| The ocean is very important to human survival,and scholars are paying more and more attention to the related research in the ocean field.In the process of Marine research,underwater acoustic sensor network has gradually developed and played an important role in many Marine fields,including resource detection,pollution and water quality monitoring,military reconnaissance,diving consumption and other application scenarios.Routing protocol plays an important role in data transmission of underwater acoustic sensor networks,ensuring the correct transmission of data information from the source node to the destination node.However,its design faces many challenges.First of all,there are many factors that affecting network performance,including nodes energy consumption,data transmission delay,nodes’ movement,distance,density,energy distribution,and so on.Most of the existing underwater routing protocols consider few factors affecting network performance when designing.Secondly,most underwater routing protocols only achieve good results in individual application scenarios,and cannot meet the requirements of different underwater applications on different network performance.In the face of the diversity of underwater applications,it is necessary to design adaptive routing protocols that can meet various network performance requirements,so as to achieve the adaptive application and network environment.In order to comprehensively consider factors that affecting network performance and improve network service quality,this paper proposed an adaptive routing protocol for static underwater sensor networks based on Q-learning(QLAR).Firstly,a routing protocol model based on Q-learning is designed to obtain information from the environment and make routing decisions according to the characteristics of underwater acoustic sensor network such as high energy consumption,high delay and high transmission error.The obtained information includes four factors: residual energy of nodes,energy distribution around nodes,distance between nodes,and density around nodes.By reducing energy consumption and making energy distribution more uniform,the lifetime of the network is extended.The transmission delay is minimized by considering the distance between nodes.By considering the density around the node,the transmission success rate of the network is improved as much as possible.Secondly,in order to solve the packet loss problem caused by the poor channel quality,the implicit confirmation method is adopted to improve the network reliability.The proposed protocol was extensively simulated on the NS-3-based Aqua-Sim-TG platform.The results show that QLAR is better than VBF and DBR in terms of transmission success rate,energy efficiency and network lifetime,and better than QELAR in terms of transmission delay.However,with the widely used of underwater mobile cluster network,the routing protocol under static network cannot meet the communication requirements of mobile cluster network.Routing decisions are affected by intermittent connections between nodes in a mobile cluster.In addition,different network applications have different requirements on network performance.Therefore,this paper proposed an adaptive routing protocol for multiple applications for underwater mobile clustering network,QLMAR).Firstly,aiming at the problem of intermittent connection between nodes,a routing protocol system model is constructed by considering the link connection time between nodes,and the link connection time prediction algorithm is derived.Secondly,in order to solve the problem of packet loss caused by poor channel quality,the explicit acknowledgement method is adopted to improve network reliability,and the re-transmission waiting time is designed.Finally,considering the different requirements of different applications on network performance,AHP is used to assist routing algorithm.By measuring the weight of influencing factors,the adaptive requirements of application and network performance are achieved.The proposed protocol QLMAR is compared with the classical underwater routing protocols VBF and DBR.The results show that it performs well in terms of transmission success rate,energy efficiency and network lifetime.The proposed protocol can be applied to various underwater network applications with mobile characteristics. |
