Research On Multi-point Cooperation Strategy Of Wireless Sensor And Actor Network Based On SDN

With the rapid development of the IoT,Wireless Sensor and Actor Network(WSANs),one of the key technologies of the IoT,plays an important role in industries,agriculture,smart cities,water monitoring and disaster prediction or rescue.However,with the wide application of WSANs,the traditional network architecture and operation mode have also exposed many problems.For example,the WSANs network architecture is specific and poorly adaptable.Due to the lack of global control and management of the distributed network structure,it is difficult to optimize the utilization of network resources.Software Defined Network(SDN)has developed rapidly in recent years.Many scholars have gradually considered introducing SDN architecture into WSANs to improve overall network performance through SDN's high flexibility and software configurability.The design of WSANs with multipoint collaboration capabilities is considered,such as intelligent sorting warehouses.In this environment,the sensor node is responsible for collecting environmental data,and the smart car or robot acts as an actuator to perform operational functions on events that need to be processed after sensing.Since the actuator makes decisions independently,only the local optimal solution can be obtained,and the global optimal or suboptimal solution cannot be obtained.Therefore,the execution efficiency is relatively low.To this end,a SDN-based WSANs three-layer network architecture is proposed,an SDN controller between the upper-layer application and the physical device is introduced,which adopts logical centralized control instead of distributed decision-making.We focus on the multi-actuator task scheduling problem in WSANs,and design a direction factor based A* path planning algorithm to find the shortest path from each actuator node to different task points.Then based on this shortest path,a genetic algorithm with dynamic mutation operator is designed to address this problem.Finally,the Mininet software is used to simulate the real environment,and compared the performance of traditional distributed networks and SDN networks with SDN controllers.The experimental results show that the SDN network shows better performance in terms of average delay and packet loss rate.