Parley as an approach to distributed detection

On Decentralized Detection With Partial Information Sharing Among Sensors

“…It is now known that likelihood ratio tests remain optimal (under the conditional independence assumption) in every directed acyclic network but, with regard to tractable computation, correctness and efficiency is retained only for sparsely-connected tree-structured networks: more specifically, threshold optimization scales exponentially with the maximal degree of the nodes [9]. Similar algorithmic developments have occurred for decentralized detection architectures with feedback [10], [11], [12], where the fusion center may communicate a preliminary decision to the peripheral sensors for them to take into consideration when forming a next message based on a next observation. References [10] and [11] consider a Bayesian formulation at a level of generality that captures a variety of feedback architectures, allowing the sensors within each stage to exchange preliminary decisions directly (i.e., "peer communication").…”

“…Experiments show that in the presence of feedback, the fusion center's probability of error decreases to zero as the number of feedback stages goes to infinity, and is never larger than that of the star architecture without feedback. Reference [12] considers a similarly general class of architectures but rather focuses on the use of feedback (of all sensor decisions to all sensors) in combination with successive retesting and rebroadcasting of the updated decisions to reach a consensus, an operation described as "parley." Two modes of "parley" are examined.…”

Dynamic Quantizer Design for Hidden Markov State Estimation Via Multiple Sensors With Fusion Center Feedback

“…However, this approach has the potential to reduce the network complexity from another point of view, i.e., in order to achieve a prescribed performance, one only needs to implement fewer sensor nodes compared to the case without feedback. This kind of feedback information pattern has been employed for performance improvement in the sensor network literature, but mainly in the context of hypothesis testing [21], [1], [25], [27], and is referred to as decision feedback. There is an underlying assumption here that the fusion center has sufficient power at its disposal in contrast to the sensor nodes which have limited energy that cannot perhaps be replenished.…”

Building Context Aware Network of Wireless Sensors Using a Scalable Distributed Estimation Scheme for Real-time Data Manipulation

“…Specifi cally, in the context of decentralized detection, cooperation allows sensor nodes to exchange information and to continuously update their local decisions until consensus is reached across the nodes (Quek, Dardari & Win, 2006c;Quek, Dardari & Win, 2006b;Quek, Dardari & Win, 2006a). For example, cooperation in decentralized detection can be accomplished via the use of Parley algorithm (Swaszek & Willett, 1995). This algorithm has been shown to converge to a global decision after sufficient number of iterations when certain conditions are met.…”