Exploiting Data-Dependent Transmission Control and MAC Timing Information for Distributed Detection in Sensor Networks

Abstract

In this paper, we propose a data-dependent transmission control policy over the slotted ALOHA MAC protocol and a cross-layered fusion rule that exploits MAC timing information for distributed detection in sensor networks. In this system, each sensor first makes a local decision at the beginning of each observation period and transmits the decision to the fusion center over a random access channel. Based on the slotted ALOHA random access protocol, we propose a class of data-dependent transmission control policies that assign to sensors their transmission probabilities according to the reliability of their local decisions. For the case with i.i.d. observations in each time slot, we show that the optimal transmission control function takes on the form of a thresholding function. That is, a sensor will transmit in a given time slot if and only if its local log-likelihood ratio exceeds a certain threshold. When observations are made only every several time slots, the message arrival time at the fusion center, which is spread over the observation period of duration K > 1, will embed the reliability of the received sensors' decisions as a result of the data-dependent transmission control. This timing information can be accounted for in the fusion rule to further enhance performance. Finally, we extend the proposed strategies to multicluster sensor network scenarios, where the sensors' local decisions are transmitted to the fusion center in a two-hop fashion. We show, through numerical simulations, that the proposed schemes outperform both conventional slotted ALOHA and TDMA-based schemes that do not adopt cross-layered transmission and fusion strategies.

Model-Based Design of Tree WSNs for Decentralized Detection

“…Decision fusion over slotted ALOHA MAC employing a collision resolution algorithm is studied in [7]. A thorough investigation of the design of MAC transmission policies to minimize the error probability has been considered in [8], where sensors are assumed non-identical and the MAC policy is assumed stochastic. The cross-layer approach is also considered in [9], where an integrated model for the physical channel and the queuing behavior for sensors is developed.…”