Sensing, allocation and trading of spectral resources in cognitive radio networks

• Main Author: Nguyen, Duy Duong
• Other Authors: A. S. Madhukumar
• Format: Theses and Dissertations
• Language: English
• Published: 2013
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Wireless communication systems; DRNTU::Engineering::Electrical and electronic engineering::Antennas, wave guides, microwaves, radar, radio
• Online Access: https://hdl.handle.net/10356/54915
• Institution: Nanyang Technological University

The invention of cognitive radio concept is to overcome the spectral scarcity issues of emerging radio systems. In cognitive radio, the secondary users can either opportunistically access the licensed spectrum in the absence of their primary users, or buy/rent the spectrum access rights from the primary users. For the former case, spectrum sensing and resource allocation are highly essential. While spectrum sensing aims to identify the unused spectrum through detecting the presence of primary users’ signal, resource allocation exploits information obtained from spectrum sensing to schedule the unused spectrum among the secondary users. For the latter case, spectrum trading is emphasized. The problems of spectrum sensing, resource allocation and spectrum trading in cognitive radio networks are focused on this thesis. For spectrum sensing, due to its sensing accuracy, the cooperative spectrum sensing rather than stand-alone spectrum sensing is studied. The challenges arising when secondary users have no a priori knowledge of primary users’ bands are discussed. These challenges are to dynamically detect the primary user’s bands and to reduce the overheads required to facilitate fusion center operation. A dynamic band clustering algorithm that uses K-means clustering technique and a novel entropy-based decision-fusion technique are proposed to address these challenges, respectively. The proposed decision-fusion technique is shown to be comparable to conventional information-fusion techniques while reducing significantly the bandwidth overheads. For resource allocation problem, two non-cooperative games named Interference Minimization game and Capacity Maximization game which reflect the targets of data radios and voice radios are proposed, respectively. From the extensive simulation, it is shown that these proposed games help improve the average SINRs and capacities of all secondary users. Finally, the problem of spectrum trading in a cognitive radio network with multiple primary users competing to sell spectrum to a secondary user is addressed. It is assumed that the spectrum requirements for the primary users’ services are time varying, and the spectrum trading process is carried out before the realization of these values. Both cases when the spectrum demand of the secondary user is a deterministic function of price, and when it is uncertain and may vary from time to time, are considered. In the first case, a Cournot game model of competition is proposed to model the spectrum trading. In this model, the primary users compete with each other by setting the size of spectrum to sell. The Nash equilibrium for a static game when the players (i.e., the primary users) have complete information on other players is studied first. The proofs for existence and uniqueness of Nash equilibrium are also provided. Next, a dynamic game, in which the players adaptively change their strategies to reach the Nash equilibrium, is also discussed. Furthermore, the trading problem is extended to a scenario which involves multiple secondary user networks. In the second case, a Supply Function Equilibrium model is proposed. In this model, the primary users compete to sell spectrum by submitting their supply functions to the secondary user. Similar to the first case, the Nash equilibrium is studied for the static game and an iterative algorithm to reach the Nash equilibrium is proposed for the dynamic game. Detailed simulation results are provided in both cases to prove the robustness of the theoretic analyses. In summary, the novel technique and algorithms for an accurate and low spectrum over-head cooperative spectrum sensing system are proposed. The sensing system can be deployed in an extreme scenario where no a prior knowledge of primary users’ bands is given. Furthermore, the innovative methods to allocate unused spectrum among the secondary users who have different objectives are also proposed. Finally, the frameworks to analyze the spectrum trading among multiple primary and secondary users under realistic network scenarios are also proposed.