| Summary: | 碩士 === 玄奘大學 === 資訊管理學系碩士班 === 98 === A design of wireless sensor networks involves deployment of hundreds or thousands of homogenous, and energy limited sensor nodes in a region from where information will be collected. Hence sensor nodes will sense their nearby environment and send the information to a BS which is not energy or computing power limited. While the sensing range is fixed, nodes will transmit the information directly to the BS or via other nodes which are nearer to the BS depend on the distance between them and the BS. Thus, in addition to the information generated by nodes in the networks, these nodes will also have to forward the data from some other distant nodes . Therefore, nodes closer to the BS will lose its battery power sooner. While many (all) nodes around the BS die, the data transmitted from other nodes could not reach the BS. This causes low data packet reception rate for the BS and leads to a bottleneck phenomenon which is also called “doughnut effect” [1]. Previous literatures [2-8] addressed the problem of energy efficient deployment of nodes and base station without considering the impact of bottleneck effect. So these proposed protocols might prolong the network lifetime, but the effectiveness of network such as data delivery rate or network availability was not necessarily good as expected. Though Padmanabh et al. [1, 9] proposed a model to evaluate the bottleneck effect, the model was too simplified to correctly reveal the loadings among nodes in different distance from the BS. Therefore the sensor deployment strategy them proposed [1] can not reach the best performance.
In this paper, we define a node is in h-tier if it is h hop counts away from the BS, and develop an analytic model to evaluate the energy consumption of nodes in different tiers. Based on the evaluations, a more load balanced sensor deployment strategy is developed. The simulation results show that the data reception rate is significantly increased by using the proposed sensor deployment strategy.
The rest of paper is organized as follows. Section 2 introduces some related work. Section 3 describes the evaluation and analysis of tier based power consumption models. Section 4 presents the simulation results in comparison with our evaluations and some previous studies. And our conclusion and future work are presented in Section 5.
|
|---|
