Communication method, device and system for distributed resources based on unlicensed frequency band
By introducing a multi-beam directional listen-before-speak protocol and deep learning technology into a communication system where 5G NR-U and WiGig coexist, and by optimizing and adjusting physical layer parameters, the problem of poor communication performance on unlicensed frequency bands was solved, the system throughput and efficiency were improved, and adaptive adjustment and performance optimization were achieved.
Patent Information
- Application Number
- CN202411035478.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-29
AI Technical Summary
Existing multi-beam communication systems that coexist with 5G NR-U and WiGig have poor communication performance on unlicensed frequency bands and lack optimization solutions.
By employing a multi-beam directional listen-before-speak protocol and deep learning technology, physical layer parameters are optimized and adjusted through a communication performance prediction model. This establishes a mapping relationship between the actual network performance indicators of user equipment and the target network performance indicators, thereby optimizing the physical layer parameters of the communication system.
It improves the throughput performance and data transmission efficiency of the communication system, enables the system to adapt and optimize its performance, and enhances the user experience and system flexibility.
Smart Images

Figure CN119052819A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless communication technology, and in particular to a communication method, device and system based on distributed resources in an unlicensed frequency band. Background Art
[0002] In the current digital social environment, with the rapid expansion of distributed energy resources and the increase in the number of terminal devices, there is a more urgent need for efficient and reliable wireless communications. As a key energy transmission and management infrastructure, the power grid urgently needs innovative communication technologies to support efficient energy data transmission and management. As a representative of future communication technology, the fifth generation mobile communication technology (5G) will play an important role in the power grid field. The continuous evolution of 5G technology marks that wireless communications are entering a new era, bringing faster data transmission speeds, lower latency and greater network capacity to power grid communications.
[0003] At the same time, as a high-speed wireless communication technology, the short-distance and high-speed data transmission characteristics of WiGig (Wireless Gigabit) provide strong support for short-distance transmission scenarios in power grids. The application of WiGig technology can not only provide fast and reliable communication connections, but also is expected to promote the further development of fields such as smart home devices in power grids. Therefore, the coexistence of the new radio in unlicensed spectrum (NR-U) technology working in the unlicensed frequency band in 5G and the high-speed and short-distance WiGig brings a wider range of spectrum resources and diversified deployment options to the power grid communication system to support the access and data transmission needs of new energy equipment in the power grid. Considering the limitations of the licensed frequency band, the introduction of the coexistence of unlicensed frequency bands and WiGig has become an inevitable choice to promote innovation in power grid communication technology.
[0004] Although multi-beam directional transmission technology is considered to be an important means to improve the communication efficiency and throughput of power grids, there is currently no communication performance optimization solution for multi-beam communication systems in which the new air interface technology NR-U and high-speed short-distance WiGig coexist in 5G operating in unlicensed frequency bands, resulting in poor communication performance of multi-beam communication systems where NR-U and WiGig coexist. Summary of the invention
[0005] In order to overcome the problem of poor communication performance of the multi-beam communication system in which NR-U and WiGig coexist, the present invention provides a communication method, device and system based on distributed resources in an unlicensed frequency band.
[0006] On the one hand, the present invention provides a communication method based on distributed resources in an unlicensed frequency band, which is applied to a target communication system, wherein the target communication system includes a plurality of distributed resource nodes, each distributed resource node corresponds to a target cell, and the plurality of distributed resource nodes include distributed resource nodes based on a new air interface operating in an unlicensed frequency band and distributed resource nodes based on wireless gigabit; the method includes:
[0007] Based on a multi-beam listen-before-talk protocol, each of the distributed resource nodes performs downlink data transmission based on an unlicensed frequency band with multiple associated user equipments through multiple main beams; the associated user equipments are user equipments in a target cell corresponding to the distributed resource node;
[0008] Determining an actual network performance indicator of each user equipment based on a communication parameter of each user equipment corresponding to each distributed resource node in the downlink data transmission;
[0009] Based on the difference between the actual network performance indicators of each user device and the target network performance indicators determined by the communication performance prediction model, the physical layer parameters of the target communication system are optimized and adjusted to increase the transmission data volume of the cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user device under the system's optimal network performance indicators.
[0010] Optionally, the target communication system is a dynamic spectrum sharing system operating in a new air interface and wireless gigabit in an unlicensed frequency band.
[0011] Optionally, the listen-before-talk protocol based on multiple beams, wherein each of the distributed resource nodes performs downlink data transmission based on an unlicensed frequency band with multiple user equipments associated with the distributed resource node through multiple main beams, includes:
[0012] For each distributed resource node, the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1;
[0013] If the channel is detected to be busy, a random backoff mechanism with a non-fixed-length contention window is used for channel access and downlink data transmission;
[0014] If the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission;
[0015] In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
[0016] Optionally, the determining, based on the communication parameters of each user equipment corresponding to each distributed resource node in the downlink signal transmission, an actual network performance indicator of each user equipment includes:
[0017] For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
[0018] Optionally, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under the channel conditions, and is implemented by the following formula:
[0019] S (c,n) =α (c,n) P suc,c T P,c B (c,n) E[log 2 (1+β (c,n) γ (c,n) )] / T ave,c
[0020] Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation. N c represents the number of user devices associated with the distributed resource node c, N B,c represents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n) is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
[0021] Optionally, the physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing and beam width.
[0022] Optionally, before optimizing and adjusting the physical layer parameters of the target communication system based on the difference between the actual network performance indicator of each user equipment and the target network performance indicator determined by the communication performance prediction model, the method further includes:
[0023] Preprocessing current physical layer parameters of the target communication system to obtain input data;
[0024] Inputting the input data into the communication performance prediction model, performing multiple convolution processing and pooling processing through the convolution module of the communication performance prediction model, and obtaining extracted features;
[0025] The extracted features are fully connected to obtain target network performance indicators of each user equipment in the target communication system.
[0026] Optionally, preprocessing current physical layer parameters of the target communication system to obtain input data includes:
[0027] Performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters;
[0028] Normalizing the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features;
[0029] Encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation;
[0030] Extracting statistical features of the cleaned physical layer parameters, including transmit power, multipath propagation parameters, channel delay, beam spacing, and beam width, respectively, to obtain corresponding statistical features;
[0031] Normalizing the vectorized representation and the statistical features to obtain corresponding normalized results;
[0032] The input data is obtained based on the input features and the normalization results.
[0033] Optionally, the training process of the communication performance prediction model includes:
[0034] Acquire historical physical layer parameters of the target communication system and network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples;
[0035] Inputting the training samples into an initial communication performance prediction model for model training to obtain a training output;
[0036] Determine a loss function based on the training output and the network performance indicator data of each user device under the system optimal network performance indicator;
[0037] With the goal of minimizing the loss function, the model parameters are updated through a back propagation algorithm to obtain a trained communication performance prediction model;
[0038] The validation sample is used to adjust the hyperparameters of the trained communication performance prediction model to complete the training process of the communication performance prediction model.
[0039] Optionally, after optimizing and adjusting the physical layer parameters of the target communication system based on the difference between the actual network performance indicator of each user equipment and the target network performance indicator determined by the communication performance prediction model, the method further includes:
[0040] Real-time monitoring of the communication performance of the target communication system to obtain real-time network performance indicators of the target communication system;
[0041] Based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model, the model parameters of the communication performance prediction model are fine-tuned to achieve feedback adjustment of the communication performance prediction model.
[0042] On the other hand, the present invention provides a communication device based on distributed resources in an unlicensed frequency band, the device is applied to a target communication system, the target communication system includes a plurality of distributed resource nodes, each of the distributed resource nodes corresponds to a target cell, and the plurality of distributed resource nodes include distributed resource nodes based on a new air interface operating in an unlicensed frequency band and distributed resource nodes based on wireless gigabit; the device includes:
[0043] A data transmission module, used for performing downlink data transmission based on an unlicensed frequency band with multiple associated user equipments through multiple main beams in a listen-before-talk protocol based on multiple beams; the associated user equipments are user equipments in a target cell corresponding to the distributed resource node;
[0044] A performance indicator determination module, configured to determine an actual network performance indicator of each user equipment based on communication parameters of each user equipment corresponding to each distributed resource node in the downlink data transmission;
[0045] The optimization and adjustment module is used to optimize and adjust the physical layer parameters of the target communication system based on the difference between the actual network performance indicators of each user equipment and the target network performance indicators determined by the communication performance prediction model to increase the transmission data volume of the cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user equipment under the system's optimal network performance indicators.
[0046] Optionally, the target communication system is a dynamic spectrum sharing system operating in a new air interface and wireless gigabit in an unlicensed frequency band.
[0047] Optionally, the data transmission module includes:
[0048] The channel detection submodule is used to detect the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1;
[0049] The channel access submodule is used to use a random backoff mechanism with a non-fixed length contention window for channel access and downlink data transmission if the channel is detected to be busy; if the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission;
[0050] In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
[0051] Optionally, the performance indicator determination module is specifically used to:
[0052] For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
[0053] Optionally, the performance indicator determination module determines the actual throughput of each user equipment by using the following formula:
[0054] S (c,n) =α (c,n) P suc,c T P,c B (c,n) E[log 2 (1+β (c,n) γ(c,n) )] / T ave,c
[0055] Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation. N c represents the number of user devices associated with the distributed resource node c, N B,c represents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n) is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
[0056] Optionally, the physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing and beam width.
[0057] Optionally, the optimization and adjustment module includes:
[0058] A preprocessing submodule, used for preprocessing the current physical layer parameters of the target communication system to obtain input data;
[0059] A feature extraction submodule, used for inputting the input data into the communication performance prediction model, performing multiple convolution and pooling processes through the convolution module of the communication performance prediction model, and obtaining extracted features;
[0060] The fully connected submodule is used to perform fully connected processing on the extracted features to obtain target network performance indicators of each user equipment in the target communication system.
[0061] Optionally, the preprocessing submodule includes:
[0062] A data cleaning subunit, used for performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters;
[0063] A normalization subunit, used to normalize the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features;
[0064] A coding subunit, used for encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation;
[0065] A feature extraction subunit is used to extract statistical features of the transmission power, multipath propagation parameters, channel delay, beam spacing, and beam width in the cleaned physical layer parameters to obtain corresponding statistical features;
[0066] The normalization subunit is further used to perform normalization processing on the vectorized representation and each of the statistical features to obtain corresponding normalization results;
[0067] A splicing subunit is used to obtain the input data based on the input features and each of the normalization results.
[0068] Optionally, the device further comprises a training module, wherein the training module comprises:
[0069] A data acquisition submodule, used to obtain the historical physical layer parameters of the target communication system and the network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples;
[0070] A training submodule, used for inputting the training samples into an initial communication performance prediction model for model training to obtain a training output;
[0071] A loss determination submodule, used to determine a loss function based on the training output and the network performance indicator data of each user equipment under the system optimal network performance indicator;
[0072] A parameter updating submodule is used to update the model parameters by a back propagation algorithm with the goal of minimizing the loss function to obtain a trained communication performance prediction model;
[0073] The hyperparameter adjustment submodule is used to adjust the hyperparameters of the trained communication performance prediction model using the verification sample to complete the training process of the communication performance prediction model.
[0074] Optionally, the device further comprises:
[0075] A monitoring module, used to monitor the communication performance of the target communication system in real time and obtain real-time network performance indicators of the target communication system;
[0076] A feedback adjustment module is used to fine-tune the model parameters of the communication performance prediction model based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model, so as to achieve feedback adjustment of the communication performance prediction model.
[0077] On the other hand, the present invention further provides a communication system, comprising: a communication device based on distributed resources in an unlicensed frequency band as described in any one of the above.
[0078] On the other hand, the present invention also provides a communication device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;
[0079] The memory is used to store one or more programs;
[0080] When the one or more programs are executed by the at least one processor, any one of the above-mentioned communication methods based on distributed resources in an unlicensed frequency band is implemented.
[0081] On the other hand, the present invention further provides a readable storage medium having an execution program stored thereon, and when the execution program is executed, the communication method based on distributed resources in an unlicensed frequency band as described in any one of the above is implemented.
[0082] Compared with the prior art, the present invention has the following beneficial effects:
[0083] The present invention provides a communication method, device and system for distributed resources based on unlicensed frequency bands. On the one hand, the present invention introduces a multi-beam directional listen-before-talk protocol in a communication system where 5G NR-U and WiGig coexist, so that each distributed resource node performs downlink data transmission based on the unlicensed frequency band with multiple user devices associated with the distributed resource node through multiple main beams. Compared with single-beam communication, the throughput performance of the communication system is effectively improved, and the data transmission efficiency is improved. On the other hand, a mapping relationship is established between the physical layer parameters and the target throughput of each user device under the maximum throughput of the system through a communication performance prediction model, and then the physical layer parameters of the target communication system are optimized and adjusted based on the communication performance prediction model, so as to achieve system communication performance optimization and adaptive adjustment, improve system communication performance and user experience, and enable rapid performance evaluation and continuous optimization, thereby improving the flexibility and efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0084] Figure 1 A flow chart of a communication method based on distributed resources in unlicensed frequency bands according to the present invention;
[0085] Figure 2 A schematic diagram of a communication scenario according to an embodiment of the present invention;
[0086] Figure 3 A schematic diagram of the training and feedback mechanism process of the communication performance prediction model of the present invention;
[0087] Figure 4 It is a schematic diagram of the structure of the communication device of the present invention. DETAILED DESCRIPTION
[0088] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.
[0089] Embodiment 1:
[0090] The present invention provides a communication method based on distributed resources in an unlicensed frequency band, which is applied to a target communication system. The method flow chart is as follows: Figure 1 As shown, including:
[0091] Step S110: Based on the multi-beam listen-before-talk protocol, each of the distributed resource nodes performs downlink data transmission based on the unlicensed frequency band with multiple associated user equipments through multiple main beams; the associated user equipment is a user equipment in a target cell corresponding to the distributed resource node, and the target cell is a cell based on a new air interface working in the unlicensed frequency band or a cell based on wireless gigabit;
[0092] Step S120, determining an actual network performance indicator of each user equipment based on the communication parameters of each user equipment corresponding to each distributed resource node in the downlink data transmission;
[0093] Step S130, based on the difference between the actual network performance indicators of each user equipment and the target network performance indicators determined by the communication performance prediction model, optimize and adjust the physical layer parameters of the target communication system to increase the transmission data volume of the cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user equipment under the system's optimal network performance indicators.
[0094] In this embodiment, the target communication system may be a power grid communication system including a plurality of distributed resource nodes, the plurality of distributed resource nodes including distributed resource nodes based on the new air interface operating in the unlicensed frequency band and distributed resource nodes based on the wireless gigabit, and each distributed resource node corresponds to a target cell. The distributed resource node may be a base station node (gNB) or an access point (Access Point, AP). Exemplarily, the target communication system is a dynamic spectrum sharing system operating in the new air interface of the unlicensed frequency band and the wireless gigabit, such as Figure 2As shown, the target communication system includes an NR-U base station / access point and a WiGig base station / access point. Each base station / access point is responsible for the wireless communication of the user (i.e., user equipment) in the corresponding cell. The cell corresponding to the NR-U base station / access point is an NR-U cell, and the cell corresponding to the WiGig base station / access point is a WiGig cell. The communication parameters may be parameters used to indicate communication performance, such as communication duration, probability of successful data transmission, transmission bandwidth, channel conditions, etc. The network performance indicators of each user equipment in the actual communication process are determined by the communication parameters. The communication performance prediction model may be various deep learning models, such as convolutional neural networks, artificial neural networks, etc. The communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user equipment under the system optimal network performance indicators. The actual network performance indicators may be system KPIs (Key Performance Indicators, key performance indicators), and the actual network performance indicators may be actual throughput or actual interference size, etc. The physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing, and beam width. Exemplarily, the communication performance prediction model is used to establish a mapping relationship between system physical layer parameters and target throughput of each user equipment under the maximum throughput of the system.
[0095] The present invention effectively improves the throughput performance of the communication system by introducing the multi-beam directional LBT (ListenBefore Talk) protocol in the 5G NR-U and WiGig dynamic spectrum sharing systems, making data transmission more efficient. The present invention uses deep learning technology to update physical layer parameters, realizes system optimization and adaptive adjustment, improves system performance and user experience, can quickly perform performance evaluation and continuous optimization, and improves the flexibility and efficiency of the system.
[0096] In one implementation, in the multi-beam listen-before-talk protocol in step S110, each of the distributed resource nodes performs downlink data transmission based on an unlicensed frequency band with multiple user equipments associated with the distributed resource node through multiple main beams, including the following process:
[0097] For each distributed resource node, the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1;
[0098] If the channel is detected to be busy, a random backoff mechanism with a non-fixed-length contention window is used for channel access and downlink data transmission;
[0099] If the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission;
[0100] In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
[0101] In this implementation, by proposing a multi-beam directional LBT protocol for downlink data transmission, the communication throughput performance can be improved. NR NR-U system with C W WiGig system with 10 cells (i.e., contention area), and C NR +C W =C tot , C tot is the total number of units in the system. Each NR-U cell or WiGig cell is controlled by a base station node (gNB) or access point (AP) for wireless communication. For a cell c, the gNB or AP in cell c is represented as node c, and the associated user equipment n of cell c is represented as node (c, n). Suppose there are N c For downlink data transmission, all cells share a broadband channel. Assume that at node c, W c,m is the backoff phase m (m=0,1,...,M c ) contention window size, M c It is the maximum backoff stage. As an extension of directional LBT with a single main beam, this example uses the multi-beam directional LBT protocol for downlink data transmission. The specific process is as follows:
[0102] Step 1: Node c adds its N B,c The main beam points to N B,c The system associates user equipment and senses the channel status through energy detection (ED) in the main beam direction.
[0103] Step 2: If any channel is sensed to be busy, node c freezes its backoff counter; otherwise, node c decreases the counter by 1. When the counter is greater than zero, node c returns to step 1; otherwise, node c proceeds to step 3.
[0104] Step 3: Node c uses N B,c The main beam is N B,c associated user equipment sends a payload packet; if NB,c <N c , random selection is used for channel access to support a fair channel access method.
[0105] Step 4: If all transmissions are successful, node c reduces the contention window size to W c,0 Otherwise, double the contention window size until the maximum backoff phase M is reached c , node c returns to the first step.
[0106] This example uses the multi-beam directional LBT protocol combined with a random backoff mechanism of a non-fixed-length contention window to perform downlink data transmission in a communication system where NR-U nodes and WiGig nodes coexist. This can enable the system to have higher spatial multiplexing, better anti-interference, larger transmission range, and better beam selection.
[0107] In another implementation, the step S120 described above, based on the communication parameters of each user equipment corresponding to each distributed resource node in the downlink signal transmission, determining the actual network performance indicator of each user equipment includes the following process:
[0108] For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
[0109] In this implementation, the communication parameters include effective communication duration, probability of successful data transmission, transmission bandwidth and channel conditions (such as signal-to-noise ratio), and the performance of the multi-cell NR-U and WiGig coexistence system is evaluated by actual throughput. Specifically, for downlink transmission from node c, the average throughput of user equipment (c, n) can be expressed by the following formula:
[0110] S (c,n) =α (c,n) P suc,c T P,c B (c,n) E[log 2 (1+β (c,n) γ (c,n) )] / T ave,c
[0111] Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation. N c represents the number of user devices associated with the distributed resource node c, N B,crepresents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n) is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
[0112] In the above formula, α (c,n) It can be estimated by the active duration of the user device (c, n), that is, the proportion of the user device (c, n) that is actually in operation during the entire time period; P suc,c It can be calculated by the success rate of the actual transmission of data packets, and reflects the transmission success of node c in the system; T P,c By recording the time it takes for a node to successfully transmit a data packet, T ave,c It is obtained by recording or estimating the average time required for a slave node to successfully transmit data once.
[0113] In some implementations, by training a communication performance prediction model, a mapping relationship is established between the physical layer parameters of the system and the target network performance indicators of each user equipment under the optimal network performance indicator of the system. The training process of the communication performance prediction model is as follows:
[0114] Acquire historical physical layer parameters of the target communication system and network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples;
[0115] Inputting the training samples into an initial communication performance prediction model for model training to obtain a training output;
[0116] Determine a loss function based on the training output and the network performance indicator data of each user device under the system optimal network performance indicator;
[0117] With the goal of minimizing the loss function, the model parameters are updated through a back propagation algorithm to obtain a trained communication performance prediction model;
[0118] The validation sample is used to adjust the hyperparameters of the trained communication performance prediction model to complete the training process of the communication performance prediction model.
[0119] In this implementation, historical operation data of the target communication system is collected, and the historical operation data includes historical physical layer parameters and corresponding system optimal network performance indicators. Historical physical layer parameters include the number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing, and beam width. The system optimal network performance indicator can be the maximum throughput of the system and the user equipment throughput (sample label) at the maximum throughput. The historical operation data can be preprocessed to ensure the stability of the model training, and the preprocessing can include one or more of normalization, coding vectorization, outlier processing, etc. The preprocessed data can be used as training data, and the training data can be divided into training samples and verification samples in proportion (such as a division ratio of 8:2). The training samples are input into the initial communication performance prediction model for model training, and the loss function (such as a mean square error loss function) is calculated using the training output and the corresponding sample labels. The model parameters (such as weights, biases and other parameters in the model) are updated through a back propagation algorithm with the goal of minimizing the loss function. For example, an optimizer such as the Adam optimization algorithm is used to minimize the loss function to obtain a trained communication performance prediction model. Afterwards, the hyperparameters of the trained communication performance prediction model are adjusted using the verification samples to complete the training process of the communication performance prediction model.
[0120] In some embodiments, the trained communication performance prediction model is applied to an actual system for system optimization, which specifically includes the following process:
[0121] Before optimizing and adjusting the physical layer parameters of the target communication system based on the difference between the actual network performance indicator of each user equipment and the target network performance indicator determined by the communication performance prediction model, the method further includes:
[0122] Preprocessing current physical layer parameters of the target communication system to obtain input data;
[0123] Inputting the input data into the communication performance prediction model, performing multiple convolution processing and pooling processing through the convolution module of the communication performance prediction model, and obtaining extracted features;
[0124] The extracted features are fully connected to obtain target network performance indicators of each user equipment in the target communication system.
[0125] In this implementation, the communication performance prediction model can be a convolutional neural network, which includes an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer, wherein the input layer is used to receive the input data of the model (the result of the current physical layer parameter preprocessing) as input features, the convolution layer is used to extract the spatial information features of the input features, the pooling layer is used to reduce the feature dimension, the fully connected layer is used to map the extracted features to the output layer, and the output layer outputs the throughput prediction results of each user device.
[0126] Exemplarily, preprocessing the current physical layer parameters of the target communication system to obtain input data includes the following process:
[0127] Performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters;
[0128] Normalizing the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features;
[0129] Encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation;
[0130] Extracting statistical features of the cleaned physical layer parameters, including transmit power, multipath propagation parameters, channel delay, beam spacing, and beam width, respectively, to obtain corresponding statistical features;
[0131] Normalizing the vectorized representation and the statistical features to obtain corresponding normalized results;
[0132] The input data is obtained based on the input features and the normalization results.
[0133] In this implementation, since different physical layer parameters have different characteristics, different preprocessing processes are required for different physical layer parameters. Specifically, data cleaning is performed first to remove outliers and supplement missing data; then, the number of antennas, antenna gain, and antenna beam width can be directly used as input features; the modulation code rate is converted into a corresponding digital representation, or the modulation code rate is uniquely encoded to be vectorized so that the model can understand different modulation methods; for parameters such as transmit power, multipath propagation parameters, channel delay, beam spacing, beam width, etc., the corresponding statistical features such as mean, standard deviation, maximum value, minimum value, etc. can be calculated to use statistical features to describe the distribution of each parameter; all physical layer parameters can be normalized, and values in different ranges can be mapped to similar ranges to ensure the stability of model training.
[0134] In some implementations, in order to ensure the prediction accuracy of the model, a feedback mechanism may be established, specifically including the following process:
[0135] Real-time monitoring of the communication performance of the target communication system to obtain real-time network performance indicators of the target communication system;
[0136] Based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model, the model parameters of the communication performance prediction model are fine-tuned to achieve feedback adjustment of the communication performance prediction model.
[0137] In this implementation, the communication performance can be throughput, and the real-time throughput of the monitoring system can be monitored to understand the current operating status and performance of the system through real-time monitoring. The real-time monitored data is compared with the prediction results of the communication performance prediction model, and the feedback adjustment model is analyzed by analyzing the difference between the two. When the difference is large, the model parameters can be retrained with new data to improve the model accuracy and generalization ability. In this way, a continuous improvement cycle is established to achieve continuous optimization and adaptive adjustment of the system.
[0138] The present invention solves the problem of spectrum congestion and low communication efficiency in a power grid environment with diversified distributed resources and a large number of terminals. The present invention combines the new radio technology (NR-U) in the unlicensed band and the high-speed, short-distance WiGig technology, and comprehensively utilizes the multi-beam directional transmission technology and the listen-before-talk (LBT) protocol to develop a multi-technology coexistence multi-beam directional communication system, which can increase the data transmission volume of users and cells. Using deep learning technology to update physical layer parameters to optimize the system can improve the communication efficiency of large urban power grids, solve the challenge of spectrum congestion, promote the flexibility of the power grid and the coordinated interaction of source, grid, load and storage, optimize system performance and maximize the data transmission capacity on the user side.
[0139] The following is a specific implementation method to illustrate the process of using deep learning to update the physical layer parameter optimization system of the present invention. Figure 3 As shown, the specific process is as follows:
[0140] Step 1: Collect historical operating data of the target communication system, including physical layer parameters such as the number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing, beam width, as well as the maximum throughput of the system and the corresponding throughput of each user device.
[0141] Step 2: Preprocess the historical operation data collected in the first step;
[0142] Step 3: Select the convolutional neural network as the communication performance prediction model. The model structure includes input layer, convolution layer, pooling layer, fully connected layer and output layer.
[0143] Step 4: Use the preprocessed data obtained in the second step to train the communication performance prediction model so that it can learn effective parameter optimization strategies to maximize system throughput. The training process uses historical operation data to train the model and updates the model parameters through the back propagation algorithm. For example, the Adam optimizer is applied to minimize the loss function and the validation set is used to adjust the model hyperparameters to prevent overfitting. The loss function can use the mean square error loss function to measure the prediction error. The L1 regularization technology is used to prevent the model from overfitting during the training process.
[0144] Step 5: Use the validation set to adjust the hyperparameters of the model obtained in step 4, such as learning rate, batch size, and activation function selection, to improve the training efficiency and generalization ability of the model; adjust the structure of the convolutional neural network according to actual needs and data characteristics, including increasing or decreasing the number and depth of convolutional layers, pooling layers, and fully connected layers. Based on the experimental results, continuously optimize the number of layers and parameters of the model to find the best model structure to ensure its efficiency and effectiveness in the training and prediction process. Improve the performance and generalization ability of deep learning models by adjusting the model structure and tuning hyperparameters.
[0145] Step 6: Deploy the deep learning model trained in step 5 to the actual target communication system, and monitor and adjust the physical layer parameters in step 1 in real time to optimize the system throughput.
[0146] Step 7: Establish a feedback mechanism to monitor system performance and throughput data in real time to understand the current operating status and performance of the system. Compare the real-time monitored system performance data with the predicted data of the communication performance prediction model, analyze the difference between the model's predicted performance and the actual data, fine-tune or retrain the model parameters, and improve the model's accuracy and generalization ability. In this way, a continuous improvement cycle is established to achieve continuous optimization and adaptability of the system.
[0147] The present invention uses deep learning to update the physical layer parameter optimization system according to network performance indicators, so that the designed unlicensed band NR-U and WiGig dynamic spectrum sharing systems have better performance, and can quickly perform performance evaluation and continuous optimization. The present invention proposes a channel access and downlink data communication method based on a multi-beam directional LBT protocol in an unlicensed band to solve the challenges of spectrum congestion and communication efficiency in a power grid environment with diversified distributed resources and a large number of terminals. By proposing a multi-beam directional listen-before-talk (LBT) protocol to obtain better system performance, and by developing new modeling and analysis methods to optimize system performance, user and regional data transmission volume is increased. The new radio technology (NR-U) and WiGig dynamic spectrum sharing system for unlicensed bands of the present invention have better performance, and can quickly perform performance evaluation and continuous optimization.
[0148] The present invention combines unlicensed frequency band technology and the LBT protocol using multi-beam directional transmission to realize dynamic spectrum sharing of the system and improve the power grid communication performance. At the same time, the introduction of deep learning technology can more accurately model and analyze the power grid communication system, achieve performance optimization and parameter update, explore new possibilities for the efficient development of power grid communication technology, and better support the continuous innovation of the power grid communication system in spectrum resource utilization and communication performance.
[0149] The present invention solves the communication performance problem that occurs when NR-U in the unlicensed frequency band and high-speed short-distance WiGig coexist in the power grid field. By introducing a deep learning model to optimize the system communication performance, the challenge of improving communication throughput can be effectively addressed, thereby promoting the further development of power grid communication technology.
[0150] Embodiment 2:
[0151] Based on the same inventive concept, the present invention also provides a communication device based on distributed resources in an unlicensed frequency band, the device is applied to a target communication system, the target communication system includes a plurality of distributed resource nodes, each of the distributed resource nodes corresponds to a target cell, and the plurality of distributed resource nodes include distributed resource nodes based on a new air interface operating in an unlicensed frequency band and distributed resource nodes based on wireless gigabit; the device includes:
[0152] A data transmission module, used for performing downlink data transmission based on an unlicensed frequency band with multiple associated user equipments through multiple main beams in a listen-before-talk protocol based on multiple beams; the associated user equipments are user equipments in a target cell corresponding to the distributed resource node;
[0153] A performance indicator determination module, configured to determine an actual network performance indicator of each user equipment based on communication parameters of each user equipment corresponding to each distributed resource node in the downlink data transmission;
[0154] The optimization and adjustment module is used to optimize and adjust the physical layer parameters of the target communication system based on the difference between the actual network performance indicators of each user equipment and the target network performance indicators determined by the communication performance prediction model to increase the transmission data volume of the cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user equipment under the system's optimal network performance indicators.
[0155] In a possible implementation manner, the target communication system is a dynamic spectrum sharing system operating in a new air interface in an unlicensed frequency band and wireless gigabit.
[0156] In a possible implementation, the data transmission module includes:
[0157] The channel detection submodule is used to detect the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1;
[0158] The channel access submodule is used to use a random backoff mechanism with a non-fixed length contention window for channel access and downlink data transmission if the channel is detected to be busy; if the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission;
[0159] In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
[0160] In a possible implementation manner, the performance indicator determination module is specifically used to:
[0161] For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
[0162] In a possible implementation manner, the performance indicator determination module determines the actual throughput of each user equipment by using the following formula:
[0163]
[0164] Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation. N c represents the number of user devices associated with the distributed resource node c, N B,c represents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n)is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
[0165] In a possible implementation, the physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing and beam width.
[0166] In a possible implementation, the optimization and adjustment module includes:
[0167] A preprocessing submodule, used for preprocessing the current physical layer parameters of the target communication system to obtain input data;
[0168] A feature extraction submodule, used for inputting the input data into the communication performance prediction model, performing multiple convolution and pooling processes through the convolution module of the communication performance prediction model, and obtaining extracted features;
[0169] The fully connected submodule is used to perform fully connected processing on the extracted features to obtain target network performance indicators of each user equipment in the target communication system.
[0170] In a possible implementation, the preprocessing submodule includes:
[0171] A data cleaning subunit, used for performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters;
[0172] A normalization subunit, used to normalize the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features;
[0173] A coding subunit, used for encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation;
[0174] A feature extraction subunit is used to extract statistical features of the transmission power, multipath propagation parameters, channel delay, beam spacing, and beam width in the cleaned physical layer parameters to obtain corresponding statistical features;
[0175] The normalization subunit is further used to perform normalization processing on the vectorized representation and each of the statistical features to obtain corresponding normalization results;
[0176] A splicing subunit is used to obtain the input data based on the input features and each of the normalization results.
[0177] In a possible implementation, the device further includes a training module, and the training module includes:
[0178] A data acquisition submodule, used to obtain the historical physical layer parameters of the target communication system and the network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples;
[0179] A training submodule, used for inputting the training samples into an initial communication performance prediction model for model training to obtain a training output;
[0180] A loss determination submodule, used to determine a loss function based on the training output and the network performance indicator data of each user equipment under the system optimal network performance indicator;
[0181] A parameter updating submodule is used to update the model parameters by a back propagation algorithm with the goal of minimizing the loss function to obtain a trained communication performance prediction model;
[0182] The hyperparameter adjustment submodule is used to adjust the hyperparameters of the trained communication performance prediction model using the verification sample to complete the training process of the communication performance prediction model.
[0183] In a possible implementation, the device further includes:
[0184] A monitoring module, used to monitor the communication performance of the target communication system in real time and obtain real-time network performance indicators of the target communication system;
[0185] A feedback adjustment module is used to fine-tune the model parameters of the communication performance prediction model based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model, so as to achieve feedback adjustment of the communication performance prediction model.
[0186] Example 3
[0187] Based on the same inventive concept, the present invention further provides a communication system, including the communication device based on the distributed resources of the unlicensed frequency band described in the above embodiment.
[0188] In this embodiment, the communication system may include multiple distributed resource nodes, each distributed resource node corresponds to a target cell, and the multiple distributed resource nodes include distributed resource nodes based on new air interfaces operating in unlicensed frequency bands and distributed resource nodes based on wireless gigabits. The communication device based on distributed resources in unlicensed frequency bands may be deployed on distributed resource nodes.
[0189] Embodiment 4:
[0190] like Figure 4 As shown, the present invention also provides a communication device, which may be a computer device, a single-chip device, an intelligent mobile device, etc. The communication device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, the processor, and the transceiver component are connected via a bus; the memory may be used to store an execution program, and an exemplary execution program may include instructions; the processor is used to execute the instructions stored in the memory. The memory may also be used to store data, which may be called and / or modified when the instructions are executed.
[0191] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in a storage medium to implement corresponding method processes or corresponding functions, so as to implement the steps of a communication method based on distributed resources in an unlicensed frequency band in the above-mentioned embodiment.
[0192] Example 5
[0193] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically a communication device readable storage medium (Memory), which is a memory device in the communication device for storing programs and data. It can be understood that the storage medium here can include both the built-in storage medium in the communication device and the extended storage medium supported by the communication device. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a communication method based on distributed resources in an unlicensed frequency band in the above embodiment.
[0194] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0195] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0196] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0197] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0198] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit its protection scope. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the protection scope of the claims to be approved.
Claims
1. A communication method based on distributed resources in unlicensed frequency bands, characterized in that: Applied to a target communication system, the target communication system includes a plurality of distributed resource nodes, each distributed resource node corresponds to a target cell, and the plurality of distributed resource nodes include distributed resource nodes based on a new air interface operating in an unlicensed frequency band and distributed resource nodes based on wireless gigabit; The method comprises: Based on a multi-beam listen-before-talk protocol, each of the distributed resource nodes performs downlink data transmission based on an unlicensed frequency band with multiple associated user equipments through multiple main beams; the associated user equipments are user equipments in a target cell corresponding to the distributed resource node; Determining an actual network performance indicator of each user equipment based on a communication parameter of each user equipment corresponding to each distributed resource node in the downlink data transmission; Based on the difference between the actual network performance indicators of each user device and the target network performance indicators determined by the communication performance prediction model, the physical layer parameters of the target communication system are optimized and adjusted to increase the transmission data volume of each cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user device under the system's optimal network performance indicators.
2. The method according to claim 1, characterized in that The target communication system is a dynamic spectrum sharing system operating in the new air interface of the unlicensed frequency band and wireless gigabit.
3. The method according to claim 1, characterized in that The listen-before-talk protocol based on multiple beams, wherein each of the distributed resource nodes performs downlink data transmission based on an unlicensed frequency band with multiple user equipments associated with the distributed resource node through multiple main beams, includes: For each distributed resource node, the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1; If the channel is detected to be busy, a random backoff mechanism with a non-fixed-length contention window is used for channel access and downlink data transmission; If the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission; In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
4. The method according to claim 1, characterized in that The determining, based on the communication parameters of each user equipment corresponding to each distributed resource node in the downlink signal transmission, an actual network performance indicator of each user equipment includes: For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
5. The method according to claim 4, characterized in that The actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions, and is implemented by the following formula: S (c,n) =a (c,n) P suc,c T P,c B (c,n) E[log2(1+β (c,n) c (c,n) )] / T ave,c Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation, N c N represents the number of user devices associated with the distributed resource node c. B,c represents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n) is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
6. The method according to claim 5, characterized in that The physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing and beam width.
7. The method according to claim 3, characterized in that Before optimizing and adjusting the physical layer parameters of the target communication system based on the difference between the actual network performance indicator of each user equipment and the target network performance indicator determined by the communication performance prediction model, the method further includes: Preprocessing current physical layer parameters of the target communication system to obtain input data; Inputting the input data into the communication performance prediction model, performing multiple convolution processing and pooling processing through the convolution module of the communication performance prediction model, and obtaining extracted features; The extracted features are fully connected to obtain target network performance indicators of each user equipment in the target communication system.
8. The method according to claim 7, characterized in that Preprocessing the current physical layer parameters of the target communication system to obtain input data includes: Performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters; Normalizing the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features; Encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation; Extracting statistical features of the cleaned physical layer parameters, including transmit power, multipath propagation parameters, channel delay, beam spacing, and beam width, respectively, to obtain corresponding statistical features; Normalizing the vectorized representation and the statistical features to obtain corresponding normalized results; The input data is obtained based on the input features and the normalization results.
9. The method according to claim 7, characterized in that: The training process of the communication performance prediction model includes: Acquire historical physical layer parameters of the target communication system and network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples; Inputting the training samples into an initial communication performance prediction model for model training to obtain a training output; Determine a loss function based on the training output and the network performance indicator data of each user device under the system optimal network performance indicator; With the goal of minimizing the loss function, the model parameters are updated through a back propagation algorithm to obtain a trained communication performance prediction model; The validation sample is used to adjust the hyperparameters of the trained communication performance prediction model to complete the training process of the communication performance prediction model.
10. The method according to claim 7, characterized in that After optimizing and adjusting the physical layer parameters of the target communication system based on the difference between the actual network performance indicator of each user equipment and the target network performance indicator determined by the communication performance prediction model, the method further includes: Real-time monitoring of the communication performance of the target communication system to obtain real-time network performance indicators of the target communication system; Based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model, the model parameters of the communication performance prediction model are fine-tuned to achieve feedback adjustment of the communication performance prediction model.
11. A communication device based on distributed resources in unlicensed frequency bands, characterized in that: Applied to a target communication system, the target communication system includes a plurality of distributed resource nodes, each of the distributed resource nodes corresponds to a target cell, and the plurality of distributed resource nodes include distributed resource nodes based on a new air interface operating in an unlicensed frequency band and distributed resource nodes based on wireless gigabit; the device includes: A data transmission module, used for performing downlink data transmission based on an unlicensed frequency band with multiple associated user equipments through multiple main beams in a listen-before-talk protocol based on multiple beams; the associated user equipments are user equipments in a target cell corresponding to the distributed resource node; A performance indicator determination module, configured to determine an actual network performance indicator of each user equipment based on communication parameters of each user equipment corresponding to each distributed resource node in the downlink data transmission; The optimization and adjustment module is used to optimize and adjust the physical layer parameters of the target communication system based on the difference between the actual network performance indicators of each user equipment and the target network performance indicators determined by the communication performance prediction model to increase the transmission data volume of the cell in the downlink data transmission; the communication performance prediction model is used to establish a mapping relationship between the physical layer parameters and the target network performance indicators of each user equipment under the system's optimal network performance indicators.
12. The device according to claim 11, characterized in that The target communication system is a dynamic spectrum sharing system operating in the new air interface of the unlicensed frequency band and wireless gigabit.
13. The device according to claim 11, characterized in that The data transmission module comprises: The channel detection submodule is used to detect the N B,c The main beam points to N B,c associated user equipment through N B,c Energy detection is performed in the direction of the main beam to determine the channel state, N B,c is an integer greater than 1; The channel access submodule is used to use a random backoff mechanism with a non-fixed length contention window for channel access and downlink data transmission if the channel is detected to be busy; if the channel is not detected to be busy, the distributed resource node uses N B,c main beam and N B,c associated user equipment for downlink data transmission; In the channel access process, if the number of main beams N of the distributed resource node B,c The number of associated user equipments is less than the number of the distributed resource nodes, and random selection is used for channel access.
14. The device according to claim 11, characterized in that The performance indicator determination module is specifically used for: For each of the distributed resource nodes, the actual throughput of each user equipment is determined by considering the effective communication duration, the probability of successful data transmission, the transmission bandwidth and the amount of successfully transmitted data under channel conditions.
15. The device according to claim 13, characterized in that The performance indicator determination module determines the actual throughput of each user equipment by the following formula: S (c,n) =a (c,n) P suc,c T P,c B (c,n) E[log2(1+β (c,n) c (c,n) )] / T ave,c Among them, S (c,n) represents the actual throughput of user equipment n associated with distributed resource node c, α (c,n) represents the effective communication duration of user equipment n associated with distributed resource node c, that is, the proportion of time actually in operation, N c N represents the number of user devices associated with the distributed resource node c. B,c represents the number of main beams of distributed resource node c; P suc,c represents the success transmission probability of distributed resource node c, which is determined based on the success rate of the actual transmission data packet of distributed resource node c; T P,c is the transmission duration of distributed resource node c, E(·) represents the expected value, B (c,n) and γ (c,n) is the bandwidth and received signal-to-noise ratio of the transmission channel formed by the user equipment n associated with the distributed resource node c; β (c,n) represents a gap function of a received signal-to-noise ratio related to a target bit error rate of a transmission channel formed by a user equipment n associated with a distributed resource node c; T ave,c It represents the average time for distributed resource node c to successfully transmit data once.
16. The device according to claim 15, characterized in that The physical layer parameters include at least one of the following: number of antennas, antenna gain, antenna beam width, modulation code rate, transmit power, multipath propagation parameters, channel delay, beam spacing and beam width.
17. The device according to claim 14, characterized in that The optimization and adjustment module comprises: A preprocessing submodule, used for preprocessing the current physical layer parameters of the target communication system to obtain input data; A feature extraction submodule, used for inputting the input data into the communication performance prediction model, performing multiple convolution and pooling processes through the convolution module of the communication performance prediction model, and obtaining extracted features; The fully connected submodule is used to perform fully connected processing on the extracted features to obtain target network performance indicators of each user equipment in the target communication system.
18. The device according to claim 17, characterized in that The preprocessing submodule includes: A data cleaning subunit, used for performing data cleaning on the current physical layer parameters to obtain cleaned physical layer parameters; A normalization subunit, used to normalize the number of antennas, antenna gain, and antenna beam width in the cleaned physical layer parameters to obtain input features; A coding subunit, used for encoding the modulation code rate in the cleaned physical layer parameters to obtain a corresponding vectorized representation; A feature extraction subunit is used to extract statistical features of the transmission power, multipath propagation parameters, channel delay, beam spacing, and beam width in the cleaned physical layer parameters to obtain corresponding statistical features; The normalization subunit is further used to perform normalization processing on the vectorized representation and each of the statistical features to obtain corresponding normalization results; A splicing subunit is used to obtain the input data based on the input features and each of the normalization results.
19. The device according to claim 17, characterized in that The device also includes a training module, which includes: A data acquisition submodule, used to obtain the historical physical layer parameters of the target communication system and the network performance indicator data of each user equipment under the system's optimal network performance indicator as training data, and divide the training data into training samples and verification samples; A training submodule, used for inputting the training samples into an initial communication performance prediction model for model training to obtain a training output; A loss determination submodule, used to determine a loss function based on the training output and the network performance indicator data of each user equipment under the system optimal network performance indicator; A parameter updating submodule is used to update the model parameters by a back propagation algorithm with the goal of minimizing the loss function to obtain a trained communication performance prediction model; The hyperparameter adjustment submodule is used to adjust the hyperparameters of the trained communication performance prediction model using the verification sample to complete the training process of the communication performance prediction model.
20. The device according to claim 17, characterized in that The device also includes: A monitoring module, used to monitor the communication performance of the target communication system in real time and obtain real-time network performance indicators of the target communication system; A feedback adjustment module is used to fine-tune the model parameters of the communication performance prediction model based on the difference between the real-time network performance indicator and the prediction result of the communication performance prediction model to achieve feedback adjustment of the communication performance prediction model.
21. A communication system, characterized in that: A communication device comprising distributed resources based on unlicensed frequency bands as described in any one of claims 11-20.
22. A communication device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the communication method based on distributed resources in an unlicensed frequency band as claimed in any one of claims 1 to 10 is implemented.
23. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, the communication method based on distributed resources of unlicensed frequency band as described in any one of claims 1 to 10 is implemented.
Citation Information
Cited By
Integrated management method and system for shared equipment based on wireless positioning, and medium
CN119815528A
Information timeliness optimization method and device of communication network, equipment and storage medium
CN122602289A
Information timeliness optimization method, device and equipment of communication network and storage medium
CN122602289B
