Machine Learning in Wireless Sensor Networks: Challenges and Opportunities

Abstract

Wireless sensor networks are made of autonomous tiny robust nodes called as motes for monitoring the environment where they are deployed. Their tiny and autonomous characteristic serves as an advantage as well as a disadvantage for their deployment and operational use. Their small size makes them resource deficient and adds a constraint on their performance and behavior in the network. Contrary to this, their small size is beneficial for many application areas where the sensors should not be visible while monitoring the environment. The wireless sensor comes with many advantages when used in a variety of environments for monitoring and tracking purposes. There are few challenges that are associated with the wireless sensor networks viz. energy deficient, fault tolerance, security, data aggregation to name a few. Machine learning algorithms play an important role to minimize the impact of these challenges to optimize the functionality of these networks by addressing networking issues and data processing issues.