Internet of Things Optimal Routing based on Markov-Reinforcement Learning Algorithm

Abstract

By increasing Internet of Things (IoT) development, it will face new challenges due to its application in various fields of science and industry. One of these challenges is the routing problem in communication which has many effects in various parts and the quality criteria of the IoT in communication. In this research, a new approach provide with an effective and optimal method for routing in the IoT, which will use 6LoWPAN as the default protocol. This protocol has qualitative weaknesses and its goal is to improve and extend it as much as possible with the Markov Reinforcement Learning (MRL). Improving power consumption and reducing energy beside quality of services criteria optimization is the main targets of this approach. The results represented the improvement of the proposed approach compared to other similar protocols such as Z-WAVE and Zig Bee and the classic 6LoWPAN protocol.

Keywords

• Internet of Things (IoT)
• Routing
• 6LoWPAN Protocol
• Markov Reinforcement Learning (MRL)