Clustering management method for improving availability of distributed multi-hop network

Abstract

The invention discloses a clustering management method for improving the availability of a distributed multi-hop network, and mainly solves the problems of disordered networking and high control overhead of large-scale network nodes in the prior art. According to the implementation scheme, the method comprises the steps of cluster generation and cluster maintenance, and the cluster generation comprises the steps that firstly, nodes randomly participate in competition of cluster heads according to the electric quantity percentage of the nodes; secondly, selecting cluster head nodes in the network through a weight optimization principle; then, periodically broadcasting cluster domain information by the cluster head node and the in-cluster node which is already clustered so as to enable the out-cluster node to be clustered; the cluster maintenance steps are as follows: 1) cluster head nodes select standby cluster head nodes according to an optimal similarity principle, and master-slave cluster head node information synchronization is realized through a differential synchronization mode; and 2) detecting the survival condition of the cluster head node by the standby cluster head node,and performing a cluster head node function as required. The method improves the network performance, ensures the stability of the network, and can be used for hierarchical division and maintenance ofa large-scale network organization structure.

Description

technical field [0001] The invention belongs to the technical field of communication, and further relates to a hierarchical network clustering management method, which can be used for hierarchical division and maintenance of large-scale network organizational structure. Background technique [0002] The traditional self-organizing network organization structure is equal, the function and status of nodes in the network are equal, and the data transmission in the network depends on node forwarding. However, in large-scale networking and multi-hop networking scenarios, traditional self-organizing network organization Chaos, high control overhead, tight time slot resources, it is difficult to adapt to large-scale network node scenarios, so the scalability is poor. In recent years, the use of clustering algorithms to construct a hierarchical network structure and optimize the application of network bandwidth can improve channel utilization and network performance, making high-per...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that the cluster head nodes selected by the minimum node ID clustering method are not necessarily reasonable, and it is easy to cause cluster head conflicts in multi-hop network topology scenarios. At the same time, the selection of assistant nodes There are also some defects. Because it only uses the weight value as the only factor for selecting assistant nodes, it may cause the selected assistant nodes to be far away from the cluster head, resulting in a large information synchronization overhead. In large-scale ad hoc network scenarios, it cannot Give full play to the advantages of clustering

However, the disadvantage of this method is that because it only uses the static distribution characteristics of nodes as a cluster division factor, it limits the mobility characteristics of nodes. In the case of dynamic node movement, it will lead to confusion in the cluster organization structure and poor information transmission between clusters. Stablize

Images