Endure - engineering durable and efficient hierarchical wireless sensor networks.

摘要

Battery powered wireless networks are becoming more and more ubiquitous in various forms such as Wireless Sensor Networks (WSN), wireless mesh networks, Wireless Body Area Networks (WBAN), and wireless distributed computing networks. With energy awareness and lifetime becoming critical design concerns in such networks, a significant amount of research has focused on energy-aware design of various network architectures and different layers of the network protocol stack. However, much less has been done in way of incorporating physical layer characteristics at the system deployment stage, applying location-aware and/or infrastructure-aware low power techniques, and analyzing the effects of these techniques on spatial energy balancing across the network (and therefore, the overall network lifetime and the corresponding quality of service).;In this work, physical layer characteristics, wireless channel access schemes, routing protocols, the overall network architecture, and relative node locations are examined and their impact on the network lifetime and the end-to-end delay are investigated. In addition, a number of techniques, which are integrated in a system called ENDURE, are introduced for cross layer design and optimization of a hierarchical wireless sensor network. The presented techniques, which focus on the physical, data link, and network layers, aim to increase the service lifetime of the network. They address both design time and run time decision making by using time varying queuing analysis and incorporating network dynamics. The effect of mobility of data collectors in a hierarchical wireless sensor network with mobile overlays is studied, and a corresponding multi-hop routing protocol is proposed. Given the real time changes in the network, the routing mechanism considers mobility and queuing delays in order to improve the network performance. In addition, a location-aware heterogeneous modulation scheme for a hierarchical wireless sensor networks is presented and its impact on the spatial distribution of energy dissipation, and hence, on the resulting network lifetime is studied. More precisely, the effects of heterogeneous modulation schemes on the time-slot arrangement and end-to-end delay (due to inherent trade-offs in power and bandwidth efficiencies) are analyzed. Finally, a new metric for quantifying the durability of a wireless network is introduced. This metric captures the remaining battery life profile of the network nodes at discrete time instances while accounting for the network functionality and specific design parameters and objectives.;Key contributions of this work may be summarized as follows: (1) Analysis of the effect of different parameters of a hierarchical architecture of a wireless network on the overall network lifetime and the associated performance of the system. (2) Development of energy and mobility-aware routing protocol (3) Development of the battery-aware cross-layer optimization techniques to maximize the life expectancy of a hierarchical network subject to an end-to-end delay constraint for data delivery. (These techniques include location-aware modulation and time slot arrangements.) (4) Analysis of the remaining battery life profile of the whole system and introduction of a novel durability metric to better capture the effective lifetime of a network. (5) Development of an event-based packet-level simulator. This simulator captures high level specification of the system along with physical properties of the wireless links for any hierarchical structure (i.e., flat and multi-tier networks comprised of a combination of mobile or fixed nodes in each level of the hierarchy).;The key intellectual merit of this research lies in addressing energy efficiency of monitoring sensor network as an intrinsic aspect of the distributed data collection tasks. This thesis includes energy-aware algorithms and techniques for wireless network design and deployment as the key enabler for cost-effective realization of many applications.