MATLAB Simulation of Intelligent Routing in Wireless Sensor Networks using Machine Learning Techniques

WSNs have demonstrated to play a key role in many applications ranging from environmental monitoring to military surveillance. The problem of efficient routing however is challenging due to resource and scalability constraints such as energy resources, dynamic network topology, etc. The final stage involves selecting an optimal path which improves the performance of a network through energy conservation and extending the network lifetime. From the previous discussion it clearly can be seen that Machine Learning (ML) could be a promising approach to learn patterns and predict the route which is best suited for WSN under changing conditions. The simulation framework for ML algorithms, implemented in MATLAB, for intelligent routing in WSNs is presented in this paper. We present a method that enhances routing with ML models selecting energy efficient paths based on packet loss and life time of the network. Results of simulations show that the ML-driven routing protocol improves on these values compared to traditional methods. The validation of the use of ML techniques for intelligent routing in WSNs and a basis for further researches in the energy efficiency of WSN communications are also presented.