This application is a national stage entry under §371 of, and claims the benefit of, PCT/CN2013/074435, which itself claims priority to CN 2012/10394910.3, the entire disclosures of which is incorporated herein by reference.
The present invention relates to a wireless communication system utilizing multiple antennas, in particular to a multi-user SDMA (Space Division Multiple Access) wireless communication system that utilizes a large-scale antenna array.
To meet the demand for extension of broadband information services to mobile terminals, the mobile communication system must support high speed grouped data transmission at a rate of hundreds of Mbps or even thousands of Mbps; under the situation that radio resources becomes tight increasingly, multi-antenna wireless transmission technology can improve spectral efficiency and power efficiency. Multi-antenna wireless transmission technology has become key technology in the 3GPP LTE (3rd Generation Partnership Project Long Term Evolution) for a new generation of mobile communication standard.
In a 3GPP LTE system, four antennas are used at the base station side. To further improve spectral efficiency and improve cell edge performance, in a high version LTE-Advanced system thereof, the number of antennas at the base station side has been increased to eight. Despite of that, the spectral efficiency and cell edge spectral efficiency achieved in a LTE system are still low, and the required transmitting power is still high. As the demand for broadband mobile communication further increases and green wireless communication is expected, it is necessary to exploit new techniques that can utilize radio resources in spatial dimensions thoroughly, so as to further significantly improve spectral efficiency and power efficiency of radio resources and meet the demand for green broadband mobile communication. To that end, the present invention provides a wireless communication technique that utilizes a large-scale antenna array.
The object of the present invention is to provide a wireless communication method utilizing a large-scale antenna array, which can thoroughly exploit radio resources in spatial dimensions and support green broadband mobile communication.
The present invention provides a wireless communication method that utilizes the beam domain characteristics of wireless channels and large-scale antenna array to achieve high-efficiency wireless communication. The method comprises:
In the large-scale antenna array, the antenna units are connected to a digital baseband processing unit for wireless communication via their sending-receiving RF unit, A-D/D-A converter unit, digital optical module, and optical fiber transmission channels, and large-scale beam coverage in the cell is implemented by digital domain multi-beam forming; alternatively, the large-scale antenna array can achieve large-scale beam coverage in the cell by means of an analog multi-beam forming network, and the beam domain sending-receiving signal ports can be connected to the digital baseband processing unit for wireless communication via A-D/D-A converter unit, digital optical module, and optical fiber transmission channels.
The wireless communication between the base station and the users is implemented in the beam domain; a beam domain digital baseband processing and control system at the base station side comprises modules like beam processing units, user processing units, exchange processing unit, and space division multi-user scheduling unit, etc. Each beam processing unit accomplishes transmission post-processing or reception pre-processing of one or more beams, each user processing unit accomplishes generation of frequency domain transmitted signals and processing of received signals of one or more users, the exchange processing unit accomplishes signal interaction between the beam processing units and the user processing units, and the space division multi-user scheduling unit accomplishes scheduling of space division multiple users.
The acquisition of said beam domain long-time channel information is accomplished in a channel detection process in the up links; in the up links, each user transmits detection signals intermittently, the detection signals from different users can be transmitted in an OFDM symbol in one time slot, different sub-carrier resources are used for the detection signals from different users, different antennas of each user transmit detection signals on different sub-carriers, the sub-carrier resources occupied by multiple antennae of each user are multiple sets of sub-carrier resources composed of adjacent sub-carriers, and each antenna uses sub-carriers with different numbers in these sub-carrier sets; on each beam of the base station, the beam domain channel parameters of each user are estimated according to the received detection signals, and thereby a channel characteristic mode energy coupling matrix (i.e., beam domain long-time channel information required for implementing multi-user space division scheduling) is calculated for each user.
The beam domain user scheduling is accomplished by a space division multi-user scheduling processing module at the base station side according to long-time channel information, i.e., the users in the cell are scheduled with the aforesaid obtained beam domain long-time channel information of each user on a criterion of maximizing system sum-rate, so as to determine the multiple users that can communicate with the same time-frequency resource and the beams to be used by each user; after the scheduling, the transmission beams of the communicated users have no overlap among them, and the users can carry out SDMA transmission in the beam domain; the user scheduling can be accomplished with a greedy algorithm or simplified greedy algorithm, i.e., all users and beams are traversed, with consideration of beams available to the current user in the remaining beam set and with consideration of the influence of the addition of the user into the selected user set on the system and rate performance, and the user that has the highest contribution to the increase of the system and rate is selected to be added into the selected user set.
The beam domain multi-user SDMA transmission utilizes the spatial angle resolution of the large-scale array antenna and the local characteristics of each user channel in the beam domain to differentiate users at different positions in different directions; different users use different beam sets to communicate with the base station with the same time-frequency resource, the beam sets have no overlap among the users, and the beams of each user don't exceed the limit of maximum beam; single-user MIMO links are formed between each user and associated multiple beam ports thereof; in the up links, the base station processes the received signals for the users on corresponding beam sets; in the down links, the base station transmits the signals for users on corresponding beam sets.
The up-link transmission process for the users involves pilot frequency training and data transmission. Each user transmits pilot frequency signals with the given time-frequency resource, the pilot frequency signals don't have to be orthogonal among different users, and can be reused; for the same user, the pilot frequency signals shall be orthogonal among different antennas; the base station utilizes the received pilot frequency signals on the corresponding beams of users in conjunction with the received data signals to estimate transient channel information and correlation matrix of interference, and thereby carries out coherent reception processing for the data signals. The down-link transmission process for each user involves pilot frequency training and data transmission. The base station transmits pilot frequency signals with the given time-frequency resource, the pilot frequency signals for each user are mapped to different beam sets for transmission; in addition, the pilot frequency signals don't have to be orthogonal among different users, and can be reused; for the same user, the pilot frequency signals shall be orthogonal among different beams; each user utilizes the received pilot frequency signals in conjunction with the received data signals to estimate transient channel information and correlation matrix of interference, and thereby carries out coherent reception processing for the data signals.
The wireless communication method utilizing a large-scale antenna array provided in the present invention has the following advantages:
To further illustrate the technical scheme in the examples of the present invention, hereunder the drawings for describing the examples or the prior art will be introduced briefly. Apparently, the drawings describe below only illustrate some examples of the present invention. Those skilled in the art can obtain drawings of other examples on the basis of these drawings without creative labor.
To make the person skilled in the art understand the technical scheme of the present invention better, hereunder the technical scheme in the examples of the present invention will be described clearly and completely with reference to the accompanying drawings. Apparently, the examples described below are only parts of the present invention, instead of all of the present invention. Those skilled in the art can obtain other embodiments without creative labor, on the basis of the examples provided here; however, all these examples shall be deemed as falling into the protected domain of the present invention.
(1) Large-Scale Antenna Array Configuration and Beam Coverage at Base Station Side
In the large-scale antenna array, the antenna units are connected to a digital baseband processing unit via their sending-receiving RF unit, A-D/D-A converter unit, digital optical module and optical fiber transmission channels, and large-scale beam coverage in the cell is implemented by means of digital domain multi-beam forming. Alternatively, the large-scale antenna array can implement large-scale beam coverage in the cell by means of an analog multi-beam forming network, and the beam sending-receiving signal ports can be connected to the digital baseband processing unit via sending-receiving RF unit, A-D/D-A converter unit, digital optical module, and optical fiber transmission channels respectively. The large-scale beam coverage is illustrated with a large number of dotted line circles in
(2) Composition of Base Station System
(3) Acquisition of Beam Domain Long-Time Channel Information
The acquisition of beam domain long-time channel information is accomplished through a channel detection process in up links. In up links, each user transmits detection signals intermittently, the detection signals from all users can be transmitted in an OFDM symbol in a time slot, the detection signals from different users use different sub-carrier resources, different antennas of each user transmit detection signals on different sub-carriers, the sub-carrier resources occupied by multiple antennas of each user are multiple sets of sub-carrier resources composed of adjacent sub-carriers, and each antenna uses sub-carriers with different numbers in these sub-carrier sets. For each beam of the base station, the beam domain channel parameters are estimated for each user according to the received detection signals, and thereby the characteristic mode energy coupling matrix is calculated for each user channel. The characteristic mode energy coupling matrix is beam domain long-time channel information required for implementing multi-user space division scheduling.
At the base station side, the characteristic mode energy coupling matrix of the up channel of each user is obtained through the following channel detection process:
Step 1: calculate the up-link channel parameters in beam domain for each user, wherein, the channel parameter of antenna n of user u on sub-carrier k in time slot t is calculated with the following formula:
Where, yu,n,t,k is the vector of received signals in the corresponding beam domain, and the element b is the received signal on beam b. The channel parameters of N antennas of user u form the following channel matrix:
Ĝu,t,kup=[ĝu,1,t,kup,ĝu,2,t,kup, . . . ,ĝu,N,t,kup] (2)
Step 2: calculate the transmitting correlation matrix of each user:
Where, superscript H represents conjugate transposition.
Step 3: carry out Eigen-value decomposition for the transmitting correlation matrix for each user:
Ruut=BuΛuVuH (4)
to obtain the transmitting characteristic matrix Vu of each user, where, Λu is a diagonal matrix composed of the Eigen values.
Step 4: calculate the characteristic mode channel parameter matrix for each user:
Hu,t,kup=Gu,t,kupVu (5)
Step 5: calculate the characteristic mode energy coupling matrix of the up channel of each user:
Where, □ is the Hadamard product of the matrix, and the superscript * represents conjugation.
Obtain the characteristic mode energy coupling matrix Ωudown=[Ωuup]T of downlink channel of each user, by utilizing the reciprocity between statistic channel information of up link and statistic channel information of down link, where, the superscript T represents transposition.
(4) User Scheduling Algorithm
Schedule the users in the cell, with the beam domain long-time channel information of each user obtained with formula (3), on the basis of a given criterion, such as a criterion of maximizing system sum-rate, to determine the multiple users that can communicate with the same time-frequency resource and the beams used by each user; after the scheduling, the transmission beams have no overlap among the users. The users can carry out SDMA transmission in the beam domain.
The user scheduling problem can be accomplished with a greedy algorithm or simplified greedy algorithm. Based on the criterion of maximizing system sum-rate, on the premise of meeting the limit of beams number of each user and no overlap among the transmission beams of different users, traverse all users and beams, with consideration of beams available for the current user in the remaining beam set and the influence of addition of the user into the selected user set on the system performance, select the user that will make the highest contribution to the increase of system sum-rate and add the user into the selected user set; terminate the scheduling if the sum-rate decreases or when all users have been searched.
For example, suppose there are U users and M beams in a cell, the set of all users in the cell is denoted as U⊂{1, 2, . . . , U}, the set of beams is denoted as B={, 1, 2, . . . , M}, the set of users selected for SDMA communication is denoted as US{u1, u2, . . . , uS} (where, S represents the number of scheduled users), the set of unselected users is denoted as Un, the beam set of user u is denoted as Bu, the free beam set is denoted as Bn, the system sum-rate when the users in the set US use their corresponding beam set Bu to communicate with the base station is denoted as R(Us, Bn
Where, Per (•) is matrix permanent, Ω1,i=[Θi,1, . . . , Θi,i−1, Θ1,i, Θi,i+1, . . . , Θi,S], Ω2,i=[Θi,1, . . . , Θi,i−1, Θi,i+1, . . . , Θi,S], and
Alternatively, R(Us, Bu
For the down link, suppose the power of frequency domain QAM modulation symbol transmitted by the base station for user u′ is Pu′down, and the noise variance in the received signals of user u is σu2, then the ratio of the power of modulation symbol transmitted by the base station for user u′ to the received noise variance of user u is ρu,u′down=Pu′down/σu2, and can R(Us, Bu
Where Ω3,t=[Θi,1, . . . , Θi,i−1, Θi,1, Θi,i+1, . . . , Θi,S], Ω4,t=[i,1, . . . , Θi,i−1, Θi,i+1, . . . , Θt,S], and Θi,j=ρu
To obtain a user scheduling result that is consistent between up link and down link, the system sum-rate can be calculated as the sum of sum-rate of up link and sum-rate of down link, or a weighted sum of them.
With the criterion of maximizing system sum-rate and the above approximate calculation method, the user scheduling algorithm is described as follows:
1) Greedy Algorithm
Traverse all users in the cell in turn, select a beam set that can attain maximum rate for a user from the free beam set, and select user that can maximize system sum-rate from the remaining user set; if the system sum-rate increases when the user is added into the communication user set, add the user, and update the user set and beam set, and keep on traversing the remaining users; otherwise terminate the scheduling. The specific implementation steps of the algorithm are as follows:
Step 1: initialize user sets and beam sets, a user set for initial users selected for communication US=ø, a remaining user set is Un={1, 2, . . . U}, a free beam set is Bn={1, 2, . . . , M}, and the number of initial users S=0.
Step 2: For each user in the remaining user set Un, select an optimal transmission beam set Bi from the free beam set Bn in a way that the system sum-rate is maximum after the user is added into the selected user set; the beam selection formula is:
Step 3: calculate the system sum-rate after the users in the remaining user set Un are added into the selected user set with their optimal transmission beam set Bi respectively:
RU
Select the users that enable maximum sum-rate.
Step 4: If the system sum-rate increases after the user is added, i.e., RU
Step 5: terminate scheduling, the selected user set is US, the scheduled number of users is S, the beam set for user communication is Bi, iεUS.
2) Simplified Greedy Algorithm
Traverse each users in the cell, with consideration of the influence of addition of each user into the user scheduling set US on the system performance; if the system sum-rate increases, select the user; otherwise, don't select the user. The implementation steps of the algorithm are as follows:
Step 1: initialize user set and beam set, initialize a user set for initial users selected for communication is US=ø, a remaining user set is Un={1, 2, . . . , U}, and a free beam set Bn={1, 2, . . . , M}, and the number of initial users is S=0; set u=1 and the maximum system sum-rate is Rmax=0.
Step 2: select an optimal transmission beam Bu for user u from the free beam set Bn in a way that the system sum-rate is maximum after the user is added into the user scheduling set US; the beam selection formula is:
Step 3: calculate the system sum-rate SU
RU
If RU
Step 4: If u≦U, return to step 2; otherwise, turn to step 5.
Step 5: terminate scheduling, the selected user set is US, the scheduled number of users is S, the beam set for user communication is Bi, iεUS.
(5) Multi-User SDMA Transmission in Beam Domain
After user scheduling, different users use different beam sets to communicate with the base station; the beams have no overlap among the users, i.e., Bt∩Bj=Ø, i≠j; in addition, the beams of each user don't exceed the limit of maximum beam, i.e., |Bi|≦Bmax. Single-user MIMO links are formed between each user and the corresponding transmission beam set. In the up link, the base station only receives the signals from user i on beam Bi. In the down link, the base station transmits the signals of user i with beam set Bi. In that way, multi-user SDMA transmission in the beam domain with the same time-frequency resource is implemented.
Where, xiup is the beam domain transmitted signals of user i, piup is the transmitting power, ni is Additive White Gaussian Noise (AWGN), and ni′ is the sum of interfering signals of other users on the beam of the user and n1. During detection and decoding for the user, the base station only utilizes the received signals yiup on beam Bi, and doesn't need to utilize signals received on other beams.
Where, xidown is the beam domain transmitted signals for user i, Pidown is the transmitting power, ni is AWGN, and n′i is the sum of the interfering signals of other users and ni. During detection and decoding for the user, the received signals yidown are used, while ni′ is treated as color interference noise.
(6) Detailed Implementation Procedure of Up-Link Transmission
After user scheduling, the up-link transmission is equivalent to a plurality of single-user MIMO links. The scheduled users communicate with the base station with the same time-frequency resource. The entire transmission process comprises pilot frequency training and data transmission. The base station receives signals from the users on corresponding beams on the basis of the scheduling result, utilizes the received pilot frequency signals in conjunction with received data signals to estimate transient channel information and correlation matrix of interference, and utilizes the correlation matrix to carry out coherent reception processing for the data signals. The transmission process is described as follows:
The base station carries out coherent detection of transmitted signals xuup with the transient information (Kuup)−1/2[Huup]B
(7) Detailed Implementation Procedure of Down-Link Transmission
Likewise, the down link transmission is equivalent to a plurality of single-user MIMO links. The base station communicates with the scheduled users with the same time-frequency resource in the down link. The down-link transmission process comprises pilot frequency training and data transmission. The base station maps the pilot frequency signals and data signals for each user to the corresponding beam set for transmission; each user uses the received pilot frequency signals in conjunction with the received data signals to estimate transient channel information and correlation matrix of interference, and finally to carry out coherent reception processing. The transmission process is described as follows:
The user carries out coherent detection of transmitted signals xudown with the transient information (Kudown)−1/2[Hudown]B
The transmission scheme is applicable to FDD and TDD systems. In a FDD system, the up-link and down-link use different frequencies; therefore, the transient channel information of up-link is different from that of down-link, and channel information has to be estimated separately, i.e., transient channel information and correlation matrix of interference have to be estimated separately for up-link and down-link; in a TDD system, channel estimation can be carried out separately for up-link and down-link, or the pilot frequency signals can be transmitted in up-link or down-link only and the channel information of up-link and down-link can be obtained by utilizing the reciprocity between up-link and down-link; however, the interference in up-link is different from the interference in down link; therefore, the correlation matrix of interference has to be estimated separately for up-link and down-link.
In the examples of the present invention, it should be appreciated that the method disclosed can be implemented in other ways, without departing from the spirit and scope of the present invention. The examples provided here are only exemplary, and shall not be deemed as limitation to the present invention, and the content described shall not be deemed as limitation to the object of the present application. For example, a plurality of units or components can be combined or integrated into another system, or some features can be omitted, or excluded from the execution.
While the present invention has been illustrated and described with reference to some detailed embodiments, the present invention is not limited to these. Those skilled in the art should recognize that various variations and modifications can be made without departing from the spirit and scope of the present invention as defined by the accompanying claims. Therefore, the protected scope of the present invention shall be only confined by the claims.
Number | Date | Country | Kind |
---|---|---|---|
2012 1 0394910 | Oct 2012 | CN | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/CN2013/074435 | 4/19/2013 | WO | 00 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2014/059774 | 4/24/2014 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
9048917 | Gomadam | Jun 2015 | B1 |
20090268685 | Chen | Oct 2009 | A1 |
20090279500 | Luo | Nov 2009 | A1 |
20100008331 | Li | Jan 2010 | A1 |
20100189189 | Hoshino | Jul 2010 | A1 |
20120020319 | Song | Jan 2012 | A1 |
20120027108 | Hong | Feb 2012 | A1 |
20120177001 | Gao et al. | Jul 2012 | A1 |
20130034040 | Priotti | Feb 2013 | A1 |
20130083770 | Kalhan | Apr 2013 | A1 |
20130089068 | Kalhan | Apr 2013 | A1 |
20130130690 | Hunukumbure | May 2013 | A1 |
20130272211 | Quan | Oct 2013 | A1 |
20130272263 | Pi | Oct 2013 | A1 |
20130322280 | Pi | Dec 2013 | A1 |
20140044044 | Josiam | Feb 2014 | A1 |
20160006122 | Seol | Jan 2016 | A1 |
Number | Date | Country |
---|---|---|
101882952 | Oct 2010 | CN |
102710309 | Mar 2012 | CN |
102916735 | Jun 2013 | CN |
Entry |
---|
International Search Report issued in corresponding PCT Patent Appln. Serial No. PCT/CN2013/074435 dated Jul. 2, 2013 (5 pgs). |
ZTE, Consideration on Multi-user beamforming, 3GPP TSG RAN WGI Meeting #56 R1-090636, Feb. 3, 2009 “2 MU scheme in TDD” (4 pgs). |
Number | Date | Country | |
---|---|---|---|
20150334726 A1 | Nov 2015 | US |