| Hi-index | 0.01 |
A wireless sensor network is an application-specific information gathering platform where sensors are required to sense their vicinity (sensing coverage) continuously, consuming highly limited resources such as energy which may not often be replenishable. Thus, an important issue in sensor networks is to design energy-aware algorithms and protocols that optimize energy consumption with a goal to extend the network lifetime while meeting the user requirements such as coverage and data reporting latency. The sensitivity to these requirements varies depending on the type of applications, implying that the designed algorithms and protocols must also be application-aware . In this dissertation, we propose a novel framework for energy and application-aware data gathering in wireless sensor networks. More specifically, our framework includes two strategies: a cluster-based delay-adaptive data gathering strategy (CD-DGS) and a coverage-adaptive data gathering strategy (CA-DGS). The first strategy, called CD-DGS, is based on a two-phase clustering scheme that requests sensors to construct two types of links: direct and relay links. The direct links are used for control and forwarding time-critical sensed data. On the other hand, the relay links are used only for data forwarding based on the user delay constraints, thus allowing the sensors to opportunistically use the most energy-saving links and forming a multi-hop path. Simulation results demonstrate that CD-DGS saves a significant amount of energy for dense sensor networks by adapting to the user delay constraints. The second strategy, called CA-DGS, is based on a trade-off between sensing coverage and data reporting latency. The basic idea is to select in each round a minimum of k data reporters (sensors) which are sufficient for the desired sensing coverage (DSC) specified by the users/applications. For selecting k reporters in a round, we make use of three efficient coverage-adaptive random sensor selection (CANSEE) schemes. These reporters form a data gathering tree and are scheduled to remain active for that round only. This process incurs some delay but saves energy. We derive a probabilistic bound on k and also estimate the probability for having almost surely k data reporters in each round. Finally, we apply the Poisson sampling technique to improve the spatial regularity of the selected k sensors and propose an enhanced selection scheme, called constrained random sensor selection (CROSS). Probabilistic analysis shows that the CROSS scheme improves the connectivity of the selected sensors and reduces the variance on the sensor covered area in each round. Simulation results demonstrate that CA-DGS results in a significant conservation of energy with a small trade-off in terms of data reporting latency. In particular, the higher the network density, the higher is the energy conservation without any additional computational overhead.