Simulation of Wireless Sensor Networks Based on Directed Diffusion Algorithm

Abstract

In order to reduce the energy consumption and data congestion caused by flooding at the initial stage of the network, the cluster head and the boundary node are used as the key nodes to reduce the data flow in the network, so that the network has better delay performance. Various clustering algorithms in wireless sensor networks and the improved method of clustering protocols are introduced. On the basis of directed diffusion algorithm, a directed diffusion protocol based on CONID clustering is studied. Through the OMNeT++ basic platform, the wireless sensor network simulation platform is built on the basis of the MiXiM module. The platform uses a hierarchical structure, and its network layer implements a directed diffusion protocol based on CONID clustering. The application layer implements the two step WLS localization algorithm. The MAC layer uses the FIexiTP protocol. The physical layer uses the simplest flat path fading model. Because the sensor has both perception and data transmission characteristics, the dual channel communication model is adopted. The platform is suitable for wireless sensor network detection, localization and tracking algorithm simulation. Through the hierarchical function of the platform, it can analyze the practicability and effectiveness of the algorithm in more detail, and has certain value. It can also be used in the simulation of cross layer protocols, and can combine several parameters to perform joint analysis of system performance. Simulation results show that the CONID based directional diffusion protocol has better delay performance than the conventional directed diffusion protocol. Finally, the performance of the two step WLS location protocol is analyzed from several perspectives, which provides an example for the platform.