Full-dimension MIMO (FD-MIMO) using planar active antenna systems (AAS) is considered a critical technology for fifth-generation (5G) cellular systems to improve network capacity. An AAS is typically subject to hardware impairments that negatively impact network capacity. Hence, this article focuses on impairments that cause phase and magnitude errors between radio frequency (RF) chains and shows why they are particularly difficult to avoid in practical AAS. Although previous investigations show these impairments to degrade performance, they are not useful in deriving measurable impairment margins for practical FD-MIMO deployments. Knowing impairment limits are critical for system designers to make hardware design tradeoffs such as AAS configuration, component selection, implementation complexity, and cost. Moreover, it also helps set conformance limits for critical lab verification. Therefore, the paper first investigates the impact of the impairments and derives their practical limits for FD-MIMO by explicitly considering the cumulative effects of the channel model, inter-cell interference, link adaptation, and channel aging due to feedback delays. It is shown that a lower number of digitized RF chains can be a better choice under lower impairments. Next, the sources of impairments are investigated by using measurements carried out in the lab and the field during live operation in a commercial LTE network. Phase drift from local oscillators (LO) and internal temperature variations are identified as two significant sources. The tradeoffs and shortcomings of some of the existing solutions in massive MIMO literature are discussed. Finally, in order to address the shortcomings, a novel and practical coherent LO distribution architecture and array calibration mechanism are proposed. This solution is shown to be applicable to both TDD and FDD FD-MIMO. Measurement results are provided to prove the high degree of coherency and stability achieved on a unique array architecture called high definition active antenna system (HDAAS).

Using a single rectangular antenna panel enhances cellular capacity via vertical beamforming (e.g. “Full Dimension MIMO”) but may impact diversity gain. We compare the receive diversity performance of several single-panel signal processing techniques, including: subarrays (with wide or narrow beamwidths), pattern diversity (with various a priori azimuth offset beam steering), and Maximum Ratio Combining. Spatial fading correlation is obtained from the 3GPP TR 25.996 model. We assume a hybrid analog-digital beamformer and derive the SNR for any analog beamformer followed by an MRC digital stage. This allows us to apply various beamforming strategies to the antenna subarrays and determine if there is a benefit. In a simple azimuth-only, single user simulation we find:4R (i.e. 4 outputs) subarrays with narrow steered beams approaches the performance of 8RMRC, while a 4 dB shortfall is found for widebeam subarrays relative to MRC; and pattern diversity is competitive with widebeam subarrays for 4R but not 8R.

In this paper, we investigate agile beamforming using Massive MIMO rectangular arrays for improving the performance of current LTE/LTE-A networks. Multi-antenna technologies like Full-dimension (FD) MIMO and the required CSI feedback mechanisms have been standardized in 3GPP Release 13/14 and 5G New Radio (NR), but their use in legacy LTE releases is limited. We show that large arrays can be leveraged to provide significant throughput improvements even in legacy LTE (Release 8) systems. Our main approach consists of switching narrow beams at the radio unit (RU) and scheduling users at the baseband unit (BBU) in such a way that users throughout the sector experience an SINR boost due to beamforming. The base station (eNB) improves the SINR in different regions of the sector by swiftly switching between different narrow beams in tight synchronization with the CSI feedback and scheduling latency mechanisms of the LTE protocol, and then schedules a user (UE) from within the high-SINR region. We consider the practically relevant case where there is no coordination between the baseband unit (BBU) and the radio unit (RU) as well as the advanced case where co-ordination is possible. Through simulation results, we demonstrate significant improvements in the average and cell-edge throughput, when compared to a conventional wide-beam system.

Massive multiple-input and multiple-output (MIMO) has been the subject of interest in both industry and academia for the past few years. Massive MIMO refers to the use of a large number of antennas typically at the base station (BS) to serve multiple user equipment (UE) simultaneously to deliver reliable and high data throughput. Two basic architectures have been considered in the literature. First is a full digital massive MIMO where all the beamforming and precoding is performed digitally in baseband and a radio frequency (RF) chain is required for every antenna element. In the second architecture referred to as hybrid massive MIMO, beamforming and precoding are done in two stages, digital precoding over fewer digital ports followed by analog beamforming across large number of antenna elements. In this paper, we provide a comparison of implementation challenges for each approach. We introduce the High Definition Active Antenna System (HDAAS), a novel scalable architecture to implement a hybrid Massive MIMO system. This design philosophy is used to build BeamCraft500, which we believe is a first commercial hybrid 3D beamforming system operating at 2 GHz. We also present results from one of the field trials in a live LTE network that validate the stability and performance of the system under real world conditions.

This paper presents two new methods for evaluating the ergodic channel capacities of cooperative non-regenerative multirelay networks in a myriad of fading environments and under three distinct source-adaptive transmission policies: i optimal rate adaptation with a fixed transmit power; ii optimal joint power-and-rate adaptation; and iii truncated channel inversion with fixed rate. In contrast to the previous related works, our proposed unified analytical frameworks that are based on the moment generating function and/or the cumulative distribution function of end-to-end signal-to-noise ratio allow us to gain insights into how power assignment during different transmission phases, relay node placement, fade distributions, and dissimilar fading statistics across the distinct communication links impact the ergodic capacity, without imposing any restrictions on the channel fading parameters. Copyright © 2013 John Wiley & Sons, Ltd.

This article investigates the efficacy of a constant-power, rate-adaptive M-QAM transmission technique in amplify-and-forward cooperative wireless networks. Tight upper and lower bounds are derived for the mean achievable spectral efficiency, outage probability, and error probability performance of non-regenerative cooperative relay networks in Nakagami-m fading environments. We observe that the spectral efficiency of a practical discrete-rate MQAM comes within a constant gap of the theoretical ergodic capacity of that channel, although this gap is smaller for an adaptive continuous-rate M-QAM.

This article analyzes the performance of CSI-assisted cooperative amplify-and-forward (CAF) relay networks that employ both the optimal power allocation strategy among collaborating nodes and adaptive M-ary quadrature amplitude modulation (MQAM) technique in Nakagami-m wireless fading environments. In particular, we advocate a simple yet unified numerical approach based on the marginal moment generating function (MGF) of the total received SNR to derive tight approximations or upper and lower bounds for the average bit error rate (ABER), mean achievable spectral efficiency, and outage probability performance metrics. The proposed analytical framework is sufficiently general to characterize the performance of adaptive-link CAF relay networks over a wide range of fading distributions (i.e., not restricted to Rayleigh fading) with independent but non-identically distributed (i.n.d) fading statistics across the spatially distributed diversity paths. Numerical results reveal that the optimal transmit power allocation among cooperative nodes in a practical CAF relay topologies could lead to a further substantial increase in the mean spectral efficiency compared to the equal power assignment case but at the expense of higher network overhead. The accuracies of our analytical results have been validated via Monte Carlo simulations.

This article analyzes the performance of CSI-assisted cooperative amplify-and-forward (CAF) relay networks that employ both the optimal power allocation strategy among collaborating nodes and adaptive M-ary quadrature amplitude modulation (M-QAM) technique in Nakagami-m wireless fading environments. In particular, we advocate a simple yet unified numerical approach based on the marginal moment generating function (MGF) of the total received SNR to derive tight approximations or upper and lower bounds for the average bit error rate (ABER), mean achievable spectral efficiency, and outage probability performance metrics. The proposed analytical framework is sufficiently general to characterize the performance of adaptive-link CAF relay networks over a wide range of fading distributions (i.e., not restricted to Rayleigh fading) with independent but non-identically distributed (i.n.d) fading statistics across the spatially distributed diversity paths. Numerical results reveal that the optimal transmit power allocation among cooperative nodes in a practical CAF relay topologies could lead to a further substantial increase in the mean spectral efficiency compared to the equal power assignment case but at the expense of higher network overhead. The accuracies of our analytical results have been validated via Monte Carlo simulations.

Upper bounds on link spectral efficiency of amplify-and-forward cooperative diversity networks with independent but non-identically distributed wireless fading statistics are studied by deriving the Shannon capacity of three distinct adaptive source transmission techniques: (i) constant power with optimal rate adaptation (ORA); (ii) optimal joint power and rate adaptation (OPRA); and (iii) fixed rate with truncated channel inversion (TCIFR). Asymptotic capacity bound is also derived which show that optimal rate adaptation with constant power policy provides roughly the same ergodic capacity as the optimal joint power and rate adaptation policy at high mean signal-to-noise ratios (SNRs). Different previous related studies, we advocate a simple numerical procedure for unified analysis of ergodic channel capacity in a myriad of fading environments. This framework allows us to gain insights as to how fade distributions and dissimilar fading statistics across the diversity paths affect the Shannon capacity, without imposing any restrictions on the fading parameters.

Tight bounds for the ergodic capacity of amplify-and-forward cooperative diversity networks are derived for three different adaptive source transmission policies in a myriad of fading environments: (i) constant power with optimal rate adaptation ( ORA); (ii) optimal joint power and rate adaptation (OPRA); and (iii) fixed rate with truncated channel inversion (TCIFR). Our unified framework based on the moment-generating function (MGF) approach allows us to gain insights as to how fade distributions and dissimilar fading statistics across the distinct communication links will affect the mean achievable rates, without imposing any restrictions on the fading parameters.

Tight bounds for the Shannon capacity of amplify-and-forward cooperative diversity networks are derived for three different adaptive source transmission policies in a myriad of fading environments: (i) constant power with optimal rate adaptation (ORA); (ii) optimal joint power and rate adaptation (OPRA); and (iii) fixed rate with truncated channel inversion (TCIFR). Our unified framework based on the moment-generating function (MGF) approach allows us to gain insights as to how fade distributions and dissimilar fading statistics across the distinct communication links will affect the Shannon capacity, without imposing any restrictions on the fading parameters.

Cooperative communication systems that exploit spatial diversity to improve the performance of end-to-end communication in ad hoc wireless networks have gained a lot of attention in recent years. In this article, we examine the energy efficiency of a simple selective decode and forward (SDF) protocol. Different from previous studies on energy efficiency our analysis accounts for the imperfect regeneration at the relay. We show that the cooperative system using the SDF protocol can achieve higher energy efficiency than a single-input-single-output (SISO) transmission when the transmit-power consumed by the power amplifiers become higher than the total power consumed by the analog circuitry at the nodes.

The distributed multiple-input-multiple-output (MIMO) system (e.g., intercluster communication via cooperating nodes in a wireless sensor network) is a topic of emerging interest. Many previous studies have assumed perfect synchronization among cooperating nodes and identically distributed communication links. Such assumptions are rarely valid in practical operating scenarios. This paper develops an analytical framework for computing the average bit error rate (ABER) of a distributed multiple-input-single-output (MISO) space-time-coded system with binary phase shift keying (BPSK) modulation affected by timing synchronization errors. The cooperating nodes use data pulse-shaping filters for transmission over generalized frequency-nonselective fading channels. As an illustrative example, the performance evaluation of a 2 x I MISO system that uses distributed orthogonal space-time block coding (OSTBC) is presented, although this approach can he readily extended to analyze distributed transmit diversity with a larger number of cooperating nodes. We show that under certain conditions, a distributed MISO system with time synchronization errors can still outperform a perfectly synchronized single-input-single-output (SISO) system.

Outage probability and ergodic capacity are important performance measures of communication systems over fading channels. Point-to-point communication systems are known to achieve tremendous improvements in channel capacity when MIMO schemes are applied. In this paper we investigate an efficient Fixed-Talbot algorithm for numerical Laplace inversion and apply it to evaluate the outage probability and ergodic capacity of OSTBC system. The framework developed can be applied to a wide range of fading distributions (including Rice, Nakagami-m, Nakagami-Hoyt and Weibull stochastic channel models) with unequal channel gains and also to independent and non-identically distributed (i.n.d) MIMO channels.

Several numerical methods have been suggested in the past to compute the outage probability and the ergodic capacity of an orthogonal space-time block coding system. In this paper we add to this body of work by describing a simpler yet general method that relies on a more efficient Fixed-Talbot algorithm for numerical Laplace inversion. The framework developed can be applied to a wide range of fading distributions (including Rice, Nakagami-m, Nakagami-Hoyt and Weibull stochastic channel models) with unequal channel gains and also to independent and non-identically distributed (i.n.d) multiple input multiple output channels.

This paper develops an efficient analytical framework for evaluating the average bit-error probability (ABER) of bandlimited coherent binary phase shift keying (BPSK) in generalized fast fading channels, where the fading rate is approximately equal to the symbol rate, subject to timing errors and asynchronous cochannel interferers. Selected simulation and computational results are presented that are of interest for outdoor microcellular and macrocellular system studies. Aside from this, our ABER results also serve as lower performance bounds for practical realizable receivers (where ideal coherent detection is difficult to implement) and as an upper performance bound for bandlimited BPSK in slow fading channels where the fading rates are much slower than the symbol rate.

This paper presents exact ABER performance analysis for selective decode and forward (SDF) cooperative diversity system with BPSK modulation under Rayleigh fading where the relay has a MAP based receiver and the retransmission is based on log-likelihood-ratio (LLR) threshold. We also derive the optimum LLR threshold that minimizes ABER performance. It is shown that the LLR relay based SDF cooperative diversity system performs better than a SNR threshold based SDF system with lower implementation complexity than lambda-MRC and C-MRC schemes.

The demand for access to information when and where you need has motivated the transition of wireless communications from a fixed infrastructure based cellular communications technology to a more pervasive adhoc wireless networking technology. Challenges still remain in wireless adhoc networks in terms of meeting higher capacity demands, improved reliability and longer connectivity before it becomes a viable widespread commercial technology. Present day wireless mesh networking uses node-tonode serial multi-hop communication to convey information from source to destination in the network. The performance of such a network depends on finding the best possible route between the source and destination nodes. However the end-to-end performance can only be as good as the weakest link within a chosen route. Unlike wired networks, the quality of point-to-point links in a wireless mesh network is subject to random fluctuations. This adversely affects the performance resulting in poor throughput and poor energy efficiency. In recent years, a new paradigm for communication called cooperative communications has been proposed for which initial information theoretic studies have shown the potential for improvements in capacity over traditional multi-hop wireless networks. Cooperative communication involves exploiting the broadcast nature of the wireless medium to form virtual antenna arrays out of independent single-antenna network nodes for transmission. In this research we explore the fundamental performance limits of

Software-defined radio system architecture must be openly structured to various system standards. It should also provide capability for distributed processing, object-oriented design, and software controllability. This implies that the software to be used in the SDR system should be independent of a given hardware platform. In order to achieve these goals, the proposed SDR system utilizes modularization to maximize hardware reuse and design flexibility, which provides the system reconfigurability. The objective of this article is to provide an open architecture of a smart antenna base station (SABS) operating in the SDR with architecture that is object-oriented and software-controlled. For this purpose, the software and hardware of a SABS is first modularized and partitioned into modules, respectively. Then the interface among the modules is specified to determine the smart antenna application programming interface proper for the SDR network. The suitability of the proposed open architecture of SABS is verified through a design example of SABS implemented in accordance with the proposed architecture. The performance of the proposed system is shown in practical signal environments of CDMA2000 1X with commercial handsets operating at various data rates ranging from 9.6 to 153.6 kb/s in terms of frame error rate and signal-to-Interference-plus-noise ratio, which is dramatically improved through the nicely shaped beam pattern.

Distributed multiple-input-multiple-output system (e.g., inter-cluster communication via cooperating nodes in a wireless network) is a topic of emerging interest. Much of previous studies assume perfect synchronization among cooperating nodes and identically distributed communication links. Such assumptions are rarely valid in practical operating scenarios. This paper develops an efficient analytical framework for computing average bit error rate of a distributed multiple-input-single- output (MISO) space-time coded system with intersymbol interference (due to imperfect location predictions and clock jitters between cooperating nodes, as well as the choice of data pulse shaping filters) over generalized fast fading channels. We show that under certain conditions distributed MISO system can outperform perfectly synchronized single-input-single-output (SISO) system.