Method, network node and system for realizing grouping based on identification extreme value

Abstract

The invention discloses a method, network node and system for realizing clustering based on identity extremums and relates to the field of network control. Network nodes divide time after appointed time into continuous time intervals; the time in each time interval is divided into N time periods with preset lengths; cluster leader selection rules based on identity extremums are allocated to the time periods; the cluster leader selection rules in the time periods of each time interval are mutually different; the cluster leader selection rules allocated for the nodes in the same time period of the same time interval are the same; a candidate node identity set is generated according to received node tables broadcasted by adjacent nodes; and the corresponding node is selected from the candidate node identity set as a cluster leader node according to the cluster leader selection rule corresponding to the current time. According to the method, the system and the network node, through utilization of different types of identity extremums, probabilities that all nodes become the group leaders are approximate, and the problem that energy is liable to exhaust due to the fact that some nodes are taken as the cluster leaders for a long time is effectively avoided.

Description

technical field [0001] The invention relates to the field of network control, in particular to a method, network node and system for realizing grouping based on an identification extreme value. Background technique [0002] The basic idea of ​​grouping is to divide the network nodes into several groups according to the geographical proximity, and provide a smaller logical organization for the network, making the network easier to manage. A group consists of a group leader and member nodes, and the group leader acts as a regional broadcast node and a local coordinator node. Regrouping requires high processing overhead and communication overhead. [0003] In the clustered large-scale mobile ad hoc network, the clustering methods mainly include the minimum identification (ID) clustering algorithm (MinID), the highest connectivity clustering algorithm (MaxDegree) and the weighted clustering algorithm (WCA). [0004] The MinID algorithm selects the node with the smallest ID in ...

AI Technical Summary

Problems solved by technology

However, due to limited node resources, the number of nodes within the transmission coverage of the cluster leader may exceed the processing capacity of the cluster leader

On the other hand, the more cluster heads, the higher the system computing overhead, the throughput may be improved, and the data packets must be transmitted through multiple hops and the delay will be greater

[0008] Disadvantages of the highest connectivity grouping algorithm: (1) The low frequency of group leader changes can easily lead to battery exhaustion; (2) The more group member nodes, the lower the throughput and the system performance declines. Since there is no upper limit on the number of member nodes , theoretically the throughput will be zero; (3) The increase or decrease of member nodes needs to reselect the highest density node

[0013] The problems of the above two schemes are: firstly, it takes a long waiting time before making a decision, and it can only be judged through complex calculations after receiving the responses of all adjacent nodes; secondly, it is judged whether to reselect the cluster leader and Residual energy is related. If the residual energy threshold is set too low, the energy consumption of the cluster leader will be too much; if the threshold is set too high, it will lead to frequent replacement of the cluster leader, especially when the energy of the member nodes is partially consumed, thereby reducing system performance.

Images