A significant cost in spectrum sensing using multiple sensors is the communication cost associated with distant transmissions to a fusion center. A new formulation for communication-efficient decentralized change detection is proposed where local sensors are memoryless, receive independent observations, and no feedback from the fusion center. Average detection delay and false alarm probability have been previously used for system design and performance assessment. In this paper, we introduce an additional constraint: the average number of communications between local sensors and the fusion center. This metric is able to reflect both the cost of establishing communication links as well as overall energy consumption over time. The proposed algorithm minimizes detection delay with constraints on both false alarm probability and average number of communications. The optimal choice of thresholds in the algorithm are determined by a combination of sequential detection analysis and constrained optimization. Performance is investigated for different scenarios through both analysis and simulation, where the effects of approximations used are evaluated.
展开▼
