Distributed Consensus-Based Weight Design for Cooperative Spectrum Sensing

Abstract

Abstract-This material is a supplement to the paper "Distributed Consensus-based Weight Design for Cooperative Spectrum Sensing". Section 1 offers related literature review on cooperative spectrum sensing and consensus algorithms. Section 2 presents related notations and models of the consensus-based graph theory. Section 3 offers further analysis of the proposed spectrum sensing scheme including detection threshold settings and convergence properties in terms of detection performance. Section 4 presents the proofs for the convergence of the proposed consensus algorithm, and discusses the convergence of the proposed algorithm under random link failure network models. Section 5 shows additional simulation results.Index Terms-Cooperative spectrum sensing, Weighted average consensus, Cognitive radio networks. ! RELATED LITERATURE REVIEW Related Work in Cooperative Spectrum SensingThe main advantage of cooperative spectrum sensing is to enhance the sensing performance by exploiting the observation diversity of spatially located SUs [1]. By cooperation, CR users can share their sensing information to make a combined decision which is more accurate than individual decisions. Cooperative sensing usually contains two stages: sensing and fusion. In the sensing stage, each SU makes the measurement using appropriate detecting techniques. Among all types of detectors, energy detector is widely applied because it requires lower design complexity and no priori knowledge of primary users, compared to other techniques such as matched filter detection or cyclostationary detection [2]. In the fusion stage, the SU network cooperatively combines the detecting statistics throughout the network and the final decision is made using global information. The key element of cooperative sensing is the cooperation scheme, which decides the SU network structure and the detecting performance. Centralized cooperative sensing and relay-assisted cooperative sensing are two major schemes in literature [1]. Centralized cooperative sensing [5] lets all SUs report their measurement information to a centralized fusion center, then a global decision is made at the fusion center according to certain measurement combining methods. Relay-assisted cooperative sensing [1][6] is a multi-hop cooperation scheme which makes use of the strong sensing channels and strong reporting channels among the SU network in order to improve the overall performance. Relay-assisted sensing can be either centralized with a fusion center, or distributed without a fusion center. Centralized cooperative spectrum sensing requires the entire received data be gathered at one place which may be difficult due to communication constraints [7]. The multi-hop communication of the relay-assisted sensing may bring extra power cost than one-hop communication, since all SUs' sensing data need to be relayed from the network nodes to the fusion center or detection node. In addition, the multi-hop communication paths may degrade the sensing data quality and affect the detection performance sig...

A Voting-Based Distributed Cooperative Spectrum Sensing Strategy for Connected Vehicles

“…Alternatively, a distant neighbor might correctly inform the vehicle about the available channels if there is a permeable obstacle that causes less fading. Zhang et al proposed the weighted distributed sensing mechanism that weights are base on signal-to-noise ratio (SNR) [37]. Although this approach includes all environmental effects such as fading and distance, it assumes all vehicles are in the same decides the same energy detection threshold that may not be true for highly dynamic vehicular environment.…”

Improved weighted average consensus in distributed cooperative spectrum sensing networks

“…Using the central limit theorem for a large number of samples, the i th statistic test T i is asymptotically normally distributed with mean and variance given by where is the noise variance, and the i th SNR of the SUs is given by …”

“…There are some techniques in distributed/decentralised cooperative sensing such as belief propagation, alternating direction method of multipliers, and consensus algorithms (CA). ()…”

“…However, the major part of existing techniques in the literature results in performance similar to the EGC centralised cooperative sensing, which is called simply average consensus (AC). Zhang et al proposed a novel consensus technique able to ensure a soft centralised cooperative sensing under the MRC rule. In the work of Ashrafi, a binary consensus technique was developed to guarantee a superior performance to the quantised AC.…”

Trust-Aware Consensus-Inspired Distributed Cooperative Spectrum Sensing for Cognitive Radio Ad Hoc Networks

“…BACKGROUND AND RELATED WORK Recently, average consensus algorithms [8] including gossip-based protocols [7] and linear iteration-based schemes [6] [10] have been exploited for the DCSS applications [11] [12] [13] [14]. However, all of the existing consensus-based DCSS schemes require the individual SUs to have some type of knowledge about the network topology.…”