Rapid progress in microelectromechanical system (MEMS) and radio frequency (RF) design has enabled the development of low-power, inexpensive, and network-enabled microsensors. These sensor nodes are capable of capturing various physical information, such as temperature, pressure, motion of an object, etc., as well as mapping such physical characteristics of the environment to quantitative measurements. A typical wireless sensor network (WSN) consists of hundreds to thousands of such sensor nodes linked by a wireless medium.

WSNs have created new paradigms for reliable monitoring. They outperform conventional sensor systems, which use large, expensive macrosensors to be placed and wired accurately to an end user. Detailed discussions of such benefits can be found in the literature [1, 13, 31-33, 43]. Some of these benefits are highlighted as follows:

• Anywhere and anytime. The coverage of a traditional macrosensor node is narrowly limited to a certain physical area due to the constraints of cost and manual deployment. In contrast, WSNs may contain a great number of physically separated nodes that do not require human attention. Although the coverage of a single node is small, the densely distributed nodes can work simultaneously and collaboratively so that the coverage of the whole network is extended. Moreover,

Quanhong Wang

Queen's University

Hossam Hassanein

Queen's University

Kenan Xu 9.6

Queen's University sensor nodes can be dropped in hazardous regions and can operate in all seasons; thus, their sensing task can be undertaken anytime.

• Greater fault-tolerance. This is achieved through the dense deployment of wireless sensor nodes. The correlated data from neighboring nodes in a given area makes WSNs more fault tolerant than single macrosensor systems. If the macrosensor node fails, the system will completely lose its functionality in the given area. On the contrary in a WSN, if a small portion of microsensor nodes fails, the WSN can continue to produce acceptable information because the extracted data are redundant enough. Furthermore, alternative communication routes can be used in case of route failure.

• Improved accuracy. Although a single macrosensor node generates more accurate measurement than one microsensor node does, the massively collected data by a large number of tiny nodes may actually reflect more of the real world. Furthermore, after processing by appropriate algorithms, the correlated and/or aggregated data enhance the common signal and reduce uncorrelated noise.

• Lower cost. WSNs are expected to be less expensive than their macrosensor system counterparts because of their reduced size and lower price, as well as the ease of their deployment.

In this chapter, Section 9.2 describes diverse applications of WSNs in various domains with examples and Section 9.3 discusses the classifications of the WSNs according to different criteria. Section 9.4 presents the characteristics of WSNs, highlights how they differ from traditional wireless ad hoc networks, and reviews the technique challenges and corresponding design directions. In Section 9.5, various technical approaches with respect to hardware design, system architectures, protocols and algorithms, and software development are illustrated. Finally, Section 9.6 concludes with emphasis on several possible open issues for future research in the area of WSNs.

Was this article helpful?

0 0

Post a comment