Research On Spectrum Sharing Strategies In Cognitive Radio Networks

Recently, with wireless communication technology developing rapidly, wireless network usersdemand for broadband service is skyrocketing. Traditional schemes of spectrum allocation, which assign spectrum for wireless networks and devices in fixed and static ways, gradually show their inherent disadvantages that in some certain frequencies (such as the ISM frequency) it is very crowded, while in another part (such as the television frequency), the spectrum utilization is very low. Thus, in order to improve the spectrum utilization and provide consumers with wonderful broadband service, cognitive radio technology arises. In cognitive radio networks, the devices equipped with cognitive ability (i.e. cognitive users) can dynamically sense the spectrum that are not occupied in time or space by the licensed users, and then intelligently adjust their center frequency and bandwidth to share the licensed spectrum with licensed users. In dynamic spectrum sharing strategies, the wireless resources that can be optimally used include frequency resource, space resource, and time resource. This paper carries out some research on dynamic spectrum sharing strategies for optimizing frequency resource and space resource. The main contents are as follows:1) Since the licensed users use the spectrum dynamically, the frequency resource available for cognitive system is fragmented unevenly, which reduces the spectrum utilization. Therefore, cognitive radio networks must be able to deal with the spectrum fragmentation problem. This paper presents two adaptive spectrum access strategies:one is based on higher-layer solutions to effectively aggregate spectrum fragments; the other suppresses the impact of spectrum fragmentation successfully at the physical layer. These two strategies not only select the best transmission channel but also efficiently solve the fragmentation problem. The simulation results show that both strategies are able to use the spectrum efficiently and improve the system throughput greatly.2) In the cognitive radio (CR) multi-user multiple-input single-output (MU-MISO) broadcast channel, eliminating multiuser interference and maximizing sum-rate is the optimization goal of the system. Precoding technology is the key to achieve the goal and realize the spectrum sharing between cognitive users and primary users. Since the MMSE-based linear precoding scheme is better to balance the computational complexity and the system performance, in this paper, we focus our research on the MMSE transmit optimization for CR MU-MISO-BC, where the cognitive users are subject to not only a sum power constraint, but also an interference power constraint. Unlike the optimization problem with single constraint, which is easy to solve, this multi-constraint optimization problem is difficult to deal with. Thus, we propose an efficient two-level iterative algorithm to solve this multi-constraint non-convex optimization problem. Firstly, this multi-constraint CR MISO-BC optimization problem is transformed into an equivalent single-constraint BC optimization problem with multiple auxiliary variables. Then, fixing these auxiliary variables, we transform this equivalent single-constraint BC problem into its dual Multiple Access Channel (MAC) optimization problem, which is able to be solved by the proposed inner Iterative Power Allocation Algorithm. Finally, the proposed outer Complete Iterative Algorithm is utilized to obtain the solution of the primal CR MU-MISO-BC problem. Our simulation results are provided to corroborate that the proposed algorithm not only eliminates the multi-user interference efficiently, but also realizes the sum-rate optimization.