Optimum consultation for parallel distributed detection systems
- PDF / 787,292 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 110 Downloads / 221 Views
(0123456789().,-volV)(0123456789(). ,- volV)
Optimum consultation for parallel distributed detection systems Nedal Al-Ababneh1
Ó Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract An optimum consultation scheme for parallel distributed detection system is proposed. Two-way communication links are assumed between a set of local sensors and the fusion center. The observations of each sensor are assumed to be statistically independent of all other sensors observations. We also show that the proposed consulting scheme can be optimized such that its performance is identical to that of the optimum parallel scheme. It should be emphasized that the proposed consulting system exhibits the advantage of conditional nonrandom consultation between the fusion center and the sensors. Therefore, the sensors are less detectable which gives the system higher chance of survivability than that of the typical parallel system. Performance results for the case of two and the case of three sensors are given and compared with those of the centralized scheme. Keywords Parallel distributed detection systems Decision fusion rule Consultation Probability of detection
1 Introduction The centralized detection systems use a single processor for detection of signals or targets. In these systems, the complete observations are assumed to be available in one place to the central detector. This requires a large bandwidth for the communication channel and requires a considerable processing time. The problem of distributed detection with multiple sensors has been, therefore, considered extensively in the literature [1–24], to mention a few. The team decision making in these multisensor systems is of great improtance as it allows less communication bandwidth in comparison with centralized systems, together with lower design complexity. Moreover, the coverage area and the number of targets they can detect are larger than those of the centralized systems. The sensors in these systems may be connected in parallel, serial, hybrid, or consulting topologies. The most successful structure using multisensor detection is the parallel fusion topology. In this parallel system
& Nedal Al-Ababneh [email protected] 1
Department of Electrical Engineering, Faculty of Engineering, Jordan University of Science and Technology, P.O. Box 3030, Irbid 22110, Jordan
each detector takes a binary decision independently of the others about the presence or absence of the target in a given geographical, and sends its local decision to a fusion center which takes the final decision. The communication between the local sensors and the fusion center is unidirectional in such away that the fusion center only recieves data form the sensors without any commcuncations from the fusion center back to the sensors. In this paper, we propose a parallel distributed system in which a two-way conditional consultation is available when necessary between the local sensors and the fusion center. The final decision of the fusion center in this scheme is either ba
Data Loading...
