Font Size: a A A

Research On Channel Modeling For Massive MIMO Systems

Posted on:2017-01-26Degree:MasterType:Thesis
Country:ChinaCandidate:H L WuFull Text:PDF
GTID:2308330488973402Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
To meet the constantly increasing demands for higher data rates, the enhanced multi-input multi-output technique known as massive MIMO has drawn considerable attention. In massive MIMO systems, antenna array of the base stations are equipped with a very large number of elements (say, tens or hundreds) to reap the MIMO transmission benefits on a greater scale. Extra antennas help by focusing energy into ever-smaller regions of space to bring huge improvements in throughput and radiated energy efficiency. In this thesis, channel modeling for massive MIMO systems is investigated.Firstly, a novel two dimensional non-stationary wideband multi-ring model for massive MIMO channels is investigated. Measurements show that massive MIMO channels are significantly different from conven-tional MIMO channels in propagation characteristics. As the antenna number of the array is very large in massive MIMO systems, the far-field assumption in conventional MIMO channels is no longer satisfied. The near-field effect makes the angles of arrival for each antenna element on the array to be unequal, which is called angle of arrival shift. Therefore, the plane wavefront should be replaced by spherical wavefront in the massive MIMO systems. The appearance and disappearance of clusters makes the array suffer a large scale fading effect which means different antenna elements on the large array have different receive power. Thus, a novel two dimensional non-stationary wideband multi-ring model for massive MIMO channels is proposed. In the proposed multi-ring model model, the spherical wavefront assumption and a birth-death process to model non-stationary properties on both time and array axes are adopted. In addition, the multi-ring geome-try relationships which are relative to the delays of multipath components (MPCs) are assumed to mimic the distribution of the clusters in massive MIMO channels. The numerical results, such as received power levels over the array and antenna correlation, show that the proposed channel model captures specific characteristics of the massive MIMO channel comparing with the measurements.Subsequently, a novel three dimensional wideband multi-spherical model for massive MIMO channel-s is investigated. One dimensional antenna array known as uniform linear array (ULA) is widely used in conventional MIMO system. However, the size of ULA trends to be huge in massive MIMO system which makes it difficult to deploy in practice. To impact the size of antenna array, multi-dimensional antenna arrays, like uniform planer array (UPA) and uniform cylindrical array (UCA), are the realized solution. Thus, we proposed a three dimensional wideband multi-spherical model for massive MIMO channels to adopt multi-dimensional antenna array. In multi-spherical model, UCA and UPA are the basic structure of antenna array, and different antenna polarization and antenna pattern are considered. Besides, Rayleigh modeling guideline is used to judge whether it is necessary to perform cluster evolution on array axis. The generation of channel coefficients consists of the generation of general parameters, generation of small scale parameters and calcu-lation of channel coefficients. Based on the multi-spherical model, we illustrate channel behavior of massive MIMO in three representative propagation conditions, and evaluate the corresponding singular value spread and sum-rate performance. The investigation shows that the modeled channels, for both array types, allow us to achieve performance close to that in i.i.d. Rayleigh channels and measured channels, which means that the multi-spherical model is accurate.
Keywords/Search Tags:Massive MIMO, Multi-ring Channel Model, Multi-spherical Channel Model, 3D MIMO, Visi- bility of Clusters, Angle of Arrival Shift, Birth-death Process
PDF Full Text Request
Related items