Link-Centric Probabilistic Coverage Model for Transceiver-Free Object Detection in Wireless Networks

• Authors: Dian Zhang;Yunhuai Liu;Lionel M. Ni
• Venue: ICDCS '10 Proceedings of the 2010 IEEE 30th International Conference on Distributed Computing Systems
• Year: 2010

Sensing coverage is essential for most applications in wireless networks. In traditional coverage problem study, the disk coverage model has been widely applied because of its simplicity. Though notable recent works point out that the disk model has many critical limitations when applied in practice, few successful works have been conducted to comprehensively study the issue. Motivated by this, in this paper we propose a new coverage model called T-R model. T-R model is derived from a real application of transceiver-free object detection. Compared with the traditional disk model, T-R model is able to describe many new coverage features such as the probabilistic coverage, the link-centric coverage units and the correlations between multiple coverage units. These new capabilities make T-R model a better abstraction of individual sensors. To evaluate the performance of T-R model, we conduct comprehensive empirical studies based on a test-bed of 30 telosB nodes. Experimental results show that the TR model can adequately describe the sensing behavior in the transceiver-free object detection applications. The average error between the model and the reality is only 8%. Moreover, T-R model presents attractive flexibility, making it more appropriate for general coverage problem studies than the transceiver-free object detection.