An Energy Efficient Cross Layer Design Scheme for Wireless Sensor Networks
Wireless Sensor Networks (WSNs) are wireless networks that have recently drawn significant research attention since they offer unique benefits and versatility with respect to sensing, allowing low-power and low-cost rapid deployment for many applications that do not need human supervision. WSNs are self-created and self-organized by the collection of a large number of sensor nodes interconnected by multi-hop wireless paths. The sensor nodes are network embedded systems with Integrated Chips (ICs) to allow signal processing and micro-sensing. Each wireless sensor node is a micro-electro-mechanical device and can only be equipped with a limited power reserve. While energy consumption occurs in sensing, data processing and communications, care should be exercised to make the most of the expendable power source for the node.
Though considerable research is being done in the area of energy saving techniques for WSNs, most of the proposed techniques have focused on energy awareness at different network layers in WSNs. Furthermore, most of the proposed techniques are based on protocols for mobile ad hoc networks that do not look into the possibility of a cross-layer design strategy that can exploit the unique features of WSNs. There still exists the need for a universal protocol that can be applied to such networks in general. In this thesis, we focus such a research on optimizing the energy consumption by suggesting a novel cross-layer architecture at the network/data-link layer for sensor networks. We have developed a scheme for better and improved energy efficiency in WSNs by combining the ideas of energy-efficient cluster formation and medium access together. Our cross-layer scheme provides good performance in terms of WSN-lifetime, scalability and minimizing network-wide energy consumption. The scheme is based on a collaborative approach supported by formation of dynamic clusters functioning with a traffic aware MAC (medium access control) scheme. Our MAC scheme incorporates a self-learning, traffic adaptive algorithm for varying traffic conditions inherent to the WSNs. The design methodology and results in this thesis aim at producing a reliable and scalable energy-aware sensing network, in spite of node failures, minimizing energy consumption at the same time.
Advisor:Dr. Sandeep K Shukla; Dr. Ira Jacobs; Dr. Amitabh Mishra
School:Virginia Polytechnic Institute and State University
School Location:USA - Virginia
Source Type:Master's Thesis
Keywords:electrical and computer engineering
Date of Publication:10/21/2003