Abstract
WSN applications demand prolonged network operation in which manually replenishing the scarce battery resources of sensor nodes are not usually possible. When the limited energy of sensor nodes is completely exhausted, this leads to reduced coverage and may cause network partitioning, which dramatically reduces the network lifetime. In this respect, network lifetime enhancement is considered the most critical aspect of WSN performance. Prolonging network lifetime requires mechanisms that provide energy consumption balancing (ECB) in addition to energy efficiency (EE). This thesis investigates network lifetime maximization problem and proposes solutions to address both EE and ECB. The scope covers two separate but equally important fronts: duty cycling mechanisms and maximum lifetime routing strategies. Duty cycling significantly reduces the energy consumption of sensor nodes resulting from idle listening. Meanwhile, maximum lifetime routing schemes aim at balancing traffic loads and hence energy consumption among sensor nodes across the network. In this regard, our contributions are in three-folds. First, a distributed sleep mechanism is proposed for the non-beacon-enabled mode of the IEEE 802.15.4 MAC protocol, to support energy-efficient operation in WSNs. In addition to achieving energy savings, our mechanism helps reshape generated traffic so that the overall channel contention is reduced. This effect in turns improves packet delivery ratio at the data sink. Second, we propose a Control-theoretic Duty Cycle Adaptation algorithm (CDCA) to adapt nodes duty cycle based on time-varying and/or spatially non-uniform data traffic loads. The proposed mechanism is distributed; hence each node adjusts its duty cycle independently. We introduce a novel concept called virtual queue, which provides better insight into actual traffic conditions in comparison to existing schemes and prevents excessive packet drop. Using NS-3 simulation, we demonstrate the performance improvements obtained from CDCA in comparison to the state of the art. Furthermore, a stability analysis is conducted to investigate system stability conditions. Third, we formulate the maximum lifetime routing problem as a minimax optimization problem, and numerically obtain the upper bound network lifetime. Moreover, we propose a Distributed Energy-aware Fuzzy-Logic based routing algorithm (DEFL). DEFL makes an appropriate trade-off between energy efficiency and energy consumption balancing and successfully extends the network lifetime under different network conditions. The simulation results demonstrate that DEFL outperforms all tested algorithms and performs very close to the upper bound. Keywords: Network Lifetime Maximization, Energy Efficiency, Energy Consumption Balancing, MAC, Routing, Wireless Sensor Networks.