Research On Multi-user Interference Alignment Algorithm In Cognitive Spectrum Sharing Networks

With the rapid development of wireless communication services in the field of information and communication, the demand of radio spectrum resources are rapidly expanding. Due to the limitation of the current level of wireless communication equipment and the complexity of the wireless communication environment, the demand of spectrum resources has become increasingly prominent. Cognitive radio(CR) technology brings historic change as an opportunistic way through sharing spectrum. In order to achieve the spectrum sharing between licensed users(primary users) and cognitive users(secondary users) in cognitive radio networks, the problem of the interference generated between primary and secondary users must be resolved. This thesis discusses the multi-user interference alignment(IA) algorithm in cognitive spectrum sharing networks. The main work includes:For multiple primary users and multiple secondary users in multiple input multiple output(MIMO) cognitive spectrum sharing networks, this thesis proposed an interference alignment algorithm, that doesn't need channel reciprocity, and thus alleviates the need to alternate between the forward and reverse network. Firstly, this scheme encodes the transmission signals of secondary users, and sets up the equivalent mode after eliminating the interference between the primary and secondary users. Secondly, this thesis establishes the cost function by maximizing the total capacity and extended to Grassmann manifold, then, a gradient method is applied to obtain the optimal precoding matrices. Finally, the receivers design the filter matrices by maximizing users' signal-to-interference-plus-noise. Simulation results show that the same spectrum efficiency provided by the proposed algorithm compared with some existing typical algorithms at low signal-noise ratio, but the spectrum efficiency are improved well by the proposed algorithm at high signal-noise ratio.Recently, there are few works on IA solutions in heterogeneous cognitive radio networks, even though some schemes only consider the simple case of macrocell and cognitive cell has one user. For the problem of downlink interference cancellation in heterogeneous cognitive radio networks, based on the analysis of multi-user heterogeneous cognitive radio network model, in this thesis, the different algorithms are designed for the macrocell and cognitive cell respectively. The cognitive cell use the cognitive ability of system, and the integrated use of linear forced zero algorithm and total minimum mean square error algorithm for cognitive system to design the cascaded of precoding matrix and the receive filter matrix, and migrate both the cross-layer interference and co-layer interference layer upon layer, thus this scheme can realize the spectrum sharing between the low power cognitive cell and the macrocell. The program has low complexity, moreover, the simulation results show that, the spectral efficiency is higher than the least squares interference alignment algorithm, and it has certain advantages for interference elimination in heterogeneous cognitive radio networks.