Semi-Persistent Medium Access Control Protocols for Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are dense clusters of sensor nodes, made up of small, intelligent, resource-constrained wireless devices that are deployed to monitor a specific phenomenon in a certain field. The sensor nodes can be constrained by limited power supply, memory capacity and/or processing capabilities, which means that the design of WSNs requires all algorithms and protocols to be lightweight and efficient, and use as little power as possible. The Medium Access Control (MAC) protocol in WSNs, defined by the IEEE 802.15.4 standard, employs the Carrier Sense Multiple Access with Collision Avoidance (CSMA-CA) algorithm to control the nodes contending for access to the communication medium. Though the performance of this protocol has been studied extensively, and several improvements to its backoff counter, superframe format and contention-free period (CFP) features have been proposed, very few studies have addressed improving the Clear Channel Assessment (CCA) feature. In this thesis, we study the impact of increasing the value of the contention window beyond the standard value of 2, on the performance of the MAC protocol. We propose a semi-persistent MAC protocol that is a hybrid form of 802.11 and 802.15.4, to achieve a favorable performance that can serve a broad range of applications over the IEEE 802.15.4-based WSNs. We build an analytical model of the proposed protocol based on Markov chain modelling and derive the analytical expressions of the performance metrics, which we then validate against the simulation result sets generated by our in-house built simulation framework. We prove analytically that the probability of collision of the semi-persistent MAC is lower than that of the standard protocol. Based on our theoretical and simulated models, we show that incorporating the semi-persistent feature into existing MAC protocols leads to significant improvement of the performance metrics, including the probability of collision, throughput, energy consumption, transmission delay and reliability, particularly for networks with a large number of sensor nodes