阅读说明：本技术 基于ris和相关信道无蜂窝系统功率设计方法 (Cellular-free system power design method based on RIS and related channels ) 是由 周猛 袁建涛 殷锐 万安平 王景霖 单添敏 颉满刚 于 2021-09-29 设计创作，主要内容包括：本发明涉及一种基于RIS和相关信道无蜂窝系统功率设计方法,包括步骤：建立基于RIS和空间相关Rician信道的无蜂窝大规模MIMO系统传输架构；根据信号传输特性,采用传统的最小均方误差估计方法获得无蜂窝大规模MIMO系统传输架构的估计信道。本发明的有益效果是：对于无蜂窝大规模MIMO系统,本发明通过联合调节功率分配系数和RISs的相移矩阵来有效地调节不同入射信号的传输相位和功率来达到提高系统信道的能力,进而可以最大化系统的总传输信号强度、降低系统干扰,这将有效地提升用户端接收信号功率、不断提高系统空间自由度和空间分集增益,最终达到最大化系统能量效率的目标。(The invention relates to a cellular-free system power design method based on RIS and related channels, comprising the following steps: establishing a cell-free large-scale MIMO system transmission architecture based on the RIS and the spatial correlation Rician channel; according to the signal transmission characteristics, an estimation channel of a transmission architecture of the cellular-free large-scale MIMO system is obtained by adopting a traditional minimum mean square error estimation method. The invention has the beneficial effects that: for a large-scale MIMO system without cells, the invention effectively adjusts the transmission phase and power of different incident signals to achieve the capability of improving the system channel by jointly adjusting the power distribution coefficient and the phase shift matrix of RISs, thereby maximizing the total transmission signal intensity of the system and reducing the system interference, effectively improving the signal receiving power of a user terminal, continuously improving the system space freedom and the space diversity gain, and finally achieving the goal of maximizing the system energy efficiency.)

1. A cellular system power design method based on RIS and associated channels, comprising the steps of:

step 1, establishing a cell-free large-scale MIMO system transmission architecture based on an RIS and a spatial correlation Rician channel;

step 2, obtaining an estimation channel of a transmission architecture of the non-cellular large-scale MIMO system by adopting an MMSE channel estimation method, and obtaining a closed expression of the performance index of the non-cellular large-scale MIMO system by using a random matrix theory and a majority theorem;

step 3, carrying out quantitative system asymptotic performance analysis on the closed expression obtained in the step 2 by adopting a Chebyshev theorem and a continuous mapping theorem;

in the above formula, R_mkRepresenting a spatial correlation matrix; the system is subjected to asymptotic performance analysis according to the property of a space correlation channel R _ mk;

step 4, according to the closed type solution obtained in the step 2, an effective compensation scheme of the performance loss of the non-cellular large-scale MIMO system caused by the spatial correlation of the channel is provided, an energy loss model is built, and the energy efficiency of the transmission architecture of the non-cellular large-scale MIMO system is obtained according to the built energy loss model;

and 5, in order to maximize the energy efficiency of the cellular-free large-scale MIMO system, the optimization problem of the system energy efficiency is solved by jointly optimizing the RIS phase shift matrix and the power distribution coefficient.

2. The RIS and associated channel based cellular system-less power design method of claim 1, wherein step 1 comprises the steps of:

step 1.1, supposing that a cellular-free large-scale MIMO system based on RIS and spatial correlation Rician channels consists of a central processing unit, M APs, R RISs and K users; wherein: each AP is provided with L antennas, each RIS is provided with S reflecting units, and each user is provided with N antennas; the channel matrix from the mth AP to the r-th RIS is denoted G_mr∈C^L×S(ii) a The channel matrix for the r-th RIS to k-th user is denoted G_rk∈C^S×N(ii) a The channel matrix between the mth AP to the kth user is denoted as G_mk∈C^L×NWherein C represents a complex field set, and L, S, N refers to L antennas, 5 reflection units, and N antennas, respectively;

step 1.2, respectively carrying out treatment on the channel from the m AP to the r RIS, the channel from the r RIS to the k user and the m user in the non-cellular large-scale MIMO system according to the Crohn's product_AModeling the channel between P and k users, then G_mr、G_rkAnd G_mkFurther expressed as:

in the above formula, H_ω1、H_ω2And H_ωRespectively representing large-scale fading matrixes from an m-th AP to an r-th RIS, an r-th RIS to a k-th user and an m-th AP to a k-th user; theta_AAnd Θ_RRespectively represents the correlation matrix theta of the mth AP and the r-th RIS sending end_TAnd Θ_DRespectively representing correlation matrixes at an r-th RIS receiving end and a k-th user;

step 1.3, channel G_mr、G_rkAnd G_mkThe distribution characteristics of (a) are expressed as:

in the above formula, the first and second carbon atoms are,representation matrixObedience mean of 0 and variance ofIs normally distributed, andrepresenting a kronecker product of a correlation matrix at the mth AP and a correlation matrix at the r-th RIS receiving end;representation matrix G_rk∈C^S×NObedience mean of 0 and variance ofIs normally distributed, andthe kronecker product of the correlation matrix at the sending end of the r-th RIS and the correlation matrix at the k-th user is expressed;representation matrix G_mk∈C^L×NObedience mean of 0 and variance ofIs normally distributed, andexpressing a kronecker product of a correlation matrix of an r-th RIS sending end and a correlation matrix of an r-th RIS receiving end; l, S, N respectively indicate L antennas, S reflection units and N antennas;

step 1.4, the channel matrix from the mth AP to the kth user is expressed as:

G_mrk＝G_mk+G_mrΘ_rG_rk (7)

in the above formula, theta_r＝diag exp(jθ_r，1)，...，exp(jθ_r，S)，Phase shift matrix, θ, representing the r-th RIS_r，sE 0,2 pi denotes the phase shift coefficient of the s-th reflection unit in the r-th RIS, andin addition to this, the present invention is,a set of all RIS's is represented,representing a feasible set of RIS reflection coefficients;

further expressed as:

in the above formula, symbolIndicating that the left side of the symbol is defined as the right side of the symbol;

step 1.5 in pilot stageSegment, projecting the received signal onto pilot matrix and performing de-spreading operation to obtain signal Y_mkThen, a minimum mean square error estimation method is adopted to obtain an estimated channel G_mrk：

In the above formula, E {. cndot } represents the desired operation,andeach represents G_mrkAnd Y_mkThe conjugate transpose operation of (1).

3. The RIS and associated channel based cellular system-less power design method of claim 2, wherein step 5 specifically comprises the steps of:

step 5.1, for the transmission architecture of the non-cellular massive MIMO system based on the RIS and the spatial correlation Rician channel, the total energy loss P of the non-cellular massive MIMO system is reduced_totalExpressed as:

in the above formula, P_mRepresenting the power, R, of the relevant circuit components for signal processing at the m-th AP_sumRepresenting the total spectral efficiency, P, of a cellular-free massive MIMO system_bt，mRepresents the front-end link power loss, P, associated with the mth AP_0，mRepresents the mth AP return link loss, and P_r(b) Represents the power loss per unit hardware at the r-th RIS in a phase-precision phase shifter with b-bit resolution;

step 5.2, according to the large-scale MIMO system P without the honeycomb_totalObtaining system energy efficiency eta^EEExpression (c):

in the above formula, B represents a transmission bandwidth, R_sumRepresenting the total rate of the system, P_totalRepresents the total power consumption of the system; eta^EERepresenting the ratio of the effective transmission rate of the system to the total power of the system signal transmission;

step 5.3, obtaining the system energy efficiency eta according to the step 5.2^EEUnder the power constraint condition, the energy efficiency of the system is maximized by jointly optimizing the RIS phase shift matrix and the power distribution scheme, and the joint optimization of the RIS phase shift matrix and the power distribution scheme is specifically modeled as follows:

in the above formula, ρ_k＝|ρ_1k，ρ_2k，...，ρ_Mk|，θ_r＝|θ_r，1，θ_r，2，...，θ_r，S|，ρ_mkRepresenting a power distribution coefficient of a kth user associated with the mth AP; r_thRepresenting minimum spectral efficiency constraint, ρ, for a cellular-free massive MIMO system_maxRepresents a maximum limit on user power;

and 5.4, solving and analyzing the joint optimization RIS phase shift matrix and the power distribution problem by adopting an alternative optimization algorithm, and providing the maximum energy efficiency for the non-cellular large-scale MIMO system.

Technical Field

The invention belongs to the field of wireless communication, and particularly relates to a large-scale cellular-free MIMO power design method based on a RIS and related channels.

Background

In order to meet the performance requirements of the system with ultra-high speed, ultra-large capacity and ultra-low delay in the future, the cellular-free large-scale multiple-input multiple-output (MIMO) technology combines the advantages of the distributed antenna technology and the centralized large-scale MIMO technology, and can greatly reduce the average distance between the access points which are dispersedly deployed in a large number of geographic locations and users, thereby effectively reducing the adverse effects of path loss and shadow fading on the system, and obtaining more macro diversity gain and larger spatial multiplexing gain through favorable channel propagation conditions. Particularly, when a Time Division Duplex (TDD) technology is adopted, channel estimation obtained by an uplink can be used in a downlink transmission process through the reciprocity of the uplink and the downlink, so that pilot overhead and system cost of a system downlink can be effectively reduced, and the method has very important research value and significance particularly for a considered large-scale MIMO antenna system without a cell.

In addition, with the rapid development of metamaterial technology, a large number of passive reflecting element impedances integrated on a super surface can be intelligently configured by a reconfigurable intelligent interface (RIS) through a software programming mode, so that the direction, the phase, the amplitude, the frequency, the polarization mode and the like of incident electromagnetic waves can be regulated and controlled by realizing the functions of wave control, wave polarization, wave absorption and the like in software, and the purposes of improving the system channel capacity and reducing the system propagation loss are achieved by intelligently reconfiguring a wireless transmission channel transmitted between transceivers. Therefore, the RIS has very important research significance and practical value in solving the problems of non-line-of-sight transmission, expanding the coverage, reducing electromagnetic pollution, reducing signal interference, sensing environment, positioning, realizing green communication and other targets and requirements.

Generally, a real channel should be composed of a semi-deterministic line of sight (LoS) path component with random phase shift and a random non-line of sight (nlos) path component, so that Rician channel is more suitable for real transmission requirements. However, in the actual transmission process, due to a large number of Access Points (APs) with randomly distributed geographic locations, the randomness of users, the non-uniform radiation direction characteristics exhibited by small antenna spacing or poor scattering conditions, and the dense deployment of a large number of reflection units in the RIS in a limited space, the spatial correlation exhibited by signals in the transmission environment can be inevitably caused in the actual wireless transmission environment. In particular, when the correlation of the channels of the system is stronger, the degree of suppression of the system performance is greater, and serious link interruption occurs when the system is serious, which causes a serious challenge to the performance of users at different geographic positions in the system.

Disclosure of Invention

The invention aims to overcome the defects in the prior art and provides a method for power design of a cellular-free massive MIMO system based on an RIS and a space-related channel.

The cellular system power design method without cells based on RIS and relative channel includes following steps:

step 1, establishing a cell-free large-scale MIMO system transmission architecture based on an RIS and a spatial correlation Rician channel;

step 2, obtaining an estimation channel of a transmission framework of the non-cellular large-scale MIMO system by adopting a traditional Minimum Mean Square Error (MMSE) method according to the signal transmission characteristics, and obtaining a closed expression of the performance index of the non-cellular large-scale MIMO system by using a random matrix theory and a majority theorem according to the signal transmission characteristics;

step 3, carrying out quantitative MIMO system asymptotic performance analysis on the closed expression obtained in the step 2 by adopting a Chebyshev theorem and a continuous mapping theorem, and deeply excavating deep influence of system parameters of the non-cellular large-scale MIMO system on the system performance;

in the above formula, R_mkRepresenting a spatial correlation matrix; the system is subjected to asymptotic performance analysis according to the property of a space correlation channel R _ mk;

step 4, according to the closed type solution obtained in the step 2, an effective compensation scheme of the performance loss of the cellular massive MIMO system caused by the spatial correlation of the channel is provided, namely parameters such as the number of antennas, an RIS phase shift matrix and transmission power in the cellular massive MIMO system are researched to compensate the factors of the performance reduction of the system caused by the correlation; constructing an energy loss model, and obtaining the energy efficiency of the transmission architecture of the non-cellular large-scale MIMO system according to the constructed energy loss model;

step 5, according to the effective compensation scheme which is provided in the step 4 and causes the performance loss of the non-cellular large-scale MIMO system due to the spatial correlation of the channel, and according to the constructed energy loss model, the energy efficiency of the transmission architecture of the non-cellular large-scale MIMO system is researched; in order to maximize the energy efficiency of a cellular-free massive MIMO system, the system energy efficiency is maximized by jointly optimizing the RIS phase shift matrix and the power distribution coefficients.

Preferably, step 1 specifically comprises the following steps:

step 1.1, supposing that a cellular-free large-scale MIMO system based on RIS and spatial correlation Rician channels consists of a central processing unit, M APs (wireless access points), R RISs and K users; wherein: each AP is provided with L antennas, each RIS is provided with S reflecting units, and each user is provided with N antennas; the channel matrix from the mth AP to the r-th RIS is denoted G_mr∈C^L×S(ii) a The channel matrix for the r-th RIS to k-th user is denoted G_rk∈C^S×N(ii) a The channel matrix between the mth AP to the kth user is denoted as G_mk∈C^L×NWherein C represents a complex field set, and L, S, N refers to L antennas, S reflection units, and N antennas, respectively; meanwhile, the channel from the m AP to the r RIS, the channel from the r RIS to the k user and the channel from the m AP to the k user are all assumed to be formed byThe antenna spacing is small, the scattering condition is poor, or the RIS reflecting elements are densely deployed and the like, so that Rician channels with spatial correlation exist in the actual wireless propagation environment, and each RIS can independently reflect incident signals;

step 1.2, respectively modeling channels from m AP to r RIS, channels from r RIS to k user and channels from m AP to k user in the cellular-free large-scale MIMO system according to Kronecker product (Kronecker), and then modeling G_mr、G_rkAnd G_mkFurther expressed as:

in the above formula, H_ω1、H_ω2And H_ωRespectively representing large-scale fading matrixes from the mth AP to the r-th RIS, from the r-th RIS to the k-th user and from the m-th AP to the k-th user, wherein the large-scale fading matrixes, the r-th RIS to the k-th user and the m-th AP to the k-th user are random variables which meet the independent and same distribution; theta_AAnd Θ_RRespectively represents the correlation matrix theta of the mth AP and the r-th RIS sending end_TAnd Θ_DRespectively representing correlation matrixes at an r-th RIS receiving end and a k-th user;

step 1.3, channel G is mapped according to the properties of the spatial correlation channel_mr、G_rkAnd G_mkThe distribution characteristics of (a) are expressed as:

in the above formula, the first and second carbon atoms are,representation matrix G_mr∈C^L×SObedience mean of 0 and variance ofIs normally distributed, andrepresenting a kronecker product of a correlation matrix at the mth AP and a correlation matrix at the r-th RIS receiving end;representation matrix G_rk∈C^S×NObedience mean of 0 and variance ofIs normally distributed, andthe kronecker product of the correlation matrix at the sending end of the r-th RIS and the correlation matrix at the k-th user is expressed;representation matrix G_mk∈C^L×NObedience mean of 0 and variance ofIs normally distributed, andrepresenting the correlation moment at the sender of the r-th RISKronecker product of the matrix and the correlation matrix of the r-th RIS receiving end; l, S, N respectively indicate L antennas, S reflection units and N antennas;

step 1.4, according to the transmission characteristics of the cellular-free massive MIMO system assisted by RIS, the channel matrix from the mth AP to the kth user is expressed as:

G_mrk＝G_mk+G_mrΘ_rG_rk (7)

in the above formula, the first and second carbon atoms are,phase shift matrix, θ, representing the r-th RIS_r,sE 0,2 pi denotes the phase shift coefficient of the s-th reflection unit in the r-th RIS, andin addition to this, the present invention is,a set of all RIS's is represented,representing a feasible set of RIS reflection coefficients;

to study the more generalized channel transmission model, it is assumed hereIs an ideal RIS, i.e. θ is associated with the RIS element_r,sBoth amplitude and phase of (a) can be controlled independently and continuously,further expressed as:

in the above formula, symbolIndicating that the left side of the symbol is defined as the right side of the symbol;

step 1.5, in the pilot stage, projecting the received signal onto the pilot matrix and performing despreading operation to obtain a signal Y_mkThen, a conventional MMSE method is adopted to obtain an estimated channel G_mrk：

In the above formula, E {. cndot } represents the desired operation,andeach represents G_mrkAnd Y_mkThe conjugate transpose operation of (1).

Preferably, step 5 specifically comprises the following steps:

step 5.1, for the transmission architecture of the non-cellular massive MIMO system based on the RIS (reconfigurable intelligent reflector) and the spatial correlation Rician channel, the total energy loss P of the non-cellular massive MIMO system_totalExpressed as:

in the above formula, P_mRepresenting the power, R, of the relevant circuit components for signal processing at the m-th AP_sumRepresenting the total spectral efficiency, P, of a cellular-free massive MIMO system_bt,mRepresents the front-end link power loss, P, associated with the mth AP_0,mRepresents the mth AP return link loss, and P_r(b) Represents the power loss per unit hardware at the r-th RIS in a phase-precision phase shifter with b-bit resolution;

step 5.2, according to the large-scale MIMO system P without the honeycomb_totalObtaining system energy efficiency eta^EEExpression (c):

in the above formula, B represents a transmission bandwidth, R_sumRepresenting the total rate of the system, P_totalRepresents the total power consumption of the system; eta^EEThe ratio of the effective transmission rate of the system to the total power of the signal emission of the system is expressed, the number of transmission bits which can be obtained by the unit energy of the system is described, the number represents the use efficiency of the system on energy resources, and the index has important research value and significance for realizing future green communication and promoting the goals of carbon peak reaching and carbon neutralization; in general, power control is an important requirement in communication systems, and can effectively adjust power distribution coefficients through different power distribution strategies to achieve the goal of optimizing system performance. Particularly, for the system model, the transmission phases and powers of different incident signals can be effectively adjusted by jointly adjusting the power distribution coefficients and the phase shift matrix of the ris to achieve the capability of improving the system channel, so that the total transmission signal strength of the system can be maximized, the system interference can be reduced, the signal receiving power of a user end can be effectively improved, the system space freedom degree and the space diversity gain are continuously improved, and the aim of maximizing the system energy efficiency is finally achieved;

step 5.3, obtaining the system energy efficiency eta according to the step 5.2^EEUnder the power constraint condition, the energy efficiency of the system is maximized by jointly optimizing the RIS phase shift matrix and the power distribution scheme, and the joint optimization of the RIS phase shift matrix and the power distribution scheme is specifically modeled as follows:

in the above formula rho_k＝[ρ_1k，ρ_2k，...，ρ_Mk]，θ_r＝[θ_r，1，θ_r，2，...，θ_r，S]，ρ_mkRepresenting a power distribution coefficient of a kth user associated with the mth AP; r_thRepresenting a large-scale MIMO system without cellsMinimum spectral efficiency constraint, ρ_maxRepresents a maximum limit on user power;

step 5.4, jointly optimizing the RIS phase shift matrix and the power distribution scheme problem, which is generally an NP-hard problem, and a non-convex constraint non-convex quadratic optimization problem which cannot be directly solved through the current software; the invention adopts an alternative optimization algorithm to solve and analyze the problem of jointly optimizing the RIS phase shift matrix and the power distribution, thereby providing the maximum energy efficiency for the non-cellular large-scale MIMO system.

The invention has the beneficial effects that:

for a large-scale MIMO system without cells, the invention effectively adjusts the transmission phase and power of different incident signals to achieve the capability of improving the system channel by jointly adjusting the power distribution coefficient and the phase shift matrix of RISs, thereby maximizing the total transmission signal intensity of the system and reducing the system interference, effectively improving the signal receiving power of a user terminal, continuously improving the system space freedom and the space diversity gain, and finally achieving the goal of maximizing the system energy efficiency.

The invention fully considers the functions of beam control, beam absorption and the like of the characteristics of the direction, the phase, the amplitude, the polarization mode and the like of the incident electromagnetic wave signals by the RIS through a software programming mode, and further can intelligently reconstruct the wireless channel transmitted by the system, thereby achieving the great system performance advantages of improving the system channel capacity and reducing the system loss.

Compared with the prior art, the method for power design of the cellular-free large-scale MIMO system based on the RIS and the space related channel provides uniform service with high speed and large capacity for the system.

Drawings

FIG. 1 is a system framework diagram of the present invention based on RIS and spatially correlated Rician channels;

fig. 2 is a flow chart of the power design method of the cellular-free massive MIMO system based on RIS and spatially correlated channels according to the present invention.

Detailed Description

The present invention will be further described with reference to the following examples. The following examples are set forth merely to aid in the understanding of the invention. It should be noted that, for a person skilled in the art, several modifications can be made to the invention without departing from the principle of the invention, and these modifications and modifications also fall within the protection scope of the claims of the present invention.

Example one

In consideration of the spatial correlation problem of signals in a transmission environment in a system due to intensive deployment of a large number of reflecting units of an RIS in a limited space under practical conditions, an embodiment of the present application proposes a transmission architecture of a cellular-free massive MIMO system based on an RIS and a space-dependent channel as shown in fig. 1, and also presents a method for power design of a cellular-free massive MIMO system based on an RIS and a space-dependent channel as shown in fig. 2:

step 1, establishing a non-cellular large-scale MIMO system transmission architecture based on an RIS (reconfigurable intelligent reflector) and a spatial correlation Rician channel;

step 2, obtaining an estimation channel of a transmission framework of the non-cellular large-scale MIMO system by adopting a traditional MMSE channel estimation method according to the signal transmission characteristics, and obtaining a closed expression of the performance index of the non-cellular large-scale MIMO system by using a random matrix theory and a majority theorem according to the signal transmission characteristics;

step 3, carrying out quantitative MIMO system asymptotic performance analysis on the closed expression obtained in the step 2 by adopting a Chebyshev theorem and a continuous mapping theorem, and deeply excavating deep influence of system parameters of the non-cellular large-scale MIMO system on the system performance;

in the above formula, R_mkRepresenting a spatial correlation matrix;

step 4, according to the closed type solution obtained in the step 3, an effective compensation scheme of the performance loss of the cellular massive MIMO system caused by the spatial correlation of the channel is provided, namely parameters such as the number of antennas, RIS phase shift and transmission power in the cellular massive MIMO system are researched to improve and compensate the factors of the performance reduction of the system caused by the correlation;

step 5, according to the effective compensation scheme which is provided in the step 4 and causes no performance loss of the cellular massive MIMO system due to the spatial correlation of the channel, the energy efficiency of the transmission architecture of the cellular massive MIMO system is researched; in order to maximize the energy efficiency of a cellular-free massive MIMO system, the system energy efficiency is maximized by jointly optimizing the RIS phase shift matrix and the power allocation scheme.

Example two

On the basis of the first embodiment, the second embodiment of the present application provides a detailed scheme of the method for power design of a cellular-free massive MIMO system based on RIS and spatially correlated channels in the first embodiment:

1. firstly, obtaining the distribution characteristics of a channel;

it is assumed that the cellular-free massive MIMO system is composed of one central processing unit, M APs (wireless access points), R RIS and K users; wherein: each AP is provided with L antennas, each RIS is provided with S reflecting units, and each user is provided with N antennas; the channel matrix from the mth AP to the r-th RIS is denoted G_mr∈C^L×S(ii) a The channel matrix for the r-th RIS to k-th user is denoted G_rk∈C^S×N(ii) a The channel matrix between the mth AP to the kth user is denoted as G_mk∈C^L×NWherein C represents a complex field set, and L, S, N refers to L antennas, S reflection units, and N antennas, respectively; meanwhile, supposing that the channel from the m-th AP to the r-th RIS, the channel from the r-th RIS to the k-th user and the channel from the m-th AP to the k-th user are all affected inevitably due to the reasons of smaller antenna spacing, poorer scattering condition or dense arrangement of RIS reflecting elements, etc., then there is a Rician channel with spatial correlation in the actual wireless propagation environment, and each RIS can independently reflect the incident signal;

respectively modeling channels from m AP to r RIS, channels from r RIS to k user and channels from m AP to k user in the cellular-free massive MIMO system according to Kronecker product (Kronecker), and then respectively modeling G_mr、G_rkAnd G_mkFurther expressed as:

in the above formula, H_ω1、H_ω2And H_ωRespectively representing large-scale fading matrixes from the mth AP to the r-th RIS, from the r-th RIS to the k-th user and from the m-th AP to the k-th user, wherein the large-scale fading matrixes, the r-th RIS to the k-th user and the m-th AP to the k-th user are random variables which meet the independent and same distribution; theta_AAnd Θ_RRespectively represents the correlation matrix theta of the mth AP and the r-th RIS sending end_TAnd Θ_DRespectively representing correlation matrixes at an r-th RIS receiving end and a k-th user;

the distribution characteristics of the obtained channel are:

in the above formula, the first and second carbon atoms are,representation matrix G_mr∈C^a×bObedience mean of 0 and variance ofIs normally distributed, andrepresenting a kronecker product of a correlation matrix at the mth AP and a correlation matrix at the r-th RIS receiving end;representation matrix G_rk∈C^a×bObedience mean of 0 and variance ofIs normally distributed, andthe kronecker product of the correlation matrix at the sending end of the r-th RIS and the correlation matrix at the k-th user is expressed;representation matrix G_mk∈C^a×bObedience mean of 0 and variance ofIs normally distributed, andexpressing a kronecker product of a correlation matrix of an r-th RIS sending end and a correlation matrix of an r-th RIS receiving end; l, S, N respectively indicate L antennas, S reflection units and N antennas;

2. then, obtaining an estimated channel by adopting a traditional minimum mean square error channel estimation method;

according to the transmission characteristics of the RIS-assisted cellular massive MIMO system, the channel matrix from the mth AP to the kth user is represented as:

G_mrk＝G_mk+G_mrΘ_rG_rk

in the above formula, the first and second carbon atoms are,phase shift matrix, θ, representing the r-th RIS_r,sE 0,2 pi denotes the phase shift coefficient of the s-th reflection unit in the r-th RIS, andin addition to this, the present invention is,a set of all RIS's is represented,representing a feasible set of RIS reflection coefficients;

to study the more generalized channel transmission model, it is assumed hereIs an ideal RIS, i.e. θ is associated with the RIS element_r,sBoth amplitude and phase of (a) can be controlled independently and continuously,further expressed as:

in the above formula, symbolIndicating that the left side of the symbol is defined as the right side of the symbol;

in the pilot phase, the received signal is projected on a pilot matrix and de-spread to obtain a signal Y_mkThen, a conventional Minimum Mean Squared Error (MMSE) method is used to obtain an estimated channel G_mrk：

In the above formula, E {. cndot } represents the desired operation,andeach represents G_mrkAnd Y_mkThe conjugate transpose operation of (1).

3. And then, according to the transmission characteristics of the signals, a closed expression of the system performance index under the system model is obtained by using a random matrix theory and a majority theorem.

4. Secondly, by using the Chebyshev theorem and the continuous mapping theorem, that isAndthe asymptotic performance of the system is deeply analyzed, and the influence of system parameters on the performance of the system is further deeply understood. Furthermore, for the proposed transmission architecture, a total energy loss model of the system is obtainedThereby achieving the overall energy efficiency of the system. Because the energy efficiency is a very important performance index of the system, the method has important research value and significance for realizing future green communication and promoting the goals of carbon peak reaching and carbon neutralization.

5. Finally, based on the obtained energy efficiency expression, the invention further proposes to maximize the energy efficiency of the novel system architecture by jointly optimizing the RIS phase shift matrix and the power distribution coefficients.