Design And Implementation Of Service-driven Ad Hoc Network Protocols

With the rapid development of wireless communication technology,mobile Ad Hoc network(MANET)has been widely used in scenarios such as emergency search and rescue,environmental monitoring,and aerial geophysical prospecting by its unique advantages such as quick deployment,strong invulnerability,and flexible communication,and its key technology has also become a key scientific research field.Due to the diversification of application scenarios and the continuous increase in the number of users,the types of services and their performance requirements are also changing.Traditional networking protocols can no longer meet the differentiated service requirements in multiservice scenarios.This thesis studies some key technologies in MANET supporting multiservice transmission,and the main work is as follows:In this thesis,we first study the hierarchical topology of MANET.With the expansion of network scale,the complexity of communication between nodes is getting higher and higher,which brings a large system management and control overhead.From the perspective of improving management efficiency and scalability,this thesis proposes a hierarchical clustering network architecture,which divides nodes into clusters,sets up a two-layer communication mechanism between upper and lower levels,and alleviates node competition through multi-frequency access.At the same time,the transmission pressure of each node is balanced to avoid communication bottlenecks caused by traffic aggregation.Secondly,the resource allocation mechanism under multi-service requirements is studied.The diversity of services and the variability of topologies lead to the disadvantages of low access efficiency,high control overhead,and difficult to ensure the quality of service(Qo S)in traditional channel access protocols.This thesis proposes a hybrid slot assignment strategy based on service priority,which divides business slot resources into fixed slots and shared slots.Fixed slots is designed to ensure low access latency,and we propose the service priority and slot state conversion mechanisms as the basis for determining shared slot occupation conflicts,the slot resources could be allocated efficiently and on demand to ensure the Qo S of high priority services while reducing the conflict probability.Finally,the routing mechanism under hierarchical cluster networking architecture is studied.Since flat routing protocols incur a huge routing overhead when the network size is large,and multi-hop transmission paths are highly susceptible to failure due to node movement,this thesis proposes an efficient hybrid routing mechanism for hierarchical clustered network architecture from the perspective of reducing protocol control overhead and improving network scalability,where an optimized link-state routing protocol is used within clusters to ensure small end-to-end delay and reduce the impact of topology changes on routing algorithm,and cross-cluster routing is established by diffusing distance vector information between clusters to avoid the transmission of non-essential link information.In order to verify the correctness and reliability of the dynamic slot allocation algorithm and the hybrid routing mechanism,this thesis builds a hardware-in-the-loop simulation platform to analyze the performance of the protocol.The results show that the service-driven dynamic slot allocation algorithm can allocate slot resources on demand,give priority to high-level services,and provide lower access delay and higher network throughput.The hybrid routing mechanism can greatly reduce routing overhead,improve network throughput,and provide lower access delay in a highly dynamic topology environment.