Cooperative spectrum sensing in cognitive radio networks

Abstract

The demand for radio spectrum has dramatically increased over the past decade due to the growing proliferation of wireless services and applications. However, under the current static spectrum allocation policy, almost all the spectrum has been allocated and that there is a shortage of spectrum for new wireless services. On the other hand, actual measurements of the spectrum usage have shown that most of the licensed spectrum is largely under-utilized. Cognitive radio (CR) is proposed as a promising technology to increase the efficiency of spectrum usage by introducing secondary (unlicensed) users to opportunistically or concurrently access the spectrum allocated to primary (licensed) users. To enable the secondary users (SUs) to utilize the under-utilized spectrum, they need to obtain the necessary observations about their surrounding radio environment. Therefore, spectrum sensing is required by the SUs to learn about the activities of the primary users (PUs). However, the performance of spectrum sensing is limited by multipath fading and shadowing which are the fundamental characteristics of wireless channels. To overcome these challenges, cooperation among SUs to perform spectrum sensing has been proposed in the literature. This thesis studies the designs of cooperative spectrum sensing schemes and exploits the spectrum sensing information to improve the performance of the cognitive radio networks (CRNs) in various scenarios. First, in a small area CRN, where the SUs are close to one another such that the detection signal-to-noise ratios (SNRs) at the various SUs do not vary much, a suitable cooperative sensing scheme is the k-out-of-N fusion rule. In this scenario, the fusion rule threshold, common energy detector threshold, and the sensing time are proposed to be optimized in order to maximize the throughput of the CRN. In this thesis, a closed-form equation for the optimal common energy detector threshold is derived and an iterative algorithm to jointly obtain the optimal fusion rule threshold and the sensing time is proposed. In a large area CRN, where the detection SNRs at various SUs can differ greatly, weightings should be given to the decisions of different SUs. Therefore, weightings of the SUs' decisions have to be designed as well under large area CRNs. In this thesis, the optimal decision fusion rule is derived and is shown to be a weighted sum of the decisions based on the likelihood-ratio test. An iterative algorithm is also proposed to compute the individual SUs' energy thresholds and the fusion rule threshold. Performing spectrum sensing and transmission of sensing data require energy. This can be detrimental to the CRN if the SUs are battery-powered wireless devices where their energies are limited. In this scenario, maximizing the energy efficiency of the CRN is important. Therefore, cooperative spectrum sensing parameters are designed to maximize the energy efficiency of the CRN. An iterative algorithm is proposed to find the optimal fusion rule threshold and the expression at which the energy threshold is optimal is derived. The impacts of the number of cooperating SUs and the length of the sensing time to the energy efficiency of the CRN are also studied in this thesis. Finally, instead of using the spectrum sensing information just for detecting the PUs, it is used to design power allocation strategies as well. Optimal power allocation strategies under both cooperative and non-cooperative spectrum sensing schemes to maximize the average data rate and to minimize the outage probability of the SU are derived in this thesis.

Keywords

Affiliated Institutions

• Georgia Institute of Technology US

Related Publications