Power control method and device, equipment and storage medium

Abstract

The invention discloses a power control method and device, equipment and a storage medium, relates to the technical field of communication, and aims to solve the problem that power control cannot be performed on an asynchronous transmission system in the prior art. The method comprises the following steps: acquiring an MMSE channel estimation result in a CF mMIMO-OFDM system under the influence of an asynchronous receiving effect; determining an AP cluster providing service for the UE according to the MMSE channel estimation result; establishing a power control model of a downlink according to the MMSE channel estimation result and the AP cluster; and solving the power control model of the downlink by adopting a power control optimization problem solving network to obtain a power control result of the downlink.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a power control method, device, equipment and storage medium. Background Art

[0002] In order to improve the spectrum efficiency of the communication system, the prior art generally uses a power control method to control the downlink communication. However, the power control method provided by the prior art is based on the synchronous transmission system setting and cannot be applied to asynchronous transmission systems such as cell-free massive multiple input multiple output-Orthogonal Frequency Division Multiplexing (CF mMIMO-OFDM) systems. Summary of the invention

[0003] The embodiments of the present application provide a power control method, apparatus, device and storage medium to solve the problem that the prior art cannot perform power control on an asynchronous transmission system.

[0004] In a first aspect, an embodiment of the present application provides a power control method, including:

[0005] Obtain the minimum mean square error (MMSE) channel estimation results in the non-cellular multiple-input multiple-output (OFDM) orthogonal frequency division multiplexing (CF) mMIMO-OFDM system under the influence of asynchronous reception effect;

[0006] Determine, according to the MMSE channel estimation result, an access point AP cluster providing services for a user equipment UE, wherein the AP cluster includes more than one AP;

[0007] Establishing a downlink power control model according to the MMSE channel estimation result and the AP cluster;

[0008] A pre-trained power control optimization problem solving network is used to solve the power control model of the downlink to obtain a power control result of the downlink.

[0009] Optionally, the MMSE channel estimation result is AP l and UE k In coherent block B r MMSE channel estimation results

[0010] Among them, AP l is the lth AP, l∈L, L is the number of APs in the CF mMIMO-OFDM system;

[0011] Among them, the UE k is the k-th UE, k ∈ K, where K is the number of single-antenna UEs in the CF mMIMO-OFDM system;

[0012] Among them, B r is the N r -th coherence block, N r ∈ N R , and N R is the number of coherence blocks for the complete time-frequency resource partitioning in the CF mMIMO-OFDM system.

[0013] Optionally, the MMSE channel estimation result of the AP l and the UE k in the coherence block B r is: is:

[0014]

[0015] Among them, n is the n-th subcarrier, n ∈ N, where N is the number of subcarriers in the CF mMIMO-OFDM system;

[0016] Among them, N sub is the number of consecutive subcarriers included in the B r ;

[0017] Among them, is the channel model of the AP l and the UE k in the coherence block B r , is the large-scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k , is the small-scale fading of the AP l and the UE k , and M is the number of antennas of the AP;

[0018] Among them, Y l is the signal sent by the UE l and received asynchronously by the AP r on the B k , is the normalized pilot signal of the UE k , Φ k is the pilot sequence sent by the UE k , τ pis the length of the pilot sequence.

[0019] Optionally, the AP l asynchronously receives on the B r the signal Y sent by the UE k which is: l is:

[0020]

[0021] where is the phase offset caused by the AP l asynchronously receiving the signal sent by the UE on the coherence block B r θ k is the phase offset caused by the signal sent by the UE l,k,n is the asynchronous phase shift of the AP l and the UE k on the nth subcarrier is the asynchronous phase shift of the AP l and the UE k on the (n + τ p -1)th subcarrier, w l is the received noise, p k is the transmit power of the UE k ;

[0022] Then, the MMSE channel estimation result of the AP l and the UE k on the coherence block B r is: is:

[0023]

[0024] where

[0025] where Λ l,k = λ l,k I M ,

[0026] Optionally, the AP cluster that provides services for the user equipment UE determined according to the MMSE channel estimation result includes:

[0027] When the UE is UE k , according to the AP l and the UE k the MMSE channel estimation result on the coherence block B r obtain the AP l and the UE kThe average channel gain therebetween, where l ∈ [0, L - 1];

[0028] Based on the average channel gain, determine the AP cluster that provides services for the UE k

[0029] Optionally, the power control model for the downlink is:

[0030]

[0031]

[0032] where N is the number of subcarriers in the CF mMIMO - OFDM system;

[0033] where γ k,n is the signal - to - interference - plus - noise ratio SINR modeling of the UE k on the nth subcarrier, and the γ k,n is obtained according to the MMSE channel estimation result and the AP cluster;

[0034] where p max is the maximum transmit power of the AP.

[0035] Optionally, the power control optimization problem solving network is a multi - agent deep deterministic policy gradient MADDPG network.

[0036] In a second aspect, an embodiment of the present application provides a power control device, including:

[0037] A first module, configured to obtain the minimum mean square error MMSE channel estimation result in a cell - free multiple - input multiple - output orthogonal frequency - division multiplexing CFmMIMO - OFDM system under the influence of asynchronous reception effects;

[0038] A second module, configured to determine, according to the MMSE channel estimation result, an access point AP cluster that provides services for a user equipment UE, where the AP cluster includes more than one AP;

[0039] A third module, configured to establish a downlink power control model according to the MMSE channel estimation result and the AP cluster;

[0040] A fourth module, configured to use a pre - trained power control optimization problem solving network to solve the downlink power control model and obtain a downlink power control result.

[0041] Optionally, the MMSE channel estimation result is for the AP l and the UE k in the coherence block B r ​MMSE channel estimation result

[0042] Among them, AP l is the l-th AP, l ∈ L, and L is the number of APs in the CF mMIMO - OFDM system;

[0043] Among them, UE k is the k-th UE, k ∈ K, and K is the number of single - antenna UEs in the CF mMIMO - OFDM system;

[0044] Among them, B r is the N r th coherence block, N r ∈ N R and N R is the number of coherence blocks for the complete time - frequency resource partitioning in the CF mMIMO - OFDM system.

[0045] Optionally, the MMSE channel estimation result of the AP l and UE k in the coherence block B r is: For:

[0046]

[0047] Among them, n is the n - th sub - carrier, n ∈ N, and N is the number of sub - carriers in the CF mMIMO - OFDM system;

[0048] Among them, N sub is the number of consecutive sub - carriers included in the B r ;

[0049] Among them, is the channel model of the AP l and the UE k in the coherence block B r , is the large - scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k , is the small - scale fading of the AP l and the UE k , and M is the number of antennas of the AP;

[0050] Among them, Y l is the AP l in the B rThe UE receives asynchronous k The signal sent, For the UE k The normalized pilot signal, Φ k For the UE k The pilot sequence sent, τ p is the length of the pilot sequence.

[0051] Optionally, the AP l In the B r The UE receives asynchronous k The signal Y sent l for:

[0052]

[0053] in, For the AP l In the coherent block B r Asynchronous reception on the UE k The phase shift caused by the transmitted signal, θ l,k,n For the AP l and the UE k The asynchronous phase shift at the nth subcarrier, For the AP l and the UE k At the n+τth p -1 subcarrier asynchronous phase shift, w l To receive noise, p k For the UE k The transmission power;

[0054] Then, the AP l and UE k In coherent block B r MMSE channel estimation results for:

[0055]

[0056] in,

[0057] Among them, Λ l,k =λ l,k I M ,

[0058] Optionally, the second module is further configured to: k In the case of AP l and UE k In coherent block B rMMSE channel estimation results, and obtain the AP l and the UE k The average channel gain between them, where l ∈ [0, L - 1]. According to the average channel gain, determine the AP cluster that provides services for the UE k

[0059] Optionally, the power control model for the downlink is:

[0060]

[0061]

[0062] where N is the number of subcarriers in the CF mMIMO - OFDM system;

[0063] where γ k,n is the signal - to - interference - plus - noise ratio SINR modeling of the UE k on the nth subcarrier, and the γ k,n is obtained according to the MMSE channel estimation results and the AP cluster;

[0064] where p max is the maximum transmit power of the AP.

[0065] Optionally, the power control optimization problem solving network is a multi - agent deep deterministic policy gradient MADDPG network.

[0066] In a third aspect, an embodiment of the present application provides a power control device, including: a processor and a transceiver;

[0067] The processor is configured to, according to the received communication data, obtain the minimum mean - square - error MMSE channel estimation results in the CF mMIMO - OFDM system under the influence of asynchronous reception effect, determine the access point AP cluster that provides services for the UE according to the MMSE channel estimation results, where the AP cluster includes more than one AP, establish a downlink power control model according to the MMSE channel estimation results and the AP cluster, and use a pre - trained power control optimization problem solving network to solve the downlink power control model to obtain the downlink power control result;

[0068] The transceiver is configured to receive the communication data.

[0069] Optionally, the MMSE channel estimation results are the MMSE channel estimation results of the AP l and the UE k in the coherence block B r ​​

[0070] Among them, AP l is the l-th AP, l ∈ L, and L is the number of APs in the CF mMIMO - OFDM system;

[0071] Among them, UE k is the k-th UE, k ∈ K, and K is the number of single - antenna UEs in the CF mMIMO - OFDM system;

[0072] Among them, B r is the N r -th coherence block, N r ∈ N R , and N R is the number of coherence blocks for the complete time - frequency resource partition in the CF mMIMO - OFDM system.

[0073] Optionally, the MMSE channel estimation result of the AP l and UE k in the coherence block B r is: For:

[0074]

[0075] Among them, n is the n - th sub - carrier, n ∈ N, and N is the number of sub - carriers in the CF mMIMO - OFDM system;

[0076] Among them, N sub is the number of consecutive sub - carriers included in the B r ;

[0077] Among them, is the channel model of the AP l and the UE k in the coherence block B r , is the large - scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k , is the small - scale fading of the AP l and the UE k , and M is the number of antennas of the AP;

[0078] Among them, Y l is the signal sent by the UE l and received asynchronously by the AP r on the B k , For the UE k 's normalized pilot signal, Φ k For the UE k The pilot sequence sent, τ p Is the length of the pilot sequence.

[0079] Optionally, the AP l Asynchronously receives on the B r The signal Y sent by the UE k Is: l As follows:

[0080]

[0081] Wherein, For the AP l On the coherent block B r Asynchronously receives the phase shift caused by the signal sent by the UE k , θ l,k,n For the AP l And the UE k The asynchronous phase shift on the nth subcarrier, For the AP l And the UE k The asynchronous phase shift on the n + τ p -1 subcarrier, w l Is the received noise, p k For the UE k 's transmit power;

[0082] Then, the AP l And UE k The MMSE channel estimation result on the coherent block B r Is: As follows:

[0083]

[0084] Wherein,

[0085] Wherein, Λ l,k = λ l,k I M ,

[0086] Optionally, the processor is further configured to, when the UE is UE k , according to the AP l And UE k The MMSE channel estimation result on the coherent block B r , obtain the AP l And the UEk The average channel gain therebetween, where l ∈ [0, L - 1], and based on the average channel gain, it is determined for the UE k The AP cluster that provides services.

[0087] Optionally, the power control model for the downlink is:

[0088]

[0089]

[0090] where N is the number of subcarriers in the CF mMIMO - OFDM system;

[0091] where γ k,n is for the UE k models the signal - to - interference - plus - noise ratio SINR on the nth subcarrier, and the γ k,n is obtained according to the MMSE channel estimation result and the AP cluster;

[0092] where p max is the maximum transmit power of the AP.

[0093] Optionally, the power control optimization problem solving network is a multi - agent deep deterministic policy gradient MADDPG network.

[0094] In a fourth aspect, an embodiment of the present application provides a communication device, including: a transceiver, a memory, a processor, and a program stored on the memory and executable on the processor; the processor is configured to read the program in the memory to implement the steps in the above - mentioned power control method.

[0095] In a fifth aspect, an embodiment of the present application provides a readable storage medium for storing a program, and when the program is executed by a processor, it implements the steps in the above - mentioned power control method.

[0096] In the embodiment of the present application, the MMSE channel estimation result is obtained based on the influence of the asynchronous reception effect, so that the power control based on the MMSE channel estimation result provided by this embodiment can be applied to asynchronous transmission systems such as CF mMIMO - OFDM systems; since the AP cluster that provides services for the UE is centered on the UE and determined according to the above - mentioned MMSE channel estimation result, the technical solution provided by this embodiment can, to a certain extent, mitigate the asynchronous reception effect, thereby enhancing the scalability of power control for the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] Figure 1 is a flowchart of the power control method provided by an embodiment of the present application;

[0098] Figure 2 is Figure 1 The diagram of the change of the MADDPG network training reward in the power control method provided by the embodiment of the present application shown in the figure;

[0099] Figure 3 is Figure 1 The schematic diagram of the average subcarrier spectral efficiency changing with the number of AP antennas in the comparison between the power control method provided by the embodiment of the present application and the solution provided by the prior art shown in the figure;

[0100] Figure 4 is Figure 1 The schematic diagram of the average subcarrier spectral efficiency changing with the SNR in the comparison between the power control method provided by the embodiment of the present application and the solution provided by the prior art shown in the figure;

[0101] Figure 5 One of the schematic diagrams of the structure of the power control device provided by the embodiment of the present application;

[0102] Figure 6 Another schematic diagram of the structure of the power control device provided by the embodiment of the present application. Detailed implementation manners

[0103] In the embodiment of the present application, the term "and / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0104] In the embodiment of the present application, the term "a plurality of" means two or more, and other quantifiers are similar thereto.

[0105] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0106] The power control method, device, equipment, and storage medium provided in this embodiment are applied in a CF mMIMO - OFDM system.

[0107] To enable those skilled in the art to have a clearer understanding and recognition of the technical solutions provided in this embodiment, in the following embodiments, a CF mMIMO-OFDM system including: L geographically distributed access points (APs), K single-antenna user equipments (UEs), and each AP is equipped with M antennas will be taken as an example for illustration.

[0108] According to the CF mMIMO-OFDM system described above, the distances between the L geographically distributed APs and the UEs are not the same. When communicating between each AP and the UE, there will be a propagation delay difference, thus generating an asynchronous reception effect.

[0109] See Figure 1 , Figure 1 is a flowchart of the power control method provided in an embodiment of the present application. As Figure 1 shown, the power control method provided in an embodiment of the present application, applied in the CF mMIMO-OFDM system described above, may include the following steps:

[0110] Step 101: Obtain the Minimum Mean Square Error (MMSE) channel estimation result in the CF mMIMO-OFDM system under the influence of the asynchronous reception effect.

[0111] In the CF mMIMO-OFDM system, the channel remains unchanged within a coherence block, and the channels between different coherence blocks are independent of each other. Based on the above channel characteristics, in this embodiment, Step 101 performs MMSE channel estimation in units of coherence blocks, so as to achieve the purpose of obtaining the MMSE channel estimation result under the influence of the asynchronous reception effect.

[0112] In this embodiment, the MMSE channel estimation result described in Step 101 may specifically be:

[0113] AP l and UE k The MMSE channel estimation result in coherence block B r is

[0114] where, AP l is the l-th AP, l ∈ L; UE k is the k-th UE, k ∈ K; B r is the N-th r coherence block, N r ∈ N R , N R is the number of coherence blocks in the complete time-frequency resource division in the CF mMIMO-OFDM system.

[0115] In this embodiment, the AP l and the UE k in the coherent block B r MMSE channel estimation result can be specifically obtained through the following formula (1):

[0116]

[0117] where n is the nth subcarrier, n ∈ N, and N is the number of subcarriers in the CF mMIMO - OFDM system; N sub is the number of consecutive subcarriers included in B r ; is the channel model of the AP l and the UE k in the coherent block B r ; is the large - scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k ; is the small - scale fading of the AP l and the UE k ; Y l is the signal sent by the UE l received by the AP asynchronously on B r , k is the normalized pilot signal of the UE , Φ k is the pilot sequence sent by the UE k , τ k is the length of the pilot sequence p .

[0118] It should be noted that, in order for the UE (such as UE k ) to be able to send a complete pilot sequence within a coherent block (such as B r ), in this embodiment, τ p = N sub , and the pilot sequence of the CF mMIMO - OFDM system is an orthogonal pilot sequence where Φ i is the i - th orthogonal pilot sequence, and ||Φ i || 2 = τ p .

[0119] Furthermore, in this embodiment, the signal sent by the UE l received by the AP asynchronously on B r k ​The transmitted signal Y l can be expressed as the following formula (2):

[0120]

[0121] where is the AP l asynchronously receives the phase shift, θ, caused by the signal transmitted by the UE r in the coherence block B k is the asynchronous phase shift of the AP l,k,n and the UE l on the nth subcarrier k is the asynchronous phase shift of the AP and the UE l on the (n + τ k - 1)th subcarrier, w p is the received noise, p l is the transmit power of the UE k k l .

[0122] In this embodiment, each component of w l can follow a Gaussian distribution with a mean of 0 and a variance of σ 2 ; where T is the length of an OFDM symbol, and Δt l,k is the propagation time delay difference from the AP l to the UE k .

[0123] Substituting the above formula (2) into formula (1), the MMSE channel estimation result of the AP l and the UE k in the coherence block B r can be obtained as shown in the following formula (3):

[0124]

[0125] where

[0126] where Λ l,k = λ l,k I M ,

[0127] According to the above formula (3), it can be seen that during the process of estimating the CSI under the influence of the asynchronous reception effect, an asynchronous phase shift component is introduced, and the channel estimation error is independent of the channel estimation value , and they respectively follow where Cl,k = Λ l,k -B l,k 。

[0128] Step 102: Determine the AP cluster that provides services to the UE according to the MMSE channel estimation result.

[0129] In this embodiment, the AP cluster includes more than one AP.

[0130] In this embodiment, step 102 may include: When the UE is the UE k case, according to the AP l and UE k in the coherent block B r MMSE channel estimation result, obtain the average channel gain between the AP l and UE k where l ∈ [0, L - 1]; Determine the AP cluster that provides services to the UE according to the average channel gain. k provide services.

[0131] In this embodiment, the average channel gain between the AP l and UE k can be expressed as: Among them, Specifically, can be calculated according to obtained.

[0132] In this embodiment, step 102 can select The higher Z APs form the AP cluster that provides services to the UE k It should be noted that the specific number of Z is not limited in this embodiment. In actual use, it can be set according to actual needs and will not be elaborated here.

[0133] It should be noted that in this embodiment, the same AP can belong to different AP clusters and provide services to different UEs.

[0134] In this embodiment, step 102 takes the UE as the center and performs dynamic clustering processing on the APs, so as to achieve the purpose of reducing the asynchronous reception effect to a certain extent and enhancing the system scalability.

[0135] Step 103: Establish a power control model for the downlink according to the MMSE channel estimation result and the AP cluster.

[0136] In this embodiment, the power control model for the downlink can be as shown in the following formula (4):

[0137]

[0138] Among them, γk,n For UE k Modeling the signal to interference plus noise ratio (SINR) on the nth subcarrier, γ k,n According to the MMSE channel estimation results and AP cluster acquisition; max The maximum transmit power of the AP.

[0139] In this embodiment, γ k,n The acquisition methods may include:

[0140] First, the UE on the nth subcarrier is obtained by the following formula (5): k The received signal:

[0141]

[0142] Among them, x l,k,n is the downlink transmission symbol; p l,k For AP l Assigned to UE k The transmission power, For UE k The received noise has zero variance. The complex Gaussian distribution of w l,k,n ∈C M×1 Maximum Ratio Transmission (MRT) precoding for units;

[0143] Secondly, according to the UE k The received signal is obtained by the following formula (6): k,n :

[0144]

[0145] Get γ k,n Afterwards, the CF mMIMO-OFDM system spectrum efficiency can be further obtained by the following formula (7):

[0146]

[0147] Where χ=(τ c -τ e ) / τ c , each coherent block contains τ c samples, the uplink training phase and the downlink data transmission phase occupy τ e and τ c -τ e samples.

[0148] Step 104: Use a pre-trained power control optimization problem-solving network to solve the power control model for the downlink and obtain the power control result for the downlink.

[0149] According to the above formula (7), it can be seen that the power control model for the downlink established in this embodiment is an optimization problem that maximizes the system spectral efficiency subject to the maximum transmit power constraint of the AP. This optimization problem is a non-convex optimization problem, and the computational complexity increases exponentially with the increase in the number of UEs and APs. In addition, due to the difficulty in obtaining an accurate environmental transmission model, model-based methods such as dynamic programming are not applicable to solving such problems.

[0150] To solve the above problems, in this embodiment, the power control optimization problem-solving network is specifically a Multi-agent Deep Deterministic Policy Gradient (MADDPG) network. Of course, this MADDPG network is only a specific example in this embodiment. In actual use, other power control optimization problem-solving networks that can solve the above technical problems can also be used, which will not be elaborated here one by one.

[0151] In this embodiment, deep reinforcement learning can be studied through a Markov decision process, which can be characterized by the tuple (S, A, P, R, ζ). S and A represent the state and action spaces respectively. P represents the transition probability matrix. R and ζ represent the reward and the discount factor of the reward respectively. At a certain moment, agent j observes the environmental state s j executes action a j , calculates the obtained reward as r j , and at the same time the environmental state transfers to s j '. Define (s j , a j , r j , s j ') as the transfer information of agent j. Store the transfer information of all agents in the replay buffer for subsequent network training.

[0152] The MADDPG network adopts an actor-critic architecture for centralized training and distributed execution. Agent j is equipped with an actor network for policy evaluation and a critic network for value evaluation The actor network of agent j observes its own state, selects an action according to the policy, and updates its own network weights through the policy gradient The critic network needs to obtain the state and action information of all agents to globally evaluate the value function. To ensure the stability of the value function, the output value of the critic network should be as close as possible to the output value of the target critic network. Therefore, the weights of the critic network can be updated by minimizing the loss function of the output values of the two networks. In addition, the target actor network and the target critic network are softly updated regularly to improve the stability of network learning.

[0153] The CF mMIMO-OFDM system is regarded as the environment, and each AP is regarded as an agent. The state information observed by the AP is the channel state information and the asynchronous phase shift matrix. The action of the AP is the optimization variable in the optimization problem, representing the transmit power allocated by the AP to each user. The reward of the AP is set as the system spectral efficiency. The network is trained iteratively multiple times to obtain the optimal network parameters.

[0154] Exemplarily, it is assumed that there are 6 APs and 3 UEs in a scenario with a radius of 500m, and the position distributions of the APs and UEs follow a PPP distribution. The total number of subcarriers in the system is 64. Figure 2 As shown, during the process of using MADDPG to solve the power control optimization problem, the curve of the average reward changing with the number of iterations. In the initial stage of training, since there is less transfer information in the replay buffer, it is difficult for the network to learn a better strategy to obtain a higher reward. Subsequently, when there is sufficient transfer information, the network gradually learns the optimal strategy and converges. When the number of iterations is about 1200, the network training converges, and at this time, the maximum spectral efficiency of the system is 121.22 bits / s / Hz.

[0155] In the embodiment of the present application, the MMSE channel estimation result is obtained based on the influence of the asynchronous reception effect, so that the power control based on this MMSE channel estimation result provided by this embodiment can be applied to asynchronous transmission systems such as CF mMIMO-OFDM systems; since the AP cluster serving the UE is centered on the UE and determined according to the above MMSE channel estimation result, the technical solution provided by this embodiment can, to a certain extent, reduce the asynchronous reception effect, thereby enhancing the scalability of the power control to the system.

[0156] Exemplarily, the average power allocation scheme is selected as the scheme provided by the prior art, that is, the AP allocates equal power to the served UE. As Figure 3The figure showing the average spectral efficiency per subcarrier varying with the number of AP antennas. As the number of antennas increases, the spectral efficiencies of both the solution provided by the prior art and the technical solution provided by the embodiments of the present application increase. Moreover, the technical solution provided by the embodiments of the present application can achieve a higher spectral efficiency, and the performance gap is getting larger. When the number of antennas is 12, the spectral efficiency of the technical solution provided by the embodiments of the present application reaches 4.12 bit / s / Hz, which is about 2.4 times that of the solution provided by the prior art. As Figure 4 The figure showing the average spectral efficiency per subcarrier varying with SNR. SNR is defined as the ratio of the maximum transmit power of the AP on each subcarrier to the noise power. Figure 4 and Figure 3 show similar changing trends. When SNR is 90 dB, the spectral efficiency of the technical solution provided by the embodiments of the present application reaches 6.99 bit / s / Hz, which is about 2.3 times that of the solution provided by the prior art. Therefore, the technical solution provided by the embodiments of the present application can improve the spectral efficiency of the CF mMIMO - OFDM system.

[0157] See Figure 5 , Figure 5 is a schematic structural diagram of a power control device provided by an embodiment of the present application. The power control device 500 may include:

[0158] A first module 501, configured to obtain the minimum mean square error (MMSE) channel estimation result in a cell - free multiple - input multiple - output orthogonal frequency - division multiplexing (CF mMIMO - OFDM) system under the influence of asynchronous reception effects;

[0159] A second module 502, configured to determine an access point (AP) cluster that provides services to a user equipment (UE) according to the MMSE channel estimation result, where the AP cluster includes more than one AP;

[0160] A third module 503, configured to establish a power control model for the downlink according to the MMSE channel estimation result and the AP cluster;

[0161] A fourth module 504, configured to solve the power control model for the downlink by using a pre - trained power control optimization problem solving network to obtain a power control result for the downlink.

[0162] Optionally, the MMSE channel estimation result is the MMSE channel estimation result of the AP l and the UE k in the coherence block B r of the MMSE channel estimation result

[0163] where the AP lis the l-th AP, where l ∈ L and L is the number of APs in the CF mMIMO-OFDM system;

[0164] where UE k is the k-th UE, where k ∈ K and K is the number of single-antenna UEs in the CF mMIMO-OFDM system;

[0165] where B r is the N r -th coherence block, where N r ∈ N R , and N R is the number of coherence blocks for the complete time-frequency resource partitioning in the CF mMIMO-OFDM system.

[0166] Optionally, the MMSE channel estimation result of the AP l and UE k in coherence block B r is:

[0167]

[0168]

[0169] where n is the n-th subcarrier, n ∈ N and N is the number of subcarriers in the CF mMIMO-OFDM system;

[0170] where N sub is the number of consecutive subcarriers included in B r

[0170] where is the channel model of the AP l and the UE k in coherence block B r , is the large-scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k , is the small-scale fading of the AP l and the UE k , and M is the number of antennas of the AP;

[0171] where Y l is the signal sent by the UE l and received by the AP r asynchronously on B k , is the UE kThe normalized pilot signal, Φ k is the pilot sequence sent by the UE k τ p is the length of the pilot sequence.

[0172] Optionally, the AP l asynchronously receives on the B r the signal Y sent by the UE k which is: l where

[0173]

[0174] where is the phase offset caused by the AP l asynchronously receiving the signal sent by the UE on the coherent block B r θ k is the asynchronous phase shift of the AP l,k,n and the UE l on the nth subcarrier k φ is the asynchronous phase shift of the AP l and the UE k on the (n + τ p - 1)th subcarrier, w l is the received noise, p k is the transmit power of the UE k ;

[0175] Then, the MMSE channel estimation result of the AP l and the UE k on the coherent block B r is: where

[0176]

[0177] where

[0178] where Λ l,k = λ l,k I M ,

[0179] Optionally, the second module 502 is further configured to, when the UE is UE k , obtain the AP l and the UE k according to the MMSE channel estimation result of the AP r and the UE l on the coherent block B kThe average channel gain therebetween, where l ∈ [0, L - 1], and based on the average channel gain, it is determined for the UE k The AP cluster that provides services to the UE

[0180] Optionally, the power control model for the downlink is as follows:

[0181]

[0182]

[0183] Where N is the number of subcarriers in the CF mMIMO - OFDM system;

[0184] Where γ k,n Is the signal - to - interference - plus - noise ratio SINR modeling for the UE k On the nth subcarrier, and the γ k,n Is obtained according to the MMSE channel estimation result and the AP cluster;

[0185] Where p max Is the maximum transmit power of the AP.

[0186] Optionally, the network for solving the power control optimization problem is a multi - agent deep deterministic policy gradient MADDPG network.

[0187] The specific implementation of the power control device described in this embodiment can be referred to the power control method provided in this embodiment shown above, and will not be elaborated here.

[0188] In the embodiments of the present application, the MMSE channel estimation result is obtained based on the influence of the asynchronous reception effect, so that the power control based on this MMSE channel estimation result provided in this embodiment can be applied to asynchronous transmission systems such as CF mMIMO - OFDM systems; since the AP cluster that provides services to the UE is centered on the UE and determined according to the above - mentioned MMSE channel estimation result, the technical solution provided in this embodiment can, to a certain extent, mitigate the asynchronous reception effect, thereby enhancing the scalability of power control for the system.

[0189] See Figure 6 , Figure 6 Is a schematic structural diagram of the power control device provided in the embodiments of the present application, which may include: a processor 601 and a transceiver 602;

[0190] The processor 601 is configured to obtain the minimum mean square error (MMSE) channel estimation result in the CF mMIMO-OFDM system under the influence of the asynchronous reception effect according to the received communication data, determine an access point (AP) cluster that provides services to the UE according to the MMSE channel estimation result, where the AP cluster includes more than one AP, establish a power control model for the downlink according to the MMSE channel estimation result and the AP cluster, and use a pre-trained power control optimization problem solving network to solve the power control model for the downlink to obtain the power control result for the downlink;

[0191] The transceiver 602 is configured to receive the communication data.

[0192] Optionally, the MMSE channel estimation result is the MMSE channel estimation result of the AP l and the UE k in the coherence block B r where the AP

[0193] is the l-th AP, l ∈ L, and L is the number of APs in the CF mMIMO-OFDM system; l where the UE

[0194] is the k-th UE, k ∈ K, and K is the number of single-antenna UEs in the CF mMIMO-OFDM system; k where B

[0195] is the N-th coherence block, N r is the N r -th coherence block, N r ∈ N R and N R is the number of coherence blocks for the complete time-frequency resource division in the CF mMIMO-OFDM system.

[0196] Optionally, the MMSE channel estimation result of the AP l and the UE k in the coherence block B r is: where n is the n-th subcarrier, n ∈ N, and N is the number of subcarriers in the CF mMIMO-OFDM system;

[0197]

[0198] where N

[0199] is the number of consecutive subcarriers included in B sub ; r where

[0200] where For the AP l and the UE k in the channel model of the coherent block B r For the AP l and the UE k large-scale fading, d l,k For the AP l and the UE k distance between them For the AP l and the UE k small-scale fading, M is the number of antennas of the AP;

[0201] Among them, Y l For the AP l asynchronously received on the B r signal sent by the UE k For the UE k normalized pilot signal, Φ k For the UE k pilot sequence sent, τ p is the length of the pilot sequence.

[0202] Optionally, the signal Y l asynchronously received by the AP r on the B k sent by the UE is l as follows:

[0203]

[0204] Among them, is the phase shift caused by the signal sent by the UE l asynchronously received by the AP r on the coherent block B k θ l,k,n is the asynchronous phase shift of the AP l and the UE k on the nth subcarrier, is the asynchronous phase shift of the AP l and the UE k on the (n + τ p - 1)th subcarrier, w l is the received noise, p k is the transmit power of the UE k

[0205] Then, the AP l and the UE k ​​​In the coherent block B r MMSE channel estimation result is as follows:

[0206]

[0207] wherein,

[0208] wherein, Λ l,k = λ l,k I M ,

[0209] Optionally, the processor 601 is further configured to, when the UE is UE k , obtain the average channel gain between the AP l and the UE k in the coherent block B r according to the MMSE channel estimation result, where l ∈ [0, L - 1], and determine an AP cluster that provides services for the UE l and the UE k according to the average channel gain. k

[0210] Optionally, the power control model for the downlink is:

[0211]

[0212]

[0213] where N is the number of subcarriers in the CF mMIMO - OFDM system;

[0214] where γ k,n models the signal - to - interference - plus - noise ratio SINR of the UE k on the nth subcarrier, and the γ k,n is obtained according to the MMSE channel estimation result and the AP cluster;

[0215] where p max is the maximum transmit power of the AP.

[0216] Optionally, the power control optimization problem solving network is a multi - agent deep deterministic policy gradient MADDPG network.

[0217] The specific implementation of the power control device described in this embodiment can be referred to the power control method provided in this embodiment shown above, and details are not described herein again.

[0218] ​In the embodiments of the present application, the MMSE channel estimation result is obtained based on the influence of the asynchronous reception effect, so that the power control based on the MMSE channel estimation result provided in this embodiment can be applied to asynchronous transmission systems such as CFmMIMO-OFDM systems. Since the AP cluster serving the UE is centered on the UE and determined according to the above MMSE channel estimation result, the technical solution provided in this embodiment can, to a certain extent, mitigate the asynchronous reception effect, thereby enhancing the scalability of power control for the system.

[0219] The embodiments of the present application also provide a communication device, including: a transceiver, a memory, a processor, and a program stored on the memory and executable on the processor. When the processor executes the program, the steps in the above-mentioned power control method are implemented.

[0220] It should be noted that the division of units in the embodiments of the present application is illustrative, only a logical function division, and there may be other division methods in actual implementation. In addition, in each embodiment of the present application, the functional units can be integrated in one processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0221] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0222] The embodiments of the present application further provide a readable storage medium, on which a program is stored. When the program is executed by a processor, it implements each process of the power control method embodiment described above and can achieve the same technical effects. To avoid repetition, it will not be elaborated here. Among them, the readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic memories (such as floppy disks, hard disks, magnetic tapes, magneto-optical discs (MO), etc.), optical memories (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor memories (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NAND FLASH), solid state drives (SSD)).

[0223] It should be noted that in this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitations, the element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0224] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, disk, optical disc) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.

[0225] The embodiments of the present application have been described above in conjunction with the accompanying drawings, but the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them fall within the protection scope of the present application.

Claims

1. A power control method, characterized in that, Including: Obtain the minimum mean square error (MMSE) channel estimation result in a cell-free multiple-input multiple-output orthogonal frequency division multiplexing (CF mMIMO-OFDM) system under the influence of asynchronous reception effects. Determine an access point (AP) cluster that provides services to a user equipment (UE) according to the MMSE channel estimation result, where the AP cluster includes more than one AP. Establish a power control model for the downlink according to the MMSE channel estimation result and the AP cluster. Use a pre-trained power control optimization problem solving network to solve the power control model for the downlink and obtain the power control result for the downlink.

2. The method according to claim 1, wherein The MMSE channel estimation result is AP l and UE k in coherence block B r of the MMSE channel estimation result Among them, AP l is the l-th AP, where l ∈ L, and L is the number of APs in the CF mMIMO-OFDM system wherein, the UE k is the k-th UE, k ∈ K, and K is the number of single-antenna UEs in the CF mMIMO-OFDM system; Among them, B r is the N r th coherent block, where N r ∈ N R , and N R is the number of coherent blocks for the complete time-frequency resource partitioning in the CF mMIMO-OFDM system.

3. The method according to claim 2, wherein The AP l and the UE k in the coherent block B r MMSE channel estimation result is as follows: Where n is the nth subcarrier, n ∈ N, and N is the number of subcarriers in the CF mMIMO-OFDM system. Among them, N sub is the number of consecutive subcarriers included in the B r ; Among them, is the channel model of the AP l and the UE k in the coherence block B r ; is the large-scale fading of the AP l and the UE k , d l,k is the distance between the AP l and the UE k ; is the small-scale fading of the AP l and the UE k , and M is the number of antennas of the AP; Wherein, Y l is the signal asynchronously received by the AP l on the B r from the UE k sent; is the normalized pilot signal of the UE k , Φ k is the pilot sequence sent by the UE k , τ p is the length of the pilot sequence.

4. The method according to claim 3, characterized in that The said AP l The said UE r asynchronously received on the said B k sends signal Y l is: Among them, is the said AP l asynchronously receives, on the said coherent block B r the phase offset, θ, caused by the signal sent by the said UE k is the asynchronous phase shift of the said AP l，k，n and the said UE l on the nth subcarrier k is the asynchronous phase shift of the said AP and the said UE l on the (n + τ k - 1)th subcarrier, w p is the received noise, p l is the transmission power of the said UE k k ;​ Then, the AP l and the UE k in the coherence block B r MMSE channel estimation result is as follows: Among them, where, Λ l，k = λ l，k I M , 5. The method according to claim 2, wherein The determining an access point (AP) cluster that provides services to a user equipment (UE) according to the MMSE channel estimation result includes: In the case where the UE is UE k According to the AP l and UE k In coherent block B r Based on the MMSE channel estimation result, obtain the average channel gain between the AP l and the UE k where l ∈ [0, L - 1]; Determine, according to the average channel gain, the AP cluster that serves the UE k ​ 6. The method according to claim 2, characterized in that, The power control model for the downlink is: Where N is the number of subcarriers in the CF mMIMO-OFDM system. where γ k，n models the signal-to-interference-plus-noise ratio (SINR) of the UE k on the n-th subcarrier, and the γ k，n is obtained according to the MMSE channel estimation result and the AP cluster where p max is the maximum transmit power of the AP.

7. The method according to claim 1, characterized in that, The power control optimization problem solving network is a multi-agent deep deterministic policy gradient (MADDPG) network.

8. A power control device, characterized in that, Including: A first module for obtaining the minimum mean square error (MMSE) channel estimation result in a cell-free multiple-input multiple-output orthogonal frequency division multiplexing (CF mMIMO-OFDM) system under the influence of asynchronous reception effects. A second module for determining an access point (AP) cluster that provides services to a user equipment (UE) according to the MMSE channel estimation result, where the AP cluster includes more than one AP. A third module for establishing a power control model for the downlink according to the MMSE channel estimation result and the AP cluster. A fourth module for using a pre-trained power control optimization problem solving network to solve the power control model for the downlink and obtain the power control result for the downlink.

9. A power control device, comprising: A processor and a transceiver; characterized in that The processor is configured to obtain the minimum mean square error (MMSE) channel estimation result in a CF mMIMO-OFDM system under the influence of asynchronous reception effects according to the received communication data, determine an access point (AP) cluster that provides services to the UE according to the MMSE channel estimation result, where the AP cluster includes more than one AP, establish a power control model for the downlink according to the MMSE channel estimation result and the AP cluster, and use a pre-trained power control optimization problem solving network to solve the power control model for the downlink and obtain the power control result for the downlink. The transceiver is configured to receive the communication data.

10. A communication device, comprising: A transceiver, a memory, a processor, and a program stored on the memory and executable on the processor; characterized in that The processor is configured to read the program in the memory to implement the steps in the power control method according to any one of claims 1-7.

11. A readable storage medium for storing a program, characterized in that, The program, when executed by the processor, implements the steps in the power control method according to any one of claims 1-7.

Images