The visual sensing capability of a Visual Sensor Network (VSN) makes it a very effective tool for applications such as large scale surveillance, environmental monitoring and object tracking. The image sensing, processing and storing functions of the VSN, combined with its function of transmitting and forwarding data towards the sink, consumes more energy and increases the well known energy-hole problem in the network. In this paper, we propose to deploy a Gaussian distributed relay network over pre existing uniform random VSN so as to avoid the energy-hole problem. By forming a heterogeneous wireless sensor network with low-cost relay nodes (RNs), lifetime of the VSN is prolonged with minimal additional cost. We use the energy model of image compression enabled VSNs and determine optimal parameters for the Gaussian deployment of the relay network.

Abstract-The most important criterion for achieving the maximum performance in a wireless mesh network (WMN) is to limit the interference within the network. For this purpose, especially in a multi-radio network, the best option is to use non-overlapping channels among different radios within the same interference range. Previous works that have considered non-overlapping channels in IEEE 802.11a as the basis for performance optimization, have considered the link quality across all channels to be uniform. In this paper, we present a measurement-based study of link quality across all channels in an IEEE 802.11a-based indoor WMN test bed. Our results show that the generalized assumption of uniform performance across all channels does not hold good in practice for an indoor environment and signal quality depends on the geometry around the mesh routers.

No abstract

No abstract

No abstract