CN-122028111-A - Load balancing method and system for wireless mesh network

CN122028111ACN 122028111 ACN122028111 ACN 122028111ACN-122028111-A

Abstract

The application provides a load balancing method and system for a wireless mesh network. The method comprises the steps of firstly obtaining real-time load data, link quality data, position data and movement history data of a plurality of nodes in a network, then generating current load distribution conditions of the network according to the real-time load data and the link quality data, identifying a node set with load higher than a preset range, analyzing the node set based on the position data and the movement history data, predicting future position change, determining an alternative path according to the future position change and the link quality data, and finally transmitting partial flow of the node set through the alternative path. The technical scheme provided by the application not only solves the problem of path quality abrupt change caused by node mobility, but also improves the long-term stability and service continuity of load transfer in the dynamic wireless mesh network.

Inventors

BAI YUNHE
PANG YUFEI
Xiao Hanlong
PANG SHIWU
QU QINGFENG
HU PING
YU CHUNYAN

Assignees

北京飞成科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260211

Claims (10)

1. A method of load balancing a wireless mesh network, comprising: Acquiring real-time load data, link quality data, position data and movement history data of a plurality of nodes in a wireless mesh network; Generating a current load distribution condition of the network according to the real-time load data and the link quality data; identifying a node set with a load higher than a preset range from the load distribution condition; Analyzing the node set based on the position data and the movement history data to obtain future position change; Determining an alternative path based on the future location change and the link quality data; and transmitting part of traffic of the node set through the alternative path so as to finish the stable migration of the load.
2. The method of claim 1, wherein generating a current load profile for the network from the real-time load data and the link quality data comprises: extracting a node load value corresponding to each node from the real-time load data; Extracting a link reliability value corresponding to each wireless communication link from the link quality data; Carrying out comprehensive operation according to the link reliability values directly connected with each node to obtain a link adjustment value of each node; Combining the node load value of each node with the corresponding link adjustment value to generate a comprehensive load value of each node; And integrating the comprehensive load values of all the nodes to obtain the current load distribution condition of the network.
3. The method of claim 1, wherein identifying a set of nodes from the load distribution that have a load above a preset range comprises: extracting comprehensive load values of all nodes from the load distribution condition; selecting a plurality of reference nodes based on the comprehensive load values of all the nodes; Calculating a preset range according to the comprehensive load value of the reference node; and comparing the comprehensive load value of each node with the preset range, screening out nodes with the comprehensive load value higher than the preset range, and forming a node set.
4. The method of claim 1, wherein analyzing the set of nodes based on the location data and the movement history data for future location changes comprises: Acquiring the current position coordinate of each node in the node set from the position data; Extracting a movement record sequence of each node in the node set from the movement history data; Calculating the movement trend parameter of each node according to the time information and the coordinate information of the continuous positions in the movement record sequence; and combining the current position coordinates of each node with the corresponding movement trend parameters to obtain the future position change.
5. The method of claim 1, wherein determining an alternative path based on the future location change and the link quality data comprises: acquiring a link reliability value of each wireless communication link from the link quality data; Calculating the connection stability of the wireless communication links between the nodes based on the future position change; Combining the connection stability with a corresponding link reliability value to generate comprehensive reliability; generating a plurality of candidate paths in the wireless mesh network according to the comprehensive reliability; and screening paths with path quality higher than a preset standard from the candidate paths as alternative paths.
6. The method of claim 5, wherein combining the connection stability with the corresponding link reliability value generates a composite reliability, comprising: acquiring the connection stability and the link reliability value of each wireless communication link; determining a first combination factor and a second combination factor according to the mobility strength of the node in the wireless mesh network; adjusting the connection stability according to the first binding factor to obtain a stability component; Adjusting the link reliability value according to the second combination factor to obtain a reliability value component; and calculating the stability component and the reliability component to obtain comprehensive reliability.
7. The method of claim 1, wherein transmitting a portion of traffic of the set of nodes through the alternative path to complete smooth migration of loads comprises: calculating a flow value to be transferred based on the comprehensive load value of the node set and the path quality of the alternative path; dividing the flow value to be transferred into a plurality of flow units according to the flow value to be transferred; And distributing the flow units to different alternative paths according to the path quality of each alternative path, and transmitting the flow units to a target node through the alternative paths so as to finish the smooth migration of the load.
8. A load balancing system for a wireless mesh network, comprising: The acquisition module is used for acquiring real-time load data, link quality data, position data and movement history data of a plurality of nodes in the wireless mesh network; the generation module is used for generating the current load distribution condition of the network according to the real-time load data and the link quality data; the identification module is used for identifying a node set with load higher than a preset range from the load distribution condition; The analysis module is used for analyzing the node set based on the position data and the movement history data to obtain future position change; a determining module for determining an alternative path based on the future location change and the link quality data; And the transmission module is used for transmitting part of traffic of the node set through the alternative path so as to finish the stable migration of the load.
9. A computing device, comprising a processing component and a storage component, the storage component storing one or more computer instructions for execution by the processing component to implement a method of load balancing a wireless mesh network according to any one of claims 1-7.
10. A computer storage medium storing a computer program which, when executed by a computer, implements a method of load balancing a wireless mesh network according to any one of claims 1 to 7.

Description

Load balancing method and system for wireless mesh network Technical Field The present application relates to the field of wireless communication networks, and in particular, to a method and system for balancing load of a wireless mesh network. Background The network is widely applied to emergency communication, mobile vehicle-mounted network, intelligent internet of things and other scenes due to the characteristics of self-organization, multi-hop transmission and flexible deployment, and nodes in the network generally have mobility, so that the network topology structure is dynamically changed, the service load distribution in the network is very unbalanced, and part of nodes become congestion bottlenecks due to excessive connection or in a critical path, so that the overall transmission efficiency and service reliability of the network are seriously affected. In the prior art, decisions are made according to real-time link quality data and node instantaneous load data of a network, and the core is that current quality indexes of all communication links and real-time load rates of nodes are monitored, and once the load of a certain node is found to exceed a preset threshold value, one or more transmission paths are reselected for the service flow according to the optimal link quality information at the current moment so as to share the load pressure. However, the scheme has inherent defects, namely a passive reaction mechanism based on the current network instantaneous state, and cannot fully consider the future influence of node mobility on a network topology structure, and because the node position in a wireless mesh network is continuously changed, a path with optimal quality at the current moment can be rapidly degraded or even interrupted in a short time due to the movement of the node, so that load migration failure carried out according to the path is caused, thereby causing service jitter and extra path switching overhead, and lacking a mechanism for predicting future connection stability, so that the load balancing effect is difficult to be durable and stable in a dynamic environment. Disclosure of Invention The application provides a load balancing method and a system of a wireless mesh network, which are used for solving the problems that in the prior art, a load balancing decision is only based on the instantaneous state of the network and cannot effectively cope with the topology dynamic change caused by node mobility, so that a selected path is likely to fail in a short time, the load migration effect is not durable and service jitter is caused. In a first aspect, the present application provides a method for load balancing a wireless mesh network, including: Acquiring real-time load data, link quality data, position data and movement history data of a plurality of nodes in a wireless mesh network; Generating a current load distribution condition of the network according to the real-time load data and the link quality data; identifying a node set with a load higher than a preset range from the load distribution condition; Analyzing the node set based on the position data and the movement history data to obtain future position change; Determining an alternative path based on the future location change and the link quality data; and transmitting part of traffic of the node set through the alternative path so as to finish the stable migration of the load. Optionally, generating the current load distribution situation of the network according to the real-time load data and the link quality data includes: extracting a node load value corresponding to each node from the real-time load data; Extracting a link reliability value corresponding to each wireless communication link from the link quality data; Carrying out comprehensive operation according to the link reliability values directly connected with each node to obtain a link adjustment value of each node; Combining the node load value of each node with the corresponding link adjustment value to generate a comprehensive load value of each node; And integrating the comprehensive load values of all the nodes to obtain the current load distribution condition of the network. Optionally, identifying a node set with a load higher than a preset range from the load distribution conditions includes: extracting comprehensive load values of all nodes from the load distribution condition; selecting a plurality of reference nodes based on the comprehensive load values of all the nodes; Calculating a preset range according to the comprehensive load value of the reference node; and comparing the comprehensive load value of each node with the preset range, screening out nodes with the comprehensive load value higher than the preset range, and forming a node set. Optionally, analyzing the node set based on the location data and the movement history data to obtain a future location change includes: Acquiring the current position coordinate of each node in the no