Coupling Calibration And MIMO Capacity Analysis For Two-Dimension Antenna Arrays

The multi-antenna techniques play an important role in enhancement of peak rates and relia-bility of data transmission, cell coverage extension, and system capacity improvement. At present, the traditional horizontal dimension based multi-antenna techniques tend to be mature, and the new network architecture and antenna techniques are driving the development of multi-antenna system-s forward onto higher dimension. In this context, the planar architecture based Full Dimension Multiple-Input Multiple-Output (FD-MIMO) technique has attracted widespread attention of the community in recent years. FD-MIMO adopts two dimensional active antenna arrays, and takes full advantage of spatial degrees of freedom in vertical dimension, thus providing a significant gain to capacity improvement and coverage extension. More importantly, since the development of active antenna techniques makes it possible to implement the large-scale arrays, FD-MIMO is regarded as one of key technologies of the next generation communication system.Although FD-MIMO holds a bright prospect, the deployment of large-scale arrays in base sta-tions will be faced with some problems, which can't be ignored due to the limitation of the available installation space. On the one hand, there exists intrinsic mutual coupling effect in antenna arrays in practice. In term of two dimensional arrays, mutual coupling effect becomes more remarkable when deployed in a limited space. If it is not taken into consideration, the practical performance of existing multi-antenna signal processing techniques will deteriorate significantly. On the other hand, when a great number of antennas are deployed in a limited space in base stations, it will cer-tainly increase the spatial correlation between subchannels, thus weaken the theoretical diversity gain of multi-antenna systems. The existence of the above problems not only poses a challenge to the robustness of traditional multi-antenna signal processing techniques, but also requires us to reexamine and evaluate the practical performance and capacity gain of FD-MIMO systems.Exactly circumventing the above topics in this thesis, we have carried out related researches in depth, and the main research works and contributions are summarized as follows:1) In term of the robustness of direction finding algorithm using two dimensional antenna arrays under mutual coupling, we propose a rank reduction based two dimensional direction of ar-rive (DOA) blind estimation algorithm and the corresponding mutual coupling calibration scheme. Based on the previous research works, we further complete the mutual coupling modeling in planar arrays. By utilizing the priori knowledge of mutual coupling, we just rearrange the perturbed array response in the transform domain, thus being able to mask the adverse effect of mutual coupling in the subsequent processing, and avoid the aperture loss as well. Then based on the rank reduction criteria, the reduced-dimension processing is performed by applying suitably designed estimation procedure, which significantly improves the computational efficiency. After we get the estimated DOAs, a decoupling matrix can be constructed for mutual coupling estimation by eigen decompo-sition. Compared to the existing method, the proposed algorithm is more numerically efficient with improved accuracy. Simulation results verify the superior performance of the proposed algorithm.2) Under the circumstance of the constrained space, we reexamine the propagation conditions and multi-user capacity of FD-MIMO systems under different channel models. Based on three dimensional spatial fading correlation model, we investigate the problem of channel's asymptotic orthogonality under different channel models, including line of sight (LOS) and Kronecker correla-tion models, and particularly derive the asymptotic lower bound of the channel vector inner product under LOS condition. The results show that in the constrained deployment space, the subchannels of different users are no longer asymptotic for both of the above models. Then resorting to the equivalent circuit model for mutual coupling in transmitting-mode antenna arrays, the down link multi-user system capacity is further evaluated comprehensively. The results indicate that under the constrained array area, as the total number of antennas in base stations increases, the system capacity will eventually reach saturation under different propagation conditions, while the behavior of mutual coupling effect depends on the specific propagation condition and array configuration, and may act as a decorrelator in compact arrays.