CN114900398A - IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information) - Google Patents

IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information) Download PDF

Info

Publication number
CN114900398A
CN114900398A CN202210460209.0A CN202210460209A CN114900398A CN 114900398 A CN114900398 A CN 114900398A CN 202210460209 A CN202210460209 A CN 202210460209A CN 114900398 A CN114900398 A CN 114900398A
Authority
CN
China
Prior art keywords
rrh
irs
matrix
bbu
user
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210460209.0A
Other languages
Chinese (zh)
Inventor
张昱
段华东
何宣宣
卢为党
黄国兴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN202210460209.0A priority Critical patent/CN114900398A/en
Publication of CN114900398A publication Critical patent/CN114900398A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H04L25/0242—Channel estimation channel estimation algorithms using matrix methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0426—Power distribution
    • H04B7/043—Power distribution using best eigenmode, e.g. beam forming or beam steering
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms
    • H04L25/0256—Channel estimation using minimum mean square error criteria
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses an IRS (interference rejection ratio) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information), wherein a BBU (baseband unit) pool of a multi-antenna baseband processing unit) communicates with a plurality of single-antenna users through a RRH (remote radio head) of the multi-antenna baseband processing unit, the BBU pool processes baseband signals through point-to-point compression or multi-element compression and sends quantization bits to the RRH through a fronthaul link, and a plurality of intelligent reflector IRSs are deployed near the RRH to assist the wireless fronthaul link and an access link in different time slots. The method and the device aim at improving the system and the rate, joint optimization is carried out on the transmitting beam forming matrix of the BBU pool and the RRH, the phase shift matrix of the reflecting surface and the forward-transmission quantization noise covariance matrix, and the downlink transmission rate of the cloud access network communication system can be effectively improved when the channel state information CSI is inaccurate.

Description

IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information)
Technical Field
The invention relates to the field of wireless communication, in particular to a communication system of a wireless downlink fronthaul link and an access link of an intelligent reflector assisted time division duplex cloud access network with inaccurate Channel State Information (CSI) so as to improve downlink and speed to enhance communication quality.
Background
With the rapid development of wireless communication technology, the number of devices accessing to a communication network is also rapidly increasing, the requirement for communication efficiency is higher, and the traditional wireless communication system cannot meet the requirement.
Intelligent reflective surfaces are an emerging technology that can effectively enhance wireless communication systems. The intelligent reflecting surface is a passive array with a large number of reprogrammable elements, the reflecting amplitude and phase of each reflecting element can be adjusted by the controller, and no radio frequency chain is transmitted, so no extra noise is introduced on the reflected signal. Compared with the traditional relay, the intelligent reflecting surface can effectively improve the performance of a wireless network. A cloud access network (C-RAN) is a promising mobile network architecture, which can improve the spectral efficiency and energy efficiency of a communication network. The C-RAN moves the baseband processing function of the traditional base station to a BBU pool, RRHs are deployed at positions close to mobile users, and the multi-antenna BBU pool communicates with a plurality of single-antenna users through a plurality of multi-antenna RRHs. Through the communication of the intelligent reflector assisted cloud access network wireless downlink fronthaul link and the access link, the RRH not only receives signals of a direct link sent by the BBU pool, but also receives signals of an IRS reflection path, and a user also receives the signals of the RRH direct link and the signals of the IRS reflection, so that the method is more advantageous compared with the cloud access network communication without the assistance of the intelligent reflector. The cloud wireless access network communication system assisted by the intelligent reflector can improve the efficiency of the communication system and realize the intellectualization of the communication network.
The performance of the intelligent reflector assisted cloud access network communication system can be further improved by performing joint optimization on a transmission beam forming matrix of a BBU pool and an RRH, a phase shift matrix of a reflector and a fronthaul quantization noise covariance matrix.
Disclosure of Invention
The invention aims to provide a method for optimizing a precoding matrix, a phase shift matrix of a reflecting surface and a forward quantization noise covariance matrix to improve the downlink rate and the speed of a system aiming at a communication system of a wireless downlink forward link and an access link of an intelligent reflecting surface auxiliary time division duplex cloud access network with inaccurate Channel State Information (CSI). Namely, the aim of maximizing the downlink and the rate of the whole system is achieved by carrying out joint optimization on a transmitting beam forming matrix of a BBU pool and an RRH, a phase shift matrix of a reflecting surface and a fronthaul quantization noise covariance matrix.
The technical scheme of the invention is as follows:
an IRS (intelligent resilient station) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information), which adopts an intelligent reflector assisted cloud access network wireless downlink fronthaul link and access link communication system, and performs joint optimization on a fronthaul link quantization noise covariance matrix, a BBU (baseband unit) pool and RRH (remote radio link) transmitting beam forming matrix and a reflector phase shift matrix with the aim of maximizing system downlink and speed, specifically comprises the following steps:
1.1) in the downlink cloud access network communication system of intelligent reflector auxiliary forward link and access link, the multi-antenna BBU pool communicates with K single-antenna users through L multi-antenna RRHs, wherein the BBU pool processes baseband signals through point-to-point compression or multi-element compression, and sends quantization bits to the RRHs through the forward link, I intelligent reflectors are deployed near the RRHs, and the IRS number of the wireless forward link and the access link at different time slots is I respectively F And I A The three IRS groups are respectively represented as I, I F And I A The number of antennas of each BBU and RRH is N B And N R The number of reflection units of the IRS is M. The RRH is a half-duplex node, operating in Time Division Duplex (TDD) mode, with each time slot divided into (1-alpha) 0 ) And alpha 0 And the two parts are respectively used for the transmission of the BBU-RRH fronthaul link and the RRH-user access link.
1.2) in the forward link, the BBU pool first encodes the downlink message of user k into a baseband signal s k Then linearly precoding the signals of all users into
Figure BDA0003621692680000021
Wherein v is k Is the transmit beamforming matrix for user k on all RRHs,
Figure BDA0003621692680000022
is the signal transmitted by RRH l. In that
Figure BDA0003621692680000023
Is transmitted to RRH
Figure BDA0003621692680000024
And (3) carrying out quantization compression:
Figure BDA0003621692680000025
E l is LN R ×N R Except for (l-1) N R +1 lines to lN R Row is of size N R The other elements except the unit matrix of (1) are 0; q. q.s l ～CN(0,Ω l,l ) Representation is independent of
Figure BDA0003621692680000026
The quantization noise of (2) is defined as q ═ q of the quantization noise vectors of all RRHs 1 ；...；q L ]Q to CN (0, Ω), Ω is a complex Gaussian distribution covariance matrix of q, Ω l,l Is the l-th dimension N on the diagonal of Ω R The identity matrix of (2). To x is l Transmitted to RRHl, and the BBU pool encodes its corresponding compression index to generate a baseband signal t l The signal transmitted by the BBU pool is
Figure BDA0003621692680000031
Wherein F l Is a transmit beamforming matrix, and has a BBU transmit power P B And (3) constraint:
Figure BDA0003621692680000032
1.3) definition of H l,B 、G i,B And G l,i Channel matrixes from the BBU pool to the RRH l, from the BBU pool to the IRS i and from the IRS i to the RRH l are respectively, and the received signals of the RRHl are as follows:
Figure BDA0003621692680000033
wherein
Figure BDA0003621692680000034
Respectively BBU pool to IRS set I F And from IRS set I F Channel matrix to RRHl. Wherein
Figure BDA0003621692680000035
Denotes IRSi (I ∈ I) F ) Assuming IRS adjusts only the phase shift, IRS I (I ∈ I) F ) Phase shift of the m-th element of (1)
Figure BDA0003621692680000036
Wherein
Figure BDA0003621692680000037
Is additive white gaussian noise for all channels BBU to RRHl,
Figure BDA0003621692680000038
is n F,l A complex gaussian distribution covariance matrix. The received signal of the r-th antenna of RRH l is:
Figure BDA0003621692680000039
wherein h is l,B,r And
Figure BDA00036216926800000310
denotes the r-th antenna from BBU pool to RRH l and from IRS set I F Channel vectors to the r-th antenna of RRH l, respectively H l,B And
Figure BDA00036216926800000311
the r-th row vector of (2),
Figure BDA00036216926800000312
is from BBU pool through IRS set I F Reflected to the r-th antenna of the RRH l,
Figure BDA00036216926800000313
is composed of
Figure BDA00036216926800000314
Vector of diagonal elements, n F,l,r Represents n F,l The r-th element of (1).
1.4) in the access link, RRH l will compress the signal x l Forwarded to all users. The maximum transmission power of RRH is P R From 1.2), the transmit power constraint of RRH l is:
Figure BDA0003621692680000041
h k,l and g k,i Representing from RRH l to user k and from IRS I (I e I), respectively A ) Channel vector to user k, G i,l Is the channel matrix from RRH l to IRS i,
Figure BDA0003621692680000042
is a phase shift matrix of IRS i, wherein
Figure BDA0003621692680000043
Represents IRS I (I ∈ I) A ) The phase shift of the mth element, the signal received by user k is:
Figure BDA0003621692680000044
wherein
Figure BDA0003621692680000045
Is composed of
Figure BDA0003621692680000046
The vector of the diagonal elements is then,
Figure BDA0003621692680000047
h k,L ＝[h k,1 ,...,h k,L ]is the channel matrix from the RRH set L (L ═ 1, 2., L,) to user k, G k,L ＝[G k,1 ,...,G k,L ]Is a concatenated channel from the set of RRHs to user k,
Figure BDA0003621692680000048
is from RRH l through IRS set I A The concatenated channel arriving at user k,
Figure BDA0003621692680000049
and
Figure BDA00036216926800000410
respectively represent slave IRS sets I A To user k and from RRH l to IRS set I A Q is the quantization noise vector of all RRHs,
Figure BDA00036216926800000411
is additive white gaussian noise at user k.
1.5) the direct channels from the BBU pool to the RRH l and from the RRH l to the user k are represented as:
Figure BDA00036216926800000412
wherein
Figure BDA00036216926800000413
And
Figure BDA00036216926800000414
is the estimated CSI of the ue,
Figure BDA00036216926800000415
and
Figure BDA00036216926800000416
is the corresponding channel estimation error(s) and,
Figure BDA00036216926800000417
for the cascade channel of BBU-IRS-RRH and RRH-IRS-user, the channel matrix is:
Figure BDA00036216926800000418
and
Figure BDA00036216926800000419
is a cascade of estimations
Figure BDA00036216926800000420
Is the corresponding channel estimation error or errors,
Figure BDA00036216926800000421
and
Figure BDA00036216926800000422
respectively represent
Figure BDA00036216926800000423
And
Figure BDA00036216926800000424
the covariance matrix of the complex gaussian distribution, the CSI error of each channel is independent of each other.
Further, the BBU pool and the RRH transmit beam forming matrix F l And v k IRS phase shift matrix theta of auxiliary access link and forward link A And Θ F And a fronthaul quantization noise covariance matrix omega, which is optimized to maximize the system and rate, the specific steps are as follows:
2.1) the communication system of the intelligent reflector assisted downlink time division duplex cloud access network wireless forward link and the access link described in the steps 1.1) -1.5) is characterized in that:
2.1.1) the lower bound of the user achievable rate is:
Figure BDA0003621692680000051
wherein the content of the first and second substances,
Figure BDA0003621692680000052
Figure BDA0003621692680000053
Figure BDA0003621692680000054
is IRS set I A Total number of reflective elements.
2.1.2) the achievable rate of the wireless fronthaul link should satisfy:
Figure BDA0003621692680000055
wherein the content of the first and second substances,
Figure BDA0003621692680000056
Figure BDA0003621692680000057
Figure BDA0003621692680000058
Figure BDA0003621692680000059
2.1.3) BBU pool pairs of precoded signals
Figure BDA00036216926800000510
Compression is performed and the output rate of the compressor cannot exceed the achievable rate of the fronthaul link. Two compression strategies, point-to-point compression and multivariate compression, are considered.
The fronthaul constraint for point-to-point compression is:
Figure BDA00036216926800000511
point-to-point compression produces independent quantization noise on the RRHs, so the quantization noise covariance matrix for all RRHs is a block diagonal matrix,i.e., Ω ═ diag ({ Ω }) l,l } l∈L )。
The forwarding constraint of multivariate compression is:
Figure BDA0003621692680000061
the multivariate compression gives correlation between the quantization noise of each RRH, so the overall quantization noise covariance matrix Ω is a complete matrix.
2.2) the optimization problem for optimizing the above system parameters when BBU pools employ multivariate compression can be expressed as:
Figure BDA0003621692680000062
Figure BDA0003621692680000063
Figure BDA0003621692680000064
Figure BDA0003621692680000065
Figure BDA0003621692680000066
Figure BDA0003621692680000067
Figure BDA0003621692680000068
Ω±0. (1h)
ω k representing the weight of each user, (1b) representing the achievable rate constraint of the fronthaul link, (1c) being the fronthaul compression constraint, (1d) and (1e) representing the transmit power constraint of the BBU pool and each RRH, respectively. (1f) And (1g) unit mode 1 constraints of passive beamforming matrices of the IRS assisted fronthaul link and access link, respectively, (1h) indicating that the fronthaul quantization noise covariance matrix is a semi-positive definite matrix.
2.3) the problem (P1) is solved after being transformed.
2.3.1) transforming the objective function of (P1) into:
Figure BDA0003621692680000071
w A,k for the introduced auxiliary variable, u A,k Is at user k from y A,k Middle estimate s k Of (2), i.e. predicting the resulting signal as
Figure BDA0003621692680000072
Wherein y is A,k For the signal received by user k, s k Base band signal obtained by coding a downlink message for user k by BBU, 1 is s k The dimension (c) of (a) is,
Figure BDA0003621692680000073
is the mean square error:
Figure BDA0003621692680000074
when w is A,k And u A,k When the following values are obtained, R sum Obtaining an optimal value:
Figure BDA0003621692680000075
Figure BDA0003621692680000076
2.3.2) similar to 2.3.1), by the MSE method, constraint (1b) can be approximated as:
Figure BDA0003621692680000077
W F,l requiring a positive half-definite, is an introduced auxiliary variable,
Figure BDA0003621692680000078
is at RRH l from y F,l Middle estimation signal t l Linear receivers, i.e. predicting the resulting signal as
Figure BDA0003621692680000079
Wherein y is F,l Is the signal received by RRH l, t l Is BBU to x l A base band signal obtained by encoding the compression index of (a), d R Is t l Of (c) is calculated.
Figure BDA00036216926800000710
Is the mean square error matrix:
Figure BDA00036216926800000711
when W is F,l And U F,l When the following values are taken, (6) the right side takes the maximum value:
Figure BDA00036216926800000712
Figure BDA00036216926800000713
2.3.3) for constraint (1c), let
Figure BDA0003621692680000081
(1c) Can be rewritten as:
Figure BDA0003621692680000082
constraint (1c) may further translate to the following approximate constraint:
Figure BDA0003621692680000083
∑ l requiring a semi-positive determination is the auxiliary variable introduced and | S | is the number of RRHs in the set S. When sigma l When the following values are taken,
Figure BDA0003621692680000084
(10) and (1c) equivalents:
Figure BDA0003621692680000085
2.3.4) the optimization problem (P1) is transformed into:
Figure BDA0003621692680000086
Figure BDA0003621692680000087
Figure BDA0003621692680000088
Figure BDA0003621692680000089
Figure BDA00036216926800000810
Figure BDA00036216926800000811
Figure BDA00036216926800000812
Ω±0. (12h)
updating the auxiliary variable w by (4), (5), (8), (9) and (11) A,k 、u A,k 、W F,l 、U F,l Sum Σ l . For a fixed w A,k 、u A,k 、W F,l 、U F,l Sum Σ l Optimizing F by solving the following problem l 、v k 、Ω、Θ A And Θ F In the formula k Weight representing each user:
Figure BDA00036216926800000813
s.t.(12b)～(12h). (13b)
2.4) the problem (P3) is decomposed into three sub-problems to solve alternately.
2.4.1) first fix Ω, Θ in the problem (P3) A And Θ F Transmit beamforming matrix F to optimize BBU pools and RRHs l And v k . The first sub-problem is given by:
Figure BDA0003621692680000091
Figure BDA0003621692680000092
Figure BDA0003621692680000093
Figure BDA0003621692680000094
Figure BDA0003621692680000095
the problem (P3.1) is convex and can be solved by some standard optimization tools (e.g. CVX).
2.4.2) second subproblem fix F l 、v k And Θ A Optimizing the fronthaul quantization noise covariance matrix omega and the IRS phase shift matrix theta for the auxiliary fronthaul link F . The second sub-problem is as follows:
Figure BDA0003621692680000096
Figure BDA0003621692680000097
Figure BDA0003621692680000098
Figure BDA0003621692680000099
Figure BDA00036216926800000910
Ω±0. (15f)
it is known that the targets (15a), constraints (15c) and (15d) are convex with respect to Ω. The constraint (15b) is transformed according to
Figure BDA00036216926800000911
And formula (6) pair
Figure BDA00036216926800000912
Rewriting (15b) to the following equation:
Figure BDA00036216926800000913
wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0003621692680000101
Figure BDA0003621692680000102
d R is the data dimension of each RRH, in the above formulae:
Figure BDA0003621692680000103
d F,l ＝diag(D F,l ),
Figure BDA0003621692680000104
the problem (P3.2) translates into the following:
Figure BDA0003621692680000105
Figure BDA0003621692680000106
Figure BDA0003621692680000107
Figure BDA0003621692680000108
Figure BDA0003621692680000109
Figure BDA00036216926800001010
Figure BDA00036216926800001011
the constraint (17f) is still non-convex, a semi-definite relaxation method (SDR) is applied, after the constraint (17f) is removed, the optimal solution is obtained by using a standard optimization tool CVX, if the optimal solution is obtained
Figure BDA00036216926800001012
The rank of the optimal solution is not 1, then
Figure BDA00036216926800001013
Randomizing to produce a feasible sub-optimal solution, and updating Ω and Θ only as the objective function value of (P3.2.1) increases F 。
2.4.3) fixing F l 、v k Omega and theta F To theta A Optimizing to obtain a third sub-problem:
Figure BDA00036216926800001014
Figure BDA00036216926800001015
the objective function is rewritten as follows:
Figure BDA0003621692680000111
wherein the content of the first and second substances,
Figure BDA0003621692680000112
in the above formula, the first and second carbon atoms are,
Figure BDA0003621692680000113
Figure BDA0003621692680000114
the sub-problem (P3.3) can be rewritten as:
Figure BDA0003621692680000115
Figure BDA0003621692680000116
Figure BDA0003621692680000117
the problem (P3.3.1) is similar to (P3.2.1), and it can also be achieved by semi-definite relaxation (SDR), using randomization to obtain a feasible sub-optimal solution, and only update Θ as the value of the objective function of (P3.3.1) increases A 。
2.5) repeat the steps in 2.4.1) -2.4.3) until convergence.
2.6) when the BBU pool adopts a point-to-point compression method, performing joint optimization on a transmission beam forming matrix of the BBU pool and the RRH, a phase shift matrix of the IRS and a fronthaul quantization noise covariance matrix, and considering fronthaul compression constraint of point-to-point compression, wherein the problem can be expressed as:
Figure BDA0003621692680000118
s.t.(1b),(1d)～(1g),(19b)
Figure BDA0003621692680000119
Figure BDA00036216926800001110
the difference between the problems (P4) and (P1) is the constraints (19c) and (19d), where (19c) is relative to Ω l,l And v k Is non-convex. Order to
Figure BDA00036216926800001111
(19c) Is rewritten as:
Figure BDA0003621692680000121
log|O l the upper limit of |:
Figure BDA0003621692680000122
S l ± 0 is an auxiliary variable, so the following convex constraint can be substituted for (19 c):
Figure BDA0003621692680000123
the objective function and other constraints in the problem (P4) are treated in the same way as the problem (P1), and then a similar method can be applied to solve the problem using a successive convex approximation method and an alternating optimization method.
2.7) for the IRS reflector phase dispersion case, first following the steps in 2.1) -2.3) until after convergence, the theta is obtained A And Θ F Its diagonal line element theta A,i,m And theta F,i,m Mapping to discrete phase point to determine theta F Scaling omega later to meet the limits of the problem (P3.2.1)Made of condition(s)
Figure BDA0003621692680000124
The invention has the beneficial effect that for the intelligent reflector assisted time division duplex cloud access network communication system, the total downlink transmission rate is obviously higher than that of the traditional cloud access network without the assistance of the intelligent reflector by carrying out combined optimization on the transmitting beam forming matrix of the BBU pool and the RRH, the phase shift matrix of the reflector and the fronthaul quantization noise covariance matrix.
Drawings
Fig. 1 is a model diagram of an intelligent reflector assisted time division duplex cloud access network wireless downlink communication system in the invention;
fig. 2 is a diagram of a relationship between downlink and rate of an intelligent reflector assisted time division duplex cloud access network wireless downlink fronthaul link and an access link communication system after the joint optimization method of the invention is adopted and the number of reflecting units of the intelligent reflector.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1-2, the multiple-antenna BBU pool in fig. 1 communicates with K single-antenna users through L multiple-antenna RRHs, the BBU pool processes baseband signals through point-to-point compression or multi-element compression, and sends quantization bits to the RRHs through a fronthaul link, I intelligent reflection surfaces are deployed near the RRHs, and the IRS numbers of the wireless fronthaul link and the access link are I respectively at different time slots F And I A . The RRH operates in time division duplex mode with each time slot divided into (1-alpha) 0 ) And alpha 0 The two parts are respectively used for the transmission of a BBU-RRH fronthaul link and an RRH-user access link; in fig. 2, the first comparison scheme shows that when the BBU pool adopts multivariate compression, joint optimization is performed on the transmit beamforming matrix and the fronthaul quantization noise covariance matrix of the BBU pool and the RRH, and the phase of the reflecting surface is randomly selected; "comparative scheme two" means no reflective surface aid; "this patent method, many, continuous" means BBU pool adopts many compression, the reflecting surface phase place is between 0-2 pi; "this patent method, multi, 2 bit" means BBU pool adoptedPerforming multi-element compression, wherein the phase level of a reflecting surface is 4; the method, point-to-point and continuous and the method, point-to-point and 2bit are respectively the conditions that the phase of a reflecting surface is between 0 and 2 pi and the phase level of the reflecting surface is 4 when a BBU pool adopts point-to-point compression.
For the transmission process of a communication system of a wireless downlink fronthaul link and an access link of an intelligent reflector assisted time division duplex cloud access network with inaccurate channel state information, before transmission begins, a BBU pool first acquires the inaccurate system channel information, and then joint optimization is performed on a fronthaul link quantization noise covariance matrix, the BBU pool, an RRH pre-coding matrix and a reflector phase shift matrix.
The transmission process comprises the following steps: the multi-antenna BBU pool communicates with K single-antenna users through L multi-antenna RRHs, I intelligent reflecting surfaces are arranged near the RRHs, and the IRS quantities of the auxiliary wireless fronthaul links and the access links in different time slots are I respectively F And I A The RRH not only receives the signal of the direct link sent by the BBU pool, but also receives the signal of the IRS reflection path, and similarly, the user also receives the sum of the signals from the RRH direct link and the IRS reflection. The method comprises the following specific steps:
1.1) in the downlink cloud access network communication system of intelligent reflector auxiliary forward link and access link, the multi-antenna BBU pool communicates with K single-antenna users through L multi-antenna RRHs, wherein the BBU pool processes baseband signals through point-to-point compression or multi-element compression, and sends quantization bits to the RRHs through the forward link, I intelligent reflectors are deployed near the RRHs, and the IRS number of the wireless forward link and the access link at different time slots is I respectively F And I A The above three IRS groups are respectively represented as I, I F And I A The number of antennas of each BBU and RRH is N B And N R The number of reflection units of the IRS is M. The RRH is a half-duplex node, operating in Time Division Duplex (TDD) mode, with each time slot divided into (1-alpha) 0 ) And alpha 0 And the two parts are respectively used for the transmission of the BBU-RRH fronthaul link and the RRH-user access link.
1.2) in the forward link, the BBU pool first encodes the downlink message of user k into a baseband signal s k Then linearly precoding the signals of all users into
Figure BDA0003621692680000141
Wherein v is k Is the transmit beamforming matrix for user k on all RRHs,
Figure BDA0003621692680000142
is the signal transmitted by RRH l. In that
Figure BDA0003621692680000143
Is transmitted to RRH
Figure BDA0003621692680000144
And (3) carrying out quantization compression:
Figure BDA0003621692680000145
E l is LN R ×N R Except for (l-1) N R +1 lines to lN R Row is of size N R The other elements except the unit matrix of (1) are 0; q. q.s l ～CN(0,Ω l,l ) Representation independent of
Figure BDA0003621692680000146
The quantization noise of (2) is defined as q ═ q of the quantization noise vectors of all RRHs 1 ；...；q L ]Q to CN (0, Ω), Ω is a complex Gaussian distribution covariance matrix of q, Ω l,l Is the l size N on the diagonal of omega R The identity matrix of (2). To x is l Transmitted to RRHl, and the BBU pool encodes its corresponding compression index to generate a baseband signal t l The signal transmitted by the BBU pool is
Figure BDA0003621692680000147
Wherein F l Is a transmit beamforming matrix, and has a BBU transmit power P B And (3) constraint:
Figure BDA0003621692680000148
1.3) definition of H l,B 、G i,B And G l,i Channel matrixes from a BBU pool to an RRH l, from the BBU pool to an IRS i and from the IRS i to the RRH l are respectively, and a receiving signal of the RRH l is as follows:
Figure BDA0003621692680000149
wherein
Figure BDA00036216926800001410
Respectively BBU pool to IRS set I F And from IRS set I F Channel matrix to RRH l. Wherein
Figure BDA00036216926800001411
Denotes IRSi (I ∈ I) F ) Assuming IRS adjusts only the phase shift, IRS I (I ∈ I) F ) Phase shift of the m-th element of (1)
Figure BDA00036216926800001412
Wherein
Figure BDA00036216926800001413
Is additive white gaussian noise for all channels BBU to RRHl,
Figure BDA00036216926800001414
is n F,l A complex gaussian distribution covariance matrix. The received signal of the r-th antenna of RRH l is:
Figure BDA0003621692680000151
wherein h is l,B,r And
Figure BDA0003621692680000152
denotes the r-th antenna from BBU pool to RRH l and from IRS set I F Channel vectors to the r-th antenna of RRH l, respectively H l,B And
Figure BDA0003621692680000153
row r of (1)The vector of the vector is then calculated,
Figure BDA0003621692680000154
is from BBU pool through IRS set I F Reflected to the r-th antenna of the RRH l,
Figure BDA0003621692680000155
is composed of
Figure BDA0003621692680000156
Vector of diagonal elements, n F,l,r Represents n F,l The r-th element of (1).
1.4) in the access link, RRH l will compress the signal x l Forwarded to all users. The maximum transmission power of RRH is P R From 1.2), the transmit power of RRH l is constrained to:
Figure BDA0003621692680000157
h k,l and g k,i Representing from RRH l to user k and from IRS I (I e I), respectively A ) Channel vector to user k, G i,l Is the channel matrix from RRH l to IRS i,
Figure BDA0003621692680000158
is a phase shift matrix of IRS i, wherein
Figure BDA0003621692680000159
Represents IRS I (I ∈ I) A ) The phase shift of the mth element, the signal received by user k is:
Figure BDA00036216926800001510
wherein
Figure BDA00036216926800001511
Is composed of
Figure BDA00036216926800001512
The vector of the diagonal elements is then,
Figure BDA00036216926800001513
h k,L ＝[h k,1 ,...,h k,L ]is the channel matrix from RRH set L (L ═ {1, 2.., L }) to user k, G k,L ＝[G k,1 ,...,G k,L ]Is a concatenated channel from the set of RRHs to user k,
Figure BDA00036216926800001514
is from RRH l through IRS set I A The concatenated channel arriving at user k,
Figure BDA00036216926800001515
and
Figure BDA00036216926800001516
respectively representing the slave IRS set I A To user k and from RRH l to IRS set I A Q is the quantization noise vector of all RRHs,
Figure BDA00036216926800001517
is additive white gaussian noise at user k.
1.5) the direct channels from the BBU pool to the RRH l and from the RRH l to the user k are represented as:
Figure BDA0003621692680000161
wherein
Figure BDA0003621692680000162
And
Figure BDA0003621692680000163
is the estimated CSI of the ue,
Figure BDA0003621692680000164
and
Figure BDA0003621692680000165
is the corresponding channel estimation error or errors,
Figure BDA0003621692680000166
for BBU-IRS-RRHAnd a concatenated channel of RRH-IRS-users, the channel matrix being:
Figure BDA0003621692680000167
and
Figure BDA0003621692680000168
is a cascade of estimations
Figure BDA0003621692680000169
Is the corresponding channel estimation error or errors,
Figure BDA00036216926800001610
and
Figure BDA00036216926800001611
respectively represent
Figure BDA00036216926800001612
And
Figure BDA00036216926800001613
the covariance matrix of the complex gaussian distribution, the CSI error of each channel is independent of each other.
Transmission beam forming matrix F of BBU pool and RRH l And v k IRS phase shift matrix theta of auxiliary access link and forward link A And Θ F And a fronthaul quantization noise covariance matrix Ω, which is optimized to maximize the system and rate, the specific steps are as follows:
2.1) the communication system of the intelligent reflecting surface auxiliary downlink time division duplex cloud access network wireless forward link and the access link described in the steps 1.1) -1.5) is characterized in that:
2.1.1) the lower bound of the user achievable rate is:
Figure BDA00036216926800001614
wherein the content of the first and second substances,
Figure BDA00036216926800001615
Figure BDA00036216926800001616
Figure BDA00036216926800001617
is IRS set I A Total number of reflective elements.
2.1.2) the achievable rate of the wireless fronthaul link should satisfy:
Figure BDA00036216926800001618
wherein the content of the first and second substances,
Figure BDA0003621692680000171
Figure BDA0003621692680000172
Figure BDA0003621692680000173
Figure BDA0003621692680000174
2.1.3) BBU pool pairs of precoded signals
Figure BDA0003621692680000175
Compression is performed and the output rate of the compressor cannot exceed the achievable rate of the fronthaul link. Two compression strategies, point-to-point compression and multivariate compression, are considered.
The fronthaul constraint for point-to-point compression is:
Figure BDA0003621692680000176
point-to-point compression produces independent quantization noise on the RRHs, so the quantization noise covariance matrix of all RRHs is a block diagonal matrix, i.e., Ω -diag ({ Ω } q } l,l } l∈L )。
The forwarding constraint of multivariate compression is:
Figure BDA0003621692680000177
the multivariate compression gives correlation between the quantization noise of each RRH, so the overall quantization noise covariance matrix Ω is a complete matrix.
2.2) the optimization problem for optimizing the above system parameters when BBU pools employ multivariate compression can be expressed as:
Figure BDA0003621692680000181
Figure BDA0003621692680000182
Figure BDA0003621692680000183
Figure BDA0003621692680000184
Figure BDA0003621692680000185
Figure BDA0003621692680000186
Figure BDA0003621692680000187
Ω±0. (1h)
ω k representing the weight of each user, (1b) representing the achievable rate constraint of the fronthaul link, (1c) being the fronthaul compression constraint, (1d) and (1e) representing the transmit power constraint of the BBU pool and each RRH, respectively. (1f) And (1g) unit mode 1 constraints of passive beamforming matrices of the IRS assisted fronthaul link and access link, respectively, (1h) indicating that the fronthaul quantization noise covariance matrix is a semi-positive definite matrix.
2.3) solving the problem (P1) after transformation.
2.3.1) transforming the objective function of (P1) into:
Figure BDA0003621692680000188
w A,k for the introduced auxiliary variable, u A,k Is at user k from y A,k Middle estimate s k Of (2), i.e. predicting the resulting signal as
Figure BDA0003621692680000189
Wherein y is A,k For signals received by user k, s k Base band signal obtained by coding a downlink message for user k by BBU, 1 is s k The dimension (c) of (a) is,
Figure BDA00036216926800001810
is the mean square error:
Figure BDA00036216926800001811
when w is A,k And u A,k When the following values are obtained, R sum Obtaining an optimal value:
Figure BDA0003621692680000191
Figure BDA0003621692680000192
2.3.2) similar to 2.3.1), by the MSE method, constraint (1b) can be approximated as:
Figure BDA0003621692680000193
W F,l requiring a positive half-definite, is an introduced auxiliary variable,
Figure BDA0003621692680000194
is at RRH l from y F,l Middle estimation signal t l Linear receivers, i.e. predicting the resulting signal as
Figure BDA0003621692680000195
Wherein y is F,l Is the signal received by RRH l, t l Is BBU to x l A base band signal obtained by encoding the compression index of (a), d R Is t l Of (c) is calculated.
Figure BDA0003621692680000196
Is the mean square error matrix:
Figure BDA0003621692680000197
when W is F,l And U F,l When the following values are taken, (6) the right side takes the maximum value:
Figure BDA0003621692680000198
Figure BDA0003621692680000199
2.3.3) for constraint (1c), let
Figure BDA00036216926800001910
(1c) Can be rewritten as:
Figure BDA00036216926800001911
constraint (1c) may further translate to the following approximate constraint:
Figure BDA00036216926800001912
∑ l requiring a semi-positive determination is the auxiliary variable introduced and | S | is the number of RRHs in the set S. When sigma l When the following values are taken,
Figure BDA00036216926800001913
(10) and (1c) equivalents:
Figure BDA00036216926800001914
2.3.4) the optimization problem (P1) is transformed into:
Figure BDA0003621692680000201
Figure BDA0003621692680000202
Figure BDA0003621692680000203
Figure BDA0003621692680000204
Figure BDA0003621692680000205
Figure BDA0003621692680000206
Figure BDA0003621692680000207
Ω±0. (12h)
updating the auxiliary variable w by (4), (5), (8), (9) and (11) A,k 、u A,k 、W F,l 、U F,l Sum Σ l . For a fixed w A,k 、u A,k 、W F,l 、U F,l Sum Σ l Optimizing F by solving the following problem l 、v k 、Ω、Θ A And Θ F In the formula k Weight representing each user:
Figure BDA0003621692680000208
s.t.(12b)～(12h). (13b)
2.4) the problem (P3) is decomposed into three sub-problems to solve alternately.
2.4.1) first fix Ω, Θ in the problem (P3) A And Θ F Transmit beamforming matrix F to optimize BBU pools and RRHs l And v k . The first sub-problem is given by:
Figure BDA0003621692680000211
Figure BDA0003621692680000212
Figure BDA0003621692680000213
Figure BDA0003621692680000214
Figure BDA0003621692680000215
the problem (P3.1) is convex and can be solved by some standard optimization tools (e.g. CVX).
2.4.2) second subproblem fix F l 、v k And Θ A Optimizing the fronthaul quantization noise covariance matrix omega and the IRS phase shift matrix theta for the auxiliary fronthaul link F . The second sub-problem is as follows:
Figure BDA0003621692680000216
Figure BDA0003621692680000217
Figure BDA0003621692680000218
Figure BDA0003621692680000219
Figure BDA00036216926800002110
Ω±0. (15f)
it is known that the targets (15a), constraints (15c) and (15d) are convex with respect to Ω. The constraint (15b) is transformed according to
Figure BDA00036216926800002111
And formula (6) pair
Figure BDA00036216926800002112
Rewriting (15b) to the following equation:
Figure BDA00036216926800002113
wherein the content of the first and second substances,
Figure BDA00036216926800002114
Figure BDA00036216926800002115
d R is the data dimension of each RRH, in the above formulae:
Figure BDA0003621692680000221
d F,l ＝diag(D F,l ),
Figure BDA0003621692680000222
the problem (P3.2) translates into the following:
Figure BDA0003621692680000223
Figure BDA0003621692680000224
Figure BDA0003621692680000225
Figure BDA0003621692680000226
Figure BDA0003621692680000227
Figure BDA0003621692680000228
Figure BDA0003621692680000229
the constraint (17f) is still non-convex, a semi-definite relaxation method (SDR) is applied, after the constraint (17f) is removed, the optimal solution is obtained by using a standard optimization tool CVX, if the optimal solution is obtained
Figure BDA00036216926800002210
The rank of the optimal solution is not 1, then
Figure BDA00036216926800002211
Randomizing to produce a feasible sub-optimal solution, and updating Ω and Θ only as the objective function value of (P3.2.1) increases F 。
2.4.3) fixing F l 、v k Omega and theta F To theta A Optimizing to obtain a third sub-problem:
Figure BDA00036216926800002212
Figure BDA00036216926800002213
the objective function is rewritten as follows:
Figure BDA00036216926800002214
wherein the content of the first and second substances,
Figure BDA00036216926800002215
in the above formula, the first and second carbon atoms are,
Figure BDA0003621692680000231
Figure BDA0003621692680000232
the sub-problem (P3.3) can be rewritten as:
Figure BDA0003621692680000233
Figure BDA0003621692680000234
Figure BDA0003621692680000235
the problem (P3.3.1) is similar to (P3.2.1), and it can also be achieved by semi-definite relaxation (SDR), using randomization to obtain a feasible sub-optimal solution, and only update Θ as the value of the objective function of (P3.3.1) increases A 。
2.5) repeat the steps in 2.4.1) -2.4.3) until convergence.
2.6) when the BBU pool adopts a point-to-point compression method, performing joint optimization on a transmission beam forming matrix of the BBU pool and the RRH, a phase shift matrix of the IRS and a fronthaul quantization noise covariance matrix, and considering fronthaul compression constraint of point-to-point compression, wherein the problem can be expressed as:
Figure BDA0003621692680000236
s.t.(1b),(1d)～(1g), (19b)
Figure BDA0003621692680000237
Figure BDA0003621692680000238
the difference between the problems (P4) and (P1) is the constraints (19c) and (19d), where (19c) is relative to Ω l,l And v k Is non-convex. Order to
Figure BDA0003621692680000239
(19c) Is rewritten as:
Figure BDA00036216926800002310
log|O l the upper limit of |:
Figure BDA00036216926800002311
S l ± 0 is an auxiliary variable, so the following convex constraint can be substituted for (19 c):
Figure BDA00036216926800002312
the objective function and other constraints in the problem (P4) are treated in the same way as the problem (P1), and then a similar method can be applied to solve the problem using a successive convex approximation method and an alternating optimization method.
2.7) for the IRS reflector phase dispersion case, first following the steps in 2.1) -2.3) until after convergence, the theta is obtained A And Θ F Its diagonal line element theta A,i,m And theta F,i,m Mapping to discrete phase point to determine theta F Scaling omega to meet the constraints of the problem (P3.2.1)
Figure BDA0003621692680000241
Computer simulation shows that for a communication system of a wireless downlink fronthaul link and an access link of an intelligent reflector assisted time division duplex cloud access network with inaccurate channel state information, the downlink and the speed of the communication system are effectively increased by adopting the joint optimization method, and the method is obviously higher than that of a traditional cloud access network without the assistance of the intelligent reflector.

Claims (2)

1. An IRS (interference rejection service) assisted cloud access network downlink beamforming method for non-ideal CSI (channel state information), which is characterized in that: when the Channel State Information (CSI) is inaccurate, a communication system of a downlink fronthaul link and an access link of an Intelligent Reflector (IRS) auxiliary cloud access network is subjected to joint optimization on a transmission beam forming matrix of a Base Band Unit (BBU) pool and a radio head unit (RRH), a phase shift matrix of a reflector and a fronthaul quantization noise covariance matrix by aiming at improving the system and the rate, and the method specifically comprises the following steps:
1.1) in the communication system of the auxiliary cloud access network downlink fronthaul link and the access link of the intelligent reflector, a multi-antenna BBU pool communicates with K single-antenna users through L multi-antenna RRHs, wherein the BBU pool processes baseband signals through point-to-point compression or multi-element compression and sends quantization bits to the RRHs through the fronthaul link, I intelligent reflectors are deployed near the RRHs, and the IRS number of the auxiliary wireless fronthaul link and the access link at different time slots is I respectively F And I A The above three IRS groups are respectively represented as I, I F And I A The number of antennas of each BBU and RRH is N B And N R The number of the reflection units of the IRS is M; the RRH is a half-duplex node, operating in time division duplex, TDD, mode, with each time slot divided into (1-alpha) 0 ) And alpha 0 The two parts are respectively used for the transmission of a BBU-RRH fronthaul link and an RRH-user access link;
1.2) in the forward link, the BBU pool first encodes the downlink message of user k into a baseband signal s k Then linearly precoding the signals of all users into
Figure FDA0003621692670000011
Wherein v is k Is the transmit beamforming matrix for user k on all RRHs,
Figure FDA0003621692670000012
is the signal transmitted by RRH l; in that
Figure FDA0003621692670000013
Is transmitted to RRH
Figure FDA0003621692670000014
And (3) carrying out quantization compression:
Figure FDA0003621692670000015
E l is LN R ×N R Except for (l-1) N R +1 lines to lN R Row is of size N R The other elements except the unit matrix of (1) are 0; q. q.s l ～CN(0,Ω l,l ) Representation is independent of
Figure FDA0003621692670000016
The quantization noise of (2) is defined as q ═ q of the quantization noise vectors of all RRHs 1 ；...；q L ]Q to CN (0, Ω), Ω is a complex Gaussian distribution covariance matrix of q, Ω l,l Is the l-th dimension N on the diagonal of Ω R For x is l Transmitted to RRHl, and the BBU pool encodes its corresponding compression index to generate a baseband signal t l The signal transmitted by the BBU pool is
Figure FDA0003621692670000021
Wherein F l Is a transmit beamforming matrix, and has a BBU transmit power P B And (3) constraint:
Figure FDA0003621692670000022
1.3) definition of H l,B 、G i,B And G l,i Channel matrixes from a BBU pool to an RRH l, from the BBU pool to an IRS i and from the IRS i to the RRH l are respectively, and a receiving signal of the RRH l is as follows:
Figure FDA0003621692670000023
wherein
Figure FDA0003621692670000024
Respectively BBU pool to IRS set I F And from IRS set I F Channel matrix to RRH l. Wherein
Figure FDA0003621692670000025
Denotes IRSi (I ∈ I) F ) Assuming IRS adjusts only the phase shift, IRSi (I ∈ I) F ) Phase shift of the m-th element of (1)
Figure FDA0003621692670000026
Wherein
Figure FDA0003621692670000027
Is additive white gaussian noise for all channels BBU to RRH l,
Figure FDA0003621692670000028
is n F,l A complex gaussian distribution covariance matrix; the received signal of the r-th antenna of RRH l is:
Figure FDA0003621692670000029
wherein h is l,B,r And
Figure FDA00036216926700000210
denotes the r-th antenna from BBU pool to RRH l and from IRS set I F Channel vectors to the r-th antenna of RRH l, respectively H l,B And
Figure FDA00036216926700000211
the r-th row vector of (2),
Figure FDA00036216926700000212
is from BBU pool through IRS set I F Reflected to the r-th antenna of RRH l,
Figure FDA00036216926700000213
is composed of
Figure FDA00036216926700000214
Vector of diagonal elements, n F,l,r Represents n F,l The r-th element of (1);
1.4) in the access link, RRH l will compress the signal x l Forwarding to all users; the maximum transmission power of RRH is P R From step 1.2), the transmit power constraint of RRHl is:
Figure FDA00036216926700000215
h k,l and g k,i Representing from RRH l to user k and from IRSi, I ∈ I, respectively A Channel vector to user k, G i,l Is the channel matrix from RRHl to IRSi,
Figure FDA0003621692670000031
is a phase shift matrix of IRSi, wherein
Figure FDA0003621692670000032
Denotes IRSi, I ∈ I A The phase shift of the mth element, the signal received by user k is:
Figure FDA0003621692670000033
wherein
Figure FDA0003621692670000034
Is composed of
Figure FDA0003621692670000035
The vector of the diagonal elements is then,
Figure FDA0003621692670000036
h k,L ＝[h k,1 ,...,h k,L ]is the channel matrix from RRH set L (L ═ {1, 2.., L }) to user k, G k,L ＝[G k,1 ,...,G k,L ]Is a concatenated channel from the set of RRHs to user k,
Figure FDA0003621692670000037
is from RRH l through IRS set I A The concatenated channel arriving at user k,
Figure FDA0003621692670000038
and
Figure FDA0003621692670000039
respectively representing the slave IRS set I A To user k and from RRH l to IRS set I A Q is the quantization noise vector of all RRHs,
Figure FDA00036216926700000310
is additive white gaussian noise at user k;
1.5) the direct channels from the BBU pool to the RRH l and from the RRH l to the user k are represented as:
Figure FDA00036216926700000311
wherein
Figure FDA00036216926700000312
And
Figure FDA00036216926700000313
is the estimated CSI of the ue,
Figure FDA00036216926700000314
and
Figure FDA00036216926700000315
is the corresponding channel estimation error or errors,
Figure FDA00036216926700000316
for the cascade channel of BBU-IRS-RRH and RRH-IRS-user, the channel matrix is:
Figure FDA00036216926700000317
and
Figure FDA00036216926700000318
is the estimated concatenated CSI that is,
Figure FDA00036216926700000319
is the corresponding channel estimation error or errors,
Figure FDA00036216926700000320
and
Figure FDA00036216926700000321
respectively represent
Figure FDA00036216926700000322
And
Figure FDA00036216926700000323
the complex gaussian distributed covariance matrix, the CSI errors of each channel are independent of each other.
2. The nonideal C of claim 1The IRS of SI assists the downstream beam forming method of the cloud access network, characterized by that, the said BBU pool and RRH transmit the beam forming matrix F l And v k IRS phase shift matrix theta of auxiliary access link and forward link A And Θ F And a fronthaul quantization noise covariance matrix omega, which is optimized to maximize the system and rate, the specific steps are as follows:
2.1) the communication system of the intelligent reflecting surface auxiliary downlink time division duplex cloud access network wireless forward link and the access link described in the steps 1.1) -1.5) is characterized in that:
2.1.1) the lower bound of the user achievable rate is:
Figure FDA0003621692670000041
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003621692670000042
Figure FDA0003621692670000043
Figure FDA0003621692670000044
is IRS set I A The total number of reflective elements of (a);
2.1.2) the achievable rate of the wireless fronthaul link should satisfy:
Figure FDA0003621692670000045
wherein the content of the first and second substances,
Figure FDA0003621692670000046
Figure FDA0003621692670000047
Figure FDA0003621692670000048
Figure FDA0003621692670000049
2.1.3) BBU pool pairs of precoded signals
Figure FDA00036216926700000410
Compressing, and the output rate of the compressor cannot exceed the reachable rate of the forwarding link; two compression strategies, namely point-to-point compression and multivariate compression, are considered;
the fronthaul constraint for point-to-point compression is:
Figure FDA00036216926700000411
point-to-point compression produces independent quantization noise on the RRHs, so the quantization noise covariance matrix of all RRHs is a block diagonal matrix, i.e., Ω -diag ({ Ω } q } l,l } l∈L )；
The forwarding constraint of multivariate compression is:
Figure FDA0003621692670000051
the multivariate compression makes the quantization noise of each RRH have correlation, so the whole quantization noise covariance matrix omega is a complete matrix;
2.2) the optimization problem for optimizing the above system parameters when BBU pools employ multivariate compression can be expressed as:
Figure FDA0003621692670000052
Figure FDA0003621692670000053
Figure FDA0003621692670000054
Figure FDA0003621692670000055
Figure FDA0003621692670000056
Figure FDA0003621692670000057
Figure FDA0003621692670000058
Ω±0. (1h)
ω k representing a weight per user, (1b) representing an achievable rate constraint for a fronthaul link, (1c) being a fronthaul compression constraint, (1d) and (1e) representing transmit power constraints for the BBU pool and each RRH, respectively; (1f) and (1g) unit mode 1 constraints of passive beamforming matrices of the IRS-assisted fronthaul link and the access link, respectively, (1h) representing that the fronthaul quantization noise covariance matrix is a semi-positive definite matrix;
2.3) solving the problem (P1) after transformation;
2.3.1) transforming the objective function of (P1) into the following by using a mean square error MSE method:
Figure FDA0003621692670000059
w A,k for the introduced auxiliary variable, u A,k Is at user k from y A,k Middle estimate s k Of (2), i.e. predicting the resulting signal as
Figure FDA0003621692670000061
Wherein y is A,k For signals received by user k, s k Base band signal obtained by coding a downlink message for user k by BBU, 1 is s k The dimension (c) of (a) is,
Figure FDA0003621692670000062
is the mean square error:
Figure FDA0003621692670000063
when w is A,k And u A,k When the following values are obtained, R sum Obtaining an optimal value:
Figure FDA0003621692670000064
Figure FDA0003621692670000065
2.3.2) similar to 2.3.1), by the MSE method, constraint (1b) can be approximated as:
Figure FDA0003621692670000066
W F,l requiring a positive half-definite, is an introduced auxiliary variable,
Figure FDA0003621692670000067
is at RRH l from y F,l Middle estimation signal t l I.e. predicting the resulting signal as
Figure FDA0003621692670000068
Wherein y is F,l Is the signal received by RRH l, t l Is BBU to x l A base band signal obtained by encoding the compression index of (a), d R Is t l Dimension (d);
Figure FDA0003621692670000069
is the mean square error matrix:
Figure FDA00036216926700000610
when W is F,l And U F,l When the following values are taken, (6) the right side takes the maximum value:
Figure FDA00036216926700000611
Figure FDA00036216926700000612
2.3.3) for constraint (1c), let
Figure FDA00036216926700000613
(1c) Can be rewritten as:
Figure FDA00036216926700000614
constraint (1c) may further translate to the following approximate constraint:
Figure FDA0003621692670000071
∑ l requiring a semi-positive definite, is an introduced auxiliary variable, | S | is the number of RRHs in the set S; when sigma l When the following values are taken as the values,
Figure FDA0003621692670000072
(10) and (1c) equivalents:
Figure FDA0003621692670000073
2.3.4) the optimization problem (P1) is transformed into:
Figure FDA0003621692670000074
Figure FDA0003621692670000075
Figure FDA0003621692670000076
Figure FDA0003621692670000077
Figure FDA0003621692670000078
Figure FDA0003621692670000079
Figure FDA00036216926700000710
Ω±0. (12h)
updating the auxiliary variable w by (4), (5), (8), (9) and (11) A,k 、u A,k 、W F,l 、U F,l Sum Σ l . For a fixed w A,k 、u A,k 、W F,l 、U F,l Sum Σ l Optimizing F by solving the following problem l 、v k 、Ω、Θ A And Θ F In the formula k Weight representing each user:
Figure FDA00036216926700000711
s.t.(12b)～(12h). (13b)
2.4) the problem (P3) is decomposed into three sub-problems to be solved alternately;
2.4.1) first fix Ω, Θ in the problem (P3) A And Θ F Transmit beamforming matrix F to optimize BBU pools and RRHs l And v k (ii) a The first sub-problem is given by:
Figure FDA0003621692670000081
Figure FDA0003621692670000082
Figure FDA0003621692670000083
Figure FDA0003621692670000084
Figure FDA0003621692670000085
the problem (P3.1) is convex and is solved by a standard optimization tool CVX;
2.4.2) second subproblem fix F l 、v k And Θ A Optimizing the fronthaul quantization noise covariance matrix omega and the IRS phase shift matrix theta for the auxiliary fronthaul link F (ii) a The second sub-problem is as follows:
Figure FDA0003621692670000086
Figure FDA0003621692670000087
Figure FDA0003621692670000088
Figure FDA0003621692670000089
Figure FDA00036216926700000810
Ω±0. (15f)
target (15a), constraints (15c) and (15d) are known to be convex with respect to Ω; the constraint (15b) is transformed according to
Figure FDA00036216926700000811
And formula (6) pair
Figure FDA00036216926700000812
Rewriting (15b) to the following equation:
Figure FDA00036216926700000813
wherein the content of the first and second substances,
Figure FDA0003621692670000091
Figure FDA0003621692670000092
d R is the data dimension of each RRH, in the above formulae:
Figure FDA0003621692670000093
d F,l ＝diag(D F,l ),
Figure FDA0003621692670000094
the problem (P3.2) translates into the following:
Figure FDA0003621692670000095
Figure FDA0003621692670000096
Figure FDA0003621692670000097
Figure FDA0003621692670000098
Figure FDA0003621692670000099
Figure FDA00036216926700000910
Figure FDA00036216926700000911
the constraint (17f) is still non-convex, a semi-definite relaxation method (SDR) is applied, after the constraint (17f) is removed, the optimal solution is obtained by using a standard optimization tool CVX, if the optimal solution is obtained
Figure FDA00036216926700000912
The rank of the optimal solution is not 1, then
Figure FDA00036216926700000913
Randomizing to produce a feasible sub-optimal solution, and updating Ω and Θ only as the objective function value of (P3.2.1) increases F ；
2.4.3) fixing F l 、v k Omega and theta F To theta A Optimizing to obtain a third sub-problem:
Figure FDA00036216926700000914
Figure FDA00036216926700000915
the objective function is rewritten as follows:
Figure FDA0003621692670000101
wherein the content of the first and second substances,
Figure FDA0003621692670000102
in the above formula, the first and second carbon atoms are,
Figure FDA0003621692670000103
Figure FDA0003621692670000104
the sub-problem (P3.3) can be rewritten as:
Figure FDA0003621692670000105
Figure FDA0003621692670000106
Figure FDA0003621692670000107
problem (P3.3.1) similar to (P3.2.1), it is also possible to obtain a feasible sub-optimal solution by semi-definite relaxation SDR, using a randomization method, and only update Θ when the value of the objective function of (P3.3.1) increases A ；
2.5) repeating the steps in 2.4.1) -2.4.3) until convergence;
2.6) when the BBU pool adopts a point-to-point compression method, performing joint optimization on a transmission beam forming matrix of the BBU pool and the RRH, a phase shift matrix of the IRS and a fronthaul quantization noise covariance matrix, and considering fronthaul compression constraint of point-to-point compression, wherein the problem can be expressed as:
Figure FDA0003621692670000108
s.t.(1b),(1d)～(1g), (19b)
Figure FDA0003621692670000109
Figure FDA00036216926700001010
the difference between the problems (P4) and (P1) is the constraints (19c) and (19d), where (19c) is relative to Ω l,l And v k Is non-convex. Order to
Figure FDA00036216926700001011
(19c) Is rewritten as:
Figure FDA0003621692670000111
log|O l the upper limit of |:
Figure FDA0003621692670000112
S l ± 0 is an auxiliary variable, so the following convex constraint can be substituted for (19 c):
Figure FDA0003621692670000113
the objective function and other constraints in the problem (P4) are treated in the same way as the problem (P1), and then a similar method can be applied to solve the problem using a successive convex approximation method and an alternating optimization method;
2.7) for the IRS reflector phase dispersion case, first following the steps in 2.1) -2.3) until after convergence, the theta is obtained A And Θ F Its diagonal line element theta A,i,m And theta F,i,m Mapping to discrete phase point to determine theta F Scaling omega to meet the constraints of the problem (P3.2.1)
Figure FDA0003621692670000114
CN202210460209.0A 2022-04-28 2022-04-28 IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information) Pending CN114900398A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210460209.0A CN114900398A (en) 2022-04-28 2022-04-28 IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210460209.0A CN114900398A (en) 2022-04-28 2022-04-28 IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information)

Publications (1)

Publication Number Publication Date
CN114900398A true CN114900398A (en) 2022-08-12

Family

ID=82720380

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210460209.0A Pending CN114900398A (en) 2022-04-28 2022-04-28 IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information)

Country Status (1)

Country Link
CN (1) CN114900398A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115834322A (en) * 2022-11-11 2023-03-21 西南交通大学 Communication system based on separation receiver and assistance of intelligent reflecting surface
CN116260490A (en) * 2023-05-16 2023-06-13 南京邮电大学 Forward compression and precoding method for cellular multi-antenna system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112600598A (en) * 2020-12-15 2021-04-02 浙江工业大学 Intelligent reflecting surface enhanced wireless forward link transmission method in cloud access network
CN113037659A (en) * 2021-02-26 2021-06-25 浙江工业大学 Multi-intelligent-reflector-assisted uplink cloud access network access link transmission method
CN113726395A (en) * 2021-08-23 2021-11-30 浙江工业大学 Intelligent reflector enhanced cloud access network multi-antenna user uplink transmission method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112600598A (en) * 2020-12-15 2021-04-02 浙江工业大学 Intelligent reflecting surface enhanced wireless forward link transmission method in cloud access network
CN113037659A (en) * 2021-02-26 2021-06-25 浙江工业大学 Multi-intelligent-reflector-assisted uplink cloud access network access link transmission method
CN113726395A (en) * 2021-08-23 2021-11-30 浙江工业大学 Intelligent reflector enhanced cloud access network multi-antenna user uplink transmission method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张昱，等: "Beamforming and fronthaul compression design for intelligent reflecting surface aided cloud radio access networks", 《FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115834322A (en) * 2022-11-11 2023-03-21 西南交通大学 Communication system based on separation receiver and assistance of intelligent reflecting surface
CN115834322B (en) * 2022-11-11 2024-04-12 西南交通大学 Communication system based on separation receiver and intelligent reflecting surface assistance
CN116260490A (en) * 2023-05-16 2023-06-13 南京邮电大学 Forward compression and precoding method for cellular multi-antenna system
CN116260490B (en) * 2023-05-16 2023-08-15 南京邮电大学 Forward compression and precoding method for cellular multi-antenna system

Similar Documents

Publication Publication Date Title
CN111314893B (en) Reflector assisted device-to-device communication system design method
RU2726850C2 (en) System and method of transmitting subset selection information
EP3510707B1 (en) Method and apparatus to enable channel compression in advanced wireless communication systems
US8098755B2 (en) Method and system for beamforming in a multiple user multiple input multiple output (MIMO) communication system using a codebook
JP5993006B2 (en) Apparatus and method for combining baseband processing and radio frequency beam steering in a wireless communication system
KR100952351B1 (en) Method and system for an alternating channel delta quantizer for 2×2 mimo pre-coders with finite rate channel state information feedback
KR100956327B1 (en) Method and system for rate reduction pre-coding matrices
CN109845133A (en) The system and method for classification beam forming and rank adaptation for hybrid antenna framework
CN111835406B (en) Robust precoding method suitable for energy efficiency and spectral efficiency balance of multi-beam satellite communication
TWI517620B (en) Method and apparatus for communicating in a mimo network
CN108736943B (en) Hybrid precoding method suitable for large-scale MIMO system
CN114900398A (en) IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information)
JP2009141957A (en) Pre-coding transmission method of mimo system
US20200136700A1 (en) Channel Prediction for Adaptive Channel State Information (CSI) Feedback Overhead Reduction
CN108566236B (en) User terminal, base station, and hybrid beamforming transmission method and system
CN114070365B (en) Intelligent reflection surface assisted low-radio-frequency-complexity multi-user MIMO uplink spectrum efficiency optimization method
CN107809275B (en) Finite feedback hybrid precoding method based on millimeter wave MIMO system
TWI538428B (en) Method for communicating in a network, secondary station and primary station
CN107171709B (en) Large-scale MIMO system precoding method applied to aggregated user scene
CN113726395A (en) Intelligent reflector enhanced cloud access network multi-antenna user uplink transmission method
CN104539392A (en) Multi-cell space interference aligning implementing method based on limited rate pre-coding feedback
WO2013113166A1 (en) User equipment, radio base station and respective methods therein for joint transmitting and receiving procedure
CN107959520B (en) Time slot interference alignment method for full duplex base station cellular network
KR20110045649A (en) Transmission apparatus and method for performing interference alignment thereof, and Reception apparatus and method for operating thereof
CN115549745B (en) RIS phase shift design method, RIS phase shift design device, RIS phase shift design computer equipment and RIS phase shift storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination