UC Irvine

With recent advances in communication, sensing, computing, and battery capacity, wireless sensor networks (WSNs) have emerged as a viable technology for monitoring and surveillance purposes in numerous applications such as precision agriculture, healthcare monitoring, and industrial monitoring. However, battery power depletion has remained the most pivotal factor in network failure since sensors are driven by battery that are infeasible to replenish, especially in hostile environments. This calls for innovative approaches for improving the energy-efficiency of WSNs and extending their lifetime.

Empirical measurements have demonstrated that wireless communication dominates the network’s energy consumption. Node deployment plays a crucial role in energy-efficiency of the WSN since electromagnetic wave propagation dampens as a power law function of the distance between the transmitter and receiver. In this dissertation, by making a resemblance between network nodes/cells and quantization points/regions, I aim to find the optimal deployment, cell partitioning, and data routing that minimizes the wireless communication power consumption of these networks. In particular, I considered the effect of both large-scale path-loss signal attenuation and small-scale signal fading and modeled the node deployment problem as an optimization problem with the total power consumption of the network as its cost function. To tackle the resulting NP-hard optimization problem, I derived the necessary conditions for optimal deployment, cell partitioning, and data routing under various network setups and environmental conditions. My theoretical results are then embedded in iterative algorithms to yield energy-efficient deployment and optimal intercommunication protocol for network nodes.

One of key contributions in this dissertation is addressing challenges that arise under various hardware settings, such as homogeneous versus heterogeneous and static versus mobile nodes, in addition to various network architectures, such as two-tier versus multi-hop. Simulation results show that, regardless of the distribution of events to be sensed by sensor nodes, the proposed deployment algorithms outperform previous state-of-the-art methods in the literature by a significant margin. In particular, the proposed algorithms improved these networks' energy efficiency and lifetime by up to a factor of two compared to existing work in the literature. This, in turn, reduces the cost of such networks and demonstrates their potential as a sustainable, rigorous, and cost-effective monitoring system.