Energy consumption minimization and throughput improvement in cognitive radio networks by joint optimization of detection threshold, sensing time and user selection Mohammad Sadeghian Kerdabadi 1 • Reza Ghazizadeh 1 • Hamid Farrokhi 1 • Maryam Najimi 2 Ó Springer Science+Business Media, LLC, part of Springer Nature 2018 Abstract Cooperative spectrum sensing schemes proposed to solve the hidden terminal problem and mitigate multipath fading and shadowing effects, which enhance the sensing performance and throughput in cognitive radio (CR) neworks. However, increasing the number of cooperative SUs leads to more communication overhead, which will increase the energy consumption of the CR network. In this paper, a new scheme is proposed to solve the joint optimization problem of the sensing time, the detection threshold and the selection of the sensing and data transmitting secondary users (SUs) for improvement of the throughput and minimization of the energy consumption of the CR network under the constrains on the global probability of detection and global probability of false alarm. For these purposes, we find the optimal values of detection threshold and sensing time such that the detection constraints are satisfied. The convex optimization methods are used to determine the sensing and data transmitting SUs. The simulation results show that there exists the optimal detection threshold and sensing time for selected sensing and transmitting SUs that can improve the average throughput and minimize the energy consumption of the CR network in comparison to other schemes. Keywords Cognitive radio  Cooperative spectrum sensing  Throughput  Energy  Detection threshold  Sensing time 1 Introduction Cognitive radio (CR) has been recently proposed as a candidate solution to alleviate spectrum shortages by allowing unlicensed users (secondary users, SUs) to opportunistically access or share the frequency bands allocated to the licensed users (primary users, PUs) when they are detected to be idle [1]. In order to avoid causing harmful interference to the PU, the CRs need to monitor the PU’s activities periodically to find a suitable spectrum band for possible utilization. Thus, effective and efficient spectrum sensing is an essential component for CR sys- tems. The spectrum sensing process can be performed via several methods, but energy detection is still the common method among them, as it can be simply implemented and requires no a priori information of PU’s signal. However, the performance of energy detection in face of multipath fading, shadowing, and unknown noise power profiles significantly degrades [2]. Cooperative spectrum sensing scheme is an effective approach to overcome these prob- lems by sharing and combining the sensing results of SUs to make the more accurate decision about the presence of the PU [3]. There are two metrics to evaluate the perfor- mance of a sensing algorithm: detection probability (p d ) that indicate the probability of a SU correctly detecting the presence of the primary signal and false alarm probability & Mohammad Sadeghian Kerdabadi Mohammad.sadeghian@birjand.ac.ir Reza Ghazizadeh rghazizade@birjand.ac.ir Hamid Farrokhi Hfarrokhi@birjand.ac.ir Maryam Najimi maryam_najimi1361@yahoo.com 1 Department of Electrical and Computer Engineering, University of Birjand, Birjand, Iran 2 The Faculty of Electrical, University of Science and Technology of Mazandaran (USTM), Behshahr, Mazandaran, Iran 123 Wireless Networks https://doi.org/10.1007/s11276-018-1797-x