Unmanned aerial vehicle (UAV) base stations (BSs) can be a promising solution to provide connectivity and quality-of-service guarantees during temporary events and after disasters. In this paper, we consider a scenario where UAV-BSs are serving a large number of mobile users in a hot spot area (e.g., in a stadium). We introduce non-orthogonal multiple-access (NOMA) transmission at UAV-BSs to serve more users simultaneously considering user distances as the available feedback for a user ordering during NOMA formulation. With millimeter-wave transmission and multi-antenna techniques, we assume that UAV-BS generates directional beams and multiple users are served simultaneously within the same beam. However, due to the limitations of physical vertical beamwidth of the UAV-BS beam, it may not be possible to cover the entire user region at UAV altitudes of practical relevance. During such situations, a beam scanning approach is proposed to maximize the achievable sum rates. We develop a comprehensive framework over which outage probabilities and respective sum rates are derived rigorously and we investigate the optimal operational altitude of UAV-BS to maximize the sum rates using our analytical framework. Our analysis shows that NOMA with distance feedback can provide better outage sum rates than orthogonal multiple access.

In this paper we consider a densely deployed phantom cell system providing a low latency network for massive connectivity. Non-orthogonal pilot designs serve as a promising solution to support a large number of users but fast collision detection at the receiver is needed for low latency networks. Recently, a nice solution based on an on-off type non-orthogonal pilot design with collision detection capability has been proposed. It can serve more users than the orthogonal pilot design but at the cost of degraded channel estimation performance compared to the orthogonal optimum pilot design. We propose a new non-orthogonal pilot design with collision detection capability and improved channel estimation performance. An optimum threshold based detection criteria is developed. We further show that a dynamically calculated optimum threshold outperforms a fixed threshold based detection. Next, we investigated non-orthogonal pilot designs for fractional bandwidth allocation. We propose two new non-orthogonal pilot designs for physical resource block (PRB) based resource allocation. Both of the new designs support fast collision detection at the receiver. Performance evaluation results show that the proposed schemes provide equivalent or better channel estimation performance and support much more users than the orthogonal pilots defined in the current 4G standards.

The overheads associated with feedback-based channel acquisition can greatly compromise the achievable rates of FDD based massive MIMO systems. Indeed, downlink (DL) training and uplink (UL) feedback overheads scale linearly with the number of base station (BS) antennas, in sharp contrast to TDD-based massive MIMO, where a single UL pilot trains the whole BS array. In this work, we propose a graph-theoretic approach to reducing DL training and UL feedback overheads in FDD massive MIMO systems. In particular, we consider a single-cell scenario involving a single BS with a massive antenna array serving to single-antenna mobile stations (MSs) in the DL. We assume the BS employs two-stage beamforming in the DL, comprising DFT pre-beamforming followed by MU-MIMO precoding. The proposed graph-theoretic approach exploits knowledge of the angular spectra of the BS-MS channels to construct DL training protocols with reduced overheads. Simulation results reveal that the proposed training-resources allocation method can provide approximately 35% sum-rate performance gain compared to conventional orthogonal training. Our analysis also sheds light into the impact of overhead reduction on channel estimation quality, and, in turn, achievable rates.

In this paper, we consider the outage performance of the M-BLAST symbol detection algorithm over uncoded quasistatic MIMO channels with spatial multiplexing. The M-BLAST algorithm is known to achieve a superior error performance over its predecessor V-BLAST with a signal-to-noise ratio (SNR) improvement of as large as 2 dB. We evaluate this performance improvement analytically, which has not been studied before, by deriving the outage probability expression over generalized complex binary alphabets with two transmitting and unlimited number of receiving antennas. Although the underlying successive interference cancellation (SIC) receiver introduces a high non-linearity which complicates the analysis, our analytical and simulation results are shown to have a very good match.

Efficient symbol detection algorithms carry critical importance for achieving the spatial multiplexing gains promised by multi-input multi-output (MIMO) systems. In this paper, we consider a maximum a posteriori probability (MAP)-based symbol detection algorithm, called M-BLAST, over uncoded quasi-static MIMO channels. Relying on the successive interference cancellation (SIC) receiver, the M-BLAST algorithm offers a superior error performance over its predecessor V-BLAST with a signal-to-noise ratio (SNR) gain of as large as 2 dB under various settings of recent interest. Performance analysis of the M-BLAST algorithm is very complicated since the proposed detection order depends on the decision errors dynamically, which makes an already complex analysis of the conventional ordered SIC receivers even more difficult. To this end, a rigorous analytical framework is proposed to analyze the outage behavior of the M-BLAST algorithm over binary complex alphabets and two transmitting antennas, which has a potential to be generalized to multiple transmitting antennas and multidimensional constellation sets. The numerical results show a very good match between the analytical and simulation data under various SNR values and modulation alphabets.

For the development of new 5G systems to operate in bands up to 100 GHz, there is a need for accurate radio propagation models at these bands that currently are not addressed by existing channel models developed for bands below 6 GHz. This document presents a preliminary overview of 5G channel models for bands up to 100 GHz. These have been derived based on extensive measurement and ray tracing results across a multitude of frequencies from 6 GHz to 100 GHz, and this document describes an initial 3D channel model which includes: 1) typical deployment scenarios for urban microcells (UMi) and urban macrocells (UMa), and 2) a baseline model for incorporating path loss, shadow fading, line of sight probability, penetration and blockage models for the typical scenarios. Various processing methodologies such as clustering and antenna decoupling algorithms are also presented.

This paper investigates the throughput performance using adaptive modulation and coding (AMC) focusing on the causes of impairment to Orthogonal Frequency Division Multiplexing (OFDM) - Multiple-Input Multiple-Output (MIMO) multiplexing. We first investigate the influence of a limited number of modulation and coding schemes (MCSs) and channel estimation error for MCS selection based on mutual information (MI). Second, we clarify the effect of MCS selection error on an increasing maximum Doppler frequency due to the round trip delay time (RTT) in comparison to the throughput upper bound that is computed from the MCS combination providing the maximum throughput considering the block error. Third, we clarify the effect of channel estimation error of maximum likelihood detection (MLD) when using reference signal (RS) based channel estimation in comparison to ideal channel estimation. Through the evaluations, we clarify the impairment factors that degrade the achievable throughput for OFDM-MIMO multiplexing using AMC in an actual multipath fading channel.

This paper investigates a beamforming control method in base station (BS) cooperative multiuser multiple-input multiple-output (MIMO) transmission assuming a block-diagonalized unitary beamforming matrix. More specifically, we comprehensively compare open-loop type random (opportunistic) beamforming control to codebook-based closed-loop type beamforming control. The proportional fair (PF)-based scheduling is assumed, and the number of users, codebook size, and delay time of the feedback signal from the user terminals are taken into account in the comparison. Based on numerical results, codebook-based beamforming control with a large codebook size achieves higher system throughput than random beamforming when the number of users is small and the feedback delay time is short. However, we show that when the number of users is large and the feedback delay time is long, random beamforming achieves higher system-level throughput than codebook-based beamforming. This implies that the transmission-rate control with random beamforming is more robust against channel variation than that with codebook-based beamforming. These results suggest that random beamforming, which requires a lower feedback overhead than the codebook-based beamforming, is a promising beamforming strategy in future radio access.

This paper proposes channel state information - reference signal (CSI-RS) multiplexing employing carrier frequency swapping for frequency division duplex (FDD) high-order multiple-input multiple-output (MIMO) spatial division multiplexing (SDM). In the proposed CSI-RS multiplexing method, the carrier frequency of a fast Fourier transform (FFT) block for multiplexing time division duplex (TDD) CSI-RSs within a subframe is swapped between the downlink and uplink. By using the proposed TDD CSI-RS multiplexing method, the CSI for performing precoding and link adaptation is estimated in the reverse link by taking advantage of the channel reciprocity feature. By using the proposed TDD CSI-RS multiplexing method, the RS overhead is decreased according to the increase in the number of simultaneously accessing sets of user equipment for multiuser MIMO. Moreover, the proposed TDD CSI-RS multiplexing method does not need fast feedback signals that track instantaneous channel response including the channel quality indicator (CQI), rank indicator (RI), and precoding matrix indicator (PMI) except for the slow feedback of the average interference level from the surrounding cells.

This paper presents indoor experimental results on the achievable throughput for 4-by-2 multi-user (MU)-MIMO using 2 mobile stations (MSs) and carrier aggregation with 5 component carriers (90-MHz bandwidth) considering various transmitter antenna arrangements in the LTE-Advanced downlink. Under two centralized and two distributed antenna arrangement (CAAs and DAAs) conditions, the throughput performance is evaluated using the implemented LTE-Advanced transceiver, where extended channel state information (CSI) feedback based on eigenvalue decomposition (EVD) is implemented for MU-MIMO operation. The experimental results show that the peak throughput of greater than 1 Gbps is achieved for the CAAs in an indoor environment. The results also showed that although the MS moving speed strongly influences the throughput for Rank-4 MU-MIMO, MU-MIMO is robust against the antenna separation, i.e., fading correlation. Furthermore, we confirm that, in the DAAs, although the peak throughput of 1 Gbps is not observed, the throughput of approximately 700 to 950 Mbps is achieved.

This paper presents laboratory experimental results on 4-by-2 multi-user (MU)-MIMO that achieves up to 1 Gbps throughput using carrier aggregation with a 100-MHz bandwidth in the LTE-Advanced downlink. The achievable throughput performance employing adaptive modulation and coding (AMC) with outer-loop threshold control based on acknowledgement (ACK) / negative ACK (NACK) feedback signaling in order to compensate for the fluctuation in the measured signal-to-interference plus noise ratio (SINR) value due to MU-MIMO multiplexing is evaluated using the implemented LTE-Advanced transceivers. Additionally, we evaluate the influence of the angle between spatially multiplexed users and the user combination in MU-MIMO operation with different received signal-to-noise ratio (SNR) conditions. Laboratory experimental results show that by performing AMC with outer-loop threshold control, the throughput of approximately 1 Gbps is achieved at the average received SNR of 30 dB. Furthermore, based on the experimental results, we confirm that angles narrower or wider than approximately 60 degrees between spatially multiplexed mobile stations (MSs) yield a lower ability to decrease the mutual interference using minimum mean square error (MMSE)-based precoding in 4-by-2 MU-MIMO. We also show that especially when the angle between spatially multiplexed MSs is narrower or wider than approximately 60 degrees, selecting the user pairs with a large difference in the average received SNR is effective in improving the total throughput.

This paper presents laboratory experimental results on the throughput performance of asymmetric carrier aggregation between the uplink and downlink using an LTE-Advanced testbed. The implemented base station (BS) and mobile station (MS) transceivers have transmission bandwidth capability of up to 100 MHz (5 component carriers (CCs)) and 40 MHz (2 CCs) in the downlink and uplink, respectively. The testbed also features CC-specific adaptive modulation and channel coding (AMC) and hybrid automatic repeat request (HARQ) functionalities based on the actually transmitted control signaling. The experimental results show that, in the uplink, CC-based AMC using up to 64QAM high-order modulation is effective in increasing the user throughput in multipath fading channels and achieves a throughput greater than 100 Mbps using carrier aggregation with 2 CCs and 1-by-2 SIMO. The results also show that when the uplink control information (UCI) corresponding to 5 downlink CCs is multiplexed onto the uplink shared channel (PUSCH), the required block error rate (BLER) of the UCI below 10-2 is almost satisfied by appropriately setting the offset value, , for adjusting the channel coding rate of UCI although AMC with high-order modulation is applied to the UCI as well. Furthermore, we show that when carrier aggregation with 5 CCs and 2-by-2 MIMO multiplexing are applied in the downlink, high throughput greater than 500 Mbps (thus, half of the 1 Gbps that can be achieved using 4-by-4 MIMO multiplexing) is achieved at the average received signal-to-noise power ratio (SNR) of approximately 25 dB when CC-specific AMC considering UCI error is applied.

This paper proposes suboptimal precoding vector (PV) selection schemes combined with frequency domain scheduling when closed-loop MIMO precoding is adopted in the LTE (Long-Term Evolution)-Advanced uplink based on single carrier-frequency division multiple access (SC-FDMA) radio access. In the proposed schemes, frequency domain channel-dependent scheduling is performed based on a tentatively selected PV for each subband, and then a suitable PV is selected that is common to the contiguously allocated resource blocks in the frequency domain for each UE in order to achieve the single carrier property, i.e., a low peak-to-average power ratio (PAPR). System-level simulation results show that by employing the proposed scheme, the sector throughput performance is increased by 1.6% and 2.3% compared to the conventional scheme in the case of 2-by-2 and 4-by-2 MIMO, respectively. Furthermore, we clarify that the losses in the achievable sector throughput by using the proposed scheme are only 0.9% and 2.4% for 2-by-2 and 4-by-2 MIMO, respectively, compared to the ideal condition using frequency selective precoding.

As the wireless channel is constantly changing, MIMO-OFDM eigenmode transmission is extremely vulnerable to the presence of feedback delay which is unavoidable in a practical system. Under such imperfect conditions, the inter-eigenmode interference can be removed through the use of the MMSE interference canceller. However, to the best of our knowledge, the performance of MIMO-OFDM eigenmode with MMSE interference canceller has not been proven by experimental results. In this paper, we present experimental results for the bit error rate (BER) performance of MIMO-OFDM eigenmode transmission with and without MMSE interference canceller. In the experiment, a time-varying channel is synthetically generated in a static environment by exploiting the inherent spatial correlation in a static environment. Our experimental results proved the robustness of MIMO-OFDM eigenmode with MMSE interference canceller in the presence of feedback delay