Design and Development of Optimized Cross Layer Framework for Wireless Sensor Networks

Abstract

Wireless networks have an edge over the wired networks in terms of mobility, deployment and ease of propagation. These advantages have led to the development of different types of wireless networks. One such type of networks is a Wireless Sensor Networks (WSN) which are widely used in many real time applications and monitoring various physical phenomena. These networks consist of sensor nodes, which are tiny devices having sensing and computational capabilities and processing powers. These networks are usually deployed in harsh physical environments and it becomes a challenging task for the networks to provide the necessary Quality of Service (QoS) to the user. Moreover these networks are having limited resources which have to be utilized in an efficient way in order to maintain the robustness of the network. Hence there has to be some technique or mechanism which can improve the efficacy of the WSN. Various Cross Layer Designs (CLD) have been proposed on the conventional layers of the WSN in order to save the network energy, increase its lifespan and side by side improve the various network parameters such as throughput, end to end delay, packet delivery ratio, jitter etc. in turn promising a better QoS. In the presented work an attempt is made to propose an optimized cross layer framework. The contributions of this thesis in the area of cross layer design in WSN are many fold. First, we deal with an introduction to WSN, various challenges faced by WSN and their applications. Apart from this, the concept of cross layer design is also introduced which touches the various aspects of interoperability between different layers of the standard OSI model. Second, we discuss and analyze the cross layer design approaches and comparison of the two. Then detailed literature review of various issues and approaches in cross layer design and techniques which are utilized in WSN is given. Apart from WSN the cross layer design proposals suggested for other wireless networks are also given. Moreover, the various cross layer frameworks designed for WSN are enumerated also. Elaborating on this, various residual battery energy models are surveyed which used to model the battery life of the WSN in order to increase the lifetime of the network. Third, an Adaptive Threshold Energy Efficient Routing (ATEER) protocol is developed for heterogeneous WSN. The protocol makes use of the cross layer interactions between transport, data link and physical layers in order to obtain the energy efficiency. The protocol is simulated and results are obtained which show a marked improvement in the network lifetime. Fourth, cross layer algorithm for efficient transmission of images over WSN is designed. This contribution is significant as this remains an unexplored arena/ area. The proposed algorithm is simulated using standard test images. The results show improvement in peak signal to noise ratio (PSNR) and structural similarity index (SSIM). Fifth, a novel optimized framework using CL is proposed. The framework makes use of the basic principle of fuzzy logic and avoids congestion thus improving upon the efficiency as well as the lifetime of the WSN. The framework is named as CLOCA (Cross-Layer Optimized Congestion Aware) framework. Sixth, we conclude with the contributions made in the thesis and put forward some outline for the future work.