Interference Alignment For MIMO Cognitive Radio Networks

The explosive growth of traffic makes the existing congestion of communication network becoming more and more serious, while the shortage of spectrum resources has become a bottleneck in the wireless and mobile communications and its sustainable development service. Given the described situations from the above, national spectrum management agencies have brought on a fierce discussion about how to improve the utilization of the limited radio spectrum to deal with the conflicts between spectrum congestion and under-utilization.Cognitive radio (CR) has been recently proposed as a smart and agile technology that allows non-legitimate users to utilize licensed bands opportunistically. CR should have the abilities to sense the unemployed spectrum, select the frequency dynamically and control transmitted power. However, the interference caused by sharing the same radio channel becomes an obstacle that limits the whole system performance such as system throughput. Thus, another aim of this system should be to consider bandwidth sharing for the cognitive network without affecting the work of primary user and maximize the throughput of the network. However, throughput could be greatly improved by multiple input multiple output (MIMO) technique and interference alignment is an important technology which eliminates mutual interference in the communication network. Given the above techniques, it’s available to apply interference alignment to MIMO cognitive radio networks to improve spectrum utilization and minimize the mutual interference within appropriate BER.From the perspective of the above problems and goals, the paper combines the advantages of MIMO and IA. An approach has been proposed to solve the bandwidth sharing between primary users (PUs) and secondary users (SUs) so as to improve spectrum efficiency, system throughput and minimize mutual interference. Main aspects of the job will be listed as follow. First, the ideological and theoretical of interference alignment algorithm is described in this paper, using mathematical formulas to prove that the algorithm in symmetric system can be converted to an equivalent eigenvalue problem. In a way, it proves the correctness of interference alignment algorithm and the existence of encoding and decoding vectors. Secondly, the feasibility analysis of IA algorithm is transformed into seeking solutions to linear algebra equations. So whether system is feasible for IA can be judged by using linear algebra method. Besides, some corresponding theorems and conclusions have been drawn. Thirdly, the concept of interference leakage has been introduced. From the energy point of view, whether the system is feasible has another method to determine. Obviously, the results of both methods are consistent. Last but not least, maximizing the SNR output of the CR network and minimizing mutual interference are conveyed by corresponding objective functions and constrains, then using convex optimization equation to solve the power allocation problem of PU and SUs. Finally, all the previous theoretical studies are applied to a specific hybrid network which contains three users. The simulation results further confirmed the correctness and feasibility of all the above theories.