The fifth-generation (5G) mobile communication system will utilize higher frequency bands with wider bandwidth for super high bit rate and large system capacity. Massive multiple-input multiple-output (Massive MIMO) beamforming (BF) technology has attracted much attention to compensate larger path-loss of a disadvantage at higher frequency bands. In low-SHF bands, approximately 100 MHz bandwidth is assumed, thus fully-digital Massive MIMO having flexibility of BF can be applied. In this paper, we modify joint processing of analog fixed BF and channel state information (CSI)-based precoding (called FBCP), which is a hybrid BF technology, to fully-digital method (called Digital-FBCP). Although conventional studies about Massive MIMO with BF do not take account of channel estimation (CE) in detail, this paper describes two-stage CE required for multi-user (MU) Massive MIMO with BF. Throughput performance is evaluated by link level computer simulation in consideration of the reference design and pilot insertion loss. This paper shows that Digital-FBCP achieves almost equal or better throughput performance than conventional precoding (PR) without fixed BF.
In order to tackle rapidly increasing traffic, distributed Massive MIMO (DM-MIMO) has been proposed for fifth-generation (5G) mobile communication systems. DM-MIMO coordinates lots of distributed transmission points (TPs) that are located in ultra-high density (UHD) and use various numbers of antenna elements for each TP. It can achieve drastic improvement of areal spectrum efficiency for 5G. It can also dynamically create user-centric virtual cells corresponding to user mobility. To design flexible antenna deployment of DM-MIMO that is applicable to various use cases in 5G, some key parameters such as the number of the distributed TPs, the number of antenna elements for each TP, and proper distance between TPs should be determined. This paper provides such discussion for 5G DM-MIMO with the flexible antenna deployment under fixed total transmission power and constant total number of antenna elements. Computer simulations show that DM-MIMO can achieve more than 2.5 times system throughput in comparison with a Massive MIMO system using 128 antenna elements.
This paper describes the performance of a state-of-the-art cell range expansion (CRE) technique: an adaptive control CRE for enhanced inter-cell interference coordination (eICIC) in Heterogeneous Network (HetNet). The features of the proposed adaptive control CRE technique are described through comparison with those of conventional methods. With eICIC in a HetNet, system-level computer simulation results such as average user throughput and 5-percentile user throughput are provided as parameters of almost blank subframes ratio for eICIC. This approach provides a remarkably effective solution for eICIC in a HetNet.
Massive multiple-input multiple-output (Massive MIMO) beamforming (BF) is an important technology to utilize higher frequency bands effectively. Fully-digital Massive MIMO BF can be applied for low-SHF-band, and joint processing of digital fixed BF and channel state information (CSI)-based precoding (Digital FBCP) is proposed as one of the fully-digital algorithms. Additionally, it is assumed that multiple sites with low-SHF-band cover continuous areas in the early fifth-generation (5G) mobile communication system. New BF algorithm for multi-site Massive MIMO is required to enable to transmit effectively multiple streams with intersite interference reduction. In this paper, we propose three types of Digital FBCP algorithm with intersite coordination. Throughput performances are shown by link level simulation, and we clarify that the proposed schemes can improve throughput performance compared to that without intersite coordination.
This paper investigates the elevation directional channel properties at a base station (BS) based on ray-tracing simulation to evaluate the performance of 3D beamforming that controls the main beam direction of a BS antenna over both the azimuth dimension and elevation dimension. Results show that the average power elevation spectrum (PES) and the distribution of the elevation angular spread at BS do not depend on the BS antenna height. Moreover, we confirmed that when a mobile station (MS) exists in the vicinity of a BS in a line of sight (LOS) situation, the more the BS antenna height decreases, the more the elevation angular spread increases. Furthermore, the dispersion of the elevation median angle at BS is small compared to that for the azimuth case.