We consider the problem of designing a path in a network that is as close as possible to most node locations in the network. In particular, we concern the case of designing mobile element tours in sensor networks, used to collect data from the network. The nodes that are not visited by the tour need to transmit their data using wireless multi-hop routing, which is expensive in terms of energy usage, and therefore limits the lifetime of the network. Nodes that happen to be placed in locations that are away from the central areas of the network can result in designing a tour that is skewed towards these out of the way' or peripheral nodes. In this work, we propose algorithmic solutions that aim to identify peripheral nodes in the network.We provide experimental evidence that the tour planning algorithms that consider peripheral nodes can significantly improve the solution.
|Number of pages||14|
|Journal||International Journal of Ad Hoc and Ubiquitous Computing|
|Publication status||Published - 1 Jan 2018|
- Travelling salesman problem
- Wireless sensor networks