A joint power allocation and phase shift design method based on distributed ris-assisted multi-user system

Abstract

The application discloses a kind of based on distributed RIS auxiliary multi-user system's joint power allocation and phase shift design method, which includes the following steps: establishing distributed RIS auxiliary multi-user system, constructing cascaded channel model and received signal model;Joint power allocation, RIS phase shift and communication scheduling design, construct the target optimization problem of maximum achievable rate maximization;Optimize power allocation by maximizing traversable achievable rate;Optimize phase shift using minimax optimization method;Optimize communication scheduling, complete the joint design of power allocation, RIS phase shift and communication scheduling.The application considers distributed RIS auxiliary multi-user system, by optimizing the phase shift of multiple RIS, power allocation and communication scheduling jointly, achievable rate maximization optimization is realized, and iterative algorithm is designed to solve, and the performance of the application is excellent, reaches high sum rate.

Description

TECHNICAL FIELD

[0001] The present application relates to wireless communications, and in particular to a joint power allocation and phase shift design method based on a distributed RIS assisted multi-user system. BACKGROUND

[0002] Reconfigurable intelligent surfaces (RIS) can achieve unprecedented high throughput and energy efficiency, and have been widely studied since their inception. Compared with traditional active beamforming / relay technology, each reflecting element of RIS can adjust the reflection direction of the incident signal, thereby effectively improving the communication quality. Due to the low hardware cost of reflecting elements, RIS can be flexibly deployed to improve the spectral efficiency of multi-user multiple-input multiple-output (MIMO) systems. Therefore, studying resource allocation schemes in RIS assisted systems to expand the range of communication services and meet the quality of service requirements of green communication is a research hotspot for new generation mobile communications.

[0003] To meet the high throughput, high connectivity and quality of service requirements in wireless communications, it is of great significance to integrate more and more RIS into microcell networks. Multiple RIS can provide more abundant macrodiversity, and the distribution of RIS in a large area can improve spatial diversity by transmitting the same information from completely unrelated RIS nodes to the receiver. Therefore, the cooperation relationship of distributed RIS cannot be ignored. Notably, existing work on multi-RIS assisted communication ignores power allocation among multiple RIS, limiting its application in the Internet of Things. These observations suggest that distributed RIS is very important in optimizing system performance, especially in providing stronger robustness to shadow fading channels.

[0004] In addition, distributed RIS provides a degree of mobility to improve the performance of wireless networks, and each user can receive signals from RIS at different locations with different propagation paths. However, each RIS is affected by independent and different degrees of large-scale fading, which further complicates communication scheduling and power allocation. And since distributed RIS is introduced, power allocation is coupled with the phase shift design of multiple RIS, which makes it impossible to directly apply existing methods. Therefore, there is a lack of research on distributed RIS assisted communication scheduling and power allocation systems. SUMMARY

[0005] In order to overcome the defects and deficiencies existing in the prior art, the present application provides a joint power allocation and phase shift design method based on a distributed RIS assisted multi-user system.

[0006] The second object of the present application is to provide a joint power allocation and phase shift design system based on a distributed RIS-assisted multi-user system.

[0007] The third object of the present application is to provide a storage medium.

[0008] The fourth object of the present application is to provide a computing device.

[0009] In order to achieve the above objects, the present application adopts the following technical solutions:

[0010] The present application provides a joint power allocation and phase shift design method based on a distributed RIS-assisted multi-user system, comprising the following steps:

[0011] A distributed RIS-assisted multi-user system is established, a cascaded channel model and a received signal model are constructed;

[0012] Based on the wake-up communication scheduling method, the index variables of the BS-user and BS-RIS-user channels are constructed, and the communication scheduling constraints are constructed. Under the constraint of the total transmit power, the target optimization problem of maximizing the achievable rate is constructed;

[0013] The power allocation is optimized by maximizing the traversable achievable rate;

[0014] The minimum-maximum optimization method is adopted to determine the phase shift of the i-th reflection coefficient in the m-th RIS by maximizing the approximate maximum of the total channel gain, and the phase shift is optimized;

[0015] The communication scheduling constraints are adjusted, the communication scheduling objective function is constructed, and the transmit power, phase shift and communication scheduling optimization of multiple RISs are completed through alternating optimization.

[0016] As a preferred technical solution, the cascaded channel model is constructed, which is specifically represented as:

[0017]

[0018] Among them, V k,m represents the cascaded BS-RIS-user channel, d k,m represents the small-scale fading channel between BS and RIS, g k,m represents the small-scale fading channel between the user and the RIS, a k,m represents the large-scale fading coefficient;

[0019] The multiple RISs are combined with the cascaded channel model, and the received pilot signal through M RISs is represented as:

[0020]

[0021] Among them, s kdenotes the transmitted symbol, p k denotes the transmit power of the BS to the kth user, h k denotes the channel between the kth user and the BS, φ k,m denotes the phase shift introduced by the mth RIS, n k denotes the noise at the kth user.

[0022] As a preferred technical solution, the target optimization problem of constructing the maximum achievable rate under the total transmit power constraint includes the following specific steps:

[0023] The index variables of the BS-user and BS-RIS-user channels are ζ a,l and ζ b,l , and the scheduling constraints are constructed;

[0024] The received signal is decomposed based on the linearity of the LOS and reflected channels;

[0025] By applying the Jensen inequality, the lower bound of the achievable rate is solved;

[0026] Under the total transmit power constraint, the achievable rate optimization problem of the distributed RIS-assisted multi-user MIMO system is represented as:

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] where P m,k denotes the transmit power of the mth antenna to the kth user, P max denotes the maximum power limit at the BS, φ k,m,i denotes the RIS phase shift matrix constraint, β k,m,i and ω k,m,i respectively denote the reflection amplitude and phase shift of the mth RIS;

[0033] The objective function is represented as:

[0034]

[0035]

[0036] where P m,k denotes the interference power of the kth user, denotes the coefficient of the variance, Vm,k represents the cascaded BS-RIS-user channel, β0represents the amplitude initial value, χ m,k represents the coefficient of shadow fading, φ k,m represents the phase shift from the mthRIS to the kthuser, represents the noise of NLOS.

[0037] As a preferred technical solution, the power allocation is optimized by maximizing the ergodic rate, and the specific steps include:

[0038] The target optimization problem of maximizing the ergodic rate is adjusted as:

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048] wherein P m,k represents the transmission power from the mthantenna to the kthuser, P m,k represents the interference power of the kthuser, P max represents the maximum power limit at the BS, X k represents the LOS variable of the kthuser, e -Xk represents the LOS exponential variable of the kthuser, e -Yk represents the NLOS exponential variable of the kthuser, V m,k represents the cascaded BS-RIS-user channel, β0represents the amplitude initial value, χ m,k represents the coefficient of shadow fading, φ m,k represents the phase shift from the mthRIS to the kthuser, and represent the noise of LOS and NOS, respectively;

[0049] The non-convex function is processed by alternating iteration based on the sum rate maximization, which is converted into:

[0050]

[0051]

[0052] wherein, is an exponential variable of , represents an arbitrary variable of LOS, Y k represents an arbitrary variable, represents a second-order Taylor expansion point of Y k , is an exponential variable of ;

[0053] The convex approximation of the optimization problem at the tth iteration is represented as:

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061] wherein, represents the LOS variable of the kth user at the tth iteration, represents the NLOS variable of the kth user at the tth iteration,

[0062] At the t-1th iteration, the Taylor expansion of is selected as the optimal value of the optimization problem;

[0063] Set the iteration end condition, and output the power allocation result.

[0064] As a preferred technical scheme, the setting of the iteration end condition is specifically represented as:

[0065]

[0066] wherein, represents a set threshold value, represents the spectral efficiency of the result of the kth user at the tth iteration, represents the spectral efficiency of the result of the kth user at the t-1th iteration.

[0067] As a preferred technical solution, the minimum-maximum optimization method is adopted, the phase shift of the i-th reflection coefficient in the m-th RIS is determined by maximizing the approximation of the total channel gain, and the phase shift is optimized, and the specific representation is as follows:

[0068] The phase shift design problem of the RIS is represented as:

[0069]

[0070]

[0071] ω k,m,i ∈(0,2π].

[0072] Where, β k,m,i' represents the amplitude of the RIS, represents the phase of the RIS, h k represents the channel between the k-th user and the BS, φ k,m,i represents the RIS phase shift matrix constraint, V k,m represents the concatenated BS-RIS-user channel, Φ k,m represents the phase shift introduced by the m-th RIS, ω k,m,i represents the phase shift of the m-th RIS;

[0073] The maximum optimization problem about the total channel gain is represented as:

[0074]

[0075] Where, V k,m,i represents the i-th path of the BS-RIS-user concatenated channel;

[0076] The phase shift of the i-th reflection coefficient in the m-th RIS is determined by maximizing the approximation of the total channel gain, and the specific representation is as follows:

[0077]

[0078] The quantized phase is represented as:

[0079]

[0080] Where, is the phase shift resolution, and B is the quantization bit.

[0081] As a preferred technical solution, the communication scheduling constraint is widened, and a communication scheduling objective function is constructed, which specifically includes:

[0082] For any given transmit power and phase shift of the m-th RIS, the communication scheduling sub-optimization problem is constructed as:

[0083]

[0084]

[0085]

[0086]

[0087]

[0088] where ζ a,l and ζ b,l denote the indicator variables of BS-user, BS-RIS-user channel, P m,k denotes the transmit power of the mth antenna to the kth user, P m,k denotes the interference power of the kth user, P max denotes the maximum power limit at the BS, V m,k denotes the cascaded BS-RIS-user channel, β0denotes the amplitude initial value, χ m,k denotes the coefficient of shadow fading, φ m,k denotes the phase shift of the mth RIS to the kth user, and denote the noise of LOS and NOS, respectively;

[0089] The indicator variables of BS-user, BS-RIS-user channel are updated continuously to obtain a local optimal solution, and the transmit power, phase shift and communication scheduling optimization of multiple RISs are completed through alternating optimization.

[0090] In order to achieve the above-mentioned second purpose, the application adopts the following technical scheme:

[0091] A joint power allocation and phase shift design system based on a distributed RIS assisted multi-user system, comprising: a multi-user system construction module, a model construction module, an optimization problem construction module, a power allocation optimization module, a phase shift optimization module and an alternating optimization module;

[0092] The multi-user system construction module is used to establish a distributed RIS assisted multi-user system;

[0093] The model construction module is used to construct a cascaded channel model and a received signal model;

[0094] The optimization problem construction module is used to construct the indicator variables of BS-user, BS-RIS-user channel based on a wake-up communication scheduling method, and to construct a communication scheduling constraint, and to construct a target optimization problem of maximizing achievable rate under total transmit power constraint;

[0095] The power allocation optimization module is used to optimize power allocation by maximizing the traversable achievable rate;

[0096] The phase shift optimization module is configured to optimize the phase shift by determining the phase shift of the i-th reflection coefficient in the m-th RIS by maximizing an approximation of the total channel gain using a minimization-maximization optimization method.

[0097] The alternating optimization module is configured to widen the communication scheduling constraint, construct a communication scheduling objective function, and complete the transmission power, phase shift, and communication scheduling optimization of the multiple RISs through alternating optimization.

[0098] To achieve the above-mentioned third object, the application adopts the following technical solutions:

[0099] A storage medium stores a program, and the program is executed by a processor to implement the joint power allocation and phase shift design method based on the distributed RIS-aided multi-user system.

[0100] To achieve the above-mentioned fourth object, the application adopts the following technical solutions:

[0101] A computing device includes a processor and a memory for storing a processor-executable program, and the processor executes the program stored in the memory to implement the joint power allocation and phase shift design method based on the distributed RIS-aided multi-user system.

[0102] Compared with the prior art, the application has the following advantages and beneficial effects:

[0103] The application constructs a distributed RIS-aided multi-user system, improves spatial diversity by transmitting the same information from completely uncorrelated channels to the receiver, and guarantees the achievable spectral efficiency of the distributed RIS-aided multi-user system using a sum-rate maximization algorithm. To solve the non-convex optimization problem, the maximization problem is decomposed into power allocation, phase shift, and communication scheduling sub-problems. In the power allocation design stage, a linear approximation of the objective function is derived using an approximate equivalent function series expansion, and then an iterative algorithm is used to solve the power optimization sub-problem. The phase shift of the multiple RISs is optimized through a weighted least mean square error and minimization-maximization method. Finally, the communication scheduling sub-problem is divided into two scalar continuous differentiable functions, which are solved in sequence, improving the effectiveness and robustness of the overall scheme. BRIEF DESCRIPTION OF DRAWINGS

[0104] Figure 1 A flowchart of the joint power allocation and phase shift design method based on the distributed RIS-aided multi-user system of the application;

[0105] Figure 2 A schematic diagram of the distributed RIS-aided multi-user system of the application;

[0106] Figure 3Fig. 1 is a schematic diagram of the convergence of the present application under different maximum power constraints with spectral efficiency as an index;

[0107] Figure 4 Fig. 2 is a schematic diagram of the achievable sum rate of the present application under different maximum power constraints as a function of the number of RISs. DETAILED DESCRIPTION

[0108] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0109] Example 1

[0110] As shown in Fig. 1, the joint power allocation and phase shift design method based on distributed RIS-aided multi-user system includes the following steps: Figure 1 Step 1: Establish a distributed RIS-aided multi-user system, and construct a cascaded channel model and a received signal model.

[0111] A distributed RIS-aided multi-user system is constructed, as shown in Fig. 1, a BS equipped with m antennas communicates with K single-antenna target users, wherein all users are simultaneously served by RISs with the same time-frequency resources. It is assumed that the channel state information is known, and all RISs can be connected to the CPU through a reliable high-speed wireless link, and the CPU can centrally complete beamforming design, scheduling and power allocation; this embodiment connects multiple RISs to a CPU, uses the CPU to cover the communication dead angle, and improves the spatial diversity by transmitting the same information from completely uncorrelated channels to the receiver;

[0112] Figure 2 In this embodiment, in order to realize the distributed RIS-aided multi-user system, an optimization method of joint power allocation, phase shift and communication scheduling is proposed, and Rayleigh fading environment is considered, wherein h and g

[0113] represent the small-scale fading channels between BS-RIS and user-RIS, respectively, and the elements thereof are independent and identically distributed, h k,i ~ CN(0, 1), g k,i ~ CN(0, 1). For the cascaded BS-RIS-user channel between the mth antenna of the BS and the kth user, it can be represented as:

[0114] wherein m is a k,m large-scale fading coefficient, and is specifically represented as:

[0115] wherein m is a k,m large-scale fading coefficient, and is specifically represented as:

[0116]

[0117] where is the shadow fading with standard deviation σ, χ k,m ~ N(0, 1), PL k,m is the path loss from the m-th antenna of the BS to the k-th user.

[0118] Assume denotes the channel between the k-th user and the BS, following a random distribution with mean 0 and variance Therefore, combining the cascaded channel model in (1) with multiple RISs, the received pilot signal through M RISs can be expressed as:

[0119]

[0120] where p k denotes the transmit power of the BS to the k-th user, s k denotes the transmit symbol and E[|s k | 2 ] = 1, n k ~ CN(0, 1) denotes the noise at the k-th user. φ k,m = [φ k,m,l , φ k,m,2 ,..., φ k,m,l ] is the phase shift introduced by the m-th RIS.

[0121] Step 2: Perform power allocation, RIS phase shift, and communication scheduling joint design to construct an objective optimization problem that maximizes the achievable rate.

[0122] Based on the wake-up communication scheduling method, define the indicator variables of the BS-user and BS-RIS-user channels as ζ a,l and ζ b,l . If ζ a,l = 1, the BS-user channel serves the k-th user, otherwise, ζ a,l = 0. Similarly, if ζ b,l = 1, the BS-RIS-user channel migrates the signal to the k-th user, and if ζ b,l = 0, it does not transmit the signal. Therefore, the following scheduling constraints can be established:

[0123]

[0124]

[0125] where L and L' correspond to the number of paths, respectively;

[0126] Using the linearity of the LOS and reflected channels, the received signal yk = [y LoS,k NLOS,k ] T LoS,k NLOS,k Assuming each user only knows the channel statistics but not the instantaneous channel gain, the received signal at the kth user can be written as:

[0127]

[0128]

[0129] where and are the desired signal, is the self-interference term due to the unknown instantaneous channel gain at the user, referred to as the beamforming gain uncertainty, I i→k = [I i→k,LOS I i→k,NLOS T is the interference from the ith user to the kth user. According to the above definition, the BS-user and BS-RIS-user channels are independent, and if the woken-up LOS channel is used to transmit the signal together with the BS-user channel, the instantaneous signal-to-interference-plus-noise ratio (SINR) of the kth user is:

[0130]

[0131] Therefore, the ergodic downlink achievable rate corresponding to the kth user is:

[0132] R k,LOS = E{log2(l + γ k,LOS )}, (9)

[0133] where E{·} is the expectation with respect to the small-scale channel fading.

[0134] On the other hand, if the woken-up NLOS (or reflected channel) channel is used to transmit the signal together with the BS-RIS-user channel, the instantaneous SINR of the kth user can be expressed as:

[0135]

[0136] When ζ a,l = 1, the ergodic downlink achievable rate of the kth user can be expressed as:

[0137] R k,NLOS = E{log2(l + γ k,NLOS )}. (11)

[0138] By applying the Jensen inequality, the lower bound of the achievable rate in (9) and (11) is derived as:​​​​

[0139]

[0140]

[0141] For the considered system, it is required to jointly design the transmit power, communication scheduling and phase shift of multiple RISs under the minimum transmit power constraint at the BS and the communication scheduling requirement of LOS and NLOS channels to maximize the overall throughput of the system. Therefore, under the total transmit power constraint, the achievable rate optimization problem of the distributed RIS-aided multi-user MIMO system can be mathematically expressed as:

[0142]

[0143]

[0144] φ k,m,i ∈F, (14c)

[0145]

[0146]

[0147] where (14b) is the transmit power constraint, P m,k represents the transmit power from the mthantenna to the kthuser, and the power sum is less than P max , P max is the maximum power limit at the BS, (14c) is the RIS phase shift matrix constraint, and The i-th reflection coefficient of the m-th RIS is denoted as β k,m,i ∈[0,1] and ω k,m,i represent the reflection amplitude and phase shift of the m-th RIS, respectively;

[0148] However, due to the constraints of self-interference and the difficulty of deriving its probability distribution, the above optimization problem is difficult directly. To this end, the objective function (14a) is re-expressed according to the following proposition:

[0149]

[0150]

[0151] where, is the interference power of the kthuser, and a is a constant equal to 1-π / 4, P m,u represents the power from the u-th user to the m-th RIS, represents the coefficient of variance, P m,kdenotes the transmit power of the mthantenna to the kthuser, β denotes the RIS amplitude, β0denotes the amplitude initial value, χ m,k denotes the coefficient of shadow fading, φ k,m denotes the phase shift of the mthRIS to the kthuser, denotes the noise of NLOS, in addition, φ k,m is the same as φ m,k denotes the same meaning, v m,k is the same as v k,m denotes the same meaning;

[0152] For the above optimization problem, multiple variables need to be optimized together, especially for the case of high-dimensional matrix and non-convex constraint, it is difficult to obtain the optimal solution. On this basis, the phase shift of multiple RIS, power allocation and communication scheduling can be considered to be handled separately, and alternating optimization is performed.

[0153] Step three: optimize the power allocation by maximizing the ergodic rate.

[0154] To complete the power allocation optimization, first optimize the transmit power in the case of fixed RIS position. Rewrite the problem P0 as:

[0155]

[0156]

[0157]

[0158]

[0159] Then, the objective function (17a) can be rewritten as

[0160]

[0161]

[0162] wherein,

[0163] Since the objective function is still difficult to solve at this time, the optimization problem P1.1 is further re-expressed as:

[0164]

[0165]

[0166]

[0167]

[0168]

[0169]

[0170]

[0171] The objective function is affine, and (20b) and (20c) are convex sets. k The LOS variable representing the k-th user, e -Xk Let e ​​represent the LOS exponent variable for the k-th user. -Yk The NLOS exponent variable represents the k-th user;

[0172] It should be pointed out that, It is p m,k The sum of the concave functions remains concave. Furthermore, the logarithm of the concave function is also concave. Therefore, (20f) and (20g) are also convex sets. For the non-convex (20d) and (20e), they can be processed using alternating iterations based on maximizing the sum rate, transforming into:

[0173]

[0174]

[0175] in, for The exponential variable, Y represents any variable in the LOS region. k Represents any variable, Y represents k The second-order Taylor expansion point, for Exponential variables;

[0176] Therefore, the convex approximation of problem P1.2 in the t-th iteration can be reformulated as:

[0177]

[0178] st(20b),(20c),(20f),(20g) (24b)

[0179]

[0180]

[0181] in, This represents the LOS variable for the k-th user in the t-th iteration. Let represent the NLOS variable for the k-th user in the t-th iteration; and in the (t-1)-th iteration, the Taylor expansion of . The optimal value of P1.3 is selected as P1.3. Thus, a power allocation optimization algorithm can be proposed, the iterative process of which continues until convergence, i.e. wherein ò is a precision threshold, wherein, denotes the spectral efficiency of the kth user at the tth iteration of the result, denotes the spectral efficiency of the kth user at the t-1th iteration of the result;

[0182] The algorithm steps are as follows:

[0183] Input: p m,k , H k , 1≤m≤M, 1≤k≤K, M represents the total number of RISs, and K represents the total number of users;

[0184] Step 1: initialization:

[0185] Step 2: update the values of {P, X k , Y k} by solving P1.3;

[0186] Step 3: update

[0187] Step 4: update

[0188] Step 5: update t = t + 1;

[0189] Step 6: if the iteration is ended, and the power allocation result is output; if return to Step 2.

[0190] Step 4: the phase shift is optimized by using the minimization-maximization optimization method.

[0191] The phase shift design of multi-RIS is the key to determine the performance of the multi-RIS assisted system. In order to find the optimal phase shift, the gain maximization It is not difficult to find that the achievable rate R k,LOS is independent of the phase shift design and can be ignored. Therefore, the phase shift design problem of RIS is equivalent to the following maximization problem:

[0192]

[0193]

[0194] ω k,m,i ∈(0,2π]. (25c)

[0195] wherein β k,m,i' represents the amplitude of RIS, denotes the phase of the RIS, the k-th user to the i-th element on the m-th RIS;

[0196] Since is a non-concave function of φ k,m , it is easy to know that problem (25) is a complex non-convex problem. In order to solve this non-convex optimization problem, define Rewrite the objective function (25a) as:

[0197]

[0198] Specifically, the first term in (26) can be further expressed as:

[0199]

[0200] where v k,m = [v k,m,l , v k,m,2 ,..., v k,m,I ] T , φ k,m = [φ k,m,1 , φ k,m,2 ,..., φ k,m,I ] T .

[0201] Since the definition of φ k,m,i is such that Therefore, (27) has no phase information. In addition, there are cross terms in the second term of (27), which can be ignored Therefore, the optimization problem about the maximum of the total channel gain can be rewritten as:

[0202]

[0203] where V k,m,i denotes the i-th path of the BS-RIS-user cascaded channel;

[0204] From problem (28), it can be seen that the phase shift of the i-th reflection coefficient in the m-th RIS can be determined by maximizing the approximation of the total channel gain:

[0205]

[0206] It is worth noting that improving system performance by increasing RIS elements is a good energy-saving scheme. However, a large number of RIS elements will lead to a significant increase in computational demand. Therefore, the phase shift of the RIS reflection coefficient is quantized, which further reduces the hardware complexity. Let The quantized phase can be calculated by

[0207]

[0208] where is the phase shift resolution, B is the quantization bits. According to the above analysis, the RIS phase shift optimization processing algorithm can be proposed, the specific steps are as follows:

[0209] Input: β k,m,i = 1, 1≤k≤K;

[0210] Step1: initialization: h k,m , v k,m ;

[0211] Step2: update the optimal

[0212] Step3: update

[0213] Step4: if convergence, stop iteration, output if not convergence, jump to Step2.

[0214] Step five: optimize the communication scheduling, complete the joint design of power allocation, RIS phase shift and communication scheduling.

[0215] In order to optimize the communication scheduling, the integer communication scheduling constraints (4) and (5) are relaxed into the following continuous constraint form:

[0216]

[0217]

[0218] For any given transmit power p k,m and the phase shift of the mth RIS φ k,m , the communication scheduling subproblem is:

[0219]

[0220] s.t.(31a)(31b). (32b)

[0221] The objective function (32a) is a scalar continuous differentiable function, and it is not difficult to get the local optimal solution by constantly updating ζ a,l and ζ b,l . It can be seen that the transmit power, phase shift and communication scheduling optimization of multiple RISs can be completed by alternating optimization.

[0222] According to the above steps, numerical verification is carried out, such as Figure 3 and Figure 4As shown, the convergence of the joint power allocation and phase shift design method in this embodiment with respect to spectral efficiency is obtained, and the achievable sum rate changes with the number of RISs under different maximum power constraints. The results show that the method of this embodiment converges well, has high spectral efficiency and achievable sum rate.

[0223] Embodiment 2

[0224] The embodiment provides a joint power allocation and phase shift design system based on a distributed RIS-aided multi-user system, comprising: a multi-user system construction module, a model construction module, an optimization problem construction module, a power allocation optimization module, a phase shift optimization module and an alternating optimization module.

[0225] In this embodiment, the multi-user system construction module is used to establish a distributed RIS-aided multi-user system.

[0226] In this embodiment, the model construction module is used to construct a cascaded channel model and a received signal model.

[0227] In this embodiment, the optimization problem construction module is used to construct index variables of BS-user and BS-RIS-user channels based on a wake-up communication scheduling method, and construct a communication scheduling constraint, and construct a target optimization problem of maximizing achievable rate under total transmit power constraint.

[0228] In this embodiment, the power allocation optimization module is used to optimize power allocation by maximizing the traversable achievable rate.

[0229] In this embodiment, the phase shift optimization module is used to adopt a minimization-maximization optimization method, determine the phase shift of the i-th reflection coefficient in the m-th RIS by maximizing the approximate total channel gain, and optimize the phase shift.

[0230] In this embodiment, the alternating optimization module is used to adjust the communication scheduling constraint, construct a communication scheduling objective function, and complete the transmission power, phase shift and communication scheduling optimization of multiple RISs through alternating optimization.

[0231] Embodiment 3

[0232] The embodiment provides a storage medium, which can be a ROM, a RAM, a magnetic disk, an optical disk or the like storage medium. The storage medium stores one or more programs, and when the programs are executed by a processor, the joint power allocation and phase shift design based on a distributed RIS-aided multi-user system of embodiment 1 is realized.

[0233] Embodiment 4

[0234] The embodiment provides a computing device which can be a desktop computer, a notebook computer, a smart phone, a PDA handheld terminal, a tablet computer or other terminal device with a display function, the computing device comprising a processor and a memory, the memory storing one or more programs, and the processor implementing the joint power allocation and phase shift design based on a distributed RIS assisted multi-user system of the embodiment 1 when executing the programs stored in the memory.

[0235] The above embodiment is a preferred embodiment of the present application, but the embodiments of the present application are not limited to the above embodiment, and any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the present application should be equivalent replacement methods, and are included in the protection scope of the present application.

Claims

1. A method for joint power allocation and phase shift design based on distributed RIS-aided multi-user systems, characterized in that, The method comprises the following steps: A distributed RIS-assisted multi-user system is established, a cascaded channel model and a received signal model are constructed; Based on the wake-up communication scheduling method, index variables of BS-user and BS-RIS-user channels are constructed, and a communication scheduling constraint is constructed, and a target optimization problem of maximizing the achievable rate is constructed under the constraint of total transmit power; Power allocation is optimized by maximizing the traversable achievable rate; By minimizing-maximizing optimization method, the phase shift of the i-th reflection coefficient in the m-th RIS is determined by maximizing the approximation of the total channel gain, and the phase shift is optimized, which is specifically represented as: The phase shift design problem of the RIS is represented as: ω k,m,i ∈(0,2π] where β k,m,i denotes the amplitude of RIS, denotes the phase of RIS, h k denotes the channel between the kth user and BS, φ k,m,i denotes the RIS phase shift matrix constraint, V k,m denotes the concatenated BS-RIS-user channel, ω k,m denotes the phase shift introduced by the mth RIS, ω k,m,i denotes the phase shift of the mth RIS; The maximum optimization problem about the total channel gain is represented as: wherein V k,m,i represents the ith path of the BS-RIS-user cascade channel; The phase shift of the i-th reflection coefficient in the m-th RIS is determined by maximizing the approximation of the total channel gain, which is specifically represented as: The quantized phase is represented as: wherein B is the quantization bit for phase resolution. The communication scheduling constraint is widened, the communication scheduling target function is constructed, and the transmit power, phase shift and communication scheduling optimization of multiple RISs are completed through alternating optimization; The communication scheduling constraint is widened, the communication scheduling target function is constructed, and the communication scheduling constraint is specifically included: For any given transmit power and phase shift of the m-th RIS, the communication scheduling sub-optimization problem is constructed as: where ζ a,l and ζ b,l are the indicator variables of BS-user, BS-RIS-user channels, P m,k is the transmit power of the mthantenna to the kthuser, PI m,k is the interference power of the kthuser, P max is the maximum power limit at the BS, V m,k is the cascaded BS-RIS-user channel, β0is the amplitude initial value, χ m,k is the coefficient of shadow fading, φ m,k is the phase shift of the mthRIS to the kthuser, and are the noise of LOS and NOS, respectively, L and L’ correspond to the number of paths, respectively. The index variables of BS-user and BS-RIS-user channels are continuously updated to obtain a local optimal solution, and the transmit power, phase shift and communication scheduling optimization of multiple RISs are completed through alternating optimization.

2. The method of claim 1, wherein, The cascaded channel model is constructed, which is specifically represented as: where V k,m denotes the cascaded BS-RIS-user channel, d k,m denotes the small-scale fading channel between the BS and the RIS, g k,m denotes the small-scale fading channel between the user and the RIS, a k,m denotes the large-scale fading coefficient; The received pilot signal through M RISs is represented by combining multiple RISs with the cascaded channel model as: where s k denotes the transmitted symbol, p k denotes the transmit power of the BS to the kth user, h k denotes the channel between the kth user and the BS, φ k,m denotes the phase shift introduced by the mth RIS, n k denotes the noise at the kth user.

3. The method of claim 1, wherein, The specific steps of constructing the target optimization problem of maximizing the achievable rate under the constraint of total transmit power include: The index variable of the BS-user, BS-RIS-user channel is ζ a,l and ζ b,l , and a scheduling constraint is constructed; The received signal is decomposed based on the linearity of the LOS and reflected channel; By applying Jensen's inequality, the lower bound of the achievable rate is solved; Under the constraint of total transmit power, the achievable rate optimization problem of the distributed RIS-assisted multi-user MIMO system is represented as: where P m,k denotes the transmit power of the mthantenna to the kthuser, P max denotes the maximum power limit at the BS, φ k,m,i denotes the RIS phase shift matrix constraint, β k,m,i and ω k,m,i denote the reflection amplitude and phase shift of the mthRIS, respectively. The target function is represented as: where PI m,k denotes the interference power of the kth user, denotes the coefficient of variance, V ,,k denotes the cascaded BS-RIS-user channel, β0denotes the amplitude initial value, χ m,k denotes the coefficient of shadow fading, φ k,, denotes the phase shift from the mth RIS to the kth user, denotes the noise of NLOS.

4. The method of claim 1, wherein, The specific steps of optimizing the power allocation by maximizing the traversable achievable rate include: The target optimization problem of maximizing the achievable rate is adjusted as: where P ,,k denotes the transmit power of the mthantenna to the kthuser, P m,k denotes the interference power of the kthuser, P max denotes the maximum power limit at the BS, X k denotes the LOS variable of the kthuser, e -Xk denotes the LOS exponential variable of the kthuser, e -Yk denotes the NLOS exponential variable of the kthuser, V m,k denotes the cascaded BS-RIS-user channel, β0denotes the amplitude initial value, χ m,k denotes the coefficient of shadow fading, φ m,k denotes the phase shift of the mthRIS to the kthuser, and denote the noise of LOS and NOS, respectively; The non-convex function is processed by alternating iteration based on the sum rate maximization, which is transformed into: wherein is an index variable, represents an arbitrary variable of the LOS, Y k represents an arbitrary variable, represents a second order Taylor expansion point of Y k is an index variable;​ The convex approximation of the optimization problem at the t-th iteration is represented as: wherein, represents the LOS variable of the kth user at the tth iteration, represents the NLOS variable of the kth user at the tth iteration, At the (t-1)th iteration, the Taylor expansion of is chosen as the optimal value of the optimization problem; The iteration end condition is set, and the power allocation result is output.

5. The method of claim 4, wherein, The iteration end condition is specifically represented as: wherein ∈ denotes a set threshold value, denotes the spectral efficiency of the result of the tth iteration of the kth user, denotes the spectral efficiency of the result of the t-1th iteration of the kth user.

6. A system for joint power allocation and phase shift design based on distributed RIS-aided multi-user systems, characterized in that, The joint power allocation and phase shift design method based on the distributed RIS-assisted multi-user system according to any one of claims 1-5 comprises a multi-user system construction module, a model construction module, an optimization problem construction module, a power allocation optimization module, a phase shift optimization module and an alternating optimization module; The multi-user system construction module is used to establish a distributed RIS-assisted multi-user system; The model construction module is used to construct a cascaded channel model and a received signal model; The optimization problem construction module is used to construct index variables of BS-user and BS-RIS-user channels based on the wake-up communication scheduling method, and to construct a communication scheduling constraint, and to construct a target optimization problem of maximizing the achievable rate under the constraint of total transmit power; The power allocation optimization module is configured to optimize the power allocation by maximizing the ergodic sum rate; The phase shift optimization module is configured to optimize the phase shift of the i-th reflection coefficient in the m-th RIS by maximizing the total channel gain using a min-max optimization method; The alternating optimization module is configured to widen the communication scheduling constraint, construct a communication scheduling objective function, and complete the optimization of the transmission power, the phase shift and the communication scheduling of the multiple RISs by alternating optimization.

7. A storage medium storing a program, characterized by comprising: The program, when executed by the processor, implements the joint power allocation and phase shift design method for the distributed RIS-aided multi-user system according to any one of claims 1-5.

8. A computing device comprising a processor and a memory for storing processor-executable programs, characterized in that, The processor executes the program stored in the memory, and implements the joint power allocation and phase shift design method for the distributed RIS-aided multi-user system according to any one of claims 1-5.

Images