| The cognitive radio(CR)has been widely concerned as an effective technology to improve the spectrum utilization.However,how to manage the serious co-channel interference in the CR network has become an urgent problem.Interference alignment(IA)technology can separate the interference subspace and the desired space,bring more multiplexing gain to the system,and thus enhance the channel capacity of the system.Therefore,it is considered as an effective way to manage interference in the CR networks.The thesis studies the IA technology in the multi user cognitive MIMO system.The thesis firstly focuses on the interference management in the CR network comprised of a single primary user(PU)and multiple secondary users(SU).The requirement that the SU generates no interference to the PU degrades the available degrees of freedom(DoF)of SUs,resulting in low system performance at high signal-to-noise-ratio(SNR)region.To address this,the thesis put forward an IA scheme which maximizes the secondary system sum rate based on dichotomy of PU's DoF,where the PU's eigenmodes are adaptively adjusted according to its rate requirement.In order to further improve the performance of the secondary system and reduce the computational complexity,the proposed algorithm firstly transforms the rate formula into two parts without the inversion operation by using the extensions of the Minkowski inequality,and then selects the two parts as the objective function.Simulations show the proposed scheme can obtain higher network rate while meeting the minimum rate of the main user.Secondly,the thesis focuses on applying IA to the CR network with multiple primary links,and respectively studies the IA scheme over reciprocity channel and non-reciprocity channel.Firstly,an interference leakage minimization algorithm in cognitive MIMO network based on interference subspaces alignment(ILM-ISA)is studied in the light of the reciprocity channel.In the light of the problem that the ILM-ISA scheme ignores the quality of the expected signal,the thesis presents an IA algorithm based on joint minimizing interference dimension and power(JMIDP).JMIDP jointly optimizes the interference subspace dimension and the interference leakage power under the constraint of the trace of the square root of the desired matrix.The presented JMIDP scheme can compress the interference signals to low dimensional space,reduce the leakage interference power,and improve the total rate of the network.Finally,the thesis focuses on the IA scheme design in the CR network with multiple primary users based on the non-reciprocity channel.To avoid the overhead of information exchanged by joint designing precoders and decoders,this thesis proposes an IA transmitter design scheme based on improved interference subspace distance minimization(IISDM).Different from the method minimizing the global interference subspace distances directly,the IISDM scheme reduces the interference dimensions via aligning the subspaces spanned by interference with the aligned subspace introduced,which greatly reduces the number of spatial distances to be optimized.Since the algorithm demands SUs to be equipped with additional antennas to eliminate the interference to the primary links,it leaves lower space dimensions to SUs and is not suitable for scenarios with limited number of antennas in secondary system.Therefore,the thesis further develops a maximum sum rate IA transmitter design based on cross layer interference matrix rank constraint(MSR-RC).MSR-RC considers the spatial distance between the desired subspace and the aligned subspace,and finds the optimal precoders that maximizing the sum rate among those achieving the minimum interference subspaces distances.Moreover,it eliminates the interference to the primary links by constraining the rank of the cross layer interference matrix to be zero.The two proposed algorithms not only have the advantage of the one side optimization of the transmitter,but also have obvious improvement in terms of the average sum rate. |
PDF Full Text Request