An active ris-assisted cell-free mmimo system energy efficiency optimization method

Abstract

The application discloses an active RIS-assisted cell-free mMIMO system energy efficiency optimization method, first, the downlink transmission data and system total power consumption model of the active RIS-assisted cell-free mMIMO system is established, and the downlink rate and energy efficiency expression is given; then, by formulating a multi-objective function optimization problem, including maximizing the downlink rate and minimizing the total system power consumption, a communication node downlink transmission power allocation scheme is designed; in order to solve the non-convex optimization problem, the weighted sum method and the arithmetic average inequality are used to convert it into a generalized geometric programming problem, and then an iterative method is used for solving; finally, the obtained optimal solution is used as the final power allocation scheme and the system energy efficiency is calculated, the application reduces the total system power consumption while improving the downlink rate and, improves the system energy efficiency, and reduces the system operation cost.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to an energy efficiency optimization method for an active RIS-assisted cellless mMIMO system. Background Technology

[0002] As a novel network architecture, cellless massive multiple-input multiple-output (mMIMO) systems no longer divide cellular ranges around large base stations. Instead, they randomly deploy a large number of small communication nodes around users, eliminating cell boundary effects and providing uniform service quality. However, due to obstacles such as large buildings and trees, the signals received by users still experience severe fading, leading to service quality degradation or even communication interruptions. To address this issue, a low-cost and flexible reconfigurable intelligent surface (RIS) has been proposed and has quickly attracted widespread research. Generally, RIS contains a large number of reflective elements with only phase-shifting circuits. Each element can independently apply a phase to the incident signal and reflect it, thus reshaping the wireless propagation channel at the electromagnetic level. Because it does not integrate complex RF components, RIS has extremely low cost, allowing for flexible deployment in designated locations to provide additional communication links. Numerous related studies have also shown that deploying RIS significantly improves the downlink rate of cellless mMIMO systems, expands communication coverage, and further ensures user fairness.

[0003] In contrast, active RIS integrates an active load for each component on top of the RIS, amplifying the signal while reflecting it and enhancing its strength. This increases the transmission rate, but also increases system power consumption. Furthermore, active RIS is constrained by amplification power budgets, making traditional optimization methods based on RIS-free power constraints unusable, thus creating an urgent need for new optimization methods. Summary of the Invention

[0004] The purpose of this invention is to provide an energy efficiency optimization method for cellless mMIMO systems with active RIS (Resonant RIS), addressing the problem of reduced system energy efficiency due to the additional power consumption caused by the use of active RIS. Considering the amplification power constraint of the active RIS, a downlink transmission power allocation scheme is designed to save signal transmission power consumption and improve system energy efficiency.

[0005] To achieve the above objectives, the present invention provides the following solution: an energy efficiency optimization method for an active RIS-assisted cellless mMIMO system, the method comprising the following steps.

[0006] Step 1: Establish an active RIS cellless mMIMO system model, specifically including a downlink data transmission model and a total system power consumption model;

[0007] Step 2: Based on the system model, give the expressions for downlink rate and energy efficiency;

[0008] Step 3: Using the transmission power control coefficient of the communication node as the optimization variable and the power budget of the communication node and the active RIS as constraints, formulate a multi-objective optimization problem with the objectives of maximizing the downlink rate and minimizing the total power consumption of the system.

[0009] Step 4: Use the weighted sum method and arithmetic mean inequality to process the objective function and polynomial constraints, transform the original optimization problem into a generalized geometric programming problem and solve iteratively to complete the downlink power allocation.

[0010] Furthermore, in the aforementioned active RIS-assisted cellless mMIMO system, the aggregation channel between the communication node and the user comprises two parts: a direct channel and a reflection channel, the mathematical expression of which is:

[0011]

[0012] in, Indicates the first The communication node and the first Aggregation channel between users Indicates the first The communication node and the first A sequence of direct-connect channels between users. Indicates active RIS and the first Channel sequences between users This represents the reflection coefficient matrix of an active RIS. Indicates active RIS and the first Channel matrix between communication nodes. This indicates the conjugate transpose operation.

[0013] Furthermore, in the downlink data transmission model, part of the signal transmitted from the communication node is transmitted through a direct channel, and part is amplified by RIS and reflected to the user. Therefore, the... Signal received by a user for

[0014]

[0015] in, Indicates the number of communication nodes. Indicates the number of users. This represents the normalized signal-to-noise ratio used for downlink symbol transmission. Indicates the first The communication node and the first Transmission power control coefficient between users Indicates the first The communication node and the first A sequence of direct-connect channels between users. Indicates the first The communication node and the first Beam sequence between users Indicates the first The data symbols of each user satisfy the following: , Indicates active RIS and the first Channel sequences between users This represents the reflection coefficient matrix of an active RIS. Indicates active RIS and the first Channel matrix between communication nodes This represents the thermal noise generated by an active RIS, where each element follows a distribution. , Indicates the first The additive Gaussian noise received by each user follows a distribution. , This indicates the absolute value operation. This represents the conjugate transpose operation. This represents the expectation operation. This represents a cyclically symmetric complex Gaussian distribution.

[0016] Furthermore, based on the data received from the user, the first Downlink speed per user Calculated as

[0017]

[0018] in, Indicates the first The communication node and the first Transmission power control coefficient between users Indicates the first The communication node and the first Beam sequence between users Indicates the first Signal-to-interference-plus-noise ratio per user.

[0019] Furthermore, based on the data expression received from the user, the first... Power consumed by each communication node when transmitting data Power consumed by active RIS signal amplification They are respectively

[0020]

[0021] in, This indicates the number of reflective elements in an active RISC. This represents Euclidean norm operations.

[0022] Furthermore, the total power consumed by an active RIS-assisted cellless mMIMO system during the downlink data transmission phase The mathematical model is

[0023]

[0024] in, Indicates the first Power amplifier efficiency of each communication node This indicates the power amplifier efficiency of an active RISC system. Indicates the first The power consumed by each communication node to transmit data. This indicates the power consumed by the active RIS amplification signal. This represents the power consumed by the radio frequency circuitry of each communication node. This represents the power consumed by the control circuitry of each active RISC element. This indicates the DC circuit power consumption of each active RIS element.

[0025] Furthermore, system energy efficiency Defined as the ratio of downlink rate to total system power consumption, its expression is:

[0026]

[0027] in, Indicates the first Download rate per user, Indicates the first Signal-to-interference-plus-noise ratio per user.

[0028] Furthermore, to improve system energy efficiency, the multi-objective optimization problem concerning maximizing downlink speed and minimizing total system power consumption can be expressed as follows:

[0029]

[0030] in, , , and Indicates intermediate variables. This represents a matrix consisting of downlink transmission power control coefficients between all communication nodes and all users. This represents the maximum power budget that the communication node can use for downlink data transmission. This represents the maximum power budget for an active RIS to amplify signals.

[0031] Furthermore, in order to represent the original problem as a generalized geometric programming problem, we first use a weighted sum method to represent the multi-objective function form as a single-objective function form by introducing weighting factors. The following single objective function optimization problem can be obtained.

[0032]

[0033] Furthermore, auxiliary variables are introduced. and The single-objective optimization problem is transformed into a generalized geometric programming problem using the arithmetic mean inequality and the first-order Taylor expansion inequality. Its specific form is as follows:

[0034]

[0035]

[0036] in,

[0037] , Indicates intermediate variables;

[0038] , Indicates intermediate variables. Indicates the first Data received by each user This represents a given initial value. is the given initial downlink power control coefficient.

[0039] Furthermore, the standard geometric programming problem is solved iteratively using MATLAB and CVX optimization tools to obtain the optimal downlink power allocation scheme. Calculate the system's energy efficiency.

[0040] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned energy efficiency optimization method for an active RIS-assisted cellless mMIMO system.

[0041] A computer-readable storage medium storing computer instructions thereon, which, when executed by a processor, implement the aforementioned energy efficiency optimization method for an active RIS-assisted cellless mMIMO system.

[0042] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0043] (1) This invention is the first to adopt a multi-objective optimization criterion in the system, that is, to improve the system rate while reducing the system power consumption. This not only solves the problem of excessive system power consumption caused by the use of active RIS in cellless mMIMO system, but also solves the contradiction between transmission rate and energy efficiency optimization, and achieves simultaneous improvement of system transmission rate and energy efficiency.

[0044] (2) This invention provides a general power optimization algorithm framework, which is applicable to the energy efficiency improvement of active RIS-assisted cellless mMIMO systems using arbitrary downlink beam methods. In addition, it can be extended to passive RIS scenarios, so this invention has a certain degree of universality.

[0045] (3) In the optimization process of this invention, only the channel statistics information that changes slowly over time, namely the channel mean and variance, are used. Thus, the result obtained by performing one optimization can be used for several consecutive coherent time intervals. Compared with the existing work that considers perfect channel information, it does not need to perform optimization measures frequently, which reduces the data processing process and reduces the system's computational pressure. Therefore, this invention has a certain degree of feasibility. Attached Figure Description

[0046] Figure 1 A graph showing the relationship between the rate and the number of active RIS reflective elements under different schemes;

[0047] Figure 2 The graph shows the relationship between energy efficiency and the number of active RIS reflective elements under different schemes. Detailed Implementation

[0048] The technical means and effects of the present invention will be further described and explained below with reference to the accompanying drawings to facilitate understanding of the invention. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0049] Example 1

[0050] This embodiment provides an energy efficiency optimization method for an active RIS-assisted cellless mMIMO system. Its purpose is to design a downlink transmission power allocation scheme for communication nodes based on multi-objective optimization criteria, thereby improving system energy efficiency. The method mainly includes the following steps:

[0051] (1) Establish an active RIS cellless mMIMO system model, specifically including a downlink data transmission model and a total system power consumption model;

[0052] (2) Based on the system model, give the expressions for downlink rate and energy efficiency;

[0053] (3) Using the transmission power control coefficient of the communication node as the optimization variable and the power budget of the communication node and the active RIS as constraints, formulate a multi-objective function optimization problem with the objective functions of maximizing the downlink rate and minimizing the total power consumption of the system;

[0054] (4) Using the weighted sum method and the arithmetic mean inequality to process the objective function and polynomial constraints, the original optimization problem is transformed into a generalized geometric programming problem and solved iteratively to complete the downlink power allocation.

[0055] Furthermore, the present invention considers an active RIS-assisted cellless mMIMO system, which includes... Multiple antenna communication nodes and A single-antenna user, active RIS includes A magnifying reflective element.

[0056] Furthermore, in the system, the first The communication node and the first Aggregation channel between users It consists of two parts: the communication node-user direct connection channel and the communication node-RIS-user reflection channel, mathematically represented as:

[0057]

[0058] in, Indicates the first The communication node and the first Aggregation channel between users Indicates the first The communication node and the first A sequence of direct-connect channels between users. Indicates active RIS and the first Channel sequences between users This represents the reflection coefficient matrix of an active RIS. Indicates active RIS and the first Channel matrix between communication nodes. This indicates the conjugate transpose operation.

[0059] Furthermore, in the downlink data transmission model, the signal transmitted from the communication node is amplified by an active RIS amplification channel and then sent to the user. Therefore, the first... Signal received by a user The expression is

[0060]

[0061] in, This represents the normalized signal-to-noise ratio used for downlink symbol transmission. Indicates the first The communication node and the first Transmission power control coefficient between users Indicates the first The communication node and the first A sequence of direct-connect channels between users. Indicates the first The communication node and the first Beam sequence between users Indicates active RIS and the first Channel sequences between users Indicates the first The communication node and the first Aggregation channel between users Indicates that the communication node sends to the first The data symbols of each user satisfy the following: , This represents a thermal noise sequence generated by an active RIS, whose elements follow a distribution. , Indicates the first The additive Gaussian noise received by each user follows a distribution. , This indicates the absolute value operation. This represents the expectation operation. This represents a cyclically symmetric complex Gaussian distribution.

[0062] Furthermore, utilizing the known characteristics of channel statistics, the first The data received by the user was rewritten as

[0063]

[0064] in, Indicates the first The communication node and the first Transmission power control coefficient between users Indicates that the communication node sends to the first The data symbols of each user satisfy the following: , Indicates the first The communication node and the first Beam sequence between users.

[0065] Furthermore, the first Downlink speed per user The expression is

[0066]

[0067]

[0068] in, Indicates the first Signal-to-interference-plus-noise ratio per user.

[0069] Furthermore, in the downlink data transmission phase... Power consumed by each communication node sending data for

[0070]

[0071] in, This represents Euclidean norm operations.

[0072] Furthermore, the power consumed by the active RIS amplification signal. The power of the output signal Subtract the power of the input signal Its expression is

[0073]

[0074] Furthermore, based on the power consumed by the communication nodes and the active RIS in transmitting and amplifying signals, the total power consumed by the active RIS-assisted cellless mMIMO system during the downlink data transmission phase is... It can be represented as

[0075]

[0076] in, Indicates the first Power amplifier efficiency of each communication node This indicates the power amplifier efficiency of an active RISC system. Indicates the first The power consumed by each communication node to transmit data. This indicates the power consumed by the active RIS amplification signal. This represents the power consumed by the radio frequency circuitry of each communication node. This represents the power consumed by the control circuitry of each active RISC element. This indicates the DC circuit power consumption of each active RIS element.

[0077] Furthermore, system energy efficiency Defined as the ratio of downlink rate to total system power consumption, its expression is:

[0078]

[0079] in, Indicates the first Download rate per user, Indicates the first Signal-to-interference-plus-noise ratio per user.

[0080] Furthermore, to improve system energy efficiency, the downlink transmission power control coefficient is used as the optimization variable. The multi-objective optimization problem of maximizing the downlink rate and minimizing the total system power consumption is planned as follows:

[0081]

[0082] in,

[0083] , and Indicates intermediate variables. This represents a matrix consisting of downlink transmission power control coefficients between all communication nodes and all users. This represents the maximum power budget that the communication node can use for downlink data transmission. This represents the maximum power budget for an active RIS to amplify signals.

[0084] Furthermore, a weighting factor is introduced. By using a weighted sum method, the above multi-objective function problem can be reprogrammed into a single-objective function problem.

[0085]

[0086] Furthermore, auxiliary variables are introduced. and The above single-objective function problem becomes

[0087]

[0088] Furthermore, for a given variable and Using the arithmetic mean inequality to process functions And approximate it using a first-order Taylor expansion. We obtain the following inequality

[0089]

[0090] in, , , and This represents an intermediate variable. In this case, the objective function can be equivalent to...

[0091]

[0092] Furthermore, given the downlink power control coefficient Under the given conditions, apply the arithmetic mean inequality again to the expression. Approximate about The monomial form, specifically:

[0093]

[0094] in,

[0095] , and Indicates intermediate variables;

[0096] At this point, the first constraint can be equivalent to:

[0097]

[0098] Furthermore, based on the above approximation, the original problem is transformed into a generalized geometric programming problem, as follows:

[0099] Then, using MATLAB and CVX optimization tools, the above geometric programming problem is solved iteratively. The above steps are repeated until the termination condition is met, at which point the optimal downlink transmission power allocation scheme is obtained. .

[0100] Furthermore, the optimal power allocation scheme is substituted into the energy efficiency expression to calculate the downlink rate and energy efficiency. (See also...) Figure 1 The system downlink rate increases with the number of active RIS reflectors. This is because the increased number of RIS reflectors provides more spatial degrees of freedom and additional communication paths, improving channel diversity gain. The method of this invention effectively reduces inter-user interference by optimizing the downlink transmission power of communication nodes, further enhancing the downlink rate. (See also...) Figure 2 It can be observed that energy efficiency initially increases and then decreases with the increase in the number of active RIS reflective elements. This is because a large number of active RIS reflective elements leads to excessively high active load power consumption. When the rate gain provided by the active RIS is insufficient to offset the increased power consumption, the energy efficiency gradually decreases. However, the method of this invention effectively reduces system power consumption and improves system energy efficiency. Therefore, by utilizing multi-objective optimization criteria, the method of this invention achieves a simultaneous improvement in downlink rate and energy efficiency in active RIS-assisted cellless mMIMO systems.

[0101] 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 an active RIS-assisted cellless mMIMO system, characterized in that, Includes the following steps: Step 1: Establish an active RIS cellless mMIMO system model, specifically including a downlink data transmission model and a total system power consumption model; Step 2: Based on the system model, give the expressions for downlink rate and energy efficiency; Step 3: Using the transmission power control coefficient of the communication node as the optimization variable and the power budget of the communication node and the active RIS as constraints, formulate a multi-objective optimization problem with the objectives of maximizing the downlink rate and minimizing the total power consumption of the system. Step 4: Using the weighted sum method and the arithmetic mean inequality to handle the objective function and polynomial constraints, the original optimization problem is transformed into a generalized geometric programming problem and solved iteratively to complete the downlink power allocation; specifically, this includes the following steps: The multi-objective function form is expressed as a single-objective function form using a weighted sum method, by introducing weighting factors. This leads to a single-objective function optimization problem. Introducing auxiliary variables and The single-objective optimization problem can be transformed into a generalized geometric programming problem using the arithmetic mean inequality and the first-order Taylor expansion inequality. The above geometric programming problem is solved iteratively using MATLAB and CVX optimization tools to obtain the optimal downlink power allocation scheme. Calculate the system's energy efficiency.

2. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 1, characterized in that, In step 1, in the downlink data transmission model, the first... Data received by individual users Represented as in, Indicates the number of communication nodes. Indicates the number of users. This represents the normalized signal-to-noise ratio used for downlink symbol transmission. Indicates the first The communication node and the first Transmission power control coefficient between users Indicates the first The communication node and the first A sequence of direct-connect channels between users. Indicates the first The communication node and the first Beam sequence between users Indicates the first The data symbols of each user satisfy the following: , Indicates active RIS and the first Channel sequences between users This represents the reflection coefficient matrix of an active RIS. Indicates active RIS and the first Channel matrix between communication nodes This represents the thermal noise generated by an active RIS, where each element follows a distribution. , Indicates the first The additive Gaussian noise received by each user follows a distribution. , This indicates the absolute value operation. This represents the conjugate transpose operation. This represents the expectation operation. This represents a cyclically symmetric complex Gaussian distribution.

3. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 1, characterized in that, In step 1, the system consumes total power. The mathematical model is in, Indicates the first Power amplifier efficiency of each communication node This indicates the power amplifier efficiency of an active RISC system. Indicates the first The power consumed by each communication node to transmit data. This indicates the power consumed by the active RIS amplification signal. This indicates the number of components in an active RISC. This represents the power consumed by the radio frequency circuitry of each communication node. This represents the power consumed by the control circuitry of each active RISC element. This represents the DC power consumption of each active RISC element. This indicates the circuit power consumption for each user.

4. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 3, characterized in that, and The expression is in, This represents Euclidean norm operations.

5. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 4, characterized in that, In step 2, the first Downlink speed per user The expression is in, Indicates the first The communication node and the first Aggregated channel sequences among users Indicates the first The communication node and the first Transmission power control coefficient between users Indicates the first The communication node and the first Beam sequence between users Indicates the first Signal-to-interference-to-noise ratio for each user This is equivalent to the system's energy efficiency. Defined as the ratio of downlink speed to total system power consumption, expressed as: ， in, Indicates the first Download rate per user, Indicates the first Signal-to-interference-plus-noise ratio per user.

6. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 5, characterized in that, In step 3, the multi-objective optimization problem P1, which maximizes the downlink rate and minimizes the total system power consumption, is established as follows: in, , , and Indicates intermediate variables. This represents a matrix consisting of downlink transmission power control coefficients between all communication nodes and all users. This represents the maximum power budget that the communication node can use for downlink data transmission. This represents the maximum power budget for an active RIS to amplify signals.

7. The energy efficiency optimization method for an active RIS-assisted cellless mMIMO system according to claim 5, characterized in that, In step 4, the specific steps for solving the original optimization problem using the weighted sum method and the arithmetic mean inequality are as follows: 4.1 Introducing weighting factors The original multi-objective problem P1 is rewritten as a single-objective problem P2 using the weighted sum method, which is expressed as follows: 4.2 Introducing Auxiliary Variables and Relax problem P2 in step 4.1 to in, Indicates the first Data received by each user Indicates the first The communication node and the first Aggregation channel between users 4.3 For a given variable and Using the arithmetic mean inequality to process functions And approximate it using a first-order Taylor expansion. We obtain the following inequality in, , , and Indicates intermediate variables; 4.4 For a given downlink power control coefficient Based on the arithmetic mean inequality, the right-hand side of the first constraint in problem P3 in step 4.2 is... in, , , Indicates intermediate variables; 4.5 Based on the inequalities in steps 4.3 and 4.4, the problem P3 in step 4.2 is rewritten as a generalized geometric programming problem, as shown below: 4.6 Use MATLAB and CVX optimization tools to solve problem P4 in step 4.5, and repeat steps 4.1-4.6 until the termination condition is met to obtain the optimal solution. ; 4.7 The optimal solution obtained As the final downlink transmission power allocation scheme, the system energy efficiency is calculated by substituting it into the energy efficiency expression.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements an energy efficiency optimization method for an active RIS-assisted cellless mMIMO system as described in any one of claims 1 to 7.

9. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by the processor, the computer instructions implement an energy efficiency optimization method for an active RIS-assisted cellless mMIMO system as described in any one of claims 1-7.

Images