Coverage Protocol for Wireless Sensor Networks Using Distance Estimates

Abstract

In wireless sensor networks, it is desirable to maintain good sensing coverage while keeping the number of active sensor nodes small to achieve long system lifetime. Existing coverage algorithms assume some form of localization and that complete coverage is needed. Our work is motivated by two observations. First, localization is expensive, error-prone and not required for a coverage algorithm. Instead, coverage algorithm can be designed using only distance information between two nodes. Second, complete coverage can lead to excessive redundancy, resulting in inefficiency in terms of active nodes needed. In this paper, we first present a scheme that estimates the distance between any two neighboring nodes using only local information. We then present <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">CCP</i> , a configurable coverage protocol that requires only neighbor distance information. CCP can be configured such that at least <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">alpha</i> portion of the area will be covered by active nodes with high probability, where <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">alpha</i> is a tunable parameter. In addition, CCP achieves similar performance to OGDC [1] in terms of coverage and active nodes required. Compare to OGDC, CCP offers the flexibility to trade-off between coverage and nodes required. By setting the coverage objective to 90%, about 22% node savings can be achieved.