Research On Interference Alignment Technology In Cognitive Bilayer Heterogeneous Networks

Femtocell is a type of small cell usually deployed by residential users. It provides better coverage and high speed data service for indoor users via internet backhaul and shares the same frequency band with the existing macrocell network. As femtocell networks and existing macrocell network share the same spectrum, it will cause inevitably high level interference when femtocell users and macrocell users are communicating over the same bandwidth. Interference alignment is an interference management method that can effectively decrease the interference in the same band.Interference alignment overlaps interference in the same signal space at receiving terminal by precoding so as to thoroughly eliminate the influence of interference on expected signals. This thesis will focus on the application of interference alignment technique in the two-tiered cognitive heterogeneous networks and attempt to solve problems in practical networks. Besides, as an efficient way to improve the availability of frequency, cognitive radio will also be discussed in this thesis.Firstly we discuss the regionalized interference alignment in two-tiered cognitive heterogeneous networks. Characteristics of interference distribution in two-tiered networks are considered to divide the cell deployed by this structure into two regions.Cognitive radio system determines which region the macrocell users locate in based on channel state information, and suppresses interference by designing their precoders to achieve interference alignment, hence it is termed regionalized interference alignment. Thereby, it can manage interference in different situations effectively.Regionalized interference alignment ensures the performance of the cognitive heterogeneous networks.A two level precoding scheme has been presented to solve the case of muti-femtocell network coexisted with macrocell network. Firstly, we define co-layer interference as the interference between femtocell and femtocell, correspondingly, we define cross-layer interference as the interference between macrocell and femtocell,and use precoders at macrocell users and femtocell users respectively. The precoders of macrocell users minimize the cross-layer interference. The precoders of macrocell users minimize the cross-layer interference, and mean square error minimization algorithm would be used to handle co-layer interference, thus it will reduce interference of femtocell at last and ensure the quality of service of femtocell users.There are two constrain conditions on using interference alignment technique in the network which contains multi-femtocell and macrocell network, one is the signal overhead and the other is the algorithm complexity of the iterative interference alignment algorithm with more and more users. To solve these problems, this thesis presents a clustering algorithm based on cognitive radio technique, it organizes the femtocell users according to their position. If the signal to interference plus noise power ratio of femtocell users to femtocell base station are better than signal to interference plus noise power ratio of marocell users to macrocell base station and the access number is not reach ceiling, then the macrocell can access the femtocell base station. Otherwise macrocell users will be viewed as interference source, it will be eliminated by interference alignment technique.In conclusion, the main innovations include the following aspects. Firstly, the regionalized interference alignment in two-tiered cognitive heterogeneous networks is presented. Secondly, using different precoder to solve multi-user interference alignment, which enhances the adaptive of interference alignment technique. At last,the clustering algorithm simplifies interference alignment algorithm, and uses cognitive radio technique to improve the availability of frequency.