Distributed cluster head election method based on wireless ad hoc network

Abstract

The invention discloses a distributed cluster head election method based on a wireless ad hoc network, and the election method is based on a highest node degree algorithm and is used under the condition that each node does not know the information of other nodes during network initialization. In a dense scene, network topology is too complex, and clustering is difficult to complete at a time through a single index, so that the cluster head election process is divided into three stages, a certain number of cluster heads are selected in each stage, and nodes subjected to three-time election become final cluster heads. Through simulation tests, the election method is high in clustering speed, the network connectivity rate reaches 100%, and a clustering result provides necessary preparation for subsequent clustering optimization, network slicing and communication resource allocation optimization.

Description

technical field [0001] The invention relates to the technical field of wireless ad hoc networks, in particular to a distributed cluster head election method based on wireless ad hoc networks. Background technique [0002] The topological types of wireless ad hoc networks can be divided into two types: one is planar structure, such as figure 1 As shown, this kind of network structure is simple, in which all nodes are equal and share the network load. There may be more than one data transmission path between each pair of source nodes and destination nodes. The disadvantage of this structure is that it has poor scalability. If As the scale of the network increases and the mobility of each node increases, maintaining dynamic routing requires a large amount of control information, so this structure is only suitable for small-scale wireless ad hoc networks. [0003] The other is a layered structure, which mainly changes the network form from the physical structure. In the hierar...

AI Technical Summary

Problems solved by technology

Typical clustering algorithms include: the lowest ID algorithm (LOWID), the highest node degree algorithm (HIGHD); weight-based clustering algorithm (WCA); comprehensively consider factors such as connectivity, node moving speed, and energy consumption rate to carry out weighting. Clustering; BIMAC, a bionic clustering algorithm based on mobile perception, can effectively improve link connectivity and cluster head survival time, but cannot guarantee data transmission between clusters; distributed clustering algorithm MOBIC, according to the movement of nodes To select the cluster head, but it is only applicable to the scene where the moving speed and direction of the nodes are the same; the clustering algorithm CEMCA based on connectivity, node energy and mobility

[0005] In the application scenario, the communication distance of network nodes is 200km, which is relatively large compared to the activity range of network nodes (500km*500km), which makes the network topology too dense (such as image 3 shown), so the above-mentioned algorithms such as the highest node degree cannot be directly applied

Images