Study On The Key Techniques Of Underwater Acoustic Communication Network Protocols

Underwater sensor networks have attracted strong attention from both researchers and industry.The unique characteristics of the underwater acoustic communication channel,such as pretty slower propagation speed of acoustic signals and high Bit Error Rate resulting in weak reliability,brings a huge challenge to the protocols for underwater acoustic communication networks.In order to provide continuous and stable data transmission and sharing services,we deeply investigate the MAC and routing protocols for underwater acoustic communication networks.By developing key techniques,including MAC layer link scheduling with large propagation delay and routing protocols with unreliable links,the channel utilization,data validity and data forwarding efficiency could be improved.The investigation is of great theoretical significance and practical value to development of underwater acoustic communication networks.The main research results and creative contributions of this thesis are given as follows:(1)In order to mitigate the influence of large propagation delay,we propose a correlation matrix based link schedule model and a minimum frame length MAC layer link scheduling algorithm.The link schedule model can not only describe the potential conflict between links,but also can ignore the constraints of scheduling priority on network performance on conflictfree scheduling.The link scheduling algorithm could get a network throughput close to the optimal link scheduling.In order to further reduce the frame length,a correlation matrix and power control combined MAC layer link scheduling algorithm is proposed.Simulation results show that the minimum frame length MAC layer link scheduling algorithm and correlation matrix and power control combined MAC layer link scheduling algorithm can effectively improve network performance with respect to throughput and average end-to-end delay.(2)We systematically explore the throughput of underwater linear acoustic communication networks,prove the upper bound throughput of some typical traffic scenarios,including unicast network with single traffic flow,bidirectional traffic flows,more general traffic flows and multicast with single source.Further,we propose two general transmission schedules for unicast network with single traffic flow and bidirectional traffic flows that can achieve the upper bound throughput,and also give some examples of topologies with the transmission schedules.The study of throughput under variable ratio between interference range and communication range is an important component for linear underwater acoustic communication networks,it can support underwater acoustic communication network nodes with different receiving processing capabilities.The study of linear underwater acoustic communication network throughput with different traffic flows can support the application of underwater environment monitoring in theory.(3)To reduce the impact of multipath interference and delay differences on data validity,we put forward a multipath routing protocol with minimum delay difference.The protocol constructs three independent path based on the characteristics of the vertical bisector,resulting a propagation delay difference between the three paths at each hop.The TDMA-based link scheduling scheme can ensure parallel multiple transmission without collision based on the propagation delay difference.Simulation results show that the multipath routing protocol with minimum delay not only can achieve a higher throughput and lower end to end delay,but also can get a small end-to-end delay difference amongst multiple paths.(4)In order to reduce the impact of unreliable links on forwarding efficiency,we propose a path diversity improved candidate selection model for opportunistic routing.It not only maximizes the packet forwarding advancements but also takes the candidates location distribution into account.Based on this model,we propose two forwarding efficiency improved opportunistic routing protocols,PICS and PRCS.Simulation results show that PICS and PRCS both can significantly improve the network performance when compared with the previous solutions in terms of packet delivery ratio and average energy consumption.