Methods, devices, equipment, media, and products for collaborative optimization of public and private networks in high-speed rail scenarios.

Abstract

This invention provides a method, apparatus, equipment, medium, and product for collaborative optimization of public and private networks in high-speed rail scenarios. The method includes: determining the network status characteristics of the public and private networks based on acquired data, wherein the acquired data includes at least measurement report data from a first terminal accessing a public network cell and a second terminal accessing a private network cell, and the network status characteristics reflect the network status of the public and private networks; inputting the network status characteristics into a trained power adjustment model, obtaining the power adjustment actions output by the power adjustment model, wherein the power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment actions include power adjustment schemes for each beam of the public network cell; and adjusting the beam power of the public network cell based on the power adjustment actions. The solution provided by this invention does not require allocating independent frequencies for the private network, but rather performs beam-level adjustment of the power of the public network cell in real time based on the actual status of the public and private networks, thereby reducing the interference of the public network on the private network.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to methods, devices, equipment, media, and products for collaborative optimization of public and private networks in high-speed rail scenarios. Background Technology

[0002] High-speed rail is a key scenario for wireless network optimization. High-speed rail users are highly concentrated and trains travel at high speeds. To ensure a good user experience, dedicated network architecture is typically used for base stations covering the high-speed rail line. To ensure coverage and capacity requirements along the line, these dedicated networks often require higher priority. However, public network users also exist in residential areas and along ordinary roads around high-speed rail lines, necessitating the construction of public network base stations as well. When high-speed rail passes through areas with dense public networks, the available frequencies in the guard bands of both the public and dedicated networks are limited. Both networks require numerous carriers simultaneously, leading to severe interference due to frequency overlap, impacting both the public and dedicated networks to varying degrees. For example, a certain operator uses the 2.6GHz band for both its 5G public network and high-speed rail dedicated network. Although the SSB frequencies are staggered, both use a 100MHz bandwidth of 2515-2615MHz for their service channels. When the public network load reaches a certain threshold, public network terminals will cause uplink co-channel interference to dedicated network cells, severely affecting wireless network performance.

[0003] In existing technologies, the most common method to address interference between high-speed rail public and private networks is to allocate dedicated frequencies or add guard bands for high-speed rail private networks. However, radio frequency resources are precious, and radio frequency carriers are limited in densely populated areas. Allocating independent frequencies for private networks is difficult. Therefore, the phenomenon of interference from public networks to high-speed rail private networks remains relatively common in existing technologies. Summary of the Invention

[0004] This invention provides a method, device, equipment, medium, and product for collaborative optimization of public and private networks in high-speed rail scenarios, which can solve the defects of high-speed rail private networks being interfered with by public networks in the prior art and reduce the interference of public networks on high-speed rail private networks.

[0005] This invention provides a method for collaborative optimization of public and private networks in high-speed rail scenarios, comprising:

[0006] The measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell are obtained. Based on the obtained data, the network status characteristics of the public and private networks are determined. The obtained data includes at least the measurement report data of the first terminal and the second terminal. The network status characteristics of the public and private networks reflect the network status of the public network and the private network.

[0007] The public and private network status features are input into the trained power adjustment model to obtain the power adjustment action output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment action includes the power adjustment scheme of each beam of the public network cell.

[0008] The beam power of the public network cell is adjusted based on the power adjustment action.

[0009] According to the present invention, a method for collaborative optimization of public and private networks in a high-speed rail scenario includes determining the network status characteristics of the public and private networks based on acquired data, comprising:

[0010] The beam performance data of the public network cell is determined based on the measurement report data of the first terminal, and the beam performance data reflects the network performance of each beam of the public network cell;

[0011] Beam interference data is determined based on the measurement report data of the second terminal, and the beam interference data reflects the interference situation of the second terminal by each beam of the public network cell;

[0012] The network status characteristics of the public and private networks are determined based on the extracted data, which includes at least the beam performance data and the beam interference data.

[0013] According to the present invention, a method for collaborative optimization of public and private networks in a high-speed rail scenario includes, before determining the network state characteristics based on extracted data, the following steps are taken:

[0014] Obtain the base station performance data of the public network base station corresponding to the public network cell and the private network base station corresponding to the private network cell; combine the beam performance data, the beam interference data and the base station performance data to form the extracted data.

[0015] According to the present invention, a method for collaborative optimization of public and private networks in a high-speed rail scenario, wherein determining beam interference data based on measurement report data from the second terminal includes:

[0016] Based on the measurement report data of the second terminal, determine whether the second terminal is subject to beam interference from the public network cell;

[0017] Obtain the proportion of terminals in the second terminal that are affected by each beam interference of the public network cell, and use each proportion as the beam interference data.

[0018] According to a method for collaborative optimization of public and private networks in a high-speed rail scenario provided by the present invention, the step of determining whether the second terminal is subject to beam interference from the public network cell based on the measurement report data of the second terminal includes:

[0019] When the second terminal accesses the private network cell, if the difference between the RSRP value of the private network cell and the RSRP value of the target beam of the public network cell in the measurement report data of the second terminal is within a preset range, it is determined that the second terminal is being interfered with by the target beam of the public network cell.

[0020] According to the present invention, a method for collaborative optimization of public and private networks in a high-speed rail scenario is provided. When training the power adjustment model based on a reinforcement learning algorithm, the reward function is determined based on the channel quality of the public and private networks after the power adjustment action of the sample output of the power adjustment model is executed.

[0021] This invention also provides a public-private network collaborative optimization device for high-speed rail scenarios, comprising:

[0022] The network status feature acquisition module is used to acquire measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell, and to determine the public and private network status features based on the acquired data. The acquired data includes at least the measurement report data of the first terminal and the second terminal, and the public and private network status features reflect the network status of the public network and the private network.

[0023] The model inference module is used to input the public and private network state features into the trained power adjustment model and obtain the power adjustment action output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment action includes the power adjustment scheme of each beam of the public network cell.

[0024] A power adjustment module is used to adjust the beam power of the public network cell based on the power adjustment action.

[0025] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the high-speed rail scenario public-private network collaborative optimization method as described above.

[0026] The present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the public-private network collaborative optimization method for high-speed rail scenarios as described above.

[0027] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the public-private network collaborative optimization method for high-speed rail scenarios as described above.

[0028] The present invention provides a method, apparatus, equipment, medium, and product for collaborative optimization of public and private networks in high-speed rail scenarios. It determines the network status characteristics of the public and private networks by acquiring data, including measurement report data from a first terminal accessing a public network cell and a second terminal accessing a private network cell. The public and private network status characteristics are input into a power adjustment model trained using a reinforcement learning algorithm. The power adjustment model outputs a beam-level power adjustment scheme for the public network cell. The method provided by this invention does not require allocating independent frequencies for the private network; instead, it adjusts the beam power of the public network cell in real time based on the actual status of the existing public and private networks, thereby reducing interference from the public network to the private network. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0030] Figure 1 This is one of the flowcharts illustrating the collaborative optimization method for public and private networks in high-speed rail scenarios provided by this invention.

[0031] Figure 2 This is one of the scenario diagrams of the public-private network collaborative optimization method for high-speed rail scenarios provided by the present invention.

[0032] Figure 3 This is the second flowchart of the public-private network collaborative optimization method for high-speed rail scenarios provided by this invention.

[0033] Figure 4 This is a schematic diagram of the power adjustment model provided by the present invention.

[0034] Figure 5 This is the second scenario illustration of the public-private network collaborative optimization method for high-speed rail scenarios provided by the present invention.

[0035] Figure 6 This is a schematic diagram of the structure of the public-private network collaborative optimization device for high-speed rail scenarios provided by the present invention.

[0036] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0038] The following is combined with Figures 1-5 This invention describes the public-private network collaborative optimization method for high-speed rail scenarios. For example... Figure 1 As shown, the public-private network collaborative optimization method for high-speed rail scenarios includes the following steps:

[0039] S110. Obtain measurement report data of the first terminal accessing the public network cell and the second terminal accessing the private network cell, and determine the network status characteristics of the public and private networks based on the obtained data. The obtained data includes at least the measurement report data of the first terminal and the second terminal. The network status characteristics of the public and private networks reflect the network status of the public network and the private network.

[0040] S120. Input the public and private network status features into the trained power adjustment model and obtain the power adjustment actions output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment actions include the power adjustment schemes of each beam of the public network cell.

[0041] S130. Adjust the beam power of the public network cell based on the power adjustment action.

[0042] like Figure 2 As shown, along railway lines, both dedicated high-speed rail network cells and public network cells exist. Dedicated high-speed rail network cells provide network access for users moving rapidly on high-speed trains, while public network cells provide network access for mobile users in ordinary scenarios (residential areas, ordinary roads). The public network may interfere with the dedicated network. The method provided by this invention obtains measurement report data from a first terminal accessing a public network cell and a second terminal accessing a dedicated network cell to determine the public and dedicated network state characteristics reflecting their respective network states. This data is then input into a power adjustment model trained using a reinforcement learning algorithm. The power adjustment model outputs a beam-level power adjustment scheme for the public network cell. This method does not require allocating independent frequencies for the dedicated network; instead, it adjusts the beam power of the public network cell in real time based on the actual state of both the public and dedicated networks, thereby reducing interference from the public network to the dedicated network.

[0043] like Figure 3As shown, the method provided by this invention generates public and private network status characteristics reflecting the status of the public and private networks by acquiring measurement report (MR) data reported by a first terminal accessing a public network cell and a second terminal accessing a private network cell. Specifically, determining the public and private network status characteristics based on the acquired data includes:

[0044] The beam performance data of the public network cell is determined based on the measurement report data of the first terminal. The beam performance data reflects the network performance of each beam of the public network cell.

[0045] Beam interference data is determined based on the measurement report data of the second terminal. The beam interference data reflects the interference situation of the second terminal by each beam of the public network cell.

[0046] The characteristics of public and private network status are determined based on the extracted data, which includes at least beam performance data and beam interference data.

[0047] The method provided by this invention describes the beam-level network status of a public network cell from two aspects: beam performance data and beam interference data. Beam performance data reflects the network performance of each beam in the public network cell, while beam interference data reflects the interference of each beam in the public network cell to dedicated networks. This accurately describes the beam-level network status of the public network cell, improving the accuracy of subsequent beam-level power adjustment actions. When a terminal accesses a wireless cell, it periodically reports a Measurement Report (MR). The MR report contains the beam number of the SSB (Synchronization Signal Block) currently accessed by the terminal, the RSRP (Reference Signal Receiving Power) value of the accessed cell, and the RSRP values ​​of each beam in neighboring cells. Beam performance data and beam interference data can be determined using the MR report data from the first and second terminals.

[0048] Specifically, the beam performance data of the public network cell is determined based on the measurement report data from the first terminal, including:

[0049] The number of first terminals accessing each beam in the public network cell and the average RSRP value of the first terminals accessing the cell are obtained as beam performance data.

[0050] Assuming the current network is configured with n SSB beams (generally, n is 8, and the value of n is related to the network configuration and is not limited), then 2n-dimensional beam performance data can be obtained.

[0051] By obtaining the number of first terminals accessing each beam in a public network cell and the average RSRP value of the first terminals accessing the cell, beam performance data that can more accurately reflect the performance of each beam in the public network cell can be obtained.

[0052] It is understandable that, in addition to using RSRP values ​​as an indicator of beam performance, other indicators that can reflect beam performance can also be used in beam performance data.

[0053] Beam interference data is determined based on measurement report data from the second terminal, including:

[0054] Based on the measurement report data from the second terminal, determine whether the second terminal is subject to beam interference from the public network cell;

[0055] Obtain the proportion of terminals in the second terminal that are affected by each beam interference from the public network cell, and use each proportion as beam interference data.

[0056] In the method provided by this invention, the interference situation of the second terminal to each beam of the public network cell is determined by measuring the report data. The interference situation of the public network cell to the private network cell is specified at the beam level, thereby improving the accuracy of the subsequent output beam-level frequency adjustment scheme for the public network cell.

[0057] More specifically, determining whether the second terminal receives beam interference from the public network cell based on the measurement report data of the second terminal includes:

[0058] When the second terminal accesses the private network cell, if the difference between the RSRP value of the private network cell and the RSRP value of the target beam of the public network cell in the measurement report data of the second terminal is within a preset range, it is determined that the second terminal is being interfered with by the target beam of the public network cell.

[0059] The method provided by this invention determines whether the private network is being interfered with by a beam from a neighboring public network cell by comparing the RSRP of the accessed private network cell with the RSRPs of each beam in a neighboring public network cell. Specifically, in the measurement report, the terminal reports the measured RSRP of the currently accessed cell and the RSRPs of each beam in the neighboring cells. When the RSRP of the private network cell measured by a second terminal accessing the private network cell is close to the measured RSRP of a beam in a public network cell, it is considered that the second terminal is being interfered with by that beam from the public network cell.

[0060] In one possible implementation, a preset range of 6dB can be set, meaning that when the difference between the RSRP of the private network cell measured by the second terminal and the RSRP of the target beam of the public network cell measured by the second terminal is within 6dB, it is determined that the second terminal is being interfered with by the target beam of the public network cell.

[0061] For each beam in a public network cell, the number of second terminals affected by that beam can be obtained. Taking the number of second terminals affected by that beam as the numerator and the total number of second terminals as the denominator, a ratio value can be obtained. Since there are n beams, n ratio values ​​can be obtained. The n ratio values ​​form n-dimensional beam interference data.

[0062] As can be seen from the preceding description, based on the measurement report data, data with 3n dimensions can be obtained. In one possible implementation, beam performance data and beam interference data can be directly combined to form the public / private network state characteristics. However, research has found that base station-side performance indicators can also describe the current network state. Therefore, further, in one embodiment of the method provided by this invention, before determining the public / private network state characteristics based on the extracted data, the following steps are included:

[0063] Obtain base station performance data for public network base stations corresponding to public network cells and private network base stations corresponding to private network cells;

[0064] The extracted data is composed of beam performance data, beam interference data, and base station performance data.

[0065] Base station performance data includes base station performance indicators and can also reflect the current network status. Including base station performance data in the public and private network status characteristics can yield more accurate public and private network status characteristics that reflect the current network status, thereby improving the accuracy of frequency adjustment actions generated based on these public and private network status characteristics.

[0066] In one possible implementation, base station performance data may include the number of access users, PDCP (Packet Data Convergence Protocol) layer rate, PDCP layer packet loss count, PDCP layer latency, and PDCP layer latency jitter. There are separate base station performance data for public network cells and private network cells, resulting in a total of 5x2 dimensions of features.

[0067] By merging beam performance data, beam interference data, and base station performance data, a (3n+5x2)-dimensional feature vector describing the current state of the public and private networks is formed. As a characteristic of public and private network status.

[0068] After obtaining the network state characteristics of the public and private networks, they are input into a trained power adjustment model to obtain the power adjustment action output by the model. Based on this power adjustment action, the beam power of the public network cells is adjusted, which can reduce the interference of the private network to the public network. To facilitate model processing, the data in the network state characteristics of the public and private networks can be normalized before being input into the power adjustment model.

[0069] In the method provided by this invention, the power adjustment model is a model trained based on a reinforcement learning algorithm. The reinforcement learning algorithm is a learning algorithm based on the Actor-Critic network, which integrates reinforcement learning with interference from high-speed rail public and private networks. The Actor-Critic network is used to learn the power adjustment, and the network status of public and private networks is fully described from both base station statistics and user reports. Sample labeling is not required, which improves learning efficiency.

[0070] Specifically, the power adjustment model is the Actor network in the Actor-Critic network, i.e., the policy network. It outputs power adjustment actions based on the current network state of the public and private networks. During the reinforcement learning process, the power adjustment model is trained together with the Critic network in the Actor-Critic network. The Critic network outputs a score for the corresponding adjustment action based on the current network state of the public and private networks.

[0071] In the method provided by this invention, the power adjustment model outputs a power adjustment scheme for each beam of the public network cell. Through reinforcement learning, the power adjustment model can learn the impact of each public network beam on the high-speed rail private network, adaptively adjust the power of beams that cause different degrees of interference to realize user migration, and not adjust beams that do not cause interference, thereby maximizing the performance of the public network.

[0072] To improve the output efficiency of the power adjustment model, the method provided in this invention predefines the action space for the power adjustment actions output by the power adjustment model. In other words, the power adjustment actions output by the power adjustment model will only be actions existing within this action space. Specifically, the power adjustment action space is defined. This describes the power adjustment actions of each public network cell beam. One possible implementation is... That is, each beam can independently increase, remain unchanged, or decrease by 1dB, with a motion space. In the diagram, 3n represents a 1dB reduction in the power of the nth beam, 3n+1 represents a unchanged power for the nth beam, and 3n+2 represents a 1dB increase in the power of the nth beam. Here, n is the beam number. The upper limit for power adjustment of each beam within the action space is controlled within ±6dB. By defining a certain range of action space, the power adjustment model can output power adjustment actions within a predetermined range, reducing the complexity of adjusting beam power in private network cells and making the solution more feasible.

[0073] In the method provided by this invention, when training the power adjustment model based on a reinforcement learning algorithm, the reward function is determined based on the channel quality of the public network and private network after executing the sample power adjustment action output by the power adjustment model. Specifically, the reward function is defined as follows: , This indicates that the current network state is as follows during the i-th request: Take power adjustment action The reward received during the process. Performing power adjustment actions. Next, the Channel Quality Indicator (CQI) of the public and private networks is obtained. The CQI reflects the current interference coordination between the public and private networks. Higher CQI values ​​for both networks indicate less interference. Therefore, the corresponding reward is calculated based on the CQI. The reward function can be set as follows: ;

[0074] in The system uses a preset benchmark CQI. When the CQI of either the public or private network is lower than the benchmark CQI, the reward is negative. The reward is positive only when the CQI of both the public and private networks is higher than the benchmark CQI. The higher the CQI value, the higher the reward. The goal of the system is to minimize the interference between the two networks while ensuring a high CQI.

[0075] In the method provided by this invention, the network state features of the public and private networks input into the power adjustment model include data reflecting the state of both the public and private networks. Furthermore, during reinforcement learning, the reward value is determined based on the CQI of both the public and private networks. Therefore, the method provided by this invention considers the network performance of both the public and private networks simultaneously, minimizing interference while maintaining the performance of the public network. Compared to traditional methods, it achieves superior public-private network coordination.

[0076] Considering that both Actor and Critic networks use public / private network state feature vectors Since the input is the same for both networks, the extraction of hidden features in the intermediate layers is identical. Therefore, this scheme adopts a method of sharing parameter weights between the two networks, differing only in the final fully connected layer. The Actor (i.e., the power adjustment model) acts as the policy network, outputting power adjustment actions, while the Critic acts as the value network, outputting scores. This approach directly reduces the number of network parameters and lowers the risk of overfitting. In one possible implementation, the network structure is as follows: Figure 4 As shown, the first three layers are shared fully connected layers, which share the public and private network state feature vectors. Mapping to higher-dimensional hidden features, the Actor policy network passes the hidden features through a fully connected layer and then through a softmax layer to output the probability of each beam power adjustment. The Critic value network passes the hidden features through a fully connected layer and outputs a 1-dimensional value, namely the evaluation value of the current state s.

[0077] The gradient update process in reinforcement learning is explained in detail below. Based on the Actor-Critic network composed of the power adjustment model in this invention, the PPO2 algorithm is used for policy gradient update, employing an optimization objective function. The following is the Actor loss (i.e., the loss of the power adjustment model):

[0078]

[0079] Where E represents expectation. It is a probability ratio, representing the current network's probability of reaching the current state. The output action probability is the same as the network's probability of the current state before the last update. The output action probability ratio; It is a small positive number, such as 0.2; clip() limits the range of the probability ratio. and Between these intervals, prevent updates from being too large or too small; It is the dominant function, let This is a discount factor, set to 0.9, which means that the reward for future actions is discounted when calculating the action reward. , This represents the evaluation value of the current state estimated by the Critic network.

[0080] Critic loss uses TD loss, TD loss Recorded as:

[0081] ;

[0082] To obtain immediate rewards after taking an action, reinforcement learning continuously improves its understanding of environmental dynamics by calculating and reducing TD errors, and makes better decisions accordingly, in order to achieve immediate rewards after taking an action.

[0083] Finally, the Actor loss and Critic loss will be divided into a certain proportion. Adding these together gives the total loss of the network. Since the Actor network needs to maximize the objective, gradient ascent is used. Therefore, the final loss can be obtained. Gradient updates are performed using the total loss.

[0084] The trained reinforcement learning module is deployed, that is, the Actor network in the trained Actor-Critic network is deployed as a trained power adjustment model. Public network base stations and private network base stations can interact with terminals that have deployed the power adjustment model to obtain measurement report data from the first terminal accessing the public network cell and the second terminal accessing the private network cell. This allows the execution of the public-private network collaborative optimization method for high-speed rail scenarios provided by this invention. After obtaining the power adjustment action output by the power adjustment model, it is sent to the public network base station to adjust the beam power of the public network cell. Figure 5 As shown, after receiving the power adjustment action, the public network base station sequentially performs power adjustment or power offset adjustment on each SSB beam of the public network cell.

[0085] The following describes the public-private network collaborative optimization device for high-speed rail scenarios provided by this invention. The public-private network collaborative optimization device described below can be referred to in correspondence with the public-private network collaborative optimization method described above. For example... Figure 6 As shown, the public-private network collaborative optimization device for high-speed rail scenarios provided by the present invention includes:

[0086] The network status feature acquisition module 610 is used to acquire measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell, and to determine the network status features of the public and private networks based on the acquired data. The acquired data includes at least the measurement report data of the first and second terminals. The network status features of the public and private networks reflect the network status of the public network and the private network.

[0087] The model inference module 620 is used to input the public and private network state features into the trained power adjustment model and obtain the power adjustment actions output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment actions include the power adjustment schemes of each beam of the public network cell.

[0088] The power adjustment module 630 is used to adjust the beam power of the public network cell based on the power adjustment action.

[0089] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7As shown, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communication bus 740. The processor 710, communications interface 720, and memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical instructions in the memory 730 to execute a public-private network collaborative optimization method for high-speed rail scenarios. This method includes: acquiring measurement report data from a first terminal accessing a public network cell and a second terminal accessing a private network cell; determining public-private network status characteristics based on the acquired data, wherein the acquired data includes at least the measurement report data from the first and second terminals, and the public-private network status characteristics reflect the network status of the public and private networks; inputting the public-private network status characteristics into a trained power adjustment model; acquiring the power adjustment actions output by the power adjustment model, wherein the power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment actions include power adjustment schemes for each beam of the public network cell; and adjusting the beam power of the public network cell based on the power adjustment actions.

[0090] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the high-speed rail scenario public-private network collaborative optimization method provided by the above methods. The method includes: acquiring measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell; determining public-private network status characteristics based on the acquired data; the acquired data includes at least the measurement report data of the first and second terminals, and the public-private network status characteristics reflect the network status of the public network and the private network; inputting the public-private network status characteristics into a trained power adjustment model; acquiring the power adjustment action output by the power adjustment model; wherein the power adjustment model is a model trained based on a reinforcement learning algorithm; the power adjustment action includes the power adjustment scheme of each beam of the public network cell; and adjusting the beam power of the public network cell based on the power adjustment action.

[0092] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the high-speed rail scenario public-private network collaborative optimization method provided by the above methods. The method includes: acquiring measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell; determining public-private network status characteristics based on the acquired data; the acquired data includes at least the measurement report data of the first and second terminals, and the public-private network status characteristics reflect the network status of the public network and the private network; inputting the public-private network status characteristics into a trained power adjustment model; acquiring the power adjustment action output by the power adjustment model; wherein the power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment action includes the power adjustment scheme of each beam of the public network cell; and adjusting the beam power of the public network cell based on the power adjustment action.

[0093] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0094] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for collaborative optimization of public and private networks in high-speed rail scenarios, characterized in that, include: The measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell are obtained. Based on the obtained data, the network status characteristics of the public and private networks are determined. The obtained data includes at least the measurement report data of the first terminal and the second terminal. The network status characteristics of the public and private networks reflect the network status of the public network and the private network. The public and private network status features are input into the trained power adjustment model to obtain the power adjustment action output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment action includes the power adjustment scheme of each beam of the public network cell. The beam power of the public network cell is adjusted based on the power adjustment action; The determination of public and private network status characteristics based on acquired data includes: The beam performance data of the public network cell is determined based on the measurement report data of the first terminal, and the beam performance data reflects the network performance of each beam of the public network cell; Beam interference data is determined based on the measurement report data of the second terminal, and the beam interference data reflects the interference situation of the second terminal by each beam of the public network cell; The network status characteristics of the public and private networks are determined based on the extracted data, and the extracted data includes at least the beam performance data and the beam interference data; The determination of beam interference data based on the measurement report data from the second terminal includes: Based on the measurement report data of the second terminal, determine whether the second terminal is subject to beam interference from the public network cell; Obtain the proportion of terminals in the second terminal that are affected by each beam interference of the public network cell, and use each proportion as the beam interference data.

2. The method for collaborative optimization of public and private networks in high-speed rail scenarios according to claim 1, characterized in that, Before determining the public / private network status characteristics based on extracted data, the following steps are included: Obtain the base station performance data of the public network base station corresponding to the public network cell and the private network base station corresponding to the private network cell; The extracted data is composed of the beam performance data, the beam interference data, and the base station performance data.

3. The method for collaborative optimization of public and private networks in high-speed rail scenarios according to claim 1, characterized in that, The determination of whether the second terminal is subject to beam interference from the public network cell based on the measurement report data of the second terminal includes: When the second terminal accesses the private network cell, if the difference between the RSRP value of the private network cell and the RSRP value of the target beam of the public network cell in the measurement report data of the second terminal is within a preset range, it is determined that the second terminal is being interfered with by the target beam of the public network cell.

4. The method for collaborative optimization of public and private networks in high-speed rail scenarios according to claim 1, characterized in that, When training the power adjustment model based on the reinforcement learning algorithm, the reward function is determined based on the channel quality of the public network and private network after the power adjustment action of the sample output of the power adjustment model is executed.

5. A public-private network collaborative optimization device for high-speed rail scenarios, characterized in that, include: The network status feature acquisition module is used to acquire measurement report data of a first terminal accessing a public network cell and a second terminal accessing a private network cell, and to determine the public and private network status features based on the acquired data. The acquired data includes at least the measurement report data of the first terminal and the second terminal, and the public and private network status features reflect the network status of the public network and the private network. The model inference module is used to input the public and private network state features into the trained power adjustment model and obtain the power adjustment action output by the power adjustment model. The power adjustment model is a model trained based on a reinforcement learning algorithm, and the power adjustment action includes the power adjustment scheme of each beam of the public network cell. A power adjustment module is used to adjust the beam power of the public network cell based on the power adjustment action; The determination of public and private network status characteristics based on acquired data includes: The beam performance data of the public network cell is determined based on the measurement report data of the first terminal, and the beam performance data reflects the network performance of each beam of the public network cell; Beam interference data is determined based on the measurement report data of the second terminal, and the beam interference data reflects the interference situation of the second terminal by each beam of the public network cell; The network status characteristics of the public and private networks are determined based on the extracted data, and the extracted data includes at least the beam performance data and the beam interference data; The determination of beam interference data based on the measurement report data from the second terminal includes: Based on the measurement report data of the second terminal, determine whether the second terminal is subject to beam interference from the public network cell; Obtain the proportion of terminals in the second terminal that are affected by each beam interference of the public network cell, and use each proportion as the beam interference data.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the high-speed rail scenario public-private network collaborative optimization method as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the public-private network collaborative optimization method for high-speed rail scenarios as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the public-private network collaborative optimization method for high-speed rail scenarios as described in any one of claims 1 to 4.

Images