Search

CN-121985353-A - Cell access optimization method, system, computer equipment and storage medium

CN121985353ACN 121985353 ACN121985353 ACN 121985353ACN-121985353-A

Abstract

The invention provides a cell access optimization method, a system, computer equipment and a storage medium, and relates to the technical field of communication, wherein the method comprises the steps of predicting the residence probability of a user in a corresponding candidate cell in each preset time period in a plurality of preset time periods in the future; the method comprises the steps of obtaining load cost values of candidate cells, obtaining access scores of the candidate cells based on residence probability and load cost of the candidate cells, outputting access score sequences of the candidate cells, screening at least one selectable target cell from the candidate cells based on preset screening rules, and screening the selectable target cell with the highest access score based on the access score sequences of the candidate cells to serve as a final target cell to be accessed by a user. The technical scheme provided by the invention realizes the dynamic optimization of access decision by deeply fusing the user behavior prediction and the cell load perception, and effectively ensures the continuity and stability of user experience while improving the utilization rate of network resources.

Inventors

  • CAI YONG
  • WU ZHENGGUANG
  • LIU ZUYING

Assignees

  • 中国联合网络通信集团有限公司

Dates

Publication Date
20260505
Application Date
20260313

Claims (10)

  1. 1. A method for optimizing cell access, comprising: predicting the residence probability of the user in each preset time period in a plurality of preset time periods in the future according to the historical track and the current position of the user, and outputting a residence probability distribution matrix of the candidate cell; Based on the current load of each candidate cell and the historical load sequences of a plurality of past moments, load cost values of each candidate cell are obtained respectively, and a load cost set of the candidate cells is output; obtaining access scores of each candidate cell based on the residence probability distribution matrix and the load cost set of the candidate cells, and outputting access score sequences of the candidate cells, and And screening at least one selectable target cell from the candidate cells based on a preset screening rule to form a selectable target cell set, and screening the selectable target cell with the highest access score from the selectable target cell set based on the access score sequence of the candidate cells to serve as a final target cell to be accessed by the user.
  2. 2. The method of claim 1, wherein predicting the camping probability of the user for each of a plurality of preset time periods in the future based on the user's historical track and current location, each preset time period corresponding to a corresponding candidate cell, comprises: Inputting the current position data of the user into a trained user track prediction model, and outputting predicted track points of the user at a plurality of moments in the future, wherein the user track prediction model is obtained by training in advance based on a user history track data set; carrying out space matching on predicted track points of the user at a plurality of future moments and a cell boundary space distribution set to obtain cells respectively corresponding to the user at the plurality of future moments as candidate cells; And mapping the future time points to a plurality of continuous and non-overlapping future preset time periods in time sequence, and obtaining the residence probability of the user in each preset time period in the future preset time periods in the corresponding candidate cell according to the candidate cell respectively corresponding to the user in the future preset time periods.
  3. 3. The method of claim 2, wherein the set of user history trace data is defined as: wherein Represents the i-th historical track point of the user, i is sequentially 1 to m, m represents the total number of the historical track points of the user, and Wherein Is a three-dimensional space position coordinate, In order to be a time stamp, A cell unique identifier for the user is then connected, The motion state data of the user; And/or the number of the groups of groups, The set of cell boundary spatial distributions is defined as: wherein Space attribute data representing a jth cell in a preset area, j sequentially taking 1 to n, wherein n represents the total number of cells in the preset area, and Wherein For the cell unique identifier of the jth cell in the preset area, For the physical cell ID of the jth cell in the preset area, The cell coverage border area data of the jth cell in the preset area, For the cell center point or reference point coordinates of the jth cell in the preset area, And the antenna engineering parameters of the jth cell in the preset area are obtained.
  4. 4. The method according to claim 1, wherein the set of load costs for the candidate cells is defined as: wherein Represents the load cost value of the jth candidate cell, J sequentially takes 1 to J, J represents the total number of candidate cells, and Wherein For the current load value of the j-th candidate cell, For the historic load fluctuation index of the j-th candidate cell, 、 To adjust the weights.
  5. 5. The method of claim 4, wherein the j-th candidate cell's historical load fluctuation index The method is calculated by adopting the following formula: ; Wherein, the Is the historic load standard deviation of the jth cell, Is the history load sequence of the jth cell, and , wherein, Representing the current time, and 1 to T 'representing the past T' preset times; the historical load average value of the jth cell; 、 To adjust the weights.
  6. 6. The method according to claim 1, wherein the sequence of access scores of the candidate cells is defined as: wherein The access score of the J candidate cells is represented, J sequentially takes 1 to J, and J represents the total number of the candidate cells; the method is calculated by adopting the following formula: ; wherein 1 to T represent T preset time periods in the future; indicating the residence probability of the user in j candidate cells in the t preset time period in the future; And the load cost value of the j candidate cell is represented.
  7. 7. The method of claim 1, wherein the set of selectable target cells is screened using the following constraints: ; Wherein, the Representing a set of selectable target cells; Representing a j-th candidate cell; an average RSRP value representing the j-th candidate cell; An average SINR value representing a j-th candidate cell; Is a first threshold factor; Is a second threshold factor; And/or the number of the groups of groups, The final target cell to be accessed by the user is obtained by screening by adopting the following formula: ; Wherein, the Representing a final target cell to be accessed by a user; Representing an access score of the j-th candidate cell; Representing a set of selectable target cells.
  8. 8. A system for optimizing cell access, comprising: A cell residence probability prediction module configured to predict residence probabilities of the user in a plurality of preset time periods in the future, each preset time period corresponding to a candidate cell, based on a historical track and a current position of the user, and output a residence probability distribution matrix of the candidate cell; the cell load cost calculation module is set to obtain load cost values of the candidate cells respectively based on current loads of the candidate cells and historical load sequences of a plurality of past moments, and output load cost sets of the candidate cells; A cell access score calculation module configured to obtain access scores of the candidate cells based on the residence probability distribution matrix and the load cost set of the candidate cells and output an access score sequence of the candidate cells, and The access cell screening module is configured to screen at least one selectable target cell from the candidate cells based on a preset screening rule and form a selectable target cell set, and screen the selectable target cell with the highest access score from the selectable target cell set based on the access score sequence of the candidate cells as a final target cell to be accessed by the user.
  9. 9. A computer device comprising a memory and a processor, the memory having stored therein a computer program, which when executed by the processor performs the cell access optimization method according to any of claims 1 to 7.
  10. 10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, performs the cell access optimization method according to any of claims 1 to 7.

Description

Cell access optimization method, system, computer equipment and storage medium Technical Field The present invention relates to the field of communications technologies, and in particular, to a cell access optimization method, a cell access optimization system, a computer device, and a computer readable storage medium. Background With rapid deployment of 5G (5 th Generation Mobile Communication Technology, fifth generation mobile communication technology) networks and gradual advancement of future 6G (6 th Generation Mobile Communication Technology, sixth generation mobile communication technology) communication architectures, cellular networks exhibit characteristics of high cell density, frequent handover, coexistence of heterogeneous networks (macro station, micro station, hot spot cell), and the like. In this context, user access optimization and cell selection policies become one of the key technologies to guarantee user quality awareness (QoE, quality of Experience) and network resource utilization. Conventional cell access control strategies typically make decisions based on radio signal strength (e.g., RSRP (REFERENCE SIGNAL RECEIVING Power, reference signal received Power), RSSI (RECEIVED SIGNAL STRENGTH Indication)), channel quality metrics (e.g., SINR (Signal to Interference plus Noise Ratio, signal to interference plus noise ratio), CQI), cell current load (number of users, PRB (Physical Resource Block, physical resource block) utilization), and access history experience rules (e.g., load balancing factor bias). Although the strategies can realize load balancing and reasonable resource scheduling to a certain extent, with the increasing complexity of user behaviors and the diversification of network service scenes, the challenges are faced that the movement track of the user has obvious periodicity and individuality, the static strategies cannot adapt to the evolution of the dynamic behaviors of the user, the user leaves rapidly after staying in a certain cell for a short time, the user experience is seriously affected by frequent switching (Ping-Pong effect), and most of access algorithms only pay attention to a single angle (load or channel quality) and ignore the deep coupling relation between the space behaviors of the user and the cell resources. Disclosure of Invention The invention is completed for at least partially solving the technical problems that the existing cell access strategy cannot adapt to the dynamic behavior evolution of the user, neglecting the coupling relation between the user space behavior and the cell resource and influencing the user experience. According to an aspect of the present invention, there is provided a cell access optimization method, including: predicting the residence probability of the user in each preset time period in a plurality of preset time periods in the future according to the historical track and the current position of the user, and outputting a residence probability distribution matrix of the candidate cell; Based on the current load of each candidate cell and the historical load sequences of a plurality of past moments, load cost values of each candidate cell are obtained respectively, and a load cost set of the candidate cells is output; obtaining access scores of each candidate cell based on the residence probability distribution matrix and the load cost set of the candidate cells, and outputting access score sequences of the candidate cells, and And screening at least one selectable target cell from the candidate cells based on a preset screening rule to form a selectable target cell set, and screening the selectable target cell with the highest access score from the selectable target cell set based on the access score sequence of the candidate cells to serve as a final target cell to be accessed by the user. Optionally, the predicting, based on the historical track and the current location of the user, the residence probability of the user in a plurality of preset time periods in the future, where each preset time period corresponds to a corresponding candidate cell includes: Inputting the current position data of the user into a trained user track prediction model, and outputting predicted track points of the user at a plurality of moments in the future, wherein the user track prediction model is obtained by training in advance based on a user history track data set; carrying out space matching on predicted track points of the user at a plurality of future moments and a cell boundary space distribution set to obtain cells respectively corresponding to the user at the plurality of future moments as candidate cells; And mapping the future time points to a plurality of continuous and non-overlapping future preset time periods in time sequence, and obtaining the residence probability of the user in each preset time period in the future preset time periods in the corresponding candidate cell according to the candidate cell respectively corres