Application Of Intelligence Algorithm In Underwater Mobile Swarm

With the attention and development of marine technology,underwater acoustic sensor network technology has become an important part of marine science and technology.Acoustic sensor network can be deployed for a long time to obtain marine data,monitorthe marine environment,detect military targets,and together with ships to complete more complex tasks.Foreign research in this area has been leading our country,for instance,the US Navy's Seaweb project,Ocean-TUNE,Europen project SUNRISE and CommsNet 12,have accumulated a lot of experience in the experiment,AUV and other mobile nodeshas been tried to be used and has achieved good results,but no research is focusing on underwater mobile swarm network yet.Mobile swarm networking technology in UAV swarm and missile swarm have been developed quickly in recent years.The key technology of unmanned mobile swarm networking is whether the node has enough intelligence,which requires a lot of intelligent algorithms.The intelligent algorithm is a kind of algorithms to solve the optimization problem,such as genetic algorithm,swarm intelligence algorithm and machine learning algorithm,which has been widely used in the unmanned platform and self-organizing network to make the platform and network efficient.Due to the high complexity of the underwater environment,underwater mobile swarm is hard to implement.The underwater acoustic channel is a time-varying channel,and there exist serious multi-way effect and Doppler shift,which has a serious impact on the communication between nodes.Besides,Acoustic communication frequency is much lower than wireless communication,the transducer is large,and the node energy is limited,It's inconvenience to replace the battery underwater.All this has become the barrier of underwater mobile swarm research.This paper will focus on the deployment of underwater mobile swarm and the data forwarding mechanism of network layer.All of the studies in this paper are based on centralized releasedswarm.For the deployment of mobile swarm,swarm intelligent algorithms like particle swarm optimization algorithm and artificial fish swarmalgorithm is suitable for mobile node deployment.These algorithmsare based on the simulation of animal's social behavior,and searching for optimal solution through iterations in the searching area.In this paper,we use the standard particle swarm algorithm,improved particle swarm algorithm and the artificial fish swarm algorithm to achieve the deployment of the mobile node,and have achieved good results.The advantage of using this algorithm is that the swarm can be automaticly deployed through the interaction between nodes without human,and the network is scalable,the network life can also be greatly extend through regular automatic adjustment of node location.In this paper,the data forwarding mechanism between nodes is also studied.Underwater data forwarding needs to consider the channel condition and the residual energy of the nodes.If some nodes consume too much energy,the node will die prematurely.In order to solve the above problems,this paper attempts to use the fuzzy logic reasoning algorithm and the decision tree algorithm to design the forwarding node selection mechanism.This mechanism will Consider both the current link state and the residual energy of the node,it is found that the above methods have achieved good result.