the block B. In other words, when the code length is generally N, a case where more than half bits are shortened or punctured does not occur, and thus a length of an auxiliary sequence may be determined to be equal to or less than N/2. Contemplating an auxiliary sequence for puncturing and an auxiliary sequence for shortening, the disclosure considers one auxiliary sequence in which the two sequences for puncturing and shortening are concatenated, for convenience. Considering longest puncturing or shortening lengths possible, the first

$R_N=T_N\left[1:\frac{N}{4}\right]=\left[s_0{s}_1\cdots s_{\frac{N}{4}-1}\right]$

of an auxiliary sequence T N is a sequence of bit indexes to be moved from the block A to the block B in a puncturing mode, and the rear

$S_N=T_N\left[\frac{N}{4}+1:\frac{N}{2}\right]=\left[s_0,s_1,\cdots ,s_{\frac{N}{4}-1}\right]$

of the sequence is a sequence of bit indexes to be moved from the block B to the block A in a shortening mode. In the disclosure, an example in which the same numbers of modulated codeword bits are allocated to different fading blocks is described, but even when different numbers of modulated codeword bits are allocated as in n_A≠n_B, the more (or fewer) number of bits may be allocated based on the auxiliary sequence defined above.

FIG. 27 is a diagram for describing a method of allocating a diversity, based on auxiliary sequences of a polar code mother code to which rate-matching is not applied and a polar code in which 6 bits are punctured according to an embodiment of the disclosure.

Referring to FIG. 27, N=16 is assumed. Also, modulated codeword bit indexes to be allocated to each resource block for a punctured or shortened polar code are determined by using an auxiliary sequence T₁₆=[11 10 14 7 4 5 1 8] pre-defined for N=16. When there is no rate-matching, a diversity-2 is guaranteed by allocating _A=set(C₁₆[0:7])={15,14,13,11,7,12,10,9}, and _B=set(C_{16 [8:15]})={6,5,3,8,4,2,1,0}. For the punctured polar code, bits to be moved from a block A to a block B are determined based on the pre-defined auxiliary sequence T₁₆. For example, when 6 bits of ={0,1,2,3,4,5} are punctured as in the above example, ′_B={0,1,2,4,8,3,5,6}\{0,1,2,3,4,5}∪{r₀=11,r₁=10, r_b2=14}={6,8,10,11,14} and ′_A={9,10,12,7,11,13,14,15}\{r₀=11, r₁=10, r₂=14}={7,9,12,13,15}.

FIG. 28 is a diagram for describing a method of allocating a diversity according to auxiliary sequences of a polar code mother code without rate-matching and a polar code in which 2 bits are shortened according to an embodiment of the disclosure.

Bits to be moved are determined by using an auxiliary sequence T N shown in FIG. 28, for both puncturing and shortening. For a shortened polar code, a bit is moved from a block B to a block A, based on a pre-defined auxiliary sequence. Referring to FIG. 28, when 2 bits having indexes of ={14,15} (: an index set of shortened bits; shortening pattern) are shortened, ′_B={0,1,2,4,8,3,5,6}\{s₀=4}={0,1,2,3,5,6,8} and ′_A={{9,10,12,7,11,13,14,15}\{14,15}}∪{4}={4,7,9,10,11,12,13}. A method of determining a priority of bits to be moved at this time will be described later. According to a user definition, one auxiliary sequence may be defined and used for each code length of N=2ⁿ (N∈{32, . . . ,1024}), or a single auxiliary sequence corresponding to N_max may be defined (N_max=1024) to correspond to rate-matching. Here, a case where a single auxiliary sequence corresponding to a plurality of modified polar codes having a length equal to or less than N_max is used may need to compromise on a certain level of performance (or diversity).

Alternatively, instead of using an auxiliary sequence and an interleaving sequence based on a polar code sequence together, a unified sequence in which an auxiliary sequence is reflected on an interleaving sequence may be defined and used. (In other words, allocation to blocks A and B when there is no code modification and reallocation through an auxiliary sequence for code modification may be defined as one unified sequence.) The unified sequence corresponds to both shortening and puncturing, and as a result, because an auxiliary sequence for puncturing and an auxiliary sequence for shortening are in a symmetric relation, an overall unified sequence may be constructed by using the symmetric relation even when only the auxiliary sequence for puncturing is known.

FIG. 29 is a diagram for describing a method of allocating modulated codeword bits to resource blocks by separately defining a sequence and an auxiliary sequence, and a method of allocating the modulated codeword bits by using a unified sequence, according to an embodiment of the disclosure.

Referring to FIG. 29, a method of using a code sequence-based interleaving sequence and auxiliary sequence together during 6-bit puncturing when N=16, and a method of allocating a modulated codeword of one unified sequence (i.e., a sequence constructed by reflecting an auxiliary sequence to an interleaving sequence) are illustrated.

When the code sequence-based interleaving sequence and the auxiliary sequence are used together, ={0,1,2,3,4,5} may be punctured when _B={0,1,2,4,8,3,5,6}, _A=9,10,12,7,11,13,14,15}, and then 3 bits ({11,10,14}) of T_N[1:3]=(r₀ r₁ r₂) may be reallocated from the auxiliary sequence when _B={6,8}, _A=7,9,10,11,12,13,14,15] . In other words, ′_A=9,12,7,13,15}, ′_B={6,8,11,10,14}. are obtained based on the allocation. On the other hand, according to the method of allocating a modulated codeword of one unified sequence, when a sequence is reconstructed such that an auxiliary sequence is reflected to an interleaving sequence, [2 8 3 6 0 1 5 4 11 10 14 15 9 12 7 13] , and when M=2 , n_A from the front excluding a punctured bit, and n_B bits thereafter may be respectively allocated to blocks A and B.

1) Method of determining auxiliary sequence for punctured polar code.

A technique for constructing an auxiliary sequence for codeword bit allocation when M=2 in a case where a polar code is punctured is proposed. The punctured polar code transmits x′ in which a length is E=N−J by puncturing J bits from a codeword x. An LLR of punctured bits is set to 0, and the number of incapable bits same as the number of punctured bits are generated in a u-domain in response to the punctured bits. In other words, _p satisfying |_p|=_p| is generated, and _p has a BD property.

FIG. 30 is a diagram for describing a diversity calculating method for a punctured polar code according to an embodiment of the disclosure.

Referring to FIG. 30, a diversity of a polar code in which _p={1,2}, _A={4}, _B={3} may be updated. When a fading stochastic variable corresponding to first punctured bits is indicated as X , an LLR corresponding to punctured bits during decoding is 0, and thus may be considered as fixed fading, and X=0. When X(=0) and A meet, an update corresponding to a check node may be updated to multiplication 0 A=0 of two stochastic variables, and an update corresponding to a variable node may be updated to addition 0+A=A of two stochastic variables.

A fading Boolean function included in u₃ is (0+B)·(0+A)=AB, and a fading Boolean function included in u₄ is (0+B)+(0+A)=A+B. In other words, n_s of u₃ is 1, n_s of u₄ is 3, and a code diversity of a (2,1) polar code where ={4} is 2.

Hereinafter, a r_i, i=0,1, . . . ,

$\frac{N}{4}-1$

sequence to be moved to another resource for a punctured polar code according to an embodiment of the disclosure may be optimized by using an algorithm. Algorithm 3 for determining a r_i value is as follows.

Algorithm 3. Select ri recursively based on depth-2 intermediate ns
1. Given information
p (puncturing pattern), h (i ∈ h; An; half-rate information set), B, ICand. = ∅
2. Obtain the set of adjacent direct path indices, gis for index i ∈ h.
gi ← {i − bj · 2j: j ∈ {0, 1, . . . , n − 1}}\{i}, ∀i ∈ h ( i = (in−1 . . . i1i0), j ∈ [0: n − 1])
If gi ∩ h = ∅, ICand. = ICand. ∪ {i} // secure candidate direct path index that
require intermediate ns test (it is assumed that bits satisfying gi ∩ h ≠ ∅
guarantee diversity-2.)
ICand. ← Sort_RelDescending(ICand.) // align bits in order of code sequence CN
by applying polarization reliability rule of u-domain for ICana. including
indexes of x-domain. (Sort_RelDescending(·) function denotes a sequence in
which · is aligned in order of code sequence CN of polar code or polarization weight.)
3. Reduce the candidate set Ti to select ri for puncturing mode, or reduce the candidate
set Si to select si for shortening mode) (prevent from degrading the depth-2
intermediate ns) (an index set of candidate bits for determining a si value is Ti
(                                                  i                          ∈                                                      [                                                                                          0                                :                                                                                                  N                                  4                                                                                            -                              1                                                        ]                                                                          )                                            )                                        .
Ti=CN[⌊❘"\[LeftBracketingBar]"𝒫❘"\[RightBracketingBar]"2⌋+N2-1:N-1]//Candidate⁢set⁢definition⁢of⁢bits⁢to⁢be⁢moved⁢to
block A from all available blocks B
U = {Ns(i): i ∈ ICand.} // Depth-2 ns set corresponding to bit index i(∈ ICand.)
(Ns(i) denotes ns value in depth -2 direct path-index i)
for j ∈ Ti // for candidates of bits to be moved from all available blocks B to block A
Switch ‘the fading block of xj’ from B to A // Reanalyze ns (diversity) when
xj is moved from block B to block A
Uj = {Ns(i): i ∈ ICand.} // Depth-2 ns set of candidate bits after switching
For every k(= 1: |Candidate|)
If U(k) > Uj(k) then Ti ← Ti\{j} // For ns of arranged information bits
(in descending order), when diversity result deteriorates through
movement of xj, j is excluded from movement candidate
endif, endfor
endfor
4. Choose si which increases the overall intermediate depth-2 ns
for j ∈ TransCan // For candidates from all updated available B → A
Switch ‘the fading block of xj’ from B to A // Same as step 3
Uj = {Ns(i): i ∈ ICand.} // Depth-2 ns set of candidate bits after switching
If Uj(k) = 3 for every k(= 1: |Candidate|)
Select j as a transition bit
If multiple bits are survived, select the bit with highest reliability or priority
If there is no survivor, go to step 5.
5. Reduce the candidate set Ti to select ri for puncturing mode, or reduce the candidate
set Si to select si for shortening mode) (prevent from degrading the depth-3 intermediate ns)
for k ∈ ICand.
If N(k) = 3 then ICand. = ICand.\{k} // Exclude xks in which ns is 3 from
index set. Search remaining bits for depth-3 ns
End
V = {Ns(i): i ∈ ICand.} // Calculate ns only for bits in which ns is equal to or
less than 2.
for j ∈ Ti
Switch ‘fading block of xj’ from B to A
Vj = {Ns(i): i ∈ ICand.}
For every k(= 1: |Candidate|)
If U(k) > Uj(k) then Ti\Ti{j} // When diversity result deteriorates
through movement of ns of information bits (arranged in descending
order), j is excluded from movement candidate.
endif, endfor
endfor
6. Choose si which increases the overall intermediate depth-3 ns
for j ∈ Ti
Switch ‘the fading block of xj’ from B to A
Vj = {Ns(i): i ∈ ICand.} // Depth-3 ns set of candidate bits after switching
If Vj(k) = 3 for every k(= 1: |Candidate|)
Select j as a transition bit
If either multiple bits are survived or there is no bit survived, select the
bit with highest reliability

As an example of an auxiliary sequence construction, a method of determining an auxiliary sequence for a PW-based auxiliary sequence will be described. For a code C₁₆=[15 14 13 11 7 12 10 9 6 5 3 8 4 2 1 0] in which N=16 , _A={15,14,13,11,7 ,12,10 ,9}, _A=6,5,3,8,4,2,1,0} is an optimum diversity allocation pattern when there is no rate-matching (for a polar code mother code). When J bits are punctured, a r_i value needs to be determined such that (N−J)/2 bits having highest reliability have diversity-2 (or n_s=3). In other words, a code diversity of an information bit _h(|_h|=(N−J)/2) when R=0.5 is 2. For example, when J=2, a r₀ value needs to be determined such that bits in which indexes are {15,14,13,11,7,12,10} is n_s , and when J=6, a r₀, r₁, r₂ value needs to be determined such that bits in which indexes are {15,14,13,11,7} is n_s=3.

A set of adjacent direct path indexes excluding a bit i(∈)=[b_n−1 . . . b_ib₀](b_i∈{0,1}) included in _h, which is encountered during a VN update process, is represented as g_i={i−b_j·2^j:j∈{0,1, . . . , n−1}}\{i}. Here, a direct path denotes a path connecting ith symbols of all encoding/decoding stages. A set of indexes of direct paths in a polarization relation with the ith symbols is g_i. Here, when g_i∩_h≠∅, it is highly likely that information bits may have a diversity-2 as a result. On the other hand, when g_i∩_h≠∅, a diversity-2 is not guaranteed, and a “depth-2 polarization” update process therefor may be traced.

TABLE 6
Adjacent path indexes of bits included in h when N = 16, | p| = 2
Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
15	1111	7	11	13	14
14	1110	6	10	12
13	1101	5	9		12
11	1011	3		9	10
7	0111		3	5	6
12	1100	4	8
10	1010	2		8

For example, an information set of a polar code in which N=16,J=2 is _h={15,14,13,11,7,12,10}, and g₁₅={7,11,13,14}, g₁₄={6,10,12}, g₁₃={5,9,12}, g₁₁={3,9,10}, g₇={3,5,6}, g₁₂={4,8},g₁₀={2,8}. Also, ={0,1},_B={2,4,8,3,5,6}∪r₀, _A={9,10,12,7 ,11,13,14,15}\r₀, and a candidate set of r₀ is R₀={9,10,12,7,11,13,14,15}=_A. For g₅, g₁₄, g₁₃, g₁₁, at least one bit is included in A h . As a result, it is expected that source bits having corresponding indexes are highly likely to have a diversity-2. However, for g₇, g₁₂, g₁₀, all bits are included in _h^c, and thus a diversity-2 is not guaranteed, and diversity trace up to depth-2 polarization is performed therefor in an order of a code sequence C_N. In embodiments of the disclosure, a r_i value is determined to maximize a diversity thereof.

For additional descriptions on an algorithm operation, a fading Boolean function for each message according to a decoding process on a bipartite graph is defined as s_i,j. Here, i∈{0,1, . . . , n−1} denotes a direct path and symbol index, and j∈{0,1, . . . , n} denotes a depth or layer index. At this time, a direct path index denotes an index i shared on a path connecting u_i and x_i. Also, a depth index j is a value obtained by subtracting a stage index from n, and denotes a degree of polarization in an actual channel. For reference, in a depth-index, j=0 corresponds to an x domain, where as a stage index starts to be indexed from a u domain.

A bit in which an index is 7(0111) performs a CN operation by encountering a bit in which an index is 15 during depth-1 polarization (s_i,j: Fading Boolean variable corresponding to a direct path index i at depth-j), and performs a VN operation by encountering a bit in which an index is 3 at a depth-2. Here, a fading function corresponding to a bit in which an index is 3 is a stochastic variable updated after a CN operation is performed with a bit in which an index is 11 at a depth-2. A fading function updated after a depth-2 polarization process is performed is represented as Equation 8.

s _7,2=(s_7,0s_15,0)+(s_3,0s_11,0)   Equation 8

When s_3,0=B and s_7,0, s_11,0, s_15,0 are not moved, all of s_7,0, s_11,0, s_15,0 have A, and thus s_7,2=A . When s_7,0 or s_15,0 is moved to B, s_7,2=AB and an intermediate diversity deteriorates compared to existing A, and thus when R_i is an index set of candidates for determining a r_i value, 7 and 15 may be excluded from R₀.

For a path in which a direct path index is 12, a fading function at a depth-2 is s_12,2=(s_12,0+s_4,0)+(s_8,0+s_0,0), s_0,0=0, s_4,0=s_8,0=B, and s_12,0 is A when not moved, and thus s_12,2=A+B(n_s=3). When s_12,0 is moved to B, s_12,2=B and the intermediate diversity deteriorates compared to existing A+B, and thus 12 may also be excluded from R₀ (R₀={9,10,11,13,14}).

A fading stochastic variable at a depth-2 corresponding to a bit in which an index is 10 is S_10,2=(s_10,0+s_2,0)(s_14,0+s_6,0), and s_2,0=s_6,0=B, s_10,1s_14,1 is A when not moved, and thus s_10,2=A+B (n_s=3). When s_10,0 or s_14,0 is moved to B, s_10,2=B, and thus intermediate diversity (depth-k diversity (1<k<n)) deteriorates compared to existing A+B. Accordingly, 10 and 14 may also be excluded from R₀ (R₀={9,11,13}). The above process is a process of excluding bits causing depth-2 deterioration for g_i satisfying g_i∩_h=0. Referring back to 7, s_7,2=(s_7,0s_15,0)+(s_3,0s_11,0)=(B·s_11,0)+A. Here, when s_11,0 is moved to B, s_7,2=A+B and n_s=3 is satisfied, and thus r₀=11.

TABLE 7
Adjacent path indexes of bits included in h when N = 16, | p| = 4
Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
15	1111	7	11	13	14
14	1110	6	10	12
13	1101	5	9		12
11	1011	3		9	10
7	0111		3	5	6
12	1100	4	8

When J=4, _p={0,1,2,4}, _B={8,3,5,6,11}∪r₁, _A={9,10,12,7,13,14,15}\r₁. An information set of a polar code in which N=16, J=4 is _h={15,14,13,11,7,12}, and g₁₅={7,11,13,14},g₁₄={6,10,12}, g₁₃={5,9,12}, g₁₁={3,9,10}, g₇={3,5,6}, g₁₂={4,8}. Also, a candidate index set for determining a bit s₁ to be moved from B to A is R₁. Here, g₁₅∪_h≠∅, g₁₄∪_h≠∅, g₁₃∪_h≠∅, but g₁₁∪_h≠∅, g₇∪_h≠∅, g₁₂∪_h≠∅, and thus an electronic device according to an embodiment of the disclosure traces a diversity for bits in which indexes are 11, 7, and 12.

First, a fading Boolean function at a depth-2 corresponding to a bit in which an index is 11 is provided as s_11,2=(s_11,0+s_3,0) (s_15,0+s_7,0). When s_11,0=s_3,0=B and s_15,0 and s_7,0 are not moved, all of s_15,0 and s_7,0 have A, and thus s_11,2=AB. Here, when even one of s_15,0 and s_7,0 is moved, deterioration of n_s does not occur, and thus the corresponding index is not excluded from a candidate set. Then, bits in which an index is 7 are s_7,2=(s_7,0s_15,0)+(s_3,0s_11,0) and s_3,0=s_11,0=B , and s_15,0, s_7,0 are A when not moved, and thus s_11,2=A+B . When one of s_15,0, s_7,0is moved, an intermediate diversity decreases to s_7,2=B+AB=B, and thus 7 and 15 may be excluded from R₁.

For a bit in which an index is 12, s_12,2=(s_12,0+s_4,0)+(s_8,0+s_0,0), and s_0,0=s_4,0=X=0, s_8,0=B and s_12,0 are A when not moved, and thus s_12,2=A+B(n_s=3). When x₁₂ is moved, a diversity deteriorates, and thus 12 is removed from R₁ (R₁={9,10,13,14}). As in a case when J=2, the above process is a process of excluding bits causing depth-2 diversity deterioration for g_i satisfying g_i∩_h=∅. Referring back to 11(1011), s_11,2=(s_11,0+s_3,0)(s_15,0+s_7,0)=AB (always n_s=1). Thus, the electronic device according to an embodiment of the disclosure may determine a bit to be moved to B in a direction an intermediate stage n_s is increased during a process after depth-3. For a bit in which an index is 11, a VN operation is performed after depth-3, and diversity update is performed by encountering a path corresponding to indexes 9 and 10. For a path index 9, a depth-2 diversity is s_9,2=(s_9,0+s_1,0)(s_13,0+s_5,0)=(s_9,0)(B+s_13,0) and for 10, s_10,2=(s_10,0+s_2,0)(s_14,0+s_6,0)=(s_10,0)(B+s_13,0). Here, when s_13,0 or s_14,0 is moved to B, n_s is damaged, and thus indexes 13 and 14 are also excluded from R₁ (R₁∈{9,10}). Any one of the indexes 9 and 10 may be selected, and at this time, one of source bits having the indexes, which has higher reliability, may be selected as r₁. Accordingly, r₁=10.

TABLE 8
Adjacent path indexes of bits included in h when
N = 16, | p| = 6
	Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
	15	1111	7	11	13	14
	14	1110	6	10	12
	13	1101	5	9		12
	11	1011	3		9	10
	7	0111		3	5	6

When J=6, _p={0,1,2,4,3,5}, _B={8,6,11,10}∪r₂, _A={9,12,7,13,14,15}\r₂. Also, R₂ is a candidate bit index set for determining a s₂ value to be moved from B to A. An information set of a polar code in which N=16, J=4 is _h={15,14,13,11,7}, and g₁₅={7,11,13,14}, g₁₄={6,10,12}, g₁₃={5,9,12}, g₁₁={3,9,10}, g₇={3,5,6}. Here, g₁₅∩_h=∅, but g₁₄∩_h=∅, g₁₃∩_h=∅, g₁₁∩_h=∅, g₇∩_h=∅, and thus the electronic device according to an embodiment of the disclosure traces a diversity for bits in which indexes are 14, 13, 11, and 7.

A fading stochastic variable at a depth-2 corresponding to a bit in which an index is 14 is s_14,2=(s_14,0+s_6,0)+(s_10,0+s_2,0), and s_2,0=X=0, s_6,0=s_10,0=B. s_14,2=A+B when not moved, and thus 14 is maintained as a candidate. Also, an intermediate fading stochastic variable at a depth-2 corresponding to a bit in which a bit index is 13 is s_13,2=(s_13,0+s_5,0)+(s_9,0+s_1,0), and s_1,0=s_5,0=X=0, and thus s_13,2=s_13,0+s_9,0·s_13,0=A when not moved, and bits do not deteriorate even when one of the bits is moved to B, and thus are maintained as candidates. (R₂={7,9,12,13,14,15}). A fading stochastic variable at a depth-2 of an index 11 is s_11,2=(s_11,0+s_3,0)(s_15,0+s_7,0). s_3,0=s_11,0=B, and thus s_11,2=B(s_15,0+s_7,0) s_11,2=AB when not moved, and bits do not deteriorate even when one of the bits is moved to B, and thus 7 and 15 are maintained as candidates. A fading Boolean function at a depth-2 of a bit line 7 is s_7,2=(s_7,0s_15,0)+(s_3,0+s_11,0). s_3,0=X=0, s_11,0=B, and thus s_7,2=s_7,0s_15,0·s_7,2=A when not moved, but deteriorates to s_7,2=AB when one bit is moved to B, and thus 7 and 15 are excluded from candidates. In other words, Cambria Math may be updated to R₂={9,12,13,14}. As in a case when J=2,4, the above process is a process of excluding bits causing depth-2 diversity deterioration for g_i satisfying g_i∩_h=0. Referring to the bit index 12 again, s_13,2=s_13,0+s_9,0=A+B. For an index 11, s_11,2=AB, and thus n_s=1. Accordingly, an intermediate diversity or n_s value may be improved during a polarization process after depth-3. A direct path index 11 encounters paths 9 and 10 respectively at depths-3 and 4, and [s_9,2=(s_9,0+0)(s_13,0+0)=s_9,0·s_13,0 and ss_10,2=[B+X][s_14,0+B]]. Accordingly, when 9 and 13 are moved, deterioration of a n_s value is caused, and thus 9 and 13 are excluded from R₂ (R₂={12,14}), and s_14,0 does not cause deterioration and thus is maintained as a candidate. Here, 14 that is an index corresponding to a source bit having highest reliability from among elements in R₂ remaining after all bits not satisfying a condition are excluded may be determined as r₂ (r₂=14).

TABLE 9
Adjacent path indexes of bits included in h when N = 16, | p| = 8
Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
15	1111	7	11	13	14
14	1110	6	10	12
13	1101	5	9		12
11	1011	3		9	10

When J=8 , _p={0,1,2,4,3, 5, 6, 8}, _A={11, 10, 14}∪r₃, _B={7, 9, 12, 13, 15}\r₃. An information set of a polar code in which N=16,J=6 is _h={15, 14, 13, 11}, and g₁₅={7, 11, 13, 14}, g₁₄={6, 10, 12}, g₁₃={5, 9, 12}, g₁₁={3, 9, 10}. Here, g₁₅∪_h≠∅, but g₁₄∪_h≠∅, g₁₃∪_h≠∅, g₁₁∪_h≠∅,and thus diversities for bits in which indexes are 14, 13, and 11 are traced. For a bit index 14, s_14,2 is always s_14,2(s_14,0+s_6,0)+(s_10,0+s_2,0)=s_10,0+s_14,0A . For 13, s_13,2(s_13,0+s_5,0)+(s_9,0+s_1,0)=s_9,0+s_13,0. Because these two bits do not cause deterioration of depth-2 intermediate diversity, the two bits are included in r₃. For 11, s_11,2(s_11,0+s_3,0)(s_15,0+s_7,0)=B(s_15,0+s_7,0). The two bits do not cause deterioration of intermediate diversity, and thus are not removed from r₃. There are not bits removed even through a process after depth-3, and thus an index corresponding to a source bit having highest reliability may be determined as r₃ (r₃=15). Accordingly, an auxiliary sequence for puncturing when N=16 is [r₀r₁r₂r₃]=[11 10 14 15].

2) Method of Constructing Auxiliary Sequence for Shortened Polar Code.

Hereinafter, a method of constructing an auxiliary sequence, according to an embodiment of the disclosure, for codeword bit reallocation when M=2 in a case where a polar code is shortened will be described. Regarding the shortened polar code, when J bits in a u domain to be shortened during encoding are selected and fixed to 0, J bits are also fixed to 0 in an x domain and thus are not transmitted. Shortening also satisfies |_s|=|_s| and _s needs to satisfy binary domination. In the 5G-NR standard, the shortening is mainly considered in an intermediate-high code rate area of R≥7/16.

FIG. 31 is a diagram for describing a diversity updating method for a shortened polar code according to an embodiment of the disclosure.

Referring to FIG. 31_s={3,4}, _B={1}, _A={2}. A fading stochastic variable of shortened bits is indicated as Y, and because the shortened bits are accurately known during decoding, the fading stochastic variable is equivalent to having an infinite LLR, and corresponding fading Y may be indicated as 1. When Y and B encounter, an update corresponding to a check node is always updated to multiplication Y·B=B of two stochastic variables (in case of puncturing, X·B=X), and an update corresponding to a variable node is always updated to OR arithmetic operation Y+B=Y of two stochastic variables (in case of puncturing, X+B=B). In FIG. 31, a fading stochastic variable included in u₁ is (A·Y)(B·Y)=AB, and a fading stochastic variable included in u₂ is (A·Y)+(B·Y)=A+B. In other words, n_s of u₁ is 1, n_s of u₂ is 3, and a diversity of a (2,1) polar code where ={2} is 2.

Hereinafter, a method of determining s_i bit

$\left(i=0,1,\cdots ,\frac{N}{4}-1\right)$

values to be allocated to different resources for a shortened polar code, according to an embodiment of the disclosure, will be described. For example, in a case of N=16, when there is no rate-matching for a code C₁₆=[15 14 13 11 7 12 10 9 6 5 3 8 4 2 1 0] , _B={1,2, 4, 8, 3, 5, 6}, _A={9, 10, 12, 7, 11, 13, 14, 15} is an optimum diversity allocation pattern. However, in a case of a polar code in which J bits are shortened, a s_i value may be determined such that (N−J)/2 bits having highest reliability have a diversity-2 (or n_s=3). Here, incapable bits corresponding to punctured bits are mostly source bits having very low reliability and thus do not affect an information set, but the shortened bits are mostly bits having very high polarization weight, and thus the information set may be greatly changed.

An electronic device according to an embodiment of the disclosure may determine bit values to be allocated to different resources such that a code diversity of an information set _h(|_h|=(N−J)/2) becomes 2 when R=0. 5. For example, an s₀ value is determined such that bits in which indexes are {13,11,7,12,10,9,6} become n_s=3 when J=2 (_s={15,14}), and s₀, s₁, s₂ values are determined such that bits in which indexes are {7, 9, 6, 5, 3} become n_s=3 when J=6 (_s={{15, 14, 13, 11, 12, 10}).

As in puncturing, a set of adjacent path indexes excluding a bit i(i∈_h)=[b_n−1 . . . b₁b₀](b_i∈{0,1}) included in _h, which is encountered during a VN update process, is represented as g_i={i−h_j·2^j: j∈{0, 1, . . . , n−1}}\{i}. As described above, when g_i∩_h=∅, a diversity-2 is not guaranteed, and a “depth-2 polarization” update process therefor is traced.

TABLE 10
Adjacent path indexes of bits included in h when N = 16, J =
2 ( s = {14, 15})
Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
13	1101	5	9		12
11	1011	3		9	10
7	0111		3	5	6
12	1100	4	8
10	1010	2		8
9	1001	1			8
6	0110		2	4

For example, an information set of a polar code in which N=16,1=2 is _h={13,11,7, 12, 10, 9, 6}, and g₁₃={5, 9, 12}, g₁₁={3, 9, 10}, g₇={3, 5, 6}, g₁₂={4, 8}, g₁₀={2, 8}, g₉={1, 8}, g₆={2,4}. First, a bit s o to be moved first is to be determined. Here, a depth-2 diversity is to be investigated and determined fora case where _s={15,14}, _B={0, 1, 2, 4, 8, 3, 5, 6}\s₀ and _A={13,11, 7, 12, 10, 9} U s o , and a candidate set of s₀ is s₀={0, 1, 2, 4, 8, 3, 5, 6}. In other words, S_i is an index set of candidate bits for determining an s_i value. Because at least one bit is included in _h for g₁₃, g₁₁, g₇, it is assumed that a diversity-2 is guaranteed. However, for g₁₂, g₁₀, g₉, g₆, all elements are included in _h^c, and thus a diversity-2 is not guaranteed, and diversity trace up to depth-2 polarization is performed therefor in an order of a code sequence C_N (e.g., (12,10,9,6)). As in the puncturing, a s_i value is determined to maximize a diversity thereof.

For a bit in which an index is 12, a fading stochastic variable updated after a depth-2 polarization process is performed is as follows.

s _12,2=(s_12,0+s_4,0)+(s_8,0+s_0,0)

Here, s_12,0=A and s_4,0, s_0,0, s_8,0 are initially B, an initial stochastic variable that is not moved is A+B. Also, n_s calculated during a depth-2 intermediate stage does not deteriorate even when any one of the bits is moved, and thus s₀(={0, 1, 2, 4, 8, 3, 5, 6}) is maintained. Next, a bit in which an index is 10 is s_10,2=(s_10,0+s_2,0)(s_14,0+s_6,0), and s_10,0=A, Y=1. As a result, s_10,2=s_10,0(=A)+s_2,0(∴Y·B=Y,Y+B=Y). s_10,2=A+B when not moved, but when 2 is moved, s_10,2=A and thus a depth-2 intermediate diversity deteriorates, and thus 2 is removed from s₀(s₀={0, 1, 4, 8, 3, 5, 6}). For a bit in which an index is s_9,2 =(s_9,0+s_1,0)(s_13,0+s_5,0), and s_9,0=s_13,0A, and thus a s_9,2 stochastic variable when not moved is s_9,2=A+(s_1,0+s_5,0)=B+A. Here, when a bit in which an index is 1 or 5 is moved, deterioration to s_9,2=A+AB=A occurs, and thus 1 and 5 are removed from s₀(s₀={0, 4, 8, 3, 6}). Lastly, a depth-2 diversity of a bit in which an index is 6 is provided as s_6,2=(s_6,0+s_14,0)(s_2,0+s_10,0), and s_14,0=Y=1s_10,0=A and thus s_9,2 when not moved is s_9,2 =s_6,2+(s_2,0)·A=B. In this case, an intermediate stage diversity does not deteriorate even when any bit is moved, and thus s₀ is maintained. The above process is a process of removing bits inducing deterioration of n_s during a depth-2 polarization process.

During a following process, a bit having a diversity-2 is searched for through fine-tuning (i.e., observation of depth-3 update or greater). Bits in which indexes are 12, 10, and 9 satisfy a diversity-2 at a depth-2. However, a bit in which an index is 6 does not guarantee a diversity-2 at a depth-2 without being moved. This is because s_6,2[s_6,0s_14,0]+[s_2,0s_10,0]=s_6,0+AB=s_6,0(=B). Thus, a depth is expanded to a depth-3. s_4,2(s_4,0+s_12,0)+(s_0,0+s_8,0) is encountered at the depth-3, and when no bit is moved, s_4,2 includes A, but when “4” is moved, A+B and thus a diversity-2 is satisfied. Accordingly, s₀=4.

TABLE 11
Adjacent path indexes of bits included in h when N = 16, J =
4 (   = {11, 13, 14, 15})
Index	Binary	Depth-1	Depth-2	Depth-3	Depth-4
7	0111		3	5	6
12	1100	4	8
10	1010	2		8
9	1001	1			8
6	0110		2	4
5	0101		1		4

For example, an information set of a polar code in which N=16, J=4 is _h={7,1 2,1 0, 9, 6, 5}, and g₇={3, 5,6}, g₁₂={4, 8}, g₁₀={2,8}, g₉={1, 8}, g₆={2, 4}, g₅={1, 4}. Also, ={15, 14, 13, 11}, _B={0,1,2,8,3,5,6}\s₁ and _A=13,11, 7, 12, 10, 9, 4}∪s₁, and a candidate set of s₁ is s₁={0, 1, 2, 8, 3, 5, 6}. Because at least one bit is included in _h for g₇, it is assumed that a diversity-2 is guaranteed. However, for g₁₂, g₁₀, g₉, g₆, g₅, all bits are included in _h^c, and thus a diversity-2 is not guaranteed, and diversity trace up to depth-2 polarization is performed therefor in an order of a code sequence C_N. First, a bit in which an index is 12 is s_12,2(s_12,0+s_4,0)+(s_8,0+s_0,0) when a depth-2 polarization process is performed. s_12,0=A and s_0,0, s_4,0, s_8,0 that is not moved is B, and thus an initial stochastic variable is A+B. Also, n_s of a depth-2 intermediate stage does not deteriorate even when any bit is moved, and thus S₁(={0, 1, 2, 8, 3, 5, 6}) is maintained. Next, a bit in which an index is 10 is s_10,2=(s_10,0+s_2,0)(s_14,0+s_6,0). s_10,0=A, s_14,0=Y−1 and s_2,0, s_16,0 that is not moved is B, and thus A+B. Deterioration occurs when s_9,0 is moved, and thus 2 is removed from s₁. A bit index 6 does not affect a change in n_s even when moved from B to A, and thus is maintained. (s₁={0, 1, 8, 3, 5, 6}). Next, a bit in which an index is 9 is s_9,2=(s_9,0+s_1,0)(s_13,0+s_5,0). s_9,0=A, s_13,0=Y and s_1,0, s_5,0=AB when not moved, and thus B+A. Deterioration occurs when s_1,0is moved, and thus 1 is removed from s₁. n_s is not affected even when 5 is moved, and thus the electronic device according to an embodiment of the disclosure maintains 5 (s₁={0, 8, 3, 5, 6}). Bits in which indexes are 6 and 5 are s_6,2=(s_6,0+s_14,0)+(s_2,0s_10,0)and s_5,2=(s_5,0s_13,0)+(s_1,0s_9,0), and because s_14,0=Y=1, s_10,0=B, s_9,0=B and s_6,0=B when not moved, s_6,2=B . There is no bit inducing deterioration of n_s , and thus 6 does not change S₁. For an index s_13,0=Y=1, s_9,0=A and s_1,0=s_5,0=B when not moved, and thus s_5,2=B. There is no bit inducing deterioration of n_s when moved, and thus the bit is maintained (s₁={0, 8, 3, 5, 6}). Through the above process, bits inducing deterioration of n_s during a depth-2 polarization process may be removed.

The electronic device according to an embodiment of the disclosure may determine a movement bit for achieving a diversity-2 by observing depth-3 update or greater during a following process. For bit indexes 7, 12, 10, and 9, a diversity-2 may be achieved at a depth-2. However, a diversity-2 is not guaranteed at a depth-2 without movement of indexes 6 and 5.

6 encounters (s_4,0s_12,0)+(s_0,0s_8,0)=A+(s_0,0·s_8,0)=A+B at a depth-3 through a VN operation, and thus n_s=3 is achieved. On the other hand, 5 (B) encounters (s_7,0s_15,0)+(s_3,0s_11,0)=A+s_3,0 at a depth-3. s_3,0 is updated to AB through a CN operation when moved to A, and thus 3 is not moved (S₁={0,8, 5, 6}). At this time, 5 is not increased to n_s=3, and thus the electronic device may observe depth-4 update. (s_4,0s_12,0)+(s_0,0s_8,0)=A+(s_0,0·s_8,0) at a depth-4, and degrade of n s occurs when one of the bits is moved, and thus 0 and 8 may be excluded from S₁. S₁ updated through the above process is S₁={5, 6}, and a bit having low reliability thereamong may be selected as s_i .

In conclusion, a bit inducing n_s deterioration is excluded from a candidate during a depth-2 polarization process, and s_i values may be determined through a method of guaranteeing n_s=3 during polarization after a depth-3. The electronic device according to an embodiment of the disclosure may recursively perform the above method to determine s₀=4, s₁=5, s₂=1, s₃=0.

To operate the puncturing of 1) and the shortening of 2) at once, an “auxiliary sequence” may be defined, and a unified auxiliary sequence may be used by concatenating two auxiliary sequences formed through the above two processes. For example, a front part of an auxiliary sequence T_N may be used for the puncturing and the rear part may be used for the shortening. When N=16 , T₁₆=[11 10 14 15 4 5 1 0]. For example, when 4 bits are punctured, bits having indexes included in T₁₆[0:1] are bits moved to a resource B, and when 6 bits are shortened, bits of T₁₆[4:6] are bits moved to a resource A. A sequence determination method for shortening may be optimized through depth-2 n_s deterioration prevention and n s upgrade method at a subsequent depth described above, or may be directly determined as the one's complement sequence of a sequence determined for puncturing.

For example, a sub-auxiliary sequence for the puncturing is [11 10 14 15], but the one's complements of the indexes are [4 5 1 0], and thus as an index sequence as the above method empirically constructed. In other words, an auxiliary sequence T_N defined for a code length N may be configured by a concatenation T_N=[R_N|S_N] of an auxiliary sequence R N for puncturing and an auxiliary sequence _{N l or shortening. In other words,}

$R_N=\left[r_0{r}_1\cdots r_{\frac{N}{4}-1}\right],S_N=\left[s_0{s}_1\cdots s_{\frac{N}{4}-1}\right].$

FIG. 32 is a graph showing a required SNR [dB } achieving BLER=0.1% according to the number of information bits, according to an interleaver type, under SC decoding. A performance of an interleaver is higher when a y value is smaller according to an embodiment of the disclosure.

Referring to FIG. 32, N=64, and a 2-block fading channel are assumed. Referring to FIG. 32, it is identified that a performance of an interleaving method according to an embodiment of the disclosure is most excellent compared to a performance of a conventional interleaver described above. In particular, an interleaver according to an embodiment of the disclosure has a high performance gain at an intermediate code rate that is a code rate area of interest, compared to a triangular interleaver that exhibits a best performance in the related art. In detail, a performance gain around 5 dB may be secured for a certain code parameter, and the interleaver according to an embodiment of the disclosure guarantees a diversity-2 in a corresponding region, whereas conventional interleavers have relatively low performances as a bit having a low diversity is included in an information set.

FIG. 33 is a graph for describing a performance of an interleaving method in an AWGNC, according to an embodiment of the disclosure.

Referring to FIG. 33, performances of SC decoding (L=1), SCL decoding (L=8), and SCL decoding (L=8) using a 11-bit CRC code are identified when N=512 and 16 QAM.

More particularly, a decoding performance indicates a required SNR [dB] achieving BLER=0.1%, and the smaller the value, the higher the performance. Through the graph of FIG. 33, it is identified that a performance of a block interleaver is slightly greater when an interleaver is not applied, but a performance of a triangular interleaver is noticeably greater. This is because the triangular interleaver satisfactorily distributes a fading effect in a symbol in overall during high-order modulation. Also, it is identified that an interleaver according to an embodiment of the disclosure maintains an almost the same level of performance as the triangular interleaver.

FIG. 34 is a graph for describing a performance of an interleaving method when rate-matching is considered in an AWGNC according to an embodiment of the disclosure.

In detail, a decoding performance is shown according to a data size (K) when R=0.33, and for example, when K=20, E=K/R=60, and thus the graph indicates a required SNR [dB] achieving a BLER 0.1% when 4 bits are punctured from N=64. Even when rate-matching is considered, it is identified that a performance of the interleaver according to an embodiment of the disclosure in the AWGNC maintains a same level as the triangular interleaver according to an embodiment of the disclosure.

FIG. 35 is a graph for describing a performance of an interleaving method according to various decoding methods according to an embodiment of the disclosure.

FIG. 35 illustrates a graph of a required SNR achieving BLER 0.1% for each code rate, according to various decoding methods. It is identified that an interleaver according to an embodiment of the disclosure has an excellent performance of SC decoding compared to conventional interleaver types, and maintains a performance even in SCL and CRC-aided SCL. Also, when a code length increases, an area of the interleaver according to an embodiment of the disclosure, having a high diversity, increases compared to a conventional triangular interleaver, and thus a performance gain width is also increased.

FIG. 36 is a diagram showing diversities included in source bits in an order of a code sequence C_N, according to an interleaver type and the number of punctured bits according to an embodiment of the disclosure.

Referring to FIG. 36, N=32, |_p|∈{0,2,4, . . . ,14}) is assumed, and a diversity and n_s included in source bits of a block channel interleaver, a triangular channel interleaver, and an interleaver according to an embodiment of the disclosure are provided in an order of a code sequence C_N.

In detail, a first column shows an index of each bit having high reliability, for each interleaver type, in a descending order and a second column shows n_s when there is no rate-matching, and third through ninth columns show n_s during rate-matching. Regarding a diversity (or n_s) analysis result |_p|=0 (i.e., no rate-matching; plain polar code), the interleaver according to an embodiment of the disclosure has a diversity-2 when R≤0.5 as described above. In other words, all bits in an information set has n_s=3 when R=0.5. Not only when the rate-matching is not considered, but also when several bits are punctured, the interleaver according to an embodiment of the disclosure may allocate high n_s to information bits most stably. For example, a triangular interleaver guarantees a diversity-2 only up to K=6 when the number of punctured bits is 12 or 14, but the interleaver according to an embodiment of the disclosure guarantees a diversity-2 up to K=8.

FIG. 37 is a graph for describing a decoding performance according to an interleaver type according to an embodiment of the disclosure.

Referring to FIG. 37, an SNR vs BLER graph is illustrated according to an interleaver type when |_p|=20,40. In particular, a left graph shows a performance of a (108,36) code and a right graph shows a performance of a (88,33) polar code. It is identified that an interleaver according to an embodiment of the disclosure, which is modified at a corresponding parameter, has a steeper curve than a conventional technique, and in particular, solely guarantees a diversity-2 for a (88,33) polar code. Here, a required SNR gain achieving BLER=0.1% is about 6 to 7 dB.

FIG. 38 is a diagram showing diversities included in source bits in an order of a code sequence C_N, according to an interleaver type and the number of shortened bits according to an embodiment of the disclosure.

Referring to FIG. 38, diversities n_s included in source bits for a block channel interleaver, a triangular channel interleaver, and an interleaver according to an embodiment of the disclosure are illustrated in an order of a code sequence C_N when N=32 and shortening of |_s|{0,2,4, ,14} is applied.

When there is no rate-matching, the interleaver according to an embodiment of the disclosure always has a diversity-2 when R=0.5. In other words, all bits in an information set _h has n_s=3 when R=0.5 . Also, the interleaver according to an embodiment of the disclosure has the maximum number K_max of information bits in various shortening environments compared to the block interleaver or triangular interleaver, and thus may secure a high performance gain.

FIG. 39 is a graph for analyzing a shortening performance of a polar code according to an embodiment of the disclosure.

Referring to FIG. 39, a shortening performance of a (20,9) polar code is illustrated for N=32, |_s|=12, and it is identified that, compared to conventional techniques, an interleaver according to an embodiment of the disclosure has a diversity-2, and thus obtains a high performance gain.

FIG. 40 is a block diagram of an electronic device for transmitting a signal to which an interleaving method is applied according to an embodiment of the disclosure.

Referring to FIG. 40, an electronic device 4000 may include a transceiver 4010, a processor 4020, and a memory 4030. The transceiver 4010, processor 4020, and memory 4030 of the electronic device 4000 may operate according to a method by which the electronic device 4000 transmits a polar code, described above. However, the components of the electronic device 4000 are not limited thereto. For example, the electronic device 4000 may be implemented by more components or fewer components than those described above. According to an embodiment of the disclosure, the transceiver 4010, processor 4020, and the memory 4030 may be implemented as a single chip. Also, the processor 4020 may include one or more processors.

The transceiver 4010 is a collective term of a receiver of the electronic device 4000 and a transmitter of the electronic device 4000, and may transmit/receive a signal to/from another electronic device or a network entity. The signal transmitted/received to/from the other electronic device or network entity may include control information and data. In this regard, the transceiver 4010 may include a RF transmitter for up-converting and amplifying a frequency of a transmitted signal, and a RF receiver for amplifying low-noise and down-converting a frequency of a received signal. However, this is only an example of the transceiver 4010 and components of the transceiver 4010 are not limited to the RF transmitter and the RF receiver.

The transceiver 4010 may perform functions for transmitting or receiving a signal through a wireless channel. For example, the transceiver 4010 may receive a signal through a wireless channel and output the signal to the processor 4020, or transmit a signal output from the processor 4020 through a wireless channel.

The memory 4030 may store a program and data required for operations of the electronic device 4000. Also, the memory 4030 may store control information or data included in a signal obtained from the other electronic device or the network entity. The memory 4030 may be a storage medium, such as read-only memory (ROM), random-access memory (RAM), a hard disk, a CD-ROM, and a DVD, or a combination of storage media. Also, the memory 4030 may not be present separately but may be included in the processor 4020. The memory 4030 may be configured in a volatile memory, a nonvolatile memory, or a combination of a volatile memory and a nonvolatile memory. Also, the memory 4030 may provide the stored data upon request by the processor 4020.

The processor 4020 may control a series of processes such that the electronic device 4000 operates according to an embodiment of the disclosure described above.

The processor 4020 may identify BD symmetry for indexes of codeword bits configuring a polar code sequence. When indexes of the codeword bits satisfy the BD symmetry, the processor 4020 may perform interleaving on codewords (x₀, x₁, . . . , x_N−1) arranged in an index ascending order, based on reliability of source bits corresponding to the codeword bits. Here, to align a high diversity to information bits of an AWGN code, a method of evaluating diversities of source bits, based on Boolean approximation, is used first for fading events of blocks, and a BD-symmetric code sequence is used as an interleaver sequence such that the information bits have high diversities. The processor 4020 may allocate interleaved codeword bits to the blocks, respectively. The processor 4020 may transmit at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocation, on at least one resource block corresponding to each of the blocks.

The processor 4020 according to an embodiment may identify codeword bit pairs breaking the BD symmetry in each bit pattern, for codeword bit patterns allocated to a block 1 and a block 2. The processor 4020 may permutate the identified codeword bits with each other. Interleaving may be performed on a polar code sequence obtained by permutating the identified codeword bits with each other.

When puncturing or shortening is applied to the polar code sequence, the processor 4020 may determine candidate codeword bits movable from a first block to a second block from among the blocks. The processor 4020 may evaluate and identify a change in a diversity value for a case where each of the determined candidate codeword bits move. The processor 4020 may determine at least one codeword bit that is a moving target from among the candidate codeword bits, based on the identified change in the diversity value.

For example, the processor 4020 may obtain a diversity value during a depth-2 polarization process for a case where codeword bits included in the determined sequence each move. The processor 4020 may compare the obtained diversity value with a diversity value during a depth d (d=1,2, . . . , n) polarization process before each codeword bit included in the determined sequence moves. A bit in which a diversity value is increased or maintained during movement, from among the codeword bits included in the determined sequence, may be determined as a bit that is the moving target, based on a result of the comparing.

Also, the processor 4020 may obtain a diversity value during a depth d+1 polarization process when a bit in which a diversity value is increased or maintained during the movement is not identified during the depth d (d=1,2, . . . , n) polarization process.

The processor 4020 according to an embodiment of the disclosure may allocate the interleaved codeword bits to the blocks, respectively, based on an auxiliary sequence including the determined at least one codeword bit. The processor 4020 according to an embodiment of the disclosure may perform interleaving using a code sequence based on reliability of at least one codeword bit that is the moving target and source bits corresponding to the codeword bits.

FIG. 41 is a block diagram of an electronic device for receiving a signal to which an interleaving method is applied according to an embodiment of the disclosure.

Referring to FIG. 41, an electronic device 4100 may include a transceiver 4110, a processor 4120, and a memory 4130. The transceiver 4110, processor 4120, and memory 4130 of the electronic device 4100 may operate according to a method by which the electronic device 4100 transmits a polar code, described above. However, the components of the electronic device 4100 are not limited thereto. For example, the electronic device 4100 may be implemented by more components or fewer components than those described above. According to an embodiment of the disclosure, the transceiver 4110, processor 4120, and the memory 4130 may be implemented as a single chip. Also, the processor 4120 may include one or more processors.

The transceiver 4110 is a collective term of a receiver of the electronic device 4100 and a transmitter of the electronic device 4100, and may transmit/receive a signal to/from another electronic device or a network entity. The signal transmitted/received to/from the other electronic device or network entity may include control information and data. In this regard, the transceiver 4110 may include a RF transmitter for up-converting and amplifying a frequency of a transmitted signal, and a RF receiver for amplifying low-noise and down-converting a frequency of a received signal. However, this is only an example of the transceiver 4110 and components of the transceiver 4110 are not limited to the RF transmitter and the RF receiver.

The transceiver 4110 may perform functions for transmitting or receiving a signal through a wireless channel. For example, the transceiver 4110 may receive a signal through a wireless channel and output the signal to the processor 4120, or transmit a signal output from the processor 4120 through a wireless channel.

The memory 4130 may store a program and data required for operations of the electronic device 4100. Also, the memory 4130 may store control information or data included in a signal obtained from the other electronic device or the network entity. The memory 4130 may be a storage medium, such as read-only memory (ROM), random-access memory (RAM), a hard disk, a CD-ROM, and a DVD, or a combination of storage media. Also, the memory 4130 may not be present separately but may be included in the processor 4120. The memory 4130 may be configured in a volatile memory, a nonvolatile memory, or a combination of a volatile memory and a nonvolatile memory. Also, the memory 4130 may provide the stored data upon request by the processor 4120.

The processor 4120 may control a series of processes such that the electronic device 4100 operates according to an embodiment of the disclosure described above.

The processor 4120 may obtain codeword bits by performing demodulation on a signal received on each of at least one resource block from another electronic device (e.g., the electronic device 4000).

The processor 4120 may perform deinterleaving on the obtained codeword bits, based on reliability of source bits corresponding to the obtained codeword bits. The processor 4120 may perform decoding on a polar code sequence including the deinterleaved codeword bits.

When puncturing or shortening is applied to the polar code sequence by the other electronic device (e.g., the electronic device 4000), the processor 4120 may identify at least one codeword bit determined as a moving target from among candidate codeword bits, based on a change in a diversity value for a case where each candidate codeword bit movable from a first block to a second block from among blocks moves. The processor 4120 may perform the deinterleaving on the codeword bits, based on the at least one identified codeword bit and the polar code sequence.

An interleaver according to an embodiment of the disclosure may supplement BICM structural vulnerability of a conventional 5G polar code, thereby obtaining a performance improvement effect according to a diversity effect of a block fading channel while maintaining a good performance in an AWGNC. Also, an interleaving method according to an embodiment of the disclosure may secure excellent performances not only for SC decoding, but also for various decoding methods, such as SCL decoding, CRC-aided SCL decoding, and the like. Also, the interleaving method according to an embodiment of the disclosure may secure an excellent performance when rate-matching is applied to not only a fading channel, but also an AWGNC.

A machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, the “non-transitory storage medium” only denotes a tangible device and does not contain a signal (for example, electromagnetic waves). This term does not distinguish a case where data is stored in the storage medium semi-permanently and a case where the data is stored in the storage medium temporarily. For example, the “non-transitory storage medium” may include a buffer where data is temporarily stored.

According to an embodiment of the disclosure, a method according to various embodiments of the disclosure in the disclosure may be provided by being included in a computer program product. The computer program products are products that can be traded between sellers and buyers. The computer program product may be distributed in a form of machine-readable storage medium (for example, a compact disc read-only memory (CD-ROM)), or distributed (for example, downloaded or uploaded) through an application store or directly or online between two user devices (for example, smart phones). In the case of online distribution, at least a part of the computer program product (for example, a downloadable application) may be at least temporarily generated or temporarily stored in a machine-readable storage medium, such as a server of a manufacturer, a server of an application store, or a memory of a relay server.

While the disclosure has been shown and described with reference to various embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims and their equivalents.

Claims

1. A method, performed by a first electronic device, of transmitting a polar code, the method comprising: performing interleaving on codeword bits, according to a polar code sequence corresponding to the codeword bits;allocating the interleaved codeword bits respectively to blocks; andtransmitting at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocating, on at least one resource block corresponding to each of the blocks.

2. The method of claim 1, further comprising: identifying codeword bits breaking binary domination (BD) symmetry between index patterns allocated to a first block and a second block, from among the codeword bits; andpermutating the identified codeword bits with each other,wherein the performing of the interleaving on the codeword bits comprises performing the interleaving on the codeword bits obtained after the permutating, according to the polar code sequence.

3. The method of claim 1, further comprising: in case that code modification of puncturing or shortening is applied to the polar code sequence, determining candidate codeword bits movable from a first block to a second block from among the blocks;identifying a change in a diversity value for a case where the determined candidate codeword bits each move; anddetermining at least one codeword bit that is a moving target from among the candidate codeword bits, based on the identified change in the diversity value.

4. The method of claim 3, wherein the allocating comprises allocating the interleaved codeword bits to different blocks, based on an auxiliary sequence including the determined at least one codeword bit.

5. The method of claim 3, wherein the performing of the interleaving comprises performing the interleaving, based on a pre-defined auxiliary sequence corresponding to the at least one codeword bit that is the moving target and the codeword bits, andwherein the pre-defined auxiliary sequence is generated by unifying the polar code sequence and an auxiliary sequence.

6. A method, performed by a second electronic device, of receiving a polar code, the method comprising: obtaining codeword bits by performing demodulation on a signal received from a first electronic device on each of at least one resource block;performing deinterleaving on the codeword bits by using a polar code sequence corresponding to the codeword bits; andperforming decoding on a sequence including the deinterleaved codeword bits.

7. The method of claim 6, wherein, in case that the first electronic device has applied code modification of puncturing or shortening to the codeword bits, at least one codeword bit that is a moving target is determined from among candidate codeword bits movable from a first block to a second block from among blocks, based on a change in a diversity value for a case where each of the candidate codeword bits moves, andwherein the codeword bits obtained as a result of the demodulation have been interleaved based on the polar code sequence and the determined at least one codeword bit.

8. The method of claim 7, wherein the performing of the deinterleaving comprises performing the deinterleaving on codeword bits allocated to different blocks, based on an auxiliary sequence including the determined at least one codeword bit.

9. The method of claim 7, wherein the performing of the deinterleaving comprises performing the deinterleaving on the codeword bits, based on a pre-defined auxiliary sequence corresponding to the at least one codeword bit that is the moving target and the codeword bits, andwherein the pre-defined auxiliary sequence is generated by unifying the polar code sequence and an auxiliary sequence.

10. A first electronic device for transmitting a polar code, the first electronic device comprising: a transceiver; andat least one processor coupled with the transceiver and configured to: perform interleaving on codeword bits, according to a polar code sequence corresponding to the codeword bits,allocate the interleaved codeword bits respectively to blocks, andtransmit at least one symbol generated as a result of performing modulation on the interleaved codeword bits for each block, based on the allocation, on at least one resource block corresponding to each of the blocks.

11. The first electronic device of claim 10, wherein the at least one processor is further configured to: identify, from among the codeword bits, codeword bits breaking binary domination (BD) symmetry between index patterns allocated to a first block and a second block;permutate the identified codeword bits with each other; andperform the interleaving on codeword bits obtained as a result of permutating the identified codeword bits with each other, according to the polar code sequence.

12. The first electronic device of claim 10, wherein the at least one processor is further configured to: in case that code modification of puncturing or shortening is applied to the polar code sequence, determine candidate codeword bits movable from a first block to a second block from among the blocks;identify a change in a diversity value for a case where the determined candidate codeword bits each move; anddetermine at least one codeword bit that is a moving target from among the candidate codeword bits, based on the identified change in the diversity value.

13. The first electronic device of claim 12, wherein the at least one processor is further configured to allocate the interleaved codeword bits to different blocks, based on an auxiliary sequence including the determined at least one codeword bit.

14. The first electronic device of claim 12, wherein the at least one processor is further configured to perform the interleaving, based on a pre-defined auxiliary sequence corresponding to the at least one codeword bit that is the moving target and the codeword bits, andwherein the pre-defined auxiliary sequence is generated by unifying the polar code sequence and an auxiliary sequence.

15. The first electronic device of claim 12, wherein the at least one processor is further configured to: obtain a diversity value during a depth-2 polarization process for a case where each of codeword bits move;compare the obtained diversity value with a diversity value during a depth d polarization process before the codeword bits included in the determined sequence each move; anddetermine, as a bit that is the moving target, a bit in which a diversity value is increased or maintained during movement, from among the codeword bits included in the determined sequence, based on a result of the comparison.

16. The first electronic device of claim 15, wherein the at least one processor is further configured to, in case that the bit in which a diversity value is increased or maintained during movement is not identified during the depth d polarization process, obtain a diversity value during a depth d+1 polarization process.

17. A second electronic device for receiving a polar code, the second electronic device comprising: a transceiver; andat least one processor connected to the transceiver and configured to: obtain codeword bits by performing demodulation on a signal received from a first electronic device on each of at least one resource block,perform deinterleaving on the codeword bits by using a polar code sequence corresponding to the codeword bits, andperform decoding on a sequence including the deinterleaved codeword bits.

18. The second electronic device of claim 17, wherein, in case that the first electronic device has applied code modification of puncturing or shortening to the codeword bits, at least one codeword bit that is a moving target is determined from among candidate codeword bits movable from a first block to a second block from among blocks, based on a change in a diversity value for a case where each of the candidate codeword bits moves, andwherein the codeword bits obtained as a result of the demodulation have been interleaved based on the polar code sequence and the determined at least one codeword bit.

19. The second electronic device of claim 18, wherein the at least one processor is further configured to perform the deinterleaving on codeword bits allocated to different blocks, based on an auxiliary sequence including the determined at least one codeword bit.

20. The second electronic device of claim 18, wherein the at least one processor is further configured to perform the deinterleaving on the codeword bits, based on a pre-defined auxiliary sequence corresponding to the at least one codeword bit that is the moving target and the codeword bits, andwherein the pre-defined auxiliary sequence is generated by unifying the polar code sequence and an auxiliary sequence.