Design and Optimization on Mobile Data Gathering in Wireless Sensor Networks

Abstract

Wireless sensor networks (WSNs) have emerged as a new information-gathering paradigm for taking spatial and temporal measurements of a given set of real-word parameters. In these applications, sensors monitor the environment and route their sensing data back to a static data sink. As the routing task depends purely on sensors themselves, the sensors near the sink need to relay much more packets than the sensors far away from the sink. As the result, it would incur substantial and non-uniform energy consumption among sensors. Therefore, how to efficiently aggregate the information from scattered sensors, generally referred to as data gathering, is an important and challenging issue in WSNs as it largely determines network lifetime. Recent studies have shown that significant benefit can be achieved in WSNs by employing mobile collectors for data gathering in WSNs via short-range communications. In such kind of mobile data gatherings, the mobile collectors roam over the sensing field with controlled mobility, perform appropriate actions to schedule data collection, and transport data back to the data sink, while sensors are engaged in sensing task and only need to relay data for local aggregation if necessary. In this way, energy can be greatly saved at sensors as mobile collectors fully or partially take the burden of routing away from sensors.This dissertation focuses on scheme design and performance optimization of mobile data gathering in WSNs. We address several important issues and propose a suite of algorithms to improve data gathering performance. First, we consider utilizing spatial-division multiple access (SDMA) to achieve concurrent data uploading from multiple sensors to the mobile collector. The moving tour of the mobile collector is determined based on the tradeoff between the shortest moving path and full utilization of SDMA among sensors. This joint design can lead to prolonged network lifetime as well as shortened data gathering latency. Second, we extend such joint design of mobility and SDMA technique to large sensor network with multiple mobile collectors. A region division and tour planning algorithm is proposed to balance the data gathering time among different regions. Third, we explore inherent tradeoff between energy saving and data gathering latency by proposing bounded relay hop mobile data gathering. In this scheme, multi-hop relay for local data aggregation is incorporated into mobile data gathering, while the relay hop count is constrained to a certain level to limit energy consumption at sensors. Fourth, we optimize the mobile data gathering performance by characterizing the data gathering strategies as a pricing mechanism, where sensors independently adjust their payment for the data uploading opportunity to the mobile collector based on the shadow prices set by the mobile collector. Fifth, we study the problem of how to achieve optimal performance of mobile data gathering based on a flow-level network model. We jointly consider data rate control at sensors, multi-hop routing for data transmissions, and sojourn time allocation for the mobile collector. We propose distributed algorithms to implement these strategies so as to achieve system-wide optimum. Finally, we propose joint design of mobile energy replenishment and mobile data gathering in wireless rechargeable sensor networks. The mobile entity plays not only as a data collector but also as an energy transporter to deliver energy to sensors via wireless energy transmissions. We present distributed algorithms to provide timely energy recharge to maintain perpetual network operations, meanwhile achieving high-performance data gatherings.