Neural Network Based Energy-Efficient Multi-hop Hierarchical Routing Protocol for Wireless Sensor Networks

Abstract-This paper presents a new neural network based routing protocol for Wireless sensor networks called Energy efficient Multi-hop Hierarchical Routing protocol using Self organizing map (EMHRS). This new protocol has an innovative quality of service (QoS)-driven routing algorithm based on artificial intelligence that can select next hop cluster head using Self Organizing Map neural network capability in multi dimensional data routing. This intelligent method optimizes the routing according to the amount of energy of each CH node in the network and its computation power while insuring more network coverage. Simulation results approve the advantages of EMHRS over two similar protocols, EMHR and EMHR-FL in terms of postponement of first node death and preserving more network coverage. As results, our method makes a better decision of next hop CH selection than classical EMHR and EMHR based fuzzy logic