Multi-user communication energy efficiency improvement method based on reconstructed intelligent surface

Abstract

The invention provides a multi-user communication energy efficiency improvement method based on a reconstructed intelligent surface, and relates to the technical field of communication. The multi-user communication energy efficiency improvement method based on the reconstructed intelligent surface comprises the following steps: S1, initializing a system; s2, initializing system parameters, an RIS reflection matrix, a base station beam forming matrix and a power distribution factor; s3, constructing an SWIPT system model assisted by the RIS; s4, introducing a target function and a constraint condition; s5, introducing an auxiliary variable to decompose the problem into three layers; s6, the RIS reflection phase is fixed, and a Dinke1bach method and a water filling algorithm are used for optimization; and S7, optimizing the phase matrix of the RIS through an SDR technology. According to the new RIS energy efficiency optimization method provided by the invention, the actual power consumption characteristic of the RIS element can be described more accurately, and the power consumption difference of the RIS element in different working states is considered, so that the accuracy and the actual effect of energy efficiency optimization are improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method for improving energy efficiency of multi-user communication based on reconstructed smart surfaces. Background Art

[0002] In current wireless communication systems, energy efficiency is an increasingly severe challenge. Especially in multi-user communication systems, how to effectively allocate energy and optimize system performance has always been the focus of research. With the rapid development of wireless communication technology, the demand for higher data transmission rates and higher quality signals is increasing, however, this also brings huge energy consumption. In order to achieve green and sustainable wireless communications, improving energy efficiency has become one of the core research goals.

[0003] In the prior art, simplified power consumption models are usually used when designing RIS systems. In these models, the power consumption of RIS components is assumed to be uniformly distributed, or the actual power consumption differences of RIS components in different working states (such as on and off) are ignored. This assumption cannot reflect the working characteristics of real hardware, because in actual applications, there are obvious differences between the switching states of RIS components and the power consumption of each component. For example, when some components are in the off state, their power consumption is significantly lower than when they are in the on state. Simplified models may lead to overestimation of energy efficiency, reducing the energy efficiency and performance of the actual system.

[0004] In addition, although RIS technology can improve signal quality and coverage, many methods ignore the energy consumption changes of RIS components during dynamic operation. For example, the switching state of RIS is not only affected by external signal changes, but also by power management strategies, environmental conditions and other factors. Existing RIS energy efficiency optimization methods do not fully consider the dynamic power consumption under different working states in the RIS system, resulting in the inability to achieve optimal energy efficiency in complex communication scenarios, thus affecting the overall performance and stability of the system. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the deficiencies in the prior art, the present invention provides a method for improving the energy efficiency of multi-user communications based on reconstructing smart surfaces, which solves the defects and deficiencies in the prior art.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for improving the energy efficiency of multi-user communication based on reconstructed smart surface, the method comprising the following steps:

[0009] S1, system initialization;

[0010] S2, initialize system parameters, RIS reflection matrix, base station beamforming matrix, power allocation factor;

[0011] S3, build RIS-assisted SWIPT system model;

[0012] S4, introduce the objective function and constraints;

[0013] S5. Introduce auxiliary variables to decompose the problem into three levels;

[0014] S6, fixed RIS reflection phase, optimized using Dinke1bach method and water filling algorithm;

[0015] S7, optimize the phase matrix of RIS through SDR technology;

[0016] S8, updating power allocation and RIS configuration by alternating optimization;

[0017] S9, judging whether the energy constraint is satisfied;

[0018] S10, if yes, generate the final RIS configuration scheme that maximizes energy efficiency, if no, return to step 6;

[0019] S11. End.

[0020] Preferably, the method studies a multi-user RIS-assisted MISO system with SWIPT, which consists of a BS equipped with a transmitting antenna and a single-antenna user. In particular, assuming that the direct BS-user link is blocked, in order to ensure the signal coverage and the communication effect of the user, a RIS is deployed to improve the communication from the base station to the user, including a reflection unit, wherein are the number of RIS units in the horizontal direction and the vertical direction respectively. Each element is controlled by binary, that is, the phase shift matrix of the RIS is given by the following formula;

[0021]

[0022] The user's received signal is represented as;

[0023] y=H H ΘGWs+n,;

[0024] Assuming that each user adopting the PS scheme consists of an energy harvesting (EH) circuit and a conventional information decoding (ID) circuit, then, the signal sent by the BS is divided into two streams: EH and ID. For the kth user, the spectral efficiency (SE) is written as;

[0025]

[0026] Furthermore, the energy harvested by the user is written as;

[0027]

[0028] The total power consumption of the system is calculated as follows;

[0029] P total =P s +P RIS +P t -E,;

[0030] Where E represents the total energy obtained by the user; P s represents the total static power consumption of the base station, RIS and users; P t Represents the base station transmit power and also includes a binary controlled RIS, where each RIS element has different power consumption in the on and off states, namely:

[0031] P RIS =||θ||0P0,θ n ∈{0,π},;

[0032] Where P0 is the power dissipated in each RIS element when a bias current is applied to turn on the corresponding PIN diode. In summary, the spectral efficiency (SE) and energy efficiency (EE) are written as;

[0033]

[0034] The total system spectral efficiency is

[0035] Preferably, the method aims to obtain the maximum EE by jointly optimizing the digital beamforming matrix and the RIS analog beamforming vector, which is expressed as:

[0036]

[0037] sttr(W H W)≤P max ,

[0038]

[0039] Preferably, the method develops an AO algorithm to solve the complex EE maximization problem by decoupling the problem into three sub-problems and then solving them alternately. In order to eliminate the interference of different channel signals, a zero-forcing digital precoder is used here to obtain a feasible solution to the precoder sub-problem, that is,

[0040]

[0041] Using the ZF precoder, the signal received by the user is rewritten as;

[0042]

[0043] Based on this, the problem is reformulated as;

[0044]

[0045]

[0046] in as well as Indicates that each user guarantees the basic communication spectrum efficiency SE min Minimum received power required.

[0047] Preferably, the method transforms the original optimization problem into a new simpler problem, and uses an AO algorithm to alternately update and solve it. According to the AO algorithm of the problem, the next step is to optimize the power allocation matrix when the RIS phase matrix is ​​fixed. The power allocation problem is given by the following formula:

[0048]

[0049] in

[0050] Since the new objective function is still non-concave, the problem is also a challenging one. A feasible method is to introduce slack variables to convert the fraction into a polynomial. Therefore, after iterative transformation using the Dinkelbach method, the following objective function can be obtained;

[0051]

[0052] in In this way, the original problem is transformed into a convex problem form, and then it is solved by class filling hydrolysis analysis.

[0053] Preferably, the method further comprises an algorithm for RIS simulation beamforming. According to the AO algorithm, the next step is to optimize using a fixed p and perform the following transformation: q = e jθ , due to θ n ∈{0,π}, we get;

[0054] q n ∈{-1, 1} and Then the original problem is stated as follows;

[0055]

[0056] As mentioned before, the objective function is not convex, but the discrete constraints make P2 difficult to solve. In general, the limitation of integer constraints makes this an NP-hard problem, and it is almost impossible to obtain the optimal solution in low computational time. Therefore, the next goal is to obtain an approximate solution in polynomial time; replace q with a quadratic variable, which is defined as Right now:

[0057]

[0058] The question is thus restated as;

[0059]

[0060] tr(E i，i X)=1,i∈{1,2,…,N+1}

[0061] X ≥ 0,

[0062] rank(X)=1,

[0063] where defined;

[0064]

[0065] Specifically:

[0066]

[0067] Next, using the SDR method, the problem is formulated as a standard convex SDP problem, which is solved by a CVX solver using a general procedure such as the interior point method.

[0068] Preferably, the method will apply the Dinkelbach method to the problem by fixing the transmission beamforming vector and the reflection beamforming vector, and then the PS ratio optimization problem is expressed in the following form:

[0069]

[0070] in Represents the total signal rate of the user, represents the energy efficiency of the kth user after the Dinkelbach method transformation.

[0071] For fixed parameters, the objective function is proved to be strictly concave and is solved by calculation, which means that the PS ratio of each user is independent of each other. Therefore, the problem is divided into independent sub-problems, and according to the constraints, ρ should be limited. k , in, Ensure that the minimum data rate constraint for each user is met, Ensure that the minimum harvested energy constraint of each user is met. In addition, In ρ k is strictly concave, so the solution is We get a unique root. In summary, the optimal PS ratio of the kth user is:

[0072]

[0073] Therefore, the optimal solution to the problem is obtained by solving sub-problems that do not interfere with each other separately.

[0074] (III) Beneficial effects

[0075] The present invention provides a method for improving energy efficiency of multi-user communication based on reconstructed smart surface. It has the following beneficial effects:

[0076] 1. The present invention provides a new RIS energy efficiency optimization method, which can more accurately describe the actual power consumption characteristics of RIS components, taking into account the power consumption differences of RIS components under different working states, thereby improving the accuracy and actual effect of energy efficiency optimization.

[0077] 2. The present invention introduces dynamic power consumption modeling, takes into account the power consumption changes of RIS components under different working states, and can effectively adjust the working mode of RIS to adapt to the changing communication environment, thereby improving the overall energy efficiency and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the system of the present invention;

[0079] Figure 2 is the first curve graph of the present invention;

[0080] Figure 3 is the second curve graph of the present invention;

[0081] Figure 4 is the third curve diagram of the present invention;

[0082] Figure 5 is the fourth curve diagram of the present invention;

[0083] Figure 6 is the fifth curve diagram of the present invention;

[0084] Figure 7 is the sixth curve diagram of the present invention;

[0085] Figure 8 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0086] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] Example:

[0088] like Figure 1-8 As shown, an embodiment of the present invention provides a method for improving the energy efficiency of multi-user communication based on reconstructed smart surface. In this method, EE (energy efficiency) is used as a communication indicator. If the scale effect of power difference is not considered when using it, the power consumption can easily rise to an unacceptable value, resulting in the failure of the EE optimization algorithm. Therefore, an accurate RIS power consumption model is crucial for the EE optimization of the SWIPT-RIS assisted communication system. In order to accurately describe and evaluate the EE of the RIS assisted communication system, the present invention establishes a more realistic power consumption model, fully considers the user's PS protocol, and proposes an alternating optimization (AO) algorithm with EE maximization as the objective function.

[0089] (1) Figure 1 As shown, the present invention studies a multi-user RIS-assisted MISO system with SWIPT, which consists of a BS equipped with a transmitting antenna and a single-antenna user. In particular, assuming that the direct BS-user link is blocked, in order to ensure the signal coverage and the communication effect of the user, a RIS is deployed to improve the communication from the base station to the user, including reflection units, where are the number of RIS units in the horizontal direction and the vertical direction respectively. Each element is controlled by binary, that is, the phase shift matrix of RIS is given by the following formula;

[0090]

[0091] The user's received signal can be expressed as;

[0092] y=H H ΘGWs+n,;

[0093] Assuming that each user adopting the PS scheme consists of an energy harvesting (EH) circuit and a conventional information decoding (ID) circuit, then, the signal sent by the BS is divided into two streams: EH and ID. For the kth user, the spectral efficiency (SE) is written as;

[0094]

[0095] Furthermore, the energy harvested by the user is written as;

[0096]

[0097] The total power consumption of the system is calculated as follows;

[0098] P total =P s +P RIS +P t -E,;

[0099] Where E represents the total energy obtained by the user; P s represents the total static power consumption of the base station, RIS and users; P t Represents the base station transmission power. In most previous studies, it is assumed that the power consumption of RIS components is a fixed value, which is inconsistent with the actual situation discussed above. Therefore, the present invention considers a binary controlled RIS, where each RIS component has different power consumption in the on and off states, namely:

[0100] P RIS =||θ||0P0,θ n ∈{0,π},;

[0101] Where P0 is the power dissipated by each RIS element when a bias current is applied to turn on the corresponding PIN diode. In summary, the spectral efficiency (SE) and energy efficiency (EE) are written as;

[0102]

[0103] The total system spectral efficiency is

[0104] The goal of the present invention is to obtain the maximum EE by jointly optimizing the digital beamforming matrix and the RIS analog beamforming vector, which is expressed as;

[0105]

[0106] sttr( W HW)≤P max ,

[0107]

[0108] (2) The present invention develops an AO algorithm to solve the complex EE maximization problem by decoupling the problem into three sub-problems and then solving them alternately. In order to eliminate the interference of different channel signals, a zero-forcing (ZF) digital precoder is used here to obtain a feasible solution to the precoder sub-problem, that is,

[0109]

[0110] Using the ZF precoder, the signal received by the user is rewritten as;

[0111]

[0112] Based on this, the question is reformulated as;

[0113]

[0114] in as well as Indicates that each user guarantees the basic communication spectrum efficiency SE min Minimum received power required.

[0115] The above discussion transforms the original optimization problem into a new and simpler problem. The present invention uses an AO algorithm to alternately update the solution.

[0116] (3) According to the AO algorithm for this problem, the next step is to optimize the power allocation matrix when the RIS phase matrix is ​​fixed. The power allocation problem is given by the following equation:

[0117]

[0118] in

[0119] Since the new objective function is still non-concave, the problem is also a challenging one. A feasible method is to introduce slack variables to convert the fraction into a polynomial. Therefore, after iterative transformation using the Dinkelbach method, the following objective function can be obtained;

[0120]

[0121] in In this way, the original problem is transformed into a convex problem form, and then it is solved by class filling hydrolysis analysis.

[0122] (4) The present invention also includes an algorithm for RIS simulation beamforming. According to the AO algorithm, the next step is to use a fixed p for optimization and perform the following transformation: q = e jθ , due to θ n ∈{0,π}, we get;

[0123] Then the original problem is stated as follows;

[0124]

[0125] As mentioned before, the objective function is not convex, but the discrete constraints make P2 difficult to solve. In general, the limitation of integer constraints makes this an NP-hard problem, and it is almost impossible to obtain the optimal solution in low computational time. Therefore, the next goal is to obtain an approximate solution in polynomial time.

[0126] In this part, the present invention replaces q with a quadratic variable, which is defined as Right now:

[0127]

[0128] The question is thus restated as;

[0129]

[0130] tr(E i，i X)=1,i∈{1,2,…,N+1}

[0131] X ≥ 0,

[0132] rank(X)=1,

[0133] where defined;

[0134]

[0135] Specifically:

[0136]

[0137] Next, using the SDR method, the problem is formulated as a standard convex SDP problem, which is solved by a CVX solver using a general procedure such as the interior point method.

[0138] (5) In the present invention, the Dinkelbach method is applied to the problem by fixing the transmission beamforming vector and the reflection beamforming vector, and then the PS ratio optimization problem is expressed in the following form:

[0139]

[0140] in Represents the total signal rate of the user, represents the energy efficiency of the kth user after the Dinkelbach method transformation.

[0141] For fixed parameters, the objective function is proved to be strictly concave and solved by calculation, which means that the PS ratio of each user is independent of each other. Therefore, the problem is divided into independent sub-problems, and according to the constraints, ρ should be limited. k , in, Ensure that the minimum data rate constraint for each user is met, Ensure that the minimum harvested energy constraint of each user is met. In addition, In ρ k is strictly concave, so the solution is We get a unique root. In summary, the optimal PS ratio of the kth user is:

[0142]

[0143] Therefore, the present invention can obtain the optimal solution to the problem by separately solving sub-problems that do not interfere with each other.

[0144] (6) In this paper, simulation results are given to verify the effectiveness of the proposed algorithm in the SWIPT-assisted MU-MISO communication environment, aiming to maximize energy efficiency. The main focus is on the variation of EE optimized by the proposed algorithm with transmit power in the Swift-assisted transmission scenario. The SE and EE curves are shown in Figure 2 and Figure 3 The proposed algorithm gives the red curve “optimized”. The black curve “Random RIS” and the green curve “all-off RIS” are the baselines, which are random RIS simulated beamforming and RIS beamforming given as a unitary matrix (i.e. all elements of RIS are configured to be off). Figure 2 and Figure 3 It can be seen that our algorithm outperforms both baselines and existing methods in both indicators SE and EE. It is worth noting that the baseline “All-OFF RIS” outperforms the baseline “Random RIS” in terms of SE and EE.

[0145] One of the reasons is that when all RIS elements are set to off, they consume very little power, which corresponds to a higher EE. For SE, an intuitive explanation is that the phase shift matrix of “All-OFF RIS” can be modeled as an IN×N, identity matrix, while the phase shift matrix of “Random RIS” is represented as a matrix with randomly distributed elements in the complex plane within a unit period. Compared with the identity matrix, the beamforming performance of the randomly distributed matrix is ​​poor, which leads to the difference in SE and EE curves.

[0146] exist Figure 4 and Figure 5 In Figure 1, SE and EE curves are plotted as a function of the number of RIS elements N. Different curves are generated in the same figure using the proposed algorithm, existing methods, and baselines. Figure 4 and Figure 5 It can be seen that both the SE and EE of the communication system increase with time. This is mainly due to the large beamforming gain of RIS. In addition, in the benchmark test, the optimization performance of this method is better than the existing methods.

[0147] like Figure 6 and Figure 7As shown in the figure, the spectral efficiency and energy efficiency have different changing trends under different SEmin values. Specifically, the red SDR scheme shows opposite trends in the two figures, showing different optimization effects. Figure 6 It can be seen that the spectral efficiency (SE) of the proposed algorithm and the existing methods are maintained at a high level and fluctuate less with SEmin. In contrast, the spectral efficiency of All-OFF and Random RIS is significantly lower. This is because these methods of the former are optimization algorithms that can effectively optimize the spectral efficiency and therefore have similar and higher performance; while these methods of the latter have not been optimized, so the spectral efficiency is lower and the change of SEmin has less impact on them. However, with the change of the minimum SE, the indicators of the proposed algorithm are lower than those of the existing algorithms. This may be because the SDR algorithm relaxes the original non-convex problem into a convex problem to solve it. Although it guarantees that the solution is close to the global optimal solution, it may be limited by computing resources and algorithms, especially the complexity in high-dimensional cases. In this case, the SDR algorithm may not be able to fully converge within a limited time, resulting in performance that is inferior to more direct and faster algorithms; another reason is that the SDR algorithm has higher requirements for spectral efficiency. This poses a higher challenge to the method, especially when the system has limited degrees of freedom. From Figure 7 It can be seen that with the increase of SEmin, the EE of SDR is generally better than the existing methods and remains stable at a high level; while the baseline method has a lower EE in comparison.

[0148] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprising a reference structure" do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0149] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for improving energy efficiency of multi-user communication based on reconstructed smart surface, characterized in that: The method comprises the following steps: S1, system initialization; S2, initialize system parameters, RIS reflection matrix, base station beamforming matrix, power allocation factor; S3, build RIS-assisted SWIPT system model; S4, introduce the objective function and constraints; S5. Introduce auxiliary variables to decompose the problem into three levels; S6, fixed RIS reflection phase, optimized using Dinke1bach method and water filling algorithm; S7, optimize the phase matrix of RIS through SDR technology; S8, updating power allocation and RIS configuration by alternating optimization; S9, judging whether the energy constraint is satisfied; S10, if yes, generate the final RIS configuration scheme that maximizes energy efficiency, if no, return to step 6; S11. End.

2. According to claim 1, a method for improving energy efficiency of multi-user communication based on reconstructed smart surface is characterized by: The method studies a multi-user RIS-assisted MISO system with SWIPT, which consists of a BS equipped with a transmitting antenna and a single-antenna user. In particular, assuming that the direct BS-user link is blocked, in order to ensure the signal coverage and the communication effect of the user, a RIS is deployed to improve the communication from the base station to the user, including reflection units, where are the number of RIS units in the horizontal direction and the vertical direction, respectively. Each element is controlled by binary, that is, the phase shift matrix of the RIS is given by the following formula; The user's received signal is represented as; y=H H ΘGWs+n,; Assuming that each user adopting the PS scheme consists of an energy harvesting (EH) circuit and a conventional information decoding (ID) circuit, then, the signal sent by the BS is divided into two streams: EH and ID. For the kth user, the spectral efficiency (SE) is written as; Furthermore, the energy harvested by the user is written as; The total power consumption of the system is calculated as follows; P total =P s +P RIS +P t -E,; Where E represents the total energy obtained by the user; P s represents the total static power consumption of the base station, RIS and users; P t Represents the base station transmit power and also includes a binary controlled RIS, where each RIS element has different power consumption in the on and off states, namely: P RIS =||θ||0P0,θ n ∈{0,π},; Where P0 is the power dissipated in each RIS element when a bias current is applied to turn on the corresponding PIN diode. In summary, the spectral efficiency (SE) and energy efficiency (EE) are written as; The total system spectral efficiency is 3. The method for improving energy efficiency of multi-user communication based on reconstructed smart surface according to claim 2, characterized in that: The goal of the method is to obtain the maximum EE by jointly optimizing the digital beamforming matrix and the RIS analog beamforming vector, which is expressed as; s.t. tr(W H W)≤P max , 4. The method for improving energy efficiency of multi-user communication based on reconstructed smart surface according to claim 3, characterized in that: The method develops an AO algorithm to solve the complex EE maximization problem by decoupling the problem into three sub-problems and then solving them alternately. In order to eliminate the interference of different channel signals, a zero-forcing digital precoder is used here to obtain a feasible solution to the precoder sub-problem, that is, Using the ZF precoder, the signal received by the user is rewritten as; Based on this, the problem is reformulated as; in as well as Indicates that each user guarantees the basic communication spectrum efficiency SE min Minimum received power required.

5. The method for improving energy efficiency of multi-user communication based on reconstructed smart surface according to claim 4, characterized in that: The method transforms the original optimization problem into a new and simpler problem, and uses an AO algorithm to alternately update and solve it. According to the AO algorithm of this problem, the next step is to optimize the power allocation matrix when the RIS phase matrix is ​​fixed. The power allocation problem is given by the following formula: in Since the new objective function is still non-concave, the problem is also a challenging one. A feasible method is to introduce slack variables to convert the fraction into a polynomial. Therefore, after iterative transformation using the Dinkelbach method, the following objective function can be obtained; in In this way, the original problem is transformed into a convex problem form, and then it is solved by class filling hydrolysis analysis.

6. The method for improving energy efficiency of multi-user communication based on reconstructed smart surface according to claim 5, characterized in that: The method also includes an algorithm for RIS simulation beamforming. According to the AO algorithm, the next step is to use a fixed p for optimization and perform the following transformation: q = e jθ , due to θ n ∈{0,π}, we get; q n ∈{-1,1} and Then the original problem is stated as follows; As mentioned before, the objective function is not convex, but the discrete constraints make P2 difficult to solve. In general, the limitation of integer constraints makes this an NP-hard problem, and it is almost impossible to obtain the optimal solution in low computational time. Therefore, the next goal is to obtain an approximate solution in polynomial time; replace q with a quadratic variable, which is defined as Right now: The question is thus restated as; tr(E i，i X)=1,i∈{1,2,…,N+1}, X≥0. rank(X)=1, where defined; Specifically: Next, using the SDR method, the problem is formulated as a standard convex SDP problem, which is solved by a CVX solver using a general procedure such as the interior point method.

7. The method for improving energy efficiency of multi-user communication based on reconstructed smart surface according to claim 6, characterized in that: The method will apply the Dinkelbach method to the problem by fixing the transmission beamforming vector and the reflection beamforming vector, and then the PS ratio optimization problem is expressed in the following form; in Represents the total signal rate of the user, represents the energy efficiency of the kth user after the Dinkelbach method transformation. For fixed parameters, the objective function is proved to be strictly concave and is solved by calculation, which means that the PS ratio of each user is independent of each other. Therefore, the problem is divided into independent sub-problems, and according to the constraints, ρ should be limited. k , in, Ensure that the minimum data rate constraint for each user is met, Ensure that the minimum harvested energy constraint of each user is met. In addition, In ρ k is strictly concave, so the solution is We get a unique root. In summary, the optimal PS ratio of the kth user is: Therefore, the optimal solution to the problem is obtained by solving sub-problems that do not interfere with each other separately.

Images