Method for energy efficiency optimization of massive MIMO system based on near field wireless power supply

Abstract

The application discloses a kind of based on near-field wireless energy supply large-scale MIMO system energy efficiency optimization method, establishes the near-field spherical wave channel model of PS to each device link, obtains the channel state information of each device to base station link;PS sends energy signal to device, calculates the energy collected by each device;Device utilizes the energy collected to base station transmission information, calculates the total throughput and total power consumption of system, and the ratio of the two is system energy efficiency;Construct the system energy efficiency maximization problem of joint optimization time resource, PS transmit beamforming vector and base station receive beamforming vector;The optimization problem is decomposed into three sub-problems, and three sub-problems are iteratively solved by alternating optimization algorithm until convergence, and the optimized system energy efficiency is obtained, and the scheme disclosed in the application significantly improves the energy efficiency of large-scale MIMO system.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to an energy efficiency optimization method for large-scale MIMO systems based on near-field wireless power supply. Background Technology

[0002] In recent years, with the widespread adoption of the Internet of Things (IoT) and smart terminals, the surge in the number of devices has brought about a dual demand for wireless power supply and efficient communication. Wireless power transfer technology, utilizing radio frequency signals to efficiently transmit energy, provides a feasible solution for the continuous power supply of a large number of passive or low-power terminals. Meanwhile, Massive Multiple Input Multiple Output (MIMO) systems, with their large-scale antenna arrays, can significantly improve spatial multiplexing capabilities and spectrum utilization efficiency, and have become an important component of future wireless networks. As the number of antennas deployed in power stations continues to increase and the operating frequency band continues to rise, the near-field effect becomes particularly significant. For example, with an antenna aperture of 0.5m and a frequency of 28GHz, the near-field region expands to 46m, causing passive devices near the power station to be distributed within its near-field range. However, when near-field wireless power transfer is combined with Massive MIMO, the system faces several new technical challenges, such as the complexity of near-field channel modeling and estimation, the coordinated scheduling of energy transfer and information transmission, and how to maximize energy efficiency while ensuring communication quality. These issues urgently need to be addressed. Therefore, designing an energy efficiency optimization method for Massive MIMO systems in near-field wireless power transmission scenarios is of great research significance and application prospects for improving the overall energy utilization of the network, extending the battery life of terminal devices, and promoting the development of green communication. Summary of the Invention

[0003] This invention proposes an energy efficiency optimization method for large-scale MIMO systems based on near-field wireless power supply, thereby maximizing energy efficiency.

[0004] The system comprises a power station (PS) equipped with a large-scale antenna array, multiple passive devices located in the near-field region of the PS's energy radiation, and a base station equipped with a large-scale receiving antenna array. The PS has M antennas, each passive device has a single antenna, and the base station has N antennas. Based on the Harvest-then-transmit (HTT) protocol, the passive devices first harvest energy from the PS's transmitted signal, then use the harvested energy to transmit information to the base station via a Non-Orthogonal Multiple Access (NOMA) protocol. The base station demodulates the signal using Successive Interference Cancellation (SIC).

[0005] The technical solution adopted in this invention includes the following steps:

[0006] Establish near-field spherical wave channel models for the PS-to-each device links and obtain channel state information for the links from each device to the base station;

[0007] PS sends energy signals to the devices and calculates the energy collected by each device.

[0008] The device uses the collected energy to transmit information to the base station, calculates the total throughput and total power consumption of the system, and the ratio of the two is the system energy efficiency;

[0009] The problem of maximizing system energy efficiency by jointly optimizing time resources, PS transmit beamforming vector, and base station receive beamforming vector;

[0010] The optimization problem is decomposed into three sub-problems. The first sub-problem optimizes the time resource allocation in the system, the second sub-problem optimizes the PS transmit beamforming vector, and the third sub-problem optimizes the base station receive beamforming vector. The three sub-problems are solved iteratively using the Alternate Optimization (AO) algorithm until convergence, and the optimized system energy efficiency is obtained.

[0011] Considering that the PS operates at a high frequency and is equipped with a large-scale antenna array, the near-field characteristics of signal propagation are more pronounced. In this case, the electromagnetic wave propagation form between the PS and the passive device changes from a plane wave to a spherical wave. Therefore, this invention uses a spherical wave propagation model to model this channel. Since the passive device is located in the near-field region of the PS, the signal propagation distance is short and there are no obvious obstacles along the path. Therefore, the near-field channel state information vector h from the PS to the k-th passive device... k There is a stable line-of-sight (LoS) propagation component. At the same time, due to the presence of scatterers in the environment, electromagnetic waves are also transmitted through reflection, refraction, and other means. Even under near-field conditions, h k There is also a non-line of sight (NLoS) propagation component. Therefore, the near-field channel model from the PS to the passive device is the sum of the line-of-sight components and the non-line-of-sight components, that is: Line-of-sight component It can be represented as:

[0012]

[0013] in, Let f represent the complex channel gain, f represent the frequency of the PS transmitted signal, and d represent the frequency of the PS transmitted signal. k The distance from the k-th passive device to the center of the PS antenna array is represented by the array response vector:

[0014] d represents the phase difference between the signals from each antenna of PS to the k-th passive device. k,m This represents the distance from the k-th passive device to the m-th antenna of the PS.

[0015] Non-line-of-sight components It can be represented as:

[0016]

[0017] Where C represents the number of scatterers in the near-field channel. d represents the complex channel gain corresponding to the signal transmitted from PS through the c-th scatterer to the passive device link. c,k This represents the distance from the c-th scatterer to the k-th passive device. d represents the phase difference between each link. c This represents the distance from the c-th scatterer to the center of the PS antenna array.

[0018] During the Wireless Energy Transmission (WET) phase, the energy signal transmitted by the PS in the system can be represented as:

[0019] x k =w k s(3),

[0020] in, Let represent the transmit beamforming vector from PS to the k-th passive device, and s represent the transmitted signal. Based on this, the received energy signal received by the k-th passive device can be expressed as:

[0021]

[0022] in, This represents the near-field channel state information vector from the PS to the k-th passive device, where n is the vector. k This represents the Gaussian white noise at the k-th passive device, which has zero mean and variance σ. 2 Gaussian distribution and denoted as In the formula (·) H This indicates the conjugate transpose.

[0023] According to the expression for the received signal of the passive device, the energy collected by the k-th passive device is expressed as:

[0024]

[0025] Where η∈(0,1) represents the energy conversion efficiency, and t0 represents the energy harvesting time. The time it takes for the passive device to send information to the base station is denoted as t1. Therefore, the power of the k-th passive device can be expressed as:

[0026]

[0027] During the Wireless Information Transmission (WIT) phase, passive devices use harvested energy to transmit signals to the base station based on the NOMA protocol. The signal received by the base station can be represented as:

[0028]

[0029] in, Let s represent the far-field channel state information vector from the k-th passive device to the base station. k This represents the signal transmitted by the k-th passive device; n represents the Gaussian white noise at the base station, which has zero mean and variance σ. 2 I N The complex Gaussian distribution is denoted as In the formula 0 N×1 Let I represent the zero vector of N×1 dimensions. N Let g represent an N×N dimensional identity matrix. Here, we consider that base stations are typically deployed at a considerable distance from passive devices, therefore g... k The traditional Ricean channel is used for modeling.

[0030] Assuming the base station uses a linear receive beamforming vector to decode the received signal, the recovered signal can be expressed as:

[0031]

[0032] in, Let represent the received beamforming vector at the base station. Without loss of generality, assume that the user's channel gain satisfies . Where |·| represents modulo. In an uplink NOMA system, a SiC receiver is configured at the base station. Co-channel interference is the sum of the channel gains of other users. The interference experienced by users mainly comes from users with good channel quality; that is, users with poor channel quality are more susceptible to strong co-channel interference. To avoid wasting resources, the system first decodes the signals from users with good channel quality and subtracts the decoded signals from the superimposed signals. Then, it decodes the signals from users with the next best channel quality, and so on, until all signals are decoded.

[0033] Therefore, the signal-to-interference-to-noise ratio of the k-th passive device can be expressed as:

[0034]

[0035] The throughput of the k-th passive device can be expressed as:

[0036] R k =t1log2(1+γ k (10),

[0037] Therefore, the throughput of the system can be expressed as:

[0038]

[0039] The total energy consumption of the system can be expressed as:

[0040]

[0041] Among them, P PS P represents the static circuit power consumption of PS. BS This indicates the static circuit power consumption of the base station.

[0042] With the goal of maximizing system energy efficiency, the optimization problem is as follows:

[0043]

[0044] stt0≥0, t1≥0, t0+t1≤1(13a),

[0045]

[0046] ||w B || 2 =1 (13c),

[0047] Among them, P max Ω represents the maximum transmit power of PS, where Ω = {t0, t1, w} B ,w1,...,w KEquation (13a) represents the energy and information transmission time constraints in the system, Equation (13b) represents the PS transmit power constraint, and Equation (13c) represents the normalization constraint of the beamforming vector of the base station received signal.

[0048] Because the objective function is non-convex and the optimization variables are strongly coupled, the original optimization problem is difficult to solve directly. Therefore, to address (P1), this invention uses the Dinkelbach algorithm to process the original optimization problem and decouples it into three sub-problems based on the AO algorithm, which are then solved iteratively.

[0049] First, since the objective function (P1) is a fraction, the variables are coupled and difficult to solve directly. Therefore, the Dinkelbach algorithm is used to handle this fractional optimization problem. By introducing the auxiliary variable λ, the objective function of the optimization problem (P1) can be expressed as:

[0050] R-λE total (14),

[0051] Where the auxiliary variable λ in the t-th iteration (t) It can be represented as:

[0052]

[0053] Therefore, the optimization problem (P1) can be transformed into:

[0054]

[0055] st(13a)-(13c)

[0056] Based on this, the problem (P1-E1) is decoupled into three sub-problems and solved separately. Given the PS transmitted signal beamforming vector {w1,...,w...} K} and the beamforming vector w of the base station received signal B In this case, optimize the PS energy transmission time t0 and the passive device information transmission time t1. The corresponding sub-problem (P2) is constructed as follows:

[0057]

[0058] st (13a)

[0059] It can be proven that the objective function can reach its maximum value when t0 + t1 = 1, thus obtaining the relationship of equation (18) with respect to t0:

[0060]

[0061] Taking the second derivative of equation (18) with respect to t0, we get:

[0062]

[0063] Since equation (19) is always less than 0, equation (18) is a convex function. Therefore, by using a one-dimensional search method on equation (18) to maximize f(t0), the optimal solution for t0 in subproblem (P2) can be obtained.

[0064] Then, given the PS energy transmission time t0, the passive device information transmission time t1, and the base station received signal beamforming vector w B In the case of optimizing the beamforming vector {w1,...,w} of the PS transmitted signal, K The corresponding sub-problem (P3) is constructed as follows:

[0065]

[0066] st(13b)

[0067] Since the problem (P3) is about {w1,...,w} K The optimization problem of} is to write the objective function as a function of w. k The expression:

[0068]

[0069] Using a low-complexity projection gradient algorithm, equation (21) applies to w k Find the gradient to obtain w k Gradient expression:

[0070]

[0071] Set initial values ​​while satisfying constraint (13b). Then update w according to the iterative formula (23). k :

[0072]

[0073] Where α represents the step size of the projection gradient iteration. When the iteration result w k If it is not in the feasible region, it is reprojected to the boundary of the feasible region according to equation (24):

[0074]

[0075] When the maximum number of iterations or f(w) is reached k When convergence occurs, we can obtain {w1,...,w} in the subproblem (P3). K The optimal solution for}.

[0076] Finally, given the PS energy transmission time t0, the passive device information transmission time t1, and the PS transmit signal beamforming vector {w1,...,w}, K}, optimize the beamforming vector w of the base station received signal B The corresponding sub-problem (P4) is constructed as follows:

[0077]

[0078] st(13c)

[0079] Question (P4) is about w B The optimization problem is such that (P4) can be represented in the following equivalent form:

[0080]

[0081] st(13c)

[0082] Based on the properties of matrix operations, the problem is transformed into:

[0083]

[0084] st(13c)

[0085] make Based on the properties of Rayleigh entropy, we obtain w B Closed-form solution:

[0086]

[0087] Where ν represents the eigenvector corresponding to the largest eigenvalue of matrix G.

[0088] The subproblems (P2), (P3), and (P4) are optimized alternately using an iterative algorithm until the objective function converges, thus obtaining the optimal solution to the original optimization problem (P1).

[0089] Table 1 shows the detailed steps of the algorithm used.

[0090] Table 1 - Iterative Algorithm of the Proposed Scheme

[0091] Attached Figure Description

[0092] Figure 1 This is a flowchart illustrating the steps of an embodiment of the present invention;

[0093] Figure 2 This is a system model diagram of an embodiment of the present invention;

[0094] Figure 3 The figure shows the simulation results of an embodiment of the present invention. Detailed Implementation

[0095] The technical solution of the present invention will be further described below with reference to the accompanying drawings:

[0096] This invention proposes an energy efficiency optimization method for a massive MIMO (Multiple Input Multiple Output) system based on near-field wireless power supply. The system comprises a power station (PS) equipped with a massive MIMO antenna array, multiple passive devices located within the near-field energy radiation region of the PS, and a base station equipped with a massive MIMO antenna array. The PS is equipped with M antennas, each passive device is equipped with a single antenna, and the base station is equipped with N antennas. Based on a pre-acquisition-then-transmission protocol, the multiple passive devices first acquire energy from the transmitted signal of the PS, and then use the acquired energy to transmit information to the base station via a non-orthogonal multiple access (NOMA) protocol. The base station demodulates the signal using successive interference cancellation (SIC). The system is described as follows: Figure 2 As shown.

[0097] Specifically, the steps include the following:

[0098] Step A: Establish near-field spherical wave channel models for the PS-to-each device links and obtain channel state information for the links from each device to the base station;

[0099] Step B: PS sends an energy signal to the devices and calculates the energy collected by each device;

[0100] Step C: The device uses the collected energy to transmit information to the base station, calculates the total throughput and total power consumption of the system, and the ratio of the two is the system energy efficiency;

[0101] Step D: Construct a system energy efficiency maximization problem that jointly optimizes time resources, PS transmit beamforming vector, and base station receive beamforming vector;

[0102] Step E: Decompose the optimization problem into three sub-problems. In the first sub-problem, optimize the time resource allocation in the system. In the second sub-problem, optimize the PS transmit beamforming vector. In the third sub-problem, optimize the base station receive beamforming vector. Iterate the solution of the three sub-problems through the Alternate Optimization (AO) algorithm until convergence, and obtain the optimized system energy efficiency.

[0103] The specific steps of step A are as follows:

[0104] Considering that the PS operates at a high frequency and is equipped with a large-scale antenna array, the near-field characteristics of signal propagation are more pronounced. In this case, the electromagnetic wave propagation form between the PS and the passive device changes from a plane wave to a spherical wave. Therefore, this invention uses a spherical wave propagation model to model this channel. Since the passive device is located in the near-field region of the PS, the signal propagation distance is short and there are no obvious obstacles along the path. Therefore, the near-field channel state information vector h from the PS to the k-th passive device... k There is a stable line-of-sight (LoS) propagation component. At the same time, due to the presence of scatterers in the environment, electromagnetic waves are also transmitted through reflection, refraction, and other means. Even under near-field conditions, h k Non-line-of-sight (Non-) also exists in the middle.

[0105] Line of Sight (NLoS) propagation component Therefore, the near-field channel model from the PS to the passive device is the sum of the line-of-sight components and the non-line-of-sight components, that is: Line-of-sight component It can be represented as:

[0106]

[0107] in, Let f represent the complex channel gain, f represent the frequency of the PS transmitted signal, and d represent the frequency of the PS transmitted signal. k The array response vector represents the distance from the k-th passive device to the center of the PS antenna array. d represents the phase difference between the signals from each antenna of PS to the k-th passive device. k,m This represents the distance from the k-th passive device to the m-th antenna of the PS.

[0108] Non-line-of-sight components It can be represented as:

[0109]

[0110] Where C represents the number of scatterers in the near-field channel. d represents the complex channel gain corresponding to the signal transmitted from PS through the c-th scatterer to the passive device link. c,k This represents the distance from the c-th scatterer to the k-th passive device. d represents the phase difference between each link. c This represents the distance from the c-th scatterer to the center of the PS antenna array.

[0111] The specific steps of step B are as follows:

[0112] During the Wireless Energy Transmission (WET) phase, the energy signal transmitted by the PS in the system can be represented as:

[0113] x k =w k s(3),

[0114] in, Let represent the transmit beamforming vector from PS to the k-th passive device, and s represent the transmitted signal. Based on this, the received energy signal received by the k-th passive device can be expressed as:

[0115]

[0116] in, This represents the near-field channel state information vector from the PS to the k-th passive device, where n is the vector. k This represents the Gaussian white noise at the k-th passive device, which has zero mean and variance σ. 2 Gaussian distribution and denoted as In the formula (·) H This indicates the conjugate transpose.

[0117] According to the expression for the received signal of the passive device, the energy collected by the k-th passive device is expressed as:

[0118]

[0119] Where η∈(0,1) represents the energy conversion efficiency, and t0 represents the energy harvesting time. The time it takes for the passive device to send information to the base station is denoted as t1. Therefore, the power of the k-th passive device can be expressed as:

[0120]

[0121] The specific steps of step C are as follows:

[0122] During the Wireless Information Transmission (WIT) phase, passive devices use harvested energy to transmit signals to the base station based on the NOMA protocol. The signal received by the base station can be represented as:

[0123]

[0124] in, Let s represent the far-field channel state information vector from the k-th passive device to the base station. k This represents the signal transmitted by the k-th passive device; n represents the Gaussian white noise at the base station, which has zero mean and variance σ. 2 I NThe complex Gaussian distribution is denoted as In the formula 0 N×1 Let I represent the zero vector of N×1 dimensions. N Let g represent an N×N dimensional identity matrix. Here, we consider that base stations are typically deployed at a considerable distance from passive devices, therefore g... k The traditional Ricean channel is used for modeling.

[0125] Assuming the base station uses a linear receive beamforming vector to decode the received signal, and mitigates inter-user interference by first decoding user signals with better channel conditions, the recovered signal can be expressed as:

[0126]

[0127] in, Let represent the received beamforming vector at the base station. Without loss of generality, assume that the user's channel gain satisfies . Where |·| represents modulo. In an uplink NOMA system, a SiC receiver is configured at the base station. Co-channel interference is the sum of the channel gains of other users. The interference experienced by users mainly comes from users with good channel quality; that is, users with poor channel quality are more susceptible to strong co-channel interference. To avoid wasting resources, the system first decodes the signals from users with good channel quality and subtracts the decoded signals from the superimposed signals. Then, it decodes the signals from users with the next best channel quality, and so on, until all signals are decoded.

[0128] Therefore, the signal-to-interference-to-noise ratio of the k-th passive device can be expressed as:

[0129]

[0130] The throughput of the k-th passive device can be expressed as:

[0131] R k =t1log2(1+γ k (10)

[0132] The throughput of the system can be expressed as:

[0133]

[0134] The total energy consumption of the system can be expressed as:

[0135]

[0136] Among them, P PS P represents the static circuit power consumption of PS. BS This indicates the static circuit power consumption of the base station.

[0137] The specific steps of step D are as follows:

[0138] With the goal of maximizing system energy efficiency, the optimization problem is as follows:

[0139]

[0140] stt0≥0, t1≥0,t0+t1≤1 (13a),

[0141]

[0142] ||w B || 2 =1 (13c),

[0143] Among them, P max Ω represents the maximum transmit power of PS, where Ω = {t0, t1, w} B ,w1,...,w K Equation (13a) represents the energy and information transmission time constraints in the system, Equation (13b) represents the PS transmit power constraint, and Equation (13c) represents the normalization constraint of the beamforming vector of the base station received signal.

[0144] The specific steps of step E are as follows:

[0145] Because the objective function is non-convex and the optimization variables are strongly coupled, the original optimization problem is difficult to solve directly. Therefore, to address (P1), this invention uses the Dinkelbach algorithm to process the original optimization problem and decouples it into three sub-problems based on the AO algorithm, which are then solved iteratively.

[0146] First, since the objective function (P1) is a fraction, the variables are coupled and difficult to solve directly. Therefore, the Dinkelbach algorithm is used to handle this fractional optimization problem. By introducing the auxiliary variable λ, the objective function of the optimization problem (P1) can be expressed as:

[0147] R-λE total (14),

[0148] Where the auxiliary variable λ in the t-th iteration (t) It can be represented as:

[0149]

[0150] Therefore, the optimization problem (P1) can be transformed into:

[0151]

[0152] st(13a)-(13c),

[0153] Based on this, the problem (P1-E1) is decoupled into three sub-problems and solved separately. Given the PS transmitted signal beamforming vector {w1,...,w...} K} and the beamforming vector w of the base station received signal B In this case, optimize the PS energy transmission time t0 and the passive device information transmission time t1. The corresponding sub-problem (P2) is constructed as follows:

[0154]

[0155] st(13a),

[0156] It can be proven that the objective function can reach its maximum value when t0 + t1 = 1, and the relationship of equation (18) with respect to t0 can be obtained:

[0157]

[0158] Taking the second derivative of equation (18) with respect to t0, we get:

[0159]

[0160] Since equation (19) is always less than 0, equation (18) is a convex function. Therefore, by using a one-dimensional search method on equation (18) to maximize f(t0), the optimal solution for t0 in subproblem (P2) can be obtained.

[0161] Then, given the PS energy transmission time t0, the passive device information transmission time t1, and the base station received signal beamforming vector w B In the case of optimizing the beamforming vector {w1,...,w} of the PS transmitted signal, K The corresponding sub-problem (P3) is constructed as follows:

[0162]

[0163] st(13b)

[0164] Since the problem (P3) is about {w1,...,w} K The optimization problem of} is to write the objective function as a function of w. k The expression:

[0165]

[0166] Using a low-complexity projection gradient algorithm, equation (21) applies to w k Find the gradient to obtain w k Gradient expression:

[0167]

[0168] Set initial values ​​while satisfying constraint (13b). Then update w according to the iterative formula (23). k :

[0169]

[0170] Where α represents the step size of the projection gradient iteration. When the iteration result w k If it is not in the feasible region, it is reprojected to the boundary of the feasible region according to equation (24):

[0171]

[0172] When the maximum number of iterations or f(w) is reached k When convergence occurs, we can obtain {w1,...,w} in the subproblem (P3). K The optimal solution for}.

[0173] Finally, given the PS energy transmission time t0, the passive device information transmission time t1, and the PS transmission signal beamforming vector {w1,...,w}, K}, optimize the beamforming vector w of the base station received signal B The corresponding sub-problem (P4) is constructed as follows:

[0174]

[0175] st(13c)

[0176] Question (P4) is about w B The optimization problem is such that (P4) can be represented in the following equivalent form:

[0177]

[0178] st(13c)

[0179] Based on the properties of matrix operations, the problem is transformed into:

[0180]

[0181] st(13c)(27a),

[0182] make Based on the properties of Rayleigh entropy, we obtain w B Closed-form solution:

[0183]

[0184] Where ν represents the eigenvector corresponding to the largest eigenvalue of matrix G.

[0185] The subproblems (P2), (P3), and (P4) are optimized alternately using an iterative algorithm until the objective function converges, thus obtaining the optimal solution to the original optimization problem (P1).

[0186] Table 1 shows the detailed steps of the algorithm used.

[0187] Table 1 - Iterative Algorithm of the Proposed Scheme

[0188]

[0189] An embodiment of the present invention provides a simulation experiment to verify the effectiveness of the method of the present invention.

[0190] An example of implementing this invention using MATLAB software simulation is shown in the system model diagram below. Figure 2 In the simulation experiment, the wireless channels were set to be independent of each other, and the channel state information was kept constant. The positions of the PS and the base station were (0,0) and (50,0) respectively, in meters. Two scatterers were randomly placed within a 3-meter range around the PS, and three passive devices were randomly placed within a 9-meter range around the PS. The channel state information g1,...,g between the passive devices and the base station was given. K Obeying Rayleigh fading, bandwidth B = 1MHz, RF signal frequency f = 28GHz, wavelength λ = c / f, where c is the speed of light, antenna array using uniform linear arrangement, antenna element spacing d = 3λ / 2, antenna aperture D = (n-1)d, PS transmit power P max =40dBm, static circuit power consumption P PS =18dBm, energy conversion efficiency η=0.9, number of base station antennas N=32, static circuit power consumption P BS =10dBm, noise power σ 2 = -80dBm.

[0191] Figure 3 This is a graph showing the relationship between system energy efficiency and the number of PS antennas when maximizing energy efficiency in an embodiment of the present invention.

[0192] First, it should be noted that, to ensure the fairness of the comparison schemes, the channel state information h1,...,h between the PS and the passive device under far-field channel conditions is used. K Assuming plane wave propagation, and that the angles between the line connecting the k-th passive device and each antenna element and the positive x-axis are all equal, the array response vector is given by the following formula. Let d represent the phase difference between each antenna and the signal from the k-th passive device, d represent the antenna element spacing, N represent the number of antenna elements, and θ represent the phase difference between each antenna and the signal from the k-th passive device. k This represents the angle between the line connecting the k-th passive device and the origin and the positive x-axis. θ represents the phase difference of the signal from the PS through the c-th scatterer to the passive device link. c This represents the angle between the line connecting the c-th passive device and the origin and the positive x-axis.

[0193] Figure 3 The relationship between system energy efficiency and the number of PS antennas is illustrated in the figure. As can be seen from the figure, the system energy efficiency shows a significant upward trend with the increase in the number of PS antennas. This is due to the greater spatial freedom brought by the increased number of antennas, which enhances the system's energy transmission capability, allowing terminal devices to collect more energy and thus improve their information transmission capability. Furthermore, compared to the OMA scheme, the NOMA scheme further improves system energy efficiency. This performance improvement mainly stems from the inherent advantages of NOMA in its resource utilization mechanism. In the OMA scheme, multiple devices exclusively occupy communication resources in an orthogonal manner. While this avoids interference between users, the utilization rate of spectrum resources is strictly limited, limiting the overall system throughput. In contrast, in the NOMA scheme, multiple devices transmit on the same time-frequency resource block. The receiver uses SIC technology to demodulate and eliminate interference signals sequentially according to the channel conditions from strongest to weakest. This allows devices with poor channel conditions to transmit collaboratively with devices with good channel conditions, achieving efficient sharing of spectrum resources, thereby significantly improving system throughput and thus enhancing system energy efficiency. This further demonstrates the advantages of considering near-field channel conditions and employing NOMA technology in the system.

[0194] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing the energy efficiency of a large-scale MIMO system based on near-field wireless power supply, characterized in that, Includes the following steps: Step 1: Establish near-field spherical wave channel models for the PS-to-each device links and obtain channel state information for the links from each device to the base station; Step two, PS sends energy signals to the devices and calculates the energy collected by each device; Step 3: The device uses the collected energy to transmit information to the base station, calculates the total throughput and total power consumption of the system, and the ratio of the two is the system energy efficiency; Step 4: Construct a system energy efficiency maximization problem by jointly optimizing time resources, PS transmit beamforming vector, and base station receive beamforming vector; Step 5: Decompose the optimization problem into three sub-problems. In the first sub-problem, optimize the allocation of time resources in the system. In the second sub-problem, optimize the PS transmit beamforming vector. In the third sub-problem, optimize the base station receive beamforming vector. Iterate the three sub-problems through an alternating optimization algorithm until convergence, and obtain the optimized system energy efficiency. During the wireless power transfer phase, the energy signal transmitted by the PS in the system can be represented as: （3）， in, This indicates that PS has reached the [number]th [number]. Transmit beamforming vector of a passive device Indicates the transmitted signal; based on this, the first... The received energy signal of a passive device is represented as follows: （4）， in, This indicates that PS has reached the [number]th [number]. Near-field channel state information vector of a passive device Indicates the first The Gaussian white noise at each passive device follows a zero mean and a variance of . Gaussian distribution and denoted as In the formula Indicates conjugate transpose; According to the expression for the received signal of a passive device, the first... The energy collected by a passive device is represented as follows: （5）， in, Indicates energy conversion efficiency. The energy harvesting time is represented by the time when the passive device sends information to the base station. Therefore, the first The power of a passive device is expressed as: （6）； During the wireless information transmission phase, passive devices use the collected energy to send signals to the base station based on the NOMA protocol. The signal received by the base station can be represented as: （7）， in, Indicates the first The far-field channel state information vector from a passive device to the base station. Indicates the first A passive device sends a signal; This represents the Gaussian white noise at the base station, which follows a zero mean and a variance of . The complex Gaussian distribution is denoted as In the formula express The zero vector of dimension, express The identity matrix is ​​of dimension 1. Here, we consider that base stations are typically deployed at a considerable distance from passive devices. Modeling is performed using a traditional Ricean channel; Assuming the base station uses a linear receive beamforming vector to decode the received signal, it mitigates inter-user interference by first decoding user signals with better channel conditions. The recovered signal is represented as follows: （8）， in, This represents the received beamforming vector at the base station. Without loss of generality, it is assumed that the user's channel gain satisfies... ,in Indicates modulo; Therefore, the first The signal-to-interference-to-noise ratio of a passive device is expressed as: （9）， No. The throughput of a passive device is expressed as: （10） The total throughput of the system is expressed as: （11）， The total energy consumption of the system is expressed as: （12）， in, This indicates the static circuit power consumption of the PS. This indicates the static circuit power consumption of the base station.

2. The energy efficiency optimization method for a large-scale MIMO system based on near-field wireless power supply according to claim 1, characterized in that, Because the passive device is located in the near-field region of the PS, the signal propagation distance is short and there are no obvious obstacles in the path, therefore the signal from the PS to the... Near-field channel state information vector of a passive device There is a stable line-of-sight propagation component. Meanwhile, scatterers in the environment can cause reflection and refraction of electromagnetic waves, even under near-field conditions. There is also a non-line-of-sight propagation component. Therefore, the near-field channel model from the PS to the passive device is the sum of the line-of-sight components and the non-line-of-sight components, that is: View distance component Represented as: （1）， in, Indicates the complex channel gain. This indicates the frequency of the PS signal transmission. Indicates the first Distance from each passive device to the center of the PS antenna array, array response vector: Indicates the distance from each PS antenna to the [missing information]. The phase difference of the signals from passive devices. Indicates the first A passive device to PS The distance between the antennas, Non-line-of-sight components Represented as: （2）， in, This indicates the number of scatterers in the near-field propagation environment. This indicates that the signal is emitted from PS and passes through the first... The scatterer to the first The complex channel gain corresponding to each passive device link Indicates the first The scatterer to the first The distance between passive devices This indicates the phase difference between each link. Indicates the first The distance from each scatterer to the center of the PS antenna array.

3. The energy efficiency optimization method for a large-scale MIMO system based on near-field wireless power supply according to claim 1, characterized in that, With the goal of maximizing system energy efficiency, the optimization problem is as follows: （P1） （13）， （13a）， （13b）， （13c）， in, This indicates the maximum transmit power of the PS. Equation (13a) represents the energy and information transmission time constraints in the system, Equation (13b) represents the PS transmit power constraint, and Equation (13c) represents the normalization constraint of the base station received signal beamforming vector.

4. The energy efficiency optimization method for a large-scale MIMO system based on near-field wireless power supply according to claim 3, characterized in that, Based on the optimization problem, the solution process is as follows: First, since the objective function (P1) is a fraction, the variables are coupled and difficult to solve directly. Therefore, the Dinkelbach algorithm is used to handle this fractional optimization problem by introducing auxiliary variables. The objective function of the optimization problem (P1) is expressed as: （14）， Among them, the first Auxiliary variables for the next iteration Represented as: （15）， Therefore, the optimization problem (P1) is transformed into: (P1-E1) (16) (13a)-(13c) Based on this, the problem (P1-E1) is decoupled into three sub-problems and solved separately, given the beamforming vector of the PS transmitted signal. and base station received signal beamforming vector In this case, optimize PS power transfer time Information transmission time with passive devices The corresponding sub-problem (P2) is constructed as follows: （P2） （17）， （13a）， Since the objective function is convex, a one-dimensional search of the objective function yields the subproblem (P2). and The optimal solution; Then, at a given PS energy transfer time Passive device information transmission time and base station received signal beamforming vector In the case of optimizing the beamforming vector of the PS transmitted signal The corresponding sub-problems (P3) are constructed as follows: （P3） （18）， （13b）， Using a low-complexity projective gradient algorithm, this subproblem is transformed into a problem about... The gradient iteration problem yields the optimized PS transmit signal beamforming vector in subproblem (P3). ; Finally, given the PS energy transfer time Passive device communication time and PS transmitted signal beamforming vector Optimize the beamforming vector of the base station received signal The corresponding sub-problem (P4) is constructed as follows: （P4） （19）， （13c）， By using the properties of Rayleigh entropy to solve this problem, the optimized beamforming vector of the base station received signal in subproblem (P4) is obtained. The closed-form solution; The subproblems (P2), (P3), and (P4) are optimized alternately using an iterative algorithm until the objective function converges, thus obtaining the optimal solution to the original optimization problem (P1). .

5. An energy efficiency optimization system for a large-scale MIMO system based on near-field wireless power supply, based on the method of any one of claims 1 to 4, characterized in that, The system comprises an energy station PS equipped with a large-scale antenna array, multiple passive devices located in the near-field region of the energy radiation from the energy station PS, and a base station equipped with a large-scale receiving antenna array. The energy station PS is equipped with... Each passive device is equipped with a single antenna, and the base station is equipped with... The antenna, based on the first-collect-then-transmit protocol, first collects energy from the transmitted signal of the power station PS, and then uses the collected energy to transmit information to the base station through a non-orthogonal multiple access protocol. The base station demodulates the signal using continuous interference cancellation.

Images