Research On Energy Efficiency Optimization For Cognitive OFDM Systems

With the evolution and development of wireless communication technologies, the number of users and service traffic of wireless communications is explosively growing. The energy consumption is simultaneously increasing. Research on energy efficiency optimization is instrumental to realize energy-saving and emission reduction and to achieve green communication. Meanwhile, cognitive radio is expected to effectively improve spectrum utilization. OFDM, as one of the key technologies for 4G/5G, enables convenient spectrum access with the characteristic of fine spectrum utilization for cognitive radio. Thus,cognitive OFDM systems have attracted much attention. Based on the traditional research of energy efficiency for OFDM systems and spectrum efficiency for cognitive OFDM systems, considering the typical characteristics of cognitive radio, this paper studies the energy efficiency optimization for cognitive OFDM systems to utilize the spectrum and energy appropriately.First, this paper investigates the energy efficiency optimization for cognitive OFDM systems with high mobility and perfect spectrum sensing.Considering the different fading experienced by different subcarriers, inter-carrier interference (ICI) and the protection for primary systems, an optimal power allocation scheme is proposed by Lagrange multiplier method. Meanwhile, an optimal subcarrier bandwidth is found to maximize the energy efficiency.Furthermore, we extend to the cognitive OFDM systems with imperfect spectrum sensing. Analysis finds that sensing duration and detection threshold indirectly affecting the energy efficiency by affecting detection probability and false alarm probability. Therefore, we try to maximize the energy efficiency by adapting the sensing duration, detection threshold and transmit power to the constraints of the interference to the primary network in a statistical manner. An alternative iteration framework is designed to obtain the near optimal solution to the original problem.Then, the case with identical detection threshold is extended to the case with individual detection threshold for each subcarrier. Simulation results show that the proposed alternating optimization framework can approach rapidly to the optimal solution and compared to the existing schemes, both cases with identical and individual detection thresholds can achieve a considerable energy efficiency gain.Then, considering the high mobile scenario with imperfect spectrum sensing, we formulate an energy efficiency model with inter-carrier interference. By referring to the non-cooperative game theory, an ICI-aware power allocation and sensing duration optimization scheme is presented. The adaptive subcarrier bandwidth is obtained by golden search.Simulation results show that, compared to the static power allocation scheme or static sensing duration, the proposed algorithm rapidly converges and improves the energy efficiency, which is feasible in practice.