This application relates to signal filtering, and more particularly, to signal filtering in wireless communications.
In electronics and telecommunications, pulse shaping is applied to signals to limit the effective bandwidth of the signal's transmission in order to minimize intersymbol interference (ISI). ISI is a form of distortion of a signal in which one symbol interferes with subsequent symbols. Pulse shaping involves shaping the transmitted signals so that they occupy less bandwidth, thereby keeping intersymbol interference under control.
Common idealized forms of pulse shaping are the sinc filter (aka boxcar filter), the Gaussian filter, and the Nyquist filter. In practice, these idealized filters can only be approximated.
A Nyquist filter is a general term used for a filter with a specific property of bring able to recover data filtered with a zero inter-symbol interference (ISI) property. One approximation of the Nyquist filter is the raised cosine (RC) filter. RC filters often suffer from poor smoothness properties that can degrade performance in pulse shaping for wireless communications. In order to improve tolerance for noise in the system, the RC filter is sometimes implemented using two cascaded square-root RC filters, one on the sender side and one on the receiver side. Because the filters' amplitude responses are pointwise square roots of an RC filter, pulse shaping is equally distributed between the sender and receiver filters.
U.S. Pat. No. 10,103,928, which issued on Oct. 16, 2018 and is incorporated herein by reference, describes the use of Nyquist and Square Root Nyquist Filters for pulse shaping in wireless communications. That patent disclosed advantages to using higher order raised cosine filters. In particular, a generalized raised cosine (GRC) filter consisting of matched transmit and receiver filters had improved time-frequency localization compared with a raised cosine filter (RC) filter. However, there has been a great deal of inertia in the industry against replacing a square root raised cosine filter at the transmitter with another filter design.
There is need to improve receiver selectivity performance in a split RC filter while enabling the industry to continue employing root raised cosine filters at the transmitter. The description herein sets forth embodiments that use asymmetric factorization of higher order Nyquist structures with square root raised cosine at the transmitter and the remaining factors of the higher order Nyquist filter at the receiver.
Accordingly, the disclosure herein distinguishes itself from that of U.S. Pat. No. 10,103,928 by focusing on a special case where the transmit filter is separated out from the rest of the composite filter response. More specifically, this disclosure considers the case where the transmit filter is a square root raised cosine filter (which the industry is wont to keep) and the receiver is something else and higher order. This disclosure demonstrates the possibility of employing root raised cosine at the transmitter while still enjoying the benefits of improved selectivity of higher order raised cosine filters at the receiver. These benefits are effectively accomplished by rolling over a cos(z/2) term from an idealized cos{circumflex over ( )}2(z/2) transmit filter response to the receive side, which results in an improved response at the receiver.
In one embodiment, an apparatus to receive wireless communications is disclosed including an antenna, receive circuitry coupled to receive input signals from the antenna that have a pulse shaping modulation, and a receive filter within the receive circuitry coupled to receive the input signals and to remove the pulse shaping modulation, where the receive filter is an asymmetric factorization of a higher order generalized Nyquist filter that has a frequency response (H(f)) represented by frequency response equations. The cascaded combination of the transmit and receive filters has a frequency response equivalent to a higher order generalized raised cosine filter response.
The asymmetric partition of a higher order Nyquist receive filter has a frequency response Rn(ƒ) given by:
wherein n represents the order of the higher order receive filter and is greater than or equal to one,
the roll-off transition region is defined by
ƒ represents frequency, T is the symbol period, β represents a roll-off factor; and
Furthermore, the coefficients An for the frequency response Rn(z) of the receive filter may satisfy a matrix M(n) for any given integer n, wherein:
In one implementation, the receive filter is implemented in the time domain. In another implementation, the receive filter is implemented in the frequency domain. In various implementations, the apparatus further comprises a decoder within the receive circuitry coupled to receive filtered signals from the receive filter and to output decoded signals and MAC (media access control) circuitry coupled to receive the decoded signals from the decoder and to output processed signals through an external interface.
In another embodiment, the receive filter comprises a linear combination of higher order Nyquist subfilters. For instance, the receive filter may comprise a combination of nth-order sub-filters, each sub-filter having a frequency response Rk(z) in a roll-off transition region, and the combination of sub-filters has a frequency response RC(z), in which
for a combination of m sub-filters, α0 through αm are coefficients that represent weights of the different order Nyquist filters, each having a value greater than or equal to 0, that add up to one; and Rk(z) represents the frequency responses of the sub-filters. Furthermore, the frequency responses of the sub-filters may be characterized by the equation
wherein each value k uniquely designates one of the nth-order sub-filters, and the remaining variables are defined as above.
In another embodiment, a higher order generalized raised cosine cascaded filter comprises both a root raised cosine transmit filter and a higher order receive filter, wherein n represents the order of the higher order receive filter and n is greater than or equal to one. In one implementation, the transmit filter is a square root raised cosine filter, and the frequency response X(z) of the transmit filter is expressed by the following:
wherein ƒ represents the frequency of a signal transmitted through the higher order generalized raised cosine cascaded filter and the remaining variables are defined as above.
In another implementation, the frequency response H(z) of the higher order generalized cosine cascaded filter is given by the formula H(z)=X(z)*R(z), where X(z) is a frequency response of the root raised cosine filter and R (z) is a frequency response of the n-order receive filter, and the remaining variables are as defined above. In yet another implementation,
In yet another embodiment, a method of wireless communication is provided comprising receiving signals that have been pulse-shape modulated by a root-raised cosine transmit filter of a transmitter and filtering the signals with a receive filter to remove the pulse shaping modulation, wherein the receive filter is a is a higher order Nyquist filter, and the cascaded combination of the transmit and receive filters has a frequency response equivalent to a higher order generalized raised cosine filter response. The method may further involve applying pulse-shaping modulation to pre-transmitted signals using the root-raised cosine transmit filter to generate transmit signals having the pulse shaping modulation, and transmitting the signals through a transmit antenna.
The method is also applicable to a receive filter comprising a linear combination of nth-order sub-filters, each sub-filter having a frequency response (Rk(z)) in a roll-off transition region, and the combination of sub-filters has a frequency response (RC(z)), as represented by the equations described above.
In one embodiment, the invention is applied to 5G New Radio, where there has been an emphasis on improved spectral occupancy. It is possible with higher order raised cosine filters to improve the spectral occupancy because it is the ability of the cascaded filters to prevent adjacent channel leakage, not just the transmit filter.
Other systems, devices, methods, features, and advantages of the disclosed product and methods for pulse shaping will be apparent or will become apparent to one with skill in the art upon examination of the following figures and detailed description. All such additional systems, devices, methods, features, and advantages are intended to be included within the description and to be protected by the accompanying claims.
The present disclosure may be better understood with reference to the following figures. Corresponding reference numerals designate corresponding parts throughout the figures, and components in the figures are not necessarily to scale.
It will be appreciated that the drawings are provided for illustrative purposes and that the invention is not limited to the illustrated embodiment. For clarity and in order to emphasize certain features, not all of the drawings depict all of the features that might be included with the depicted embodiment. The invention also encompasses embodiments that combine features illustrated in multiple different drawings; embodiments that omit, modify, or replace some of the features depicted; and embodiments that include features not illustrated in the drawings. Therefore, it should be understood that there is no restrictive one-to-one correspondence between any given embodiment of the invention and any of the drawings.
The present invention discloses embodiments of an imbalanced split RC filter comprising a lower-order filter, such as a root-raised cosine (RC) filter, at the transmitter paired with a higher order filter, such as a Nyquist filter, at the receiver. While one embodiment presents a general transmit/receive architecture for wireless communications, it will be understood that the Nyquist filter structures described herein can be used in a wide variety of applications and system implementations. It is noted that different and/or additional features can also be implemented, as desired, and related systems and methods can be utilized as well. It is further noted that example Nyquist filters and raised cosine filters are described, for example, in U.S. Pat. Nos. 6,628,728, 5,487,089, and 4,731,800, each of which is hereby incorporated by reference in its entirety.
Any reference to “invention” within this document is a reference to an embodiment of a family of inventions, with no single embodiment including features that are necessarily included in all embodiments, unless otherwise stated. Furthermore, although there may be references to “advantages” provided by some embodiments, other embodiments may not include those same advantages, or may include different advantages. Any advantages described herein are not to be construed as limiting to any of the claims.
Specific quantities (e.g., spatial dimensions) may be used explicitly or implicitly herein as examples only and are approximate values unless otherwise indicated. Discussions pertaining to specific compositions of matter, if present, are presented as examples only and do not limit the applicability of other compositions of matter, especially other compositions of matter with similar properties, unless otherwise indicated.
In describing preferred and alternate embodiments of the technology described herein, specific terminology is employed for the sake of clarity. Technology described herein, however, is not intended to be limited to the specific terminology so selected, and it is to be understood that each specific element includes all technical equivalents that operate similarly to accomplish similar functions.
The transmit (TX) path circuitry 109/159 for the access point 102 and the user device 152 includes a modulator/coder (Mod/Cod) 122/172 that modulates and/or codes data 120/170 onto a carrier signal that is combined with pilot signal 118/168 by a multiplexer (MUX) 116/166 and provided to output circuitry. The output circuitry further provides rate converter 114/164, pulse shape filter 110/160, and/or other functions to the transmit signals (e.g., carrier plus pilot) prior to the signal being transmitted through an antenna 126/176 to the receive section 129/179 of the other system. As described herein, a Nyquist filter 112/162 and more particularly a square root raised cosine filter is used to add the pulse shaping modulation to the transmit signals. Further, output circuitry for the transmit path circuitry 109/159 optionally includes a radio frequency front-end (RFFE) 124/174 that provides digital-to-analog converters, up-conversion mixers, power amplifiers, and/or other output circuitry that facilitates the transmission of the wireless signals through the antenna 126/176 that have been modulated using the pulse shaping function applied by the pulse shape filter 112/162.
Each receive (RX) path circuitry 129/179 for the access point 102 and the user device 152 receives the transmitted wireless signals (e.g., carrier and pilot signal) through an antenna 144/194 and provides it to a receive filter 130/180. The receive filter 130/180 is an asymmetric partition of a higher order (n>=1) generalized Nyquist filter, examples of which are provided further below, that removes the pulse shaping modulation provided by the pulse shape filter 112/162 and outputs a filtered signal to a frequency domain equalization (FDE) module 138/188. A synchronization (SYNC) block 134/184 also receives the filtered signal and provides a synchronization signal to the channel estimator (Chan Est) 136/186 and the FDE module 138/188. The channel estimator 136/186 provides a received signal strength estimate to the FDE module 138/188. The output of the FDE module 138/188 is provided to the decoder (DECODE) 140/180 which decodes and demodulates the received signals. Further, input circuitry for the receive path circuitry 129/179 optionally includes a radio frequency front-end (RFFE) 142/192 that provides analog-to-digital converters, down-conversion mixers, low noise amplifiers, and/or other input circuitry that facilitates the reception of the wireless signals through the antenna 144/194.
The MAC processing circuitry 104/154 for the access point 102 and the user device 152 includes one or more processing features such as mobile communication services (MCS), beam steering, uplink (UL) control, downlink (DL) control, and/or other MAC processing features. The MAC processing circuitry 104/154 also provides an external communication interface 106/156, for example, to other processing circuits such as host processors. The TX/RX signal processing circuitry 108/158 and MAC processing circuitry 104/154 optionally communicate information with each other such as MAC parameters, payload information, PHY measurements, decoded data, and/or other desired information.
The Nyquist filters described below can be used, for example, with respect to the pulse shape filter 110/160 in the transmit (TX) path circuitry 109/159 and/or with respect to the receive filter 130/180 in the receive (RX) path circuitry 129/179 for the example embodiment provided in
In various embodiments of the invention, a higher order generalized raised cosine filter is provided that comprises a square root raised cosine filter at the transmitter paired a high order Nyquist filter at the receiver. The following paragraphs construct a mathematical model of the frequency response of the cascaded filter for different orders of receiver filters. Each combination constitutes a separate embodiment of the invention.
The following paragraphs characterize the frequency response H(z) of the cascaded filter in the roll-off transition region:
where ƒ represents the frequency of the signal, T is the symbol period, and # represents a roll-off factor for the frequency response equations.
Because the cascaded filter comprises a transmitter filter paired with a receiver filter, the cascaded filter transitional region's frequency response H(z) equals the product of the frequency response X(z) of the transmitter filter and the frequency response R(z) of the receiver filter in the transition region, as in:
H(z)=X(z)R(z) (2)
H(z) is further characterized by the following formula:
where n represents the order of an nth order filter. The value n designates the Fourier series components and their amplitude coefficients—or, to state it another way, An are coefficients and cos(nz) are the frequency components of the Fourier series. The value z is defined by the following formula:
H(z) only defines the frequency response in the transition region. More generally, the frequency behavior or response H(ƒ) of the raised-cosine filter for all values of ƒ is defined as:
This filter could equivalently be implemented in the time domain by taking the inverse Fourier transform of H(f).
In the following paragraphs, the frequency response H(z) of the cascaded filter which includes a root raised cosine filter at the transmitter—is constructed for the cases of n=0, 1 and 2 which represent the order of the receive filter.
For the trivial case n=0 where direct square root factorization applies, R0(z) constitutes the zero-order receive filter. The value for T was set to 1 without loss of generality. The values H(z) and R(z) for this case are represented as follows:
For n=1 in the transition region of the frequency domain:
characterizes the frequency response of the first order receive filter.
For n=2:
The equations (6), (8), and (10) above are derived from trigonometric identities of equation (3). Accordingly, the asymmetric partition of a higher order Nyquist receive filter has a frequency response Rn(ƒ) given by:
wherein n represents the order of the higher order Nyquist receive filter and is greater than or equal to one;
the roll-off transition region is defined by
represents frequency; T is the symbol period; β represents a roll-off factor; and:
As seen above, for all cases of n and for all n-order filters of this type, the root raised cosine transmitter filter has a frequency response X(ƒ) given by the following equation:
wherein ƒ represents the frequency of a signal transmitted through the higher order generalized raised cosine cascaded filter; T is the symbol period; β represents a roll-off factor; and z is given by the equation
Unlike the transmit filter, the receive filter is characterized differently for each value of n. More specifically, the values of the coefficients An for the frequency response Rn(z) of the receive filter satisfy a square Vandermonde matrix M(n) for the order n of the receive filter, where n is an integer of at least 1 and:
These values are determined by solving the system of equations generated by eq. (13). For example, for n=1, a 2×2 matrix is used where A1+A3=1 and A1+9A3=0, the solution for which is A1=9/8 and A3=−1/8, which is consistent with eq. (8).
For the case of n=3, there are four simultaneous equations where:
1A1+1A3+1A5+1A7=1
12A1+32A3+52A5+72A7=A1+9A3+25A5+49A7=0
12*2A1+32*2A3+52*2A5+72*2A7=A1+81A3+625A5+2402A7=0
12*3A1+32*3A3+52*3A5+72*3A7=A1+729A3+15625A5+117649A7=0
Solving for these four variables yields the following:
By rearranging eq. (2) and substituting cos (for X(z), one can solve for R(z). Accordingly, the higher order Nyquist receive filter has a frequency response (R(z)) in a roll-off transition region given by:
wherein n represents the order of the higher order receive filter and is greater than or equal to one,
the roll-off transition region is defined by
ƒ represents frequency; T is the symbol period, and β represents a roll-off factor.
In other embodiments, the frequency response of the receive filter is given by a linear combination of higher order terms with this factorization for a full design space of filter options. The composite frequency response RC(z) of such receive filters is expressed as follows:
=α0R0(z)+α1R1(z)+α2R2(z)+ . . . +αmRm(z)
wherein for any given coefficient αk, αk≥0, and the sum of those coefficients equals one, as in α0+α1+α2+ . . . +αm=1.
The frequency responses of the sub-filters are characterized by the equation:
where n represents the order of the higher order receive sub-filter and is greater than or equal to one, each value k uniquely designates one of the nth-order sub-filters, A0, A1, A3, . . . A2n+1 represent cosine coefficients for each sub-filter,
ƒ represents frequency; T is a symbol period, β represents a roll-off factor; and the roll-off transition region is defined by:
Moreover, the cosine coefficients A0, A1, A3 of each subfilter satisfy the matrix M(n) set forth in eq. (13).
It is noted that the functional blocks described herein can be implemented using hardware, software, or a combination of hardware and software, as desired. In addition, one or more processors or processing circuitry running software and/or firmware can also be used, as desired, to implement the disclosed embodiments. It is further understood that one or more of the operations, tasks, functions, or methodologies described herein may be implemented, for example, as software or firmware and/or other program instructions that are embodied in one or more non-transitory tangible computer readable mediums (e.g., memory) and that are executed by one or more controllers, microcontrollers, microprocessors, hardware accelerators, and/or other processors or processing circuitry to perform the operations and functions described herein.
It is further noted that the functional blocks, devices, and/or circuitry described herein can be implemented using hardware, software, or a combination of hardware and software. In addition, one or more processors (e.g., central processing units (CPUs), controllers, microcontrollers, microprocessors, hardware accelerators, programmable integrated circuitry, FPGAs (field programmable gate arrays), ASICs (application specific integrated circuits), and/or other programmable processing circuitry) can be programmed to perform the operations, tasks, functions, or actions described herein for the disclosed embodiments. For example, the one or more electronic circuits can be configured to execute or otherwise be programmed with software, firmware, logic, and/or other program instructions stored in one or more non-transitory tangible computer-readable mediums (e.g., data storage devices, flash memory, random access memory, read only memory, programmable memory devices, reprogrammable storage devices, hard drives, floppy disks, DVDs, CD-ROMs, and/or any other tangible data storage medium) to perform the operations, tasks, functions, or actions described herein for the disclosed embodiments.
Further modifications and alternative embodiments of this invention will be apparent to those skilled in the art in view of this description. It will be recognized, therefore, that the present invention is not limited by these example arrangements. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the manner of carrying out the invention. It is to be understood that the forms of the invention herein shown and described are to be taken as the presently preferred embodiments. Various changes may be made in the implementations and architectures. For example, equivalent elements may be substituted for those illustrated and described herein, and certain features of the invention may be utilized independently of the use of other features, all as would be apparent to one skilled in the art after having the benefit of this description of the invention.
Number | Name | Date | Kind |
---|---|---|---|
4731800 | Motley et al. | Mar 1988 | A |
5487089 | Misaizu et al. | Jan 1996 | A |
6373955 | Hooley | Apr 2002 | B1 |
6628728 | McCarty, Jr. | Sep 2003 | B1 |
8325755 | Lee et al. | Dec 2012 | B2 |
9191161 | Li et al. | Nov 2015 | B2 |
10103928 | McCoy et al. | Oct 2018 | B2 |
20070281633 | Papadopoulos | Dec 2007 | A1 |
20150326286 | Wong et al. | Nov 2015 | A1 |
20150326291 | Wong et al. | Nov 2015 | A1 |
20150326383 | Wong et al. | Nov 2015 | A1 |
20170290013 | McCoy et al. | Oct 2017 | A1 |
Entry |
---|
Alagha et al. “Generalized Raised-cosine filters”; Jul. 1999, IEEE transcations on communications. vol. 47. No. 7 pp. 989-997. |
Vieira et al., “Reciprocity Calibration Methods for Massive MIMO Based on Antenna Coupling”, Department of Electrical Engineering, Lund University, Jan. 1, 2014, pp. 1-5. |
Harris et al., “University of Bristol and Lund University Partner with NI to Set World Records in 5G Wireless Spectral Efficiency Using Massive MIMO”, dowloaded from www.ni.com, May 2016, pp. 1-7. |
“LTE in a Nutshell—Physical Layer”, Telesystem Innovations Inc. White Paper, 2010, pp. 1-18. |
“An Introduction to Orthogonal Frequency Division Multiplex Technology”, Keithley Instruments, Inc., 2008, pp. 1-66. |
HUAWEI, “WF on CSI Acquisition in NR”, 3GPP TSG RAB WG1 Meeting #85, May 23-27, 2016, pp. 1-4. |
HUAWEI et al., “WF on CSI Acquisition Framework in NR”, 3GPP TSG RAN WG1 Meeting #85, May 23-27, 2016, pp. 1-3. |
“Study on Latency Reduction Techniques for LTE”, 3GPP TR 36.881 Vo.6.0 Technical Report, Feb. 2016, pp. 1-92. |
“Study on Scenarios and Requirements for Next Generation Access Technologies”, 3GPP TR 38.913 Vo.2.0, Feb. 2016, pp. 1-19. |
Vieira et al., “A Flexible 100-Antenna Testbed for Massive MIMO”, Department of Electrical Information Technology at Lund University, Jan. 1, 2014, pp. 1-7. |
“Bristol and Lund Set a New World Record in 5G Wireless Spectrum Efficiency” Bristol Univeristy, News, March: Massive MIMO demonstration, Mar. 23, 2016, pp. 1-6. |
“Study on NR New Radio Access Technology”, 3GPP TSG RAN Meeting #71, Goteborg, Sweden, Mar. 7-10, 2016, pp. 1-8. |
“LTE Small Cell Enhancement by Dual Connectivity”, Wireless World Research Forum, Outlook Visions and Research Directions for the Wireless World, Whitepaper, Nov. 2014, pp. 1-22. |
“LTE in a Nutshell: Protocol Architecture” Telesystem Innovations Inc. White Paper, 2010, pp. 1-12. |
“LTE Random Access Procedure” http://www.eventhelix.com/eventstudio/, 2015, pp. 1-5. |
Luther, “5G Massive MIMO Testbed: From Theory to Reality”, National Instruments, Jun. 8, 2016, pp. 1-11. |
“Introduction to the NI mmWave Transceiver System Hardware”, National Instruments, Apr. 20, 2016, pp. 1-8. |
“National Instruments Wireless Research Handbook”, National Instruments, May 2016, pp. 1-37. |
Shepard et al. “Argos: Practical Many-Antenna Base Stations” MobiCom 12, Aug. 22-26, 2012, Istanbul, Turkey, pp. 53-64. |
Khoolenjani, et al. “Distribution of the Ratio of Normal and Rice Random Variables” Digital Commons @WayneState, Journal of Modern Applied Statistical Methods, vol. 12, Issue 2, Article 27, Nov. 1, 2013, pp. 1-15. |
Shen et al. “Channel Estimation in OFDM Systems”, Freescale Semiconductor, Inc. Application Note, Jan. 2006, pp. 1-16. |