|Abstract:||As wireless devices are more widely used, it is clear that security and energy consumption are major concerns. From a energy perspective, it is increasingly evident that marginal gains in battery energy density necessitate energy efficient protocols. In the security realm, growth in the value and amount of information being transmitted over wireless channels demands confidentiality and integrity.
In the energy efficiency domain, this dissertation focuses on the wireless interface since this has been identified as a major source of energy consumption on devices such as sensors. Within this domain, many previous approaches propose using fixed listening and sleeping intervals regardless of the network conditions. We propose adaptive listening and sleeping techniques where these intervals are adjusted based on observations of traffic patterns and channel state. Another shortcoming of many power save protocols is that they wastefully listen for entire packets as a wake-up signal. In this dissertation, we propose carrier sensing techniques that reduce the cost of checking for such signals.
In the security domain, this dissertation looks at key distribution in wireless sensor networks. Because such devices may face severe resource constraints, symmetric keys are used since public-key cryptography may be infeasible. Previous approaches to this problem include key predistribution, and broadcasting plaintext keys, under the assumption that few eavesdroppers are present during key discovery. However, drawbacks to these approaches include poor secure connectivity or degraded security when several eavesdroppers are in the network. Our work exploits the underlying wireless channel diversity to address the problem. In doing so, our key distribution protocol effectively addresses the drawbacks of previous techniques.
The major contributions of this dissertation are: (1) leveraging multiple channels to improve the connectivity and security of key distribution, (2) proposing adaptive power save mechanisms to reduce energy consumption, and (3) improving power save protocols by using carrier sensing to enhance their energy efficiency