A relay subset selection scheme for Wireless Sensor Networks based on channel state information
Seyyed Hamed Mousavi, Javad Haghighat, Walaa Hamouda · 2016
We propose a relay subset selection method for two-hop Wireless Sensor Networks (WSNs) where the source-relay links are modeled as time-varying fading channels. Assuming perfect channel estimation at relays, the channel state is quantized to binary levels, called as Good and Bad states, and is modeled by a Gilbert-Elliott channel. The relays compress their quantized channel state information and transmit to a fusion centre. The fusion centre selects the smallest possible subset such that each transmitted source bit is received in a Good channel state by at least one of the relays in that subset. We show through simulations that selecting this subset for relaying the information to the source, reduces the transmission power compared to a conventional all-relay transmit scheme, while maintaining the end-to-end bit error rate below a desired threshold.