Abstract:

     We collect the received signal strength values of a node at different positions in a wireless sensor network. We then employ Elman neural networks to construct a mapping model relationship between received signal strength and radio wave propagation path distance. We therefore can train such neural networks with the collected signal strength before detect distance with the well trained neural networks. Experimental results show that this approach has better ranging capability.

Key words: Elman neural networks, wireless sensor networks, ranging model