Decision fusion with censored sensors

Motivated by the sensor censoring idea, we consider a canonical decentralized detection problem with each sensor employing an on/off signaling scheme. Novel to the current work is the integration of fading transmission channels in the fusion algorithm design. The on/off signaling, in addition to its low communication overhead which is crucial to bandwidth limited applications, also enables the decision fusion to be carried out without the knowledge of channel phase. Resorting to incoherent detection schemes, we develop optimal fusion rules for the following two scenarios: 1) when the fading channel envelopes are available at the fusion center; 2) when only the fading statistic is available. Under the low signal-to-noise ratio regime, we further reduce the optimal fusion rule into simple nonlinearities that are both easy to implement and are not subject to prior knowledge constraints.