Discrete Hughes‐Hartogs optimized dual vector handoff for spectrum management in cognitive radio networks

• Published in: International journal of communication systems 2021-11, Vol.34 (16), p.n/a
• Main Authors: Kalpana Devi, M., Umamaheswari, K.
• Format: Article
• Language: English
• Subjects: Algorithms; Channels; Cognitive radio; cognitive radio networks; Constraint modelling; expected Q channel maximization; Hughes‐Hartogs; Optimization; Q learning; support vector machine; swarm intelligence
• ISSN: 1074-5351; 1099-1131
• DOI: 10.1002/dac.4950

Summary Cognitive radio network (CRN) has occupied rich awareness on amazing potentiality of spectrum utilization for the secondary users. In CRN, effective spectrum management has received more attention by increasing number of communication devices and several techniques have been evolved in recent years. However, efficient spectrum sensing with minimum delay reducing number of spectrum handoffs and maximum throughputs are some of the major challenges to be addressed. In this proposed work, discrete Hughes‐Hartogs optimal and dual vector spectrum management (DHHO‐DVSM) in CRN are used for effective utilization of channel selection. The expected maximization Q channel selection is designed to obtain an efficient idle channels for improved transmission of the secondary user. The optimal channel is attained by the number of idle channels maintained from the Q channel selection list. Based on each SU requests, discrete Hughes‐Hartogs particle swarm optimal channel selection is proposed to find the best channel by computing the fitness function for discrete factor. Finally, dual constraints‐based vector spectrum handoff model is proposed to produce the hyper plane based on the probability conditional factors with the objective of minimizing the number of handoffs for secondary users. The simulation result shows that the proposed method has transparent edges over conventional channel selection models. In addition, results are verified to learn the optimal channel selection with minimal sensing delay, increasing throughput and minimal handoffs. The proposed algorithm DHHO‐DVSM is 22.8% comparatively better than E‐CRNs and 41% improvement than SpecPSO. The intelligent network focuses on cognitive radio network, which consist of two users, namely, primary user (licensed) and secondary user (unlicensed). The free bandwidth of the primary user is utilized by the secondary user for smooth transmission. The idle primary user channels are sensed using proposed expected Q channel selection (EQCS) model. This model is based on Q learning concept. Due to handoff, the secondary user will request for the idle channel. So the corresponding idle channel selection is sensed with discrete Hughes‐Hartogs particle swarm optimal channel selection. The identified optimal channel is used for the secondary user. Here, number of handoff occurrence is reduced by using the support vector machine. This work mainly concentrates on enhancing the throughput, to procure minimum delay for se