Energy Efficient Routing Algorithm For Wireless Heterogeneous Networks Under Statistical Delay Constraints

In recent years, wireless sensor networks (WSN) have seen a tremendous growth in their applications for environments such as industry, home automation, healthcare, etc. Multi-hop sensor networks are largely being deployed in industries in order to increase the coverage of the system. This has led to the need for evaluating the performance of such networks so that a quantifiable end-to-end quality of service (QoS) metric can be defined. One such metric is the end-to-end probabilistic delay bound, which defines the probability that a packet will not satisfy the required end-to-end delay constraints. This thesis presents a routing algorithm which finds optimum paths using two metrics: the end-to-end delay violation probability, and the total energy consumption of the entire path for transmitting a packet. We propose a distributed as well as a centralized version of routing algorithm. The routing algorithm is integrated within the routing protocol for low-power and lossy networks (RPL). The presented routing algorithm relies on a novel framework for performance analysis of wireless multi-hop networks using the theory of stochastic network calculus. This novel framework demonstrates that the end-to-end delay bound for wireless heterogeneous networks is recursive in nature. The proposed routing algorithm utilizes the recursive nature, which makes the routing algorithm first of its kind.