Power consumption optimization method for honeycomb-removed large-scale multiple-input-output common inductance integrated system

Abstract

The invention discloses a power consumption optimization method for a cellular-removal large-scale multiple-input-multiple-output integrated system. The method comprises the following steps: firstly, constructing a cellular-removal large-scale multiple-input-multiple-output integrated system model; selecting a communication and inductance integrated signal pre-coding scheme; and then a non-convex optimization problem model of the cellular-removed large-scale multiple-input-multiple-output sensing integrated system is constructed, and the total power consumption of the system is minimized by controlling a wireless access point switch and the transmitting power in consideration that the transmitting power of an access point is limited and the spectral efficiency and the sensing spectral efficiency of a communication user meet basic requirements. Processing non-convex constraints by adopting a feasible solution tracking-successive convex approximation method, and converting a non-convex problem into a convex problem; solving an optimization problem by adopting a branch and bound method; and finally, obtaining an optimal solution of a corresponding convex problem by utilizing a convex optimization solving tool. According to the invention, the total power consumption of the system is minimized under the condition that communication and sensing requirements are met. And more power consumption can be saved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and particularly relates to a power consumption optimization method for a de-cellular massive multiple-input multiple-output integrated sensing and communication system. Background Art

[0002] With the development of communication technology, more and more application scenarios require both communication and sensing capabilities, such as autonomous driving vehicles, intelligent industries, smart homes, wearable electronic devices, etc. Integrated Sensing and Communication (ISAC) and de-cellular massive multiple-input multiple-output systems can achieve the above application scenarios. On the one hand, the introduction of ISAC technology into the de-cellular massive multiple-input multiple-output system can endow it with sensing capabilities. On the other hand, distributed wireless access points can perform joint sensing to improve sensing accuracy.

[0003] However, due to the introduction of ISAC technology, the de-cellular massive multiple-input multiple-output system needs to transmit additional sensing signals, which will inevitably lead to an increase in power consumption. In addition, the communication and sensing performance of the system will be affected by various interference factors (such as sensing clutter and interference caused by imperfect communication channel state information). Therefore, if the system wants to ensure normal communication and sensing services, it needs a larger transmit power, thus increasing the system power consumption. In the current era of energy shortage, reducing energy consumption and achieving green development are particularly important. Especially with the wide commercialization of the fifth-generation network, the power consumption has increased significantly. Data shows that the full-load power consumption of the fifth-generation network's BS is about three to four times that of the fourth-generation network. According to the current planning of the sixth-generation network, it is expected that the overall power consumption of a single station may exceed four times that of the fifth-generation network. Therefore, reducing power consumption and improving energy efficiency have become key issues that need to be solved urgently. Summary of the Invention

[0004] Object of the Invention: The object of the present invention is to provide a power consumption optimization method for a de-cellular massive multiple-input multiple-output integrated sensing and communication system. The aim is to minimize the total power consumption of the de-cellular massive multiple-input multiple-output integrated sensing and communication system while meeting communication and sensing requirements.

[0005] Technical Solution: A power consumption optimization method for a de-cellular massive multiple-input multiple-output integrated sensing and communication system of the present invention includes the following steps:

[0006] Step S1: Construct a de-cellular massive multiple-input multiple-output integrated sensing and communication system model;

[0007] The de-cellular massive multiple-input multiple-output integrated sensing and communication system model includes a channel estimation model, an integrated sensing and communication signal model, a communication model, and a sensing model;

[0008] Step S2: Select a joint communication and sensing signal precoding scheme;

[0009] Step S3: Construct a non-convex optimization problem model for a cell-free massive MIMO communication and sensing integrated system, considering the limited transmit power of the access point, the spectral efficiency of communication users and the sensing spectral efficiency meeting the basic requirements, and minimizing the total system power consumption by controlling the wireless access point switch and transmit power;

[0010] Step S4: Use the feasible solution tracking-successive convex approximation method to handle non-convex constraints and transform the non-convex problem into a convex problem;

[0011] Step S5: Use the branch and bound method to solve the optimization problem;

[0012] Step S6: Use a convex optimization solver to obtain the optimal solution of the corresponding convex problem.

[0013] Further, in step S1, constructing a cell-free massive MIMO communication and sensing integrated system model specifically includes:

[0014] Construct a channel estimation model: Define τ p as the length of the pilot sequence, M as the number of transmitting wireless access points, K as the number of single-antenna communication users, and h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user. is the pilot sequence allocated to the k-th communication user, and the signal received by the m-th transmitting wireless access point is:

[0015]

[0016] where p k represents the pilot transmit power of the k-th communication user, N m represents additive noise, (·) H represents the conjugate transpose operation, represents taking the square root;

[0017] Estimate the channel using the minimum mean square error estimation method. The channel estimate between the m-th transmitting wireless access point and the k-th communication user is:

[0018]

[0019] where the pilot signal processed by the wireless access point R mk represents the spatial correlation matrix of the channel h mk and the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t and k' represents the k'-th communication user. denotes the subset of users using the same pilot sequence as the k-th communication user, M t denotes the number of antennas of the transmitting wireless access point, (·) -1 denotes the inverse operation, denotes the mean value;

[0020] Channel estimation value The spatial correlation matrix of is expressed as:

[0021]

[0022] Construct an integrated communication and sensing signal model: Define s k denotes the communication signal of the k-th communication user, s0 denotes the sensing signal, ρ mk and ρ m0 respectively denote the transmit power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal, w mk and w m0 respectively denote the precoding vectors of the m-th transmitting wireless access point with respect to the k-th communication user and the sensing signal, then the integrated communication and sensing signal is defined as:

[0023]

[0024] Among them, the precoding matrix W of the m-th transmitting wireless access point m =[w m1 ,...,w mK ,w m0 , the signal matrix D s [m]=diag(s1,...,s K ,s0), the power allocation vector of the m-th transmitting wireless access point diag(·) denotes the diagonal matrix, (·) T denotes the transpose operation;

[0025] Construct a communication model: Define the subset denotes the set of active transmitting wireless access points, and the signal received by the k-th communication user is:

[0026]

[0027] Among them, n k denotes the additive noise at the communication user k, and K / {k} denotes the set of K communication users excluding the k-th communication user;

[0028] Then, the signal-to-interference-plus-noise ratio of the k-th communication user is expressed as:

[0029]

[0030] Among them, is the desired signal,

[0031] represents the uncertainty gain caused by beamforming, represents the interference between users, represents the interference caused by the sensing signal, represents the noise power at the communication user k, and ∣·∣ represents the modulo operation;

[0032] Therefore, the spectral efficiency of the k-th communication user is:

[0033]

[0034] where τ c represents the coherence interval length;

[0035] Construct a sensing model: Define G m as the reflection path from the m-th transmitting radio access point through the sensing target to the receiving radio access point, and H L,m represents the line-of-sight path between the m-th transmitting radio access point and the receiving radio access point, and H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting radio access point and the receiving radio access point. The signal received by the receiving radio access point is expressed as:

[0036]

[0037] where n s represents the sensing noise. The sensing channel matrix A of the active transmitting radio access point = [G1W1Ds[1],...,G|A|W|A|Ds[|A|]], the sensing interference channel matrix B of the active transmitting radio access point = [H1W1Ds[1],...,H|A|W|A|Ds[|A]], and the power allocation vector of the active transmitting radio access point represents the subset the number of elements of;

[0038] Then, the sensing signal-to-interference-plus-noise ratio is expressed as:

[0039]

[0040] where M r represents the number of antennas of the receiving radio access point, represents the sensing channel noise power. Therefore, the sensing spectral efficiency is:

[0041]

[0042] In summary, the total spectral efficiency is expressed as:

[0043]

[0044] Furthermore, in step S2, a joint communication and sensing signal precoding scheme is selected:

[0045] M is the number of transmitting wireless access points, and K represents the number of single-antenna communication users. denotes the set of active transmitting wireless access points, and ρ mk and ρ m0 respectively represent the transmit power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal. h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user. represents the channel estimate value between the m-th transmitting wireless access point and the k-th communication user. a(θ m ) represents the antenna array response vector of the m-th transmitting wireless access point, and θ m represents the angle from the sensing target to the m-th transmitting wireless access point. Assuming that the precoding of the m-th transmitting wireless access point for the k-th communication user is The precoding of the m-th transmitting wireless access point for the sensing signal is ||·|| represents the Euclidean norm. represents the mean. Therefore, the effective signal-to-interference-plus-noise ratio expression for the k-th communication user is derived as:

[0046]

[0047] where the desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient the pilot contamination coefficient the sensing signal interference coefficient d mk = tr(R mk ), represents the communication user noise power, R mk is the spatial correlation matrix of the channel h mk , the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value represents the subset of users using the same pilot sequence as the k-th communication user. represents the subset excluding the k-th communication user, τ p is the length of the pilot sequence, and p kdenotes the pilot transmit power of the k-th communication user, k′ denotes the k'-th communication user, and M t denotes the number of antennas of the transmitting wireless access point, and tr(·) denotes the trace of a matrix. denotes taking the square root.

[0048] Furthermore, in step S3, a non-convex optimization problem model of the cellular-free massive MIMO communication and sensing integrated system is constructed. Considering the limited transmit power of the access point, the spectral efficiency of the communication users and the sensing spectral efficiency meet the basic requirements, and the total power consumption of the system is minimized by controlling the wireless access point switch and transmit power.

[0049] Let M be the number of transmitting wireless access points, and K denote the number of single-antenna communication users. denotes the set of active transmitting wireless access points. denotes a subset the number of elements of, and ρ mk and ρ m0 respectively denote the transmit power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal. h mk denotes the channel vector between the m-th transmitting wireless access point and the k-th communication user. denotes the channel estimate value between the m-th transmitting wireless access point and the k-th communication user, and P m and P bt,m respectively denote the power consumption unrelated to information transmission and the power consumption related to information transmission of the m-th transmitting wireless access point. B denotes the transmission bandwidth. and respectively denote the total spectral efficiency, the spectral efficiency of the k-th communication user, and the sensing spectral efficiency. The problem is

[0050]

[0051] where P max,m denotes the maximum transmit power of the m-th transmitting wireless access point, ξ k and γ respectively represent the spectral efficiency requirement and the sensing spectral efficiency requirement of each k-th communication user. The power allocation vector of the m-th transmitting wireless access point (·) T denotes the transpose operation, ||·|| denotes the Euclidean norm. denotes taking the square root;

[0052] τ c denotes the coherence interval length, and τ p is the length of the pilot sequence. Let the signal-to-interference-plus-noise ratio requirement of each k-th communication user and the sensing signal-to-interference-plus-noise ratio requirement The problem becomes:

[0053]

[0054]

[0055] Among them, denotes the signal-to-interference-plus-noise ratio (SINR) of the k-th communication user, SINR (s) denotes the sensing signal-to-interference-plus-noise ratio, and the hardware power consumption generated by the m-th transmitting radio access point

[0056] For the non-convex constraints in the problem, the non-convex signal-to-interference-plus-noise ratio constraint of the communication user is transformed into a second-order cone form:

[0057]

[0058] Among them, denotes the noise power at the communication user k, and the power vector allocated by the active transmitting radio access point to the k-th communication user The power vector allocated by the active transmitting radio access point to the sensing signal The vector of expected signal coefficients of the active transmitting radio access point with respect to the k-th communication user The vector of inter-user interference coefficients of the active transmitting radio access point with respect to the k-th communication user The vector of pilot contamination coefficients c of the active transmitting radio access point with respect to the k-th communication user kk′A =[c 1k′k ,...,c ∣A∣k′k T , and the vector of sensing signal interference coefficients of the active transmitting radio access point with respect to the k-th communication user The expected signal coefficient a mk =tr(Q mk ), the inter-user interference coefficient The pilot contamination coefficient The sensing signal interference coefficient d mk =tr(R mk ), R mk denotes the spatial correlation matrix of the channel h mk , and the inverse matrix of the minimum mean square error estimator denotes a diagonal matrix of dimension M t , and the spatial correlation matrix of the channel estimation value k' represents the k'-th communication user, denotes the subset of users using the same pilot sequence as the k-th communication user, p k denotes the pilot transmission power of the k-th communication user, denotes the subset ​​The k-th communication user is removed from M t represents the number of antennas of the transmitting wireless access point, and tr(·) represents the trace of a matrix. represents the Hadamard product.

[0059] Furthermore, in step S4, the feasible solution tracking-successive convex approximation method is used to handle non-convex constraints, transforming the non-convex problem into a convex problem:

[0060] Define M as the number of transmitting wireless access points, and K represents the number of single-antenna communication users. represents the set of active transmitting wireless access points. represents a subset. the number of elements of, M r represents the number of antennas of the receiving wireless access point, s k represents the communication signal of the k-th communication user, s0 represents the sensing signal, w mk and w m0 respectively represent the precoding vectors of the m-th transmitting wireless access point with respect to the k-th communication user and the sensing signal, ρ mk and ρ m0 respectively represent the transmit powers allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal, h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user. represents the channel estimate value between the m-th transmitting wireless access point and the k-th communication user, G m is the reflected path from the m-th transmitting wireless access point through the sensing target to the receiving wireless access point, H L,m represents the line-of-sight path between the m-th transmitting wireless access point and the receiving wireless access point, H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting wireless access point and the receiving wireless access point, η represents the sensing signal-to-interference-plus-noise requirement, and the precoding matrix W of the m-th transmitting wireless access point m =[w m1 ,...,w mK ,w m0 , the signal matrix D s [m]=diag(s1,...,s K ,s0), the sensing interference channel H between the m-th transmitting wireless access point and the receiving wireless access point m =H L,m +H NL,m , the power allocation vector of the m-th transmitting wireless access point The sensing channel matrix of the active transmitting wireless access point The perceived interference channel matrix B of the active transmitting wireless access point is B = [H1W1Ds[1],...,H|A||W|A|Ds[|A|]] denotes the perceived channel noise power, denotes taking the square root, (·) T denotes the transpose operation, diag(·) denotes the diagonal matrix, and the feasible solution tracking-successive convex approximation method is used to handle the non-convex perceived signal-to-interference-plus-noise ratio constraint, and it is transformed into:

[0061]

[0062] where χ≥0 represents the relaxation variable, ρ (i) denotes the optimal solution of the i-th iteration, and the perceived channel coefficient after the active transmitting wireless access point takes real number processing the perceived interference channel coefficient after the active transmitting wireless access point takes real number processing the power allocation vector of the active transmitting wireless access point (·) H denotes the conjugate transpose operation, denotes taking the real part;

[0063] Define as the noise power at the communication user k, and the power vector allocated by the active transmitting wireless access point to the k-th communication user the power vector allocated by the active transmitting wireless access point to the perceived signal the desired signal coefficient vector of the active transmitting wireless access point with respect to the k-th communication user the inter-user interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user the pilot contamination coefficient vector of the active transmitting wireless access point with respect to the k-th communication user the perceived signal interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user the desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient the pilot contamination coefficient the perceived signal interference coefficient d mk = tr(R mk ), R mk denotes the spatial correlation matrix of the channel h mk the inverse matrix of the minimum mean square error estimator denotes the diagonal matrix of dimension M t the channel estimate value the spatial correlation matrix of Denote the subset of users using the same pilot sequence as the \(k\)-th communication user, denote the subset excluding the \(k\)-th communication user, where \(\tau\) p is the length of the pilot sequence, and \(p\) k denotes the pilot transmission power of the \(k\)-th communication user, \(k'\) denotes the \(k'\)-th communication user, and \(M\) t denotes the number of antennas of the transmitting radio access point, and \(\nu\) k represents the signal-to-interference-plus-noise ratio requirement for each \(k\)-th communication user. \(tr(\cdot)\) represents the trace of a matrix. Then the problem becomes:

[0064]

[0065]

[0066] where \(\lambda>0\) is the penalty coefficient, \(P\) total denotes the total system power consumption, and the hardware power consumption generated by the \(m\)-th transmitting radio access point \(B\) represents the transmission bandwidth, \(P\) m and \(P\) bt,m represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the \(m\)-th transmitting radio access point respectively. \(P\) max,m denotes the maximum transmission power of the \(m\)-th transmitting radio access point, and \(\xi\) k represents the spectral efficiency requirement for each \(k\)-th communication user. \(\|\cdot\|\) represents the Euclidean norm, denotes the Hadamard product;

[0067] The specific steps are as follows:

[0068] 4.1 Initialization: Set arbitrary initial power values \(\rho\) (0) \(\geq0\) and convergence accuracy \(\epsilon>0\), \(\epsilon\) χ \(>0\), and penalty coefficient \(\lambda>0\). Let the iteration index \(i = 0\) and the maximum number of iterations \(i\) max \(= 100\);

[0069] 4.2 Start the loop \(i\leq i\) max ;

[0070] 4.3 \(i = i + 1\);

[0071] 4.4 Use \(\rho\) (i) to solve the above optimization problem to obtain \(\rho\) (i+1) ;

[0072] 4.5 If \(\chi<\epsilon\) χ , in the next loop, let \(\chi = 0\);

[0073] 4.6 If end the loop and execute 4.8;

[0074] 4.7 End the loop;

[0075] 4.8 If χ ≠ 0, it is considered that the problem is infeasible;

[0076] 4.9 If χ = 0, output the optimal transmit power;

[0077] Among them, represents the total system power consumption in the i-th iteration.

[0078] Furthermore, in the step S5, the branch and bound method is used to solve the optimization problem:

[0079] Define M as the number of transmitting wireless access points, K represents the number of single-antenna communication users, M r represents the number of antennas of the receiving wireless access point, s k represents the communication signal of the k-th communication user, s0 represents the sensing signal, w mk and w m0 respectively represent the precoding vectors of the m-th transmitting wireless access point with respect to the k-th communication user and the sensing signal, ρ mk and ρ m0 respectively represent the transmit power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal, G m is the reflection path from the m-th transmitting wireless access point through the sensing target to the receiving wireless access point, H L,m represents the line-of-sight path between the m-th transmitting wireless access point and the receiving wireless access point, H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting wireless access point and the receiving wireless access point, η represents the sensing signal-to-interference-plus-noise requirement, and the precoding matrix W of the m-th transmitting wireless access point m = [w m1 ,..., w mK , w m0 , the signal matrix D s [m] = diag(s1,..., s K , s0), the power allocation vector of the m-th transmitting wireless access point ρ (i) represents the optimal solution in the i-th iteration, represents the sensing channel noise power, P m and P bt,m respectively represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the m-th transmitting wireless access point, and the hardware power consumption generated by the m-th transmitting wireless access point B represents the transmission bandwidth, P max,m represents the maximum transmit power of the m-th transmitting wireless access point, ξ kRepresents the spectral efficiency requirement of each k-th communication user, ν k Represents the signal-to-interference-plus-noise ratio requirement of each k-th communication user, λ is the penalty coefficient, χ represents the slack variable Denotes taking the square root, (·) T Denotes the transpose operation, diag(·) denotes the diagonal matrix, using the binary variable α m Denotes the switch of the transmitting wireless access point, the problem becomes

[0080]

[0081]

[0082] Among them, the power vector allocated by M transmitting wireless access points to the k-th communication user The power vector allocated by M transmitting wireless access points to the sensing signal The desired signal coefficient vector of M transmitting wireless access points with respect to the k-th communication user The inter-user interference coefficient vector of M transmitting wireless access points with respect to the k-th communication user The pilot contamination coefficient vector c of M transmitting wireless access points with respect to the k-th communication user kk′M =[c 1k′k ,...,c Mk′k T , the sensing signal interference coefficient vector of M transmitting wireless access points with respect to the k-th communication user The power allocation vector ρ′ of M transmitting wireless access points = [ρ1,..., ρ M T , the sensing channel coefficients of M transmitting wireless access points after taking real numbers The sensing interference channel coefficients of M transmitting wireless access points after taking real numbers The sensing interference channel H between the m-th transmitting wireless access point and the receiving wireless access point m =H L,m +H NL,m , the sensing channel matrix A′ of M transmitting wireless access points = [G1W1D s [1],...,G M W M D s [M]], the sensing interference channel matrix B′ of M transmitting wireless access points = [H1W1D s [1],...,H M W M D s [M]], the desired signal coefficient a mk =tr(Q mk ), the inter-user interference coefficient ​​Pilot pollution coefficient Perceived signal interference coefficient d mk = tr(R mk ), where R mk represents the spatial correlation matrix of the channel h mk , and the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value k' represents the k'-th communication user, represents the subset of users using the same pilot sequence as the k-th communication user, represents the subset excluding the k-th communication user, τ p is the length of the pilot sequence, p k represents the pilot transmission power of the k-th communication user, M t represents the number of antennas of the transmitting wireless access point, (·) H represents the conjugate transpose operation, represents taking the real part, tr(·) represents the trace of a matrix, ||·|| represents the Euclidean norm, represents the Hadamard product.

[0083] Furthermore, in step S6, the optimal solution of the corresponding convex problem is obtained by using a convex optimization solver:

[0084] The MOSEK solver and the CVX solver are used to solve the above problem to obtain the optimal solution.

[0085] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored on the memory, and the processor executes the computer program to implement the steps of the method of the present invention.

[0086] The present invention also discloses a computer-readable storage medium, on which a computer program / instructions are stored, and when the computer program / instructions are executed by a processor, the steps of the method of the present invention are implemented.

[0087] The present invention also discloses a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method of the present invention are implemented.

[0088] Beneficial effects: Compared with the prior art, the present invention has the following remarkable advantages:

[0089] 1. The present invention proposes a power consumption optimization method based on a de-cellular massive multiple-input multiple-output communication and sensing integrated system. While meeting the communication and sensing requirements of the system, by controlling the switches of the wireless access points and their transmission powers, the total power consumption is minimized.​

[0090] 2. The present invention takes into account the impact of the line-of-sight path between the transmitting wireless access point and the receiving wireless access point and the non-line-of-sight path caused by obstacles in the environment on the sensing performance, as well as the impact of imperfect communication channel state information on the communication performance, making the present invention more practical.

[0091] 3. The present invention uses the method of feasible solution tracking - successive convex approximation to handle non-convex constraints for problem solving, and uses the branch and bound method to replace the exhaustive search method, reducing the computational complexity. Brief Description of the Drawings

[0092] Figure 1 is the overall flowchart of the present invention;

[0093] Figure 2 is the diagram of the de-cellularized massive multiple-input multiple-output communication and sensing integrated system of the present invention;

[0094] Figure 3 is the flowchart of the feasible solution tracking - successive convex approximation method of the present invention;

[0095] Figure 4 is the comparison diagram of the total power consumption of the algorithms of the present invention;

[0096] Figure 5 is the comparison diagram of the average number of active transmitting wireless access points required by the algorithms of the present invention. Detailed Embodiment

[0097] The technical solution of the present invention will be further described below with reference to the drawings.

[0098] As Figures 1-3 shown, an embodiment of the present invention proposes a power consumption optimization method based on a de-cellularized massive multiple-input multiple-output communication and sensing integrated system, including the following steps:

[0099] Step S1: Construct a de-cellularized massive multiple-input multiple-output communication and sensing integrated system model;

[0100] The de-cellularized massive multiple-input multiple-output communication and sensing integrated system model mainly includes a channel estimation model, a communication and sensing integrated signal model, a communication model, and a sensing model;

[0101] Step S2: Select a communication and sensing integrated signal precoding scheme;

[0102] Step S3: Construct a non-convex optimization problem model for the de-cellularized massive multiple-input multiple-output communication and sensing integrated system, considering the limited transmit power of the access point, and the spectral efficiency of the communication user and the sensing spectral efficiency meet the basic requirements, and by controlling the wireless access point switch and transmit power, to minimize the total power consumption of the system;

[0103] Step S4: Process the non-convex constraints using the feasible solution tracking - successive convex approximation method to transform the non-convex problem into a convex problem;

[0104] Step S5: Solve the optimization problem using the branch and bound method;

[0105] Step S6: Use the convex optimization solver to obtain the optimal solution of the corresponding convex problem;

[0106] Specifically, as Figure 1 shown, a power consumption optimization method for a de-cellularized massive multiple-input multiple-output communication and sensing integrated system, which specifically includes the following steps:

[0107] In step S1, construct a de-cellularized massive multiple-input multiple-output communication and sensing integrated system model; the de-cellularized massive multiple-input multiple-output communication and sensing integrated system model mainly includes a channel estimation model, a communication and sensing integrated signal model, a communication model, and a sensing model;

[0108] The de-cellularized massive multiple-input multiple-output communication and sensing integrated system model, as Figure 2 shown. The system uses multi-static sensing, which can not only avoid interference between the sensing echo signal and the transmitted signal, but also provide diversity gain and expand the sensing area.

[0109] Construct the channel estimation model: Define τ p as the length of the pilot sequence, M as the number of transmitting radio access points, K as the number of single-antenna communication users, h mk represents the channel vector between the m-th transmitting radio access point and the k-th communication user, as the pilot sequence assigned to the k-th communication user, and the signal received by the m-th transmitting radio access point is:

[0110]

[0111] where p k represents the pilot transmission power of the k-th communication user, N m represents the additive noise, (·) H represents the conjugate transpose operation, represents taking the square root;

[0112] Estimate the channel using the minimum mean square error estimation method, and the channel estimation value between the m-th transmitting radio access point and the k-th communication user is:

[0113]

[0114] where the pilot signal processed by the radio access point R mk represents the channel h mkSpatial correlation matrix, inverse matrix of the minimum mean square error estimator Denote a diagonal matrix of dimension M t , where k′ represents the k'th communication user Denote the subset of users using the same pilot sequence as the k'th communication user, M t Denote the number of antennas of the transmitting wireless access point, (·) -1 Denote the inverse operation Denote the mean

[0115] Channel estimate The spatial correlation matrix of is expressed as:

[0116]

[0117] Construct the integrated communication and sensing signal model: Define s k Denote the communication signal of the k'th communication user, s0 denotes the sensing signal, ρ mk And ρ m0 Denote the transmission powers allocated by the m'th transmitting wireless access point to the k'th communication user and the sensing signal respectively, w mk And w m0 Denote the precoding vectors of the m'th transmitting wireless access point with respect to the k'th communication user and the sensing signal respectively. Then the integrated communication and sensing signal is defined as:

[0118]

[0119] Among them, the precoding matrix W of the m'th transmitting wireless access point m =[w m1 ,...,w mK ,w m0 , the signal matrix D s [m]=diag(s1,...,s K ,s0), the power allocation vector of the m'th transmitting wireless access point diag(·) denotes a diagonal matrix, (·) T Denote the transpose operation

[0120] Construct the communication model: Define the subset Denote the set of active transmitting wireless access points. The signal received by the k'th communication user is:

[0121]

[0122] Among them, n k Denote the additive noise at communication user k, K / {k} denotes the set of K communication users excluding the k'th communication user

[0123] Then, the signal-to-interference-plus-noise ratio (SINR) of the k-th communication user is expressed as:

[0124]

[0125] Where, is the desired signal,

[0126] represents the uncertainty gain caused by beamforming, represents the inter-user interference, represents the interference caused by the sensing signal, represents the noise power at the communication user k, and |·| represents the modulus operation;

[0127] Therefore, the spectral efficiency of the k-th communication user is:

[0128]

[0129] Where, τ c represents the coherence interval length;

[0130] Construct a sensing model: Define G m as the reflection path from the m-th transmitting radio access point to the receiving radio access point through the sensing target, and H L,m represents the line-of-sight path between the m-th transmitting radio access point and the receiving radio access point, and H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting radio access point and the receiving radio access point. The signal received by the receiving radio access point is expressed as:

[0131]

[0132] Where, n s represents the sensing noise, and the sensing channel matrix of the active transmitting radio access point The sensing interference channel matrix of the active transmitting radio access point The power allocation vector of the active transmitting radio access point represents the number of elements in the subset ;

[0133] Then, the sensing signal-to-interference-plus-noise ratio (SINR) is expressed as:

[0134]

[0135] Where, M r represents the number of antennas of the receiving radio access point, represents the sensing channel noise power. Therefore, the sensing spectral efficiency is:

[0136]

[0137] In summary, the total spectral efficiency is expressed as:

[0138]

[0139] In step S2, a communication and sensing integrated signal precoding scheme is selected:

[0140] M represents the number of transmitting wireless access points, and K represents the number of single-antenna communication users. represents the set of active transmitting wireless access points, and ρ mk and ρ m0 respectively represent the transmit power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal. h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user. represents the channel estimate value between the m-th transmitting wireless access point and the k-th communication user. a(θ m ) represents the antenna array response vector of the m-th transmitting wireless access point, and θ m represents the angle from the sensing target to the m-th transmitting wireless access point. Assume that the precoding of the m-th transmitting wireless access point for the k-th communication user is The precoding of the m-th transmitting wireless access point for the sensing signal is ||·|| represents the Euclidean norm. represents the mean. Therefore, the effective signal-to-interference-plus-noise ratio expression for the k-th communication user is derived as:

[0141]

[0142] Among them, the desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient the pilot contamination coefficient the sensing signal interference coefficient d mk = tr(R mk ), represents the communication user noise power, and R mk is the spatial correlation matrix of the channel h mk . The inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t . The spatial correlation matrix of the channel estimate value represents the subset of users using the same pilot sequence as the k-th communication user. represents the subset excluding the k-th communication user, and τ pis the length of the pilot sequence, p k represents the pilot transmission power of the k-th communication user, k' represents the k'-th communication user, M t represents the number of antennas of the transmitting wireless access point, tr(·) represents the trace of a matrix, represents taking the square root.

[0143] In step S3, a non-convex optimization problem model of a de-cellular massive MIMO integrated communication and sensing system is constructed. Considering the limited transmission power of the access point, the spectral efficiency of communication users and the sensing spectral efficiency meet the basic requirements, and by controlling the wireless access point switch and transmission power, the total power consumption of the system is minimized;

[0144] Let M be the number of transmitting wireless access points, and K represent the number of single-antenna communication users, represents the set of active transmitting wireless access points, represents a subset the number of elements of, ρ mk and ρ m0 respectively represent the transmission power allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal, h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user, represents the channel estimate value between the m-th transmitting wireless access point and the k-th communication user, P m and P bt,m respectively represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the m-th transmitting wireless access point, B represents the transmission bandwidth, and respectively represent the total spectral efficiency, the spectral efficiency of the k-th communication user, and the sensing spectral efficiency. The problem is

[0145]

[0146]

[0147] where, P max,m represents the maximum transmission power of the m-th transmitting wireless access point, ξ k and γ respectively represent the spectral efficiency requirement and sensing spectral efficiency requirement of each k-th communication user. The power allocation vector of the m-th transmitting wireless access point (·) T represents the transpose operation, ||·|| represents the Euclidean norm, represents taking the square root;

[0148] τ c represents the coherence interval length, τ p is the length of the pilot sequence. Let the signal-to-interference-plus-noise ratio requirement of each k-th communication user and the perceived signal-to-interference-plus-noise ratio requirement The problem becomes:

[0149]

[0150] where, denotes the signal-to-interference-plus-noise ratio of the k-th communication user, SINR (s) denotes the perceived signal-to-interference-plus-noise ratio, and the hardware power consumption generated by the m-th transmitting radio access point

[0151] For the non-convex constraints in the problem, the non-convex signal-to-interference-plus-noise ratio constraint of the communication user is transformed into a second-order cone form:

[0152]

[0153] where, denotes the noise power at the communication user k, and the power vector allocated by the active transmitting radio access point to the k-th communication user The power vector allocated by the active transmitting radio access point to the sensing signal The vector of desired signal coefficients of the active transmitting radio access point with respect to the k-th communication user The vector of inter-user interference coefficients of the active transmitting radio access point with respect to the k-th communication user The vector of pilot contamination coefficients of the active transmitting radio access point with respect to the k-th communication user The vector of sensing signal interference coefficients of the active transmitting radio access point with respect to the k-th communication user The desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient The pilot contamination coefficient The sensing signal interference coefficient d mk = tr(R mk ), R mk denotes the spatial correlation matrix of the channel h mk The inverse matrix of the minimum mean square error estimator denotes a diagonal matrix of dimension M t The channel estimate The spatial correlation matrix of k′ denotes the k'-th communication user, denotes the subset of users using the same pilot sequence as the k-th communication user, p k denotes the pilot transmission power of the k-th communication user, denotes the subset excluding the k-th communication user from, M tIndicates the number of antennas of the transmitting wireless access point, and tr(·) represents the trace of a matrix. Indicates the Hadamard product;

[0154] In step S4, the feasible solution tracking - successive convex approximation method is used to handle non - convex constraints, converting the non - convex problem into a convex problem:

[0155] Define M as the number of transmitting wireless access points, and K represents the number of single - antenna communication users. Indicates the set of active transmitting wireless access points, Indicates a subset The number of elements of, M r Indicates the number of antennas of the receiving wireless access point, s k Indicates the communication signal of the k - th communication user, s0 represents the sensing signal, w mk And w m0 Respectively represent the precoding vectors of the m - th transmitting wireless access point with respect to the k - th communication user and the sensing signal, ρ mk And ρ m0 Respectively represent the transmit power allocated by the m - th transmitting wireless access point to the k - th communication user and the sensing signal, h mk Represents the channel vector between the m - th transmitting wireless access point and the k - th communication user. Represents the channel estimate value between the m - th transmitting wireless access point and the k - th communication user, G m Is the reflection path of the m - th transmitting wireless access point through the sensing target to the receiving wireless access point, H L,m Represents the line - of - sight path between the m - th transmitting wireless access point and the receiving wireless access point, H NL,m Represents the non - line - of - sight path caused by obstacles between the m - th transmitting wireless access point and the receiving wireless access point, η represents the sensing signal - to - interference - plus - noise requirement, and the precoding matrix W of the m - th transmitting wireless access point m =[w m1 ,...,w mK ,w m0 , the signal matrix D s [m]=diag(s1,...,s K ,s0), the sensing interference channel H between the m - th transmitting wireless access point and the receiving wireless access point m =H L,m +H NL,m , the power allocation vector of the m - th transmitting wireless access point The sensing channel matrix of the active transmitting wireless access point The sensing interference channel matrix of the active transmitting wireless access point Represents the sensing channel noise power. Indicates taking the square root, (·)T represents the transpose operation, diag(·) represents the diagonal matrix, and the feasible solution tracking-successive convex approximation method is used to handle the non-convex sensing signal-to-interference-plus-noise ratio constraint, and it is transformed into:

[0156]

[0157] where, χ≥0 represents the slack variable, ρ (i) represents the optimal solution of the i-th iteration, and the sensing channel coefficient after the active transmitting wireless access point takes real numbers The sensing interference channel coefficient after the active transmitting wireless access point takes real numbers The power allocation vector of the active transmitting wireless access point (·) H represents the conjugate transpose operation, represents taking the real part;

[0158] Define as the noise power at the communication user k, and the power vector allocated by the active transmitting wireless access point to the k-th communication user The power vector allocated by the active transmitting wireless access point to the sensing signal The desired signal coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The inter-user interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The pilot contamination coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The sensing signal interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user Desired signal coefficient a mk = tr(Q mk ), inter-user interference coefficient Pilot contamination coefficient Sensing signal interference coefficient d mk = tr(R mk ), R mk represents the spatial correlation matrix of the channel h mk The inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value represents the subset of users using the same pilot sequence as the k-th communication user, represents the subset excluding the k-th communication user, τ p is the length of the pilot sequence, p kDenote the pilot transmit power of the \(k\)th communication user, \(k'\) represents the \(k'\)th communication user, \(M\) t Denote the number of antennas of the transmitting wireless access point, \(\nu\) k Represent the signal-to-interference-plus-noise ratio requirement of each \(k\)th communication user, \(tr(\cdot)\) represents the trace of a matrix, then the problem becomes:

[0159]

[0160] where \(\lambda>0\) is the penalty coefficient, \(P\) total Denote the total system power consumption, the hardware power consumption generated by the \(m\)th transmitting wireless access point \(B\) represents the transmission bandwidth, \(P\) m and \(P\) bt,m respectively represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the \(m\)th transmitting wireless access point, \(P\) max,m Denote the maximum transmit power of the \(m\)th transmitting wireless access point, \(\xi\) k Represent the spectral efficiency requirement of each \(k\)th communication user, \(\|\cdot\|\) represents the Euclidean norm, Denote the Hadamard product;

[0161] Such as Figure 3 As shown, the specific steps are as follows:

[0162] 4.1 Initialization: Set any initial power value \(\rho\) (0) \(\geq0\) and the convergence accuracy \(\epsilon>0\), \(\epsilon\) χ \(>0\), and the penalty coefficient \(\lambda>0\), let The iteration index \(i = 0\) and the maximum number of iterations \(i\) max \(= 100\);

[0163] 4.2 Start the loop \(i\leq i\) max ;

[0164] 4.3 \(i = i + 1\);

[0165] 4.4 Use \(\rho\) (i) To solve the above optimization problem to obtain \(\rho\) (i+1) ;

[0166] 4.5 If \(\chi<\epsilon\) χ , in the next loop let \(\chi = 0\);

[0167] 4.6 If End the loop and execute 4.8;

[0168] 4.7 End the loop;

[0169] 4.8 If \(\chi\neq0\), then the problem is considered infeasible;

[0170] 4.9 If \(\chi = 0\), output the optimal transmit power;

[0171] Among them, represents the total system power consumption at the i-th iteration;

[0172] In step S5, the branch and bound method is used to solve the optimization problem:

[0173] Define M as the number of transmitting wireless access points, K represents the number of single-antenna communication users, M r represents the number of antennas of the receiving wireless access point, s k represents the communication signal of the k-th communication user, s0 represents the sensing signal, w mk and w m0 respectively represent the precoding vectors of the m-th transmitting wireless access point with respect to the k-th communication user and the sensing signal, ρ mk and ρ m0 respectively represent the transmission powers allocated by the m-th transmitting wireless access point to the k-th communication user and the sensing signal, G m is the reflection path from the m-th transmitting wireless access point through the sensing target to the receiving wireless access point, H L,m represents the line-of-sight path between the m-th transmitting wireless access point and the receiving wireless access point, H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting wireless access point and the receiving wireless access point, η represents the sensing signal-to-interference-plus-noise requirement, and the precoding matrix W of the m-th transmitting wireless access point m =[w m1 ,...,w mK ,w m0 , the signal matrix D s [m]=diag(s1,...,s K ,s0), the power allocation vector of the m-th transmitting wireless access point ρ (i) represents the optimal solution at the i-th iteration, represents the sensing channel noise power, P m and P bt,m respectively represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the m-th transmitting wireless access point, and the hardware power consumption generated by the m-th transmitting wireless access point B represents the transmission bandwidth, P max,m represents the maximum transmission power of the m-th transmitting wireless access point, ξ k represents the spectral efficiency requirement of each k-th communication user, ν k represents the signal-to-interference-plus-noise ratio requirement of each k-th communication user, λ is the penalty coefficient, χ represents the slack variable, represents taking the square root, (·) T represents the transpose operation, diag(·) represents the diagonal matrix, and the binary variable α is usedm Indicates the switch of the transmitting wireless access point, and the problem becomes:

[0174]

[0175] Among them, the power vector of the M transmitting wireless access points allocated to the k-th communication user The power vector of the M transmitting wireless access points allocated to the sensing signal The vector of expected signal coefficients of the M transmitting wireless access points with respect to the k-th communication user The vector of inter-user interference coefficients of the M transmitting wireless access points with respect to the k-th communication user The pilot contamination coefficient vector c of the M transmitting wireless access points with respect to the k-th communication user kk′M = [c 1k′k ,..., c Mk′k T , the vector of sensing signal interference coefficients of the M transmitting wireless access points with respect to the k-th communication user The power allocation vector ρ′ of the M transmitting wireless access points = [ρ1,..., ρ M T , the sensing channel coefficients of the M transmitting wireless access points after real number processing The sensing interference channel coefficients of the M transmitting wireless access points after real number processing The sensing interference channel H between the m-th transmitting wireless access point and the receiving wireless access point m = H L,m + H NL,m , the sensing channel matrix A′ of the M transmitting wireless access points = [G1W1D s [1],..., G M W M D s [M]], the sensing interference channel matrix B′ of the M transmitting wireless access points = [H1W1D s [1],..., H M W M D s [M]], the expected signal coefficient a mk = tr(Q mk ) Inter-user interference coefficient Pilot contamination coefficient mk = tr(R mk ) mk R mk represents the spatial correlation matrix of the channel h represents a diagonal matrix of dimension M t Channel estimate Spatial correlation matrix k′ represents the k'-th communication user, represents the user subset using the same pilot sequence as the k-th communication user, represents the subset after removing the k-th communication user from τ p is the length of the pilot sequence, p k represents the pilot transmit power of the k-th communication user, M t represents the number of antennas of the transmitting radio access point, (·) H represents the conjugate transpose operation, represents taking the real part, tr(·) represents the trace of a matrix, ||·|| represents the Euclidean norm, represents the Hadamard product;

[0176] In step S6, the optimal solution of the corresponding convex problem is obtained by using a convex optimization solver:

[0177] The MOSEK solver and the CVX solver are used to solve the above problem to obtain the optimal solution;

[0178] The present invention proposes a power consumption optimization method based on a de-cellular massive multiple-input multiple-output communication and sensing integrated system, which realizes the minimum total power consumption by controlling the switch of the transmitting radio access point and its transmission power; and compares it with the total power consumption when all transmitting radio access points are turned on and only the transmission power of the transmitting radio access point is optimized.

[0179] As Figure 4 shown, regardless of how the number of antennas of each transmitting radio access point changes, the total power consumption required by the proposed algorithm is much less than that of the benchmark algorithm where all transmitting radio access points are turned on and only the transmission power of the transmitting radio access point is optimized.

[0180] Figure 5 It shows that regardless of how the number of antennas of the transmitting access point changes, the number of active transmitting radio access points required by the proposed algorithm is much less than that of the benchmark algorithm where all transmitting radio access points are turned on and only the transmission power of the transmitting radio access point is optimized. And as the number of antennas of each transmitting radio access point increases, the number of active radio access points required by the proposed algorithm becomes less and less.

Claims

1. A power consumption optimization method for a de-cellularized massive multiple-input multiple-output (MIMO) communication and sensing integrated system, characterized in that, It includes the following steps: Step S1: Construct a de-cellularized massive multiple-input multiple-output (mMIMO) communication and sensing integrated system model; The de-cellularized mMIMO communication and sensing integrated system model includes a channel estimation model, a communication and sensing integrated signal model, a communication model, and a sensing model; Step S2: Select a communication and sensing integrated signal precoding scheme; Step S3: Construct a non-convex optimization problem model for the de-cellularized mMIMO communication and sensing integrated system, considering the limited transmit power of access points, and ensuring that the spectral efficiency of communication users and the sensing spectral efficiency meet basic requirements. Minimize the total system power consumption by controlling the wireless access point switches and transmit powers; Step S4: Use the feasible solution tracking-successive convex approximation method to handle non-convex constraints and transform the non-convex problem into a convex problem; Step S5: Use the branch and bound method to solve the optimization problem; Step S6: Use a convex optimization solver to obtain the optimal solution of the corresponding convex problem.

2. The power consumption optimization method of a de-cellularized massive multiple-input multiple-output integrated communication and sensing system according to claim 1, characterized in that In step S1, constructing the de-cellularized mMIMO communication and sensing integrated system model specifically involves: Construct a channel estimation model: Define τ p as the length of the pilot sequence, M as the number of transmitting wireless access points, K as the number of single-antenna communication users, and h mk represents the channel vector between the m-th transmitting wireless access point and the k-th communication user. is the pilot sequence assigned to the k-th communication user, and the signal received by the m-th transmitting wireless access point is: Among them, p k represents the pilot transmission power of the k-th communication user, N m represents additive noise, (·) H represents the conjugate transpose operation, represents taking the square root; Using the minimum mean square error (MMSE) estimation method to estimate the channel. The channel estimate between the m-th transmitting wireless access point and the k-th communication user is: Among them, the pilot signal processed by the wireless access point R mk represents the spatial correlation matrix of the channel h mk and the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t where k′ represents the k'th communication user represents the subset of users using the same pilot sequence as the k'th communication user, and M t represents the number of antennas of the transmitting wireless access point, (·) -1 represents the inverse operation represents the mean value; Channel estimation value The spatial correlation matrix is expressed as: Construct a synesthesia integrated signal model: Define s k to represent the communication signal of the k-th communication user, s0 represents the sensing signal, and ρ mk and ρ m0 respectively represent the transmission power allocated by the m-th transmitting radio access point to the k-th communication user and the sensing signal, w mk and w m0 respectively represent the precoding vectors of the m-th transmitting radio access point with respect to the k-th communication user and the sensing signal. Then the synesthesia integrated signal is defined as: Among them, the precoding matrix W of the m-th transmitting wireless access point m = [w m1 ,..., w mK , w m0 , the signal matrix D s [m] = diag(s1,..., s K , s0), the power allocation vector of the m-th transmitting wireless access point diag(·) represents a diagonal matrix, (·) T represents the transpose operation; Construct a communication model: Define subsets Denote the set of active transmitting wireless access points. The signal received by the k-th communication user is as follows: where n k represents the additive noise at communication user k, and K / {k} represents the set of K communication users excluding the k-th communication user; Then, the signal-to-interference-plus-noise ratio (SINR) of the k-th communication user is expressed as: Among them, is the desired signal, represents the uncertainty gain caused by beamforming, represents the interference between users, represents the interference caused by the sensing signal, while represents the noise power at the communication user k, and ∣·∣ represents the modulus operation; Therefore, the spectral efficiency of the k-th communication user is: Among them, τ c represents the coherence interval length; Construct a perception model: Define G m is the reflection path from the m-th transmitting wireless access point through the perception target to the receiving wireless access point, H L,m represents the line-of-sight path between the m-th transmitting wireless access point and the receiving wireless access point, H NL,m represents the non-line-of-sight path caused by obstacles between the m-th transmitting wireless access point and the receiving wireless access point. The signal received by the receiving wireless access point is expressed as: where n s represents the perceived noise, the sensing channel matrix of the active transmitting wireless access point \(A = [G_1W_1Ds[1],\ldots,G_{|A|}W_{|A|}Ds[|A|]]\), the sensing interference channel matrix of the active transmitting wireless access point \(B = [H_1W_1Ds[1],\ldots,H_{|A|}W_{|A|}Ds[|A|]]\), and the power allocation vector of the active transmitting wireless access point represents a subset and the number of elements of Then, the sensing SINR is expressed as: Among them, M r represents the number of antennas of the receiving wireless access point, represents the perceived channel noise power. Therefore, the sensing spectral efficiency is as follows: In summary, the total spectral efficiency is expressed as:

3. The power consumption optimization method for a de-cellularized massive multiple-input multiple-output integrated communication and sensing system according to claim 1, characterized in that In step S2, select the communication and sensing integrated signal precoding scheme: Let \(M\) be the number of transmitting wireless access points, and \(K\) represent the number of single - antenna communication users. Denote the set of active transmitting wireless access points as \(\varPhi\), and \(\rho\) mk and \(\rho\) m0 respectively represent the transmit power allocated by the \(m\) - th transmitting wireless access point to the \(k\) - th communication user and the sensing signal. Let \(h\) mk denote the channel vector between the \(m\) - th transmitting wireless access point and the \(k\) - th communication user. Let \(\hat{h}\) denote the channel estimate value between the \(m\) - th transmitting wireless access point and the \(k\) - th communication user. Let \(a(\theta\) m ) denote the antenna array response vector of the \(m\) - th transmitting wireless access point, and \(\theta\) m represent the angle from the sensing target to the \(m\) - th transmitting wireless access point. Assume that the precoding of the \(m\) - th transmitting wireless access point with respect to the \(k\) - th communication user is and the precoding of the \(m\) - th transmitting wireless access point with respect to the sensing signal is Let \(\|\cdot\|\) denote the Euclidean norm. Let \(\mathbb{E}\{\cdot\}\) denote the mean. Thus, the effective signal - to - interference - plus - noise ratio expression of the \(k\) - th communication user is derived as: Among them, the desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient pilot contamination coefficient the sensing signal interference coefficient d mk = tr(R mk ), represents the communication user noise power, R mk is the spatial correlation matrix of the channel h mk , the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value represents the subset of users using the same pilot sequence as the k-th communication user, represents the subset excluding the k-th communication user, τ p is the length of the pilot sequence, p k represents the pilot transmission power of the k-th communication user, k' represents the k'-th communication user, M t represents the number of antennas of the transmitting wireless access point, tr(·) represents the trace of a matrix, represents taking the square root.​ 4. The power consumption optimization method of a de-cellularized massive multiple-input multiple-output integrated communication and sensing system according to claim 1, characterized in that In step S3, construct a non-convex optimization problem model for the de-cellularized mMIMO communication and sensing integrated system, considering the limited transmit power of access points, and ensuring that the spectral efficiency of communication users and the sensing spectral efficiency meet basic requirements. Minimize the total system power consumption by controlling the wireless access point switches and transmit powers; Let \(M\) be the number of wireless access points transmitting, and \(K\) denote the number of single - antenna communication users. Denote the set of active wireless access points transmitting. Denote the subset The number of elements of, \(\rho\) mk And \(\rho\) m0 Respectively represent the transmit power allocated by the \(m\) - th transmitting wireless access point to the \(k\) - th communication user and the transmit power of the sensing signal. \(h\) mk Denote the channel vector between the \(m\) - th transmitting wireless access point and the \(k\) - th communication user. Denote the channel estimate value between the \(m\) - th transmitting wireless access point and the \(k\) - th communication user. \(P\) m And \(P\) bt,m Respectively represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the \(m\) - th transmitting wireless access point. \(B\) represents the transmission bandwidth. And Respectively represent the total spectral efficiency, the spectral efficiency of the \(k\) - th communication user, and the sensing spectral efficiency. The problem is: Among them, P max,m represents the maximum transmission power of the m-th transmitting wireless access point, ξ k and γ respectively represent the spectral efficiency requirement and the perceived spectral efficiency requirement of each k-th communication user. The power allocation vector of the m-th transmitting wireless access point (·) T represents the transpose operation, ||·|| represents the Euclidean norm, represents taking the square root; τ c is expressed as the coherence interval length, τ p is the length of the pilot sequence. Let the signal-to-interference-plus-noise ratio requirement and the sensing signal-to-interference-plus-noise ratio requirement of each k-th communication user be such that the problem becomes: SINR (s) ≥η, Among them, represents the signal-to-interference-plus-noise ratio of the k-th communication user, SINR (s) represents the perceived signal-to-interference-plus-noise ratio, and the hardware power consumption generated by the m-th transmitting radio access point For the non-convex constraints in the problem, transform the non-convex SINR constraint of communication users into a second-order cone form: Among them, denotes the noise power at the communication user k, and the power vector allocated by the active transmitting wireless access point to the k-th communication user The power vector allocated by the active transmitting wireless access point to the sensing signal The desired signal coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The inter-user interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The pilot contamination coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The sensing signal interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient The pilot contamination coefficient The sensing signal interference coefficient d mk = tr(R mk ), R mk denotes the spatial correlation matrix of the channel h mk , the inverse matrix of the minimum mean square error estimator denotes a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value k′ represents the k'-th communication user, denotes the subset of users using the same pilot sequence as the k-th communication user, p k denotes the pilot transmission power of the k-th communication user, denotes the subset excluding the k-th communication user, M t denotes the number of antennas of the transmitting wireless access point, tr(·) represents the trace of a matrix, denotes the Hadamard product.

5. The power consumption optimization method of a de-cellular massive multiple-input multiple-output integrated communication and sensing system according to claim 1, wherein In step S4, use the feasible solution tracking-successive convex approximation method to handle non-convex constraints and transform the non-convex problem into a convex problem: Define \(M\) as the number of transmitting wireless access points, and \(K\) as the number of single-antenna communication users. Denote the set of active transmitting wireless access points as Denote the number of elements in the subset as \(M\). r Denote the number of antennas of the receiving wireless access point as \(s\). k Denote the communication signal of the \(k\)th communication user as \(s_k\), \(s_0\) as the sensing signal, and \(w\). mk and \(w\). m0 Denote the precoding vectors of the \(m\)th transmitting wireless access point with respect to the \(k\)th communication user and the sensing signal as \(\mathbf{w}_{mk}\) and \(\mathbf{w}_{m0}\) respectively, and \(\rho\). mk and \(\rho\). m0 Denote the transmit powers allocated by the \(m\)th transmitting wireless access point to the \(k\)th communication user and the sensing signal as \(\rho_{mk}\) and \(\rho_{m0}\) respectively, and \(h\). mk Denote the channel vector between the \(m\)th transmitting wireless access point and the \(k\)th communication user as \(\mathbf{h}_{mk}\). Denote the channel estimate between the \(m\)th transmitting wireless access point and the \(k\)th communication user as \(\mathbf{G}_{mk}\). m Denote the reflected path from the \(m\)th transmitting wireless access point through the sensing target to the receiving wireless access point as \(\mathbf{H}_{m1}\), and \(H\). L,m Denote the line-of-sight path between the \(m\)th transmitting wireless access point and the receiving wireless access point as \(\mathbf{H}_{m2}\), and \(H\). NL,m Denote the non-line-of-sight path caused by obstacles between the \(m\)th transmitting wireless access point and the receiving wireless access point as \(\mathbf{H}_{m3}\), \(\eta\) as the sensing signal-to-interference-plus-noise requirement, and the precoding matrix \(\mathbf{W}_m\) of the \(m\)th transmitting wireless access point m \(= [\mathbf{w}_{m1},...,\mathbf{w}_{mK},\mathbf{w}_{m0}]\), the signal matrix \(\mathbf{D}[m] = \text{diag}(s_1,...,s_K,s_0)\), the sensing interference channel \(\mathbf{H}_m\) between the \(m\)th transmitting wireless access point and the receiving wireless access point m1 ,...,\mathbf{w}_{mK} mK ,\mathbf{w}_{m0} m0 \), s \(\mathbf{D}[m] = \text{diag}(s_1,...,s_K,s_0)\), the sensing interference channel \(\mathbf{H}_m\) between the \(m\)th transmitting wireless access point and the receiving wireless access point K ,s_0)\), m \(= \mathbf{H}_{m1} + \mathbf{H}_{m3}\), L,m +\mathbf{H}_{m3}\), NL,m The power allocation vector \(\mathbf{\rho}_m\) of the \(m\)th transmitting wireless access point The sensing channel matrix \(\mathbf{A}\) of the active transmitting wireless access points \(= [\mathbf{G}_1\mathbf{W}_1\mathbf{D}[1]\mathbf{s}[1],...,\mathbf{G}_{|\mathcal{A}|}\mathbf{W}_{|\mathcal{A}|}\mathbf{D}[|\mathcal{A}|]\mathbf{s}[|\mathcal{A}|]]\), the sensing interference channel matrix \(\mathbf{B}\) of the active transmitting wireless access points \(= [\mathbf{H}_1\mathbf{W}_1\mathbf{D}[1]\mathbf{s}[1],...,\mathbf{H}_{|\mathcal{A}|}\mathbf{W}_{|\mathcal{A}|}\mathbf{D}[|\mathcal{A}|]\mathbf{s}[|\mathcal{A}|]]\). Denote the sensing channel noise power as \(\sigma^2\). Denote taking the square root as \(\sqrt{(\cdot)}\), \((\cdot)^T\) represents the transpose operation, \(\text{diag}(\cdot)\) represents the diagonal matrix. For the non-convex sensing signal-to-interference-plus-noise ratio constraint, the feasible solution tracking-successive convex approximation method is used to handle it, and it is transformed into: T \((\cdot)^T\) represents the transpose operation, \(\text{diag}(\cdot)\) represents the diagonal matrix. For the non-convex sensing signal-to-interference-plus-noise ratio constraint, the feasible solution tracking-successive convex approximation method is used to handle it, and it is transformed into: where χ≥0 represents the slack variable, ρ (i) represents the optimal solution of the i-th iteration, and the sensed channel coefficient after the active transmitting wireless access point takes real numbers The sensed interference channel coefficient after the active transmitting wireless access point takes real numbers The power allocation vector of the active transmitting wireless access point (·) H represents the conjugate transpose operation, represents taking the real part; Definition is the noise power at communication user k, and the power vector allocated by the active transmitting wireless access point to the k-th communication user The power vector allocated by the active transmitting wireless access point to the sensing signal The desired signal coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The inter-user interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The pilot contamination coefficient vector of the active transmitting wireless access point with respect to the k-th communication user The sensing signal interference coefficient vector of the active transmitting wireless access point with respect to the k-th communication user Desired signal coefficient a mk = tr(Q mk ), the inter-user interference coefficient Pilot contamination coefficient Sensing signal interference coefficient d mk = tr(R mk ), R mk represents the spatial correlation matrix of the channel h mk The inverse matrix of the minimum mean square error estimator Represents a diagonal matrix of dimension M t The spatial correlation matrix of the channel estimate value Represents the subset of users using the same pilot sequence as the k-th communication user, Represents the subset excluding the k-th communication user, τ p is the length of the pilot sequence, p k represents the pilot transmission power of the k-th communication user, k' represents the k'-th communication user, M t represents the number of antennas of the transmitting wireless access point, ν k represents the signal-to-interference-plus-noise ratio requirement for each k-th communication user, tr(·) represents the trace of the matrix, then the problem becomes:​ χ≥0 where λ > 0 is the penalty coefficient, P total represents the total system power consumption, and the hardware power consumption generated by the m-th transmitting wireless access point B represents the transmission bandwidth, P m and P bt,m represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the m-th transmitting wireless access point, respectively, P max,m represents the maximum transmission power of the m-th transmitting wireless access point, ξ k represents the spectral efficiency requirement of each k-th communication user, ||·|| represents the Euclidean norm, represents the Hadamard product; The specific steps are: 4.1 Initialization: Set an arbitrary initial power value ρ (0) ≥ 0 and a convergence accuracy ∈ > 0, ∈ χ > 0, and a penalty coefficient λ > 0, and let the iteration index i = 0 and the maximum number of iterations i max = 100; 4.2 Start the loop where i ≤ i max ; 4.3 i = i + 1; 4.4 Solve using ρ (i) to solve the above optimization problem to obtain ρ (i+1) ; 4.5 If χ < ∈ χ , in the next loop, let χ = 0; 4.6 If End the loop and execute 4.8; 4.7 End the loop; 4.8 If χ≠0, it is considered that the problem is infeasible; 4.9 If χ = 0, output the optimal transmit power; Among them, represents the total system power consumption at the i-th iteration.

6. The power consumption optimization method of a de-cellularized massive multiple-input multiple-output (MIMO) communication and sensing integrated system according to claim 1, wherein In step S5, use the branch and bound method to solve the optimization problem: Define \(M\) as the number of transmitting wireless access points, \(K\) represents the number of single-antenna communication users, and \(M\) r represents the number of antennas of the receiving wireless access point, \(s\) k represents the communication signal of the \(k\)th communication user, \(s_0\) represents the sensing signal, \(w\) mk and \(w\) m0 represent the precoding vectors of the \(m\)th transmitting wireless access point with respect to the \(k\)th communication user and the sensing signal, respectively. \(\rho\) mk and \(\rho\) m0 represent the transmit powers allocated by the \(m\)th transmitting wireless access point to the \(k\)th communication user and the sensing signal, respectively. \(G\) m is the reflection path from the \(m\)th transmitting wireless access point through the sensing target to the receiving wireless access point, \(H\) L,m represents the line-of-sight path between the \(m\)th transmitting wireless access point and the receiving wireless access point, \(H\) NL,m represents the non-line-of-sight path caused by obstacles between the \(m\)th transmitting wireless access point and the receiving wireless access point. \(\eta\) represents the sensing signal-to-interference-plus-noise requirement. The precoding matrix \(W\) of the \(m\)th transmitting wireless access point m \( = [w\) m1 ,\(\cdots\), \(w\) mK , \(w\) m0 , the signal matrix \(D\) s [m] \(=\text{diag}(s_1,\cdots,s\) K , \(s_0)\), the power allocation vector of the \(m\)th transmitting wireless access point \(\rho\) (i) represents the optimal solution of the \(i\)th iteration, represents the sensing channel noise power, \(P\) m and \(P\) bt,m represent the power consumption unrelated to information transmission and the power consumption related to information transmission of the \(m\)th transmitting wireless access point, respectively. The hardware power consumption generated by the \(m\)th transmitting wireless access point \(B\) represents the transmission bandwidth, \(P\) max,m represents the maximum transmit power of the \(m\)th transmitting wireless access point, \(\xi\) k represents the spectral efficiency requirement of each \(k\)th communication user, \(\nu\) k represents the signal-to-interference-plus-noise ratio requirement of each \(k\)th communication user, \(\lambda\) is the penalty coefficient, \(\chi\) represents the slack variable, represents taking the square root, \((\cdot)\) T represents the transpose operation, \(\text{diag}(\cdot)\) represents the diagonal matrix. Use the binary variable \(\alpha\) m to represent the switch of the transmitting wireless access point. The problem becomes: χ≥0 Among them, the power vector allocated by M transmitting wireless access points to the k-th communication user The power vector allocated by M transmitting wireless access points to the sensing signal The desired signal coefficient vector of M transmitting wireless access points with respect to the k-th communication user The inter-user interference coefficient vector of M transmitting wireless access points with respect to the k-th communication user The pilot contamination coefficient vector c of M transmitting wireless access points with respect to the k-th communication user kk′M =[c 1k′k ,...,c Mk′k T , the sensing signal interference coefficient vector of M transmitting wireless access points with respect to the k-th communication user The power allocation vector ρ′ of M transmitting wireless access points = [ρ1,..., ρ M T , the sensing channel coefficients of M transmitting wireless access points after taking real numbers The sensing interference channel coefficients of M transmitting wireless access points after taking real numbers The sensing interference channel H between the m-th transmitting wireless access point and the receiving wireless access point m =H L,m +H NL,m , the sensing channel matrix A′ of M transmitting wireless access points = [G1W1D s [1],..., G M W M D s [M]], the sensing interference channel matrix B′ of M transmitting wireless access points = [H1W1D s [1],..., H M W M D s [M]], the desired signal coefficient a mk =tr(Q mk ), the inter-user interference coefficient Pilot contamination coefficient Sensing signal interference coefficient d mk =tr(R mk ), R mk represents the spatial correlation matrix of the channel h mk , the inverse matrix of the minimum mean square error estimator represents a diagonal matrix of dimension M t , the spatial correlation matrix of the channel estimate value k′ represents the k'-th communication user, represents the subset of users using the same pilot sequence as the k-th communication user,​​​ Indicates a subset in which the k-th communication user is removed, τ p is the length of the pilot sequence, p k represents the pilot transmission power of the k-th communication user, M t represents the number of antennas of the transmitting wireless access point, (·) H represents the conjugate transpose operation, represents taking the real part, tr(·) represents the trace of a matrix, ||·|| represents the Euclidean norm, represents the Hadamard product.

7. The power consumption optimization method of a de-cellularized massive multiple-input multiple-output communication and sensing integrated system according to claim 1, characterized in that In step S6, use a convex optimization solver to obtain the optimal solution of the corresponding convex problem: Use the MOSEK solver and the CVX solver to solve the above problem to obtain the optimal solution.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the steps of the method described in claim 1.

9. A computer-readable storage medium having computer programs / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method described in claim 1 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method described in claim 1 are implemented.

Images