Channel ranking based on packet delivery ratio estimation in wireless sensor networks

Abstract

Wireless sensor networks (WSNs) operating in 2.4 GHz unlicensed bands must explore favorable channels in order to mitigate the effects of induced interference by co-existing wireless systems and frequency selective fading. In this context, we develop a packet delivery ratio (PDR) estimation method for channel ranking in WSNs. The PDR, in general, is defined as a function of signal-to-noise ratio (SNR) and signal-to-interferenceplus-noise ratio (SINR) at the sensor and the packet collision-time distribution of the sensor link. The collision-time distribution depends on the packet size and packet inter-arrival time distributions of both networks. Under limited channel measurements, the collision-time cannot be estimated satisfactorily. In order to bypass the collision-time estimation process, the proposed PDR estimation method utilizes signal level, interference and noise characteristics identified by spectrum measurements adjusted to the intended traffic pattern of the sensor link. The proposed method is validated against the empirical PDR using off-the-shelf sensor platform in emulated multi path wireless fading channels. The results reveal that the method is accurate in modeling the empirical PDR with limited channel energy measurements. In addition, we used the estimated PDR as a metric for channel ranking and verified its effectiveness by ranking the available channels to a WSN under interference from multiple WLANs in a real environment.