A stochastic approach for spectrum sensing and sensor selection in dynamic cognitive radio sensor networks

摘要

Cognitive Radio Sensor Networks (CRSN) is currently demanding to deal with spectrum scarcity through opportunistic spectrum access solutions. Opportunistic access to the available spectrum is achieved by determining spectrum holes, which in turn demands to run further signal processing operations on network nodes. Consequently, energy consumption to support these processing algorithms is increased and thus, it remains a major concern in CRSN. Some solutions address this issue via sensor selection during spectrum sensing in static CRSN. In this case, certain nodes participate in cooperative spectrum sensing (CSS) to guarantee proper performance, and the remaining nodes go to sleep to extend network battery lifetime. However, this strategy becomes difficult to apply to mobile sensor networks, where nodes change positions dynamically. This is the case of mobile nodes on Cognitive Radio Internet of Things (CR-IoT) networks, where sensor nodes consume significant energy to support CR operations. Due to the random displacement of nodes, new solutions must be developed in contrast to previous strategies based on on-off nodes from static CRSN, this to cooperate between nodes and also to reduce consumed energy. This paper reports a novel energy-efficient sensor selection technique applicable to dynamic CRSN. Stochastic approaches are developed to describe and determine the minimum total number of awake sensors to participate in CSS. This approach is particularly suited for solutions in which conditions, such as the position of nodes, change randomly. Proposed solution is obtained through the use of "here-and-now" approach considering statistic features of the random distance between each sensor node and a given fusion center. The methodology achieves energy consumption levels comparable to "wait-and-see" approach for networks of reduced size. Additionally, we provide further insights into the statistical nature of the problem to state a proper problem formulation, then to devise solutions accordingly. The analysis and performance of the proposed solution are discussed and illustrated with the aid of simulations. (c) 2019 Elsevier B.V. All rights reserved.