Zhendong Zhou

MIMO systems with adaptive modulation

Adaptive modulation (AM) schemes in multiple input multiple output (MIMO) systems with a perfect or imperfect channel state information (CSI) at both the transmitter and receiver were investigated. Under an average transmit power and instantaneous bit error rate (BER) constraint, the transmit parameters including the subchannel transmit power and/or spectral efficiency are optimally adapted in the spatial and/or temporal domain to maximize the average spectral efficiency (ASE). Two categories, the continuous rate and discrete rate, of adaptive systems were considered, where the derived asymptotic closed form expressions for the former provided much insight into the latter. Analytical and numerical results showed that a full multiplexing gain was achieved in variable rate variable power (VRVP) systems and variable rate (VR) systems. Variable power (VP) systems with unequal numbers of transmit and receive antennas also achieved the full multiplexing gain, unlike VP systems with equal number of transmit and receive antennas. The effect of CSI imperfection on the ASE and BER was evaluated for VR systems and closed form expressions for the ASE and BER were obtained. They prove to be a useful tool to assess the system performance without taking time consuming AM MIMO system simulations.

Transmit antenna selection schemes with reduced feedback rate

In this paper, we propose and analyze three new transmit antenna selection schemes with reduced feedback rate requirement compared with the conventional scheme. In scheme 1, Lt available transmit antennas are divided as equally as possible into two groups with consecutive antennas. The best single antenna within each group is selected. In scheme 2, only the best one among Lt antennas is made known to the transmitter, and the other one is selected at random. In Scheme 3, Lt antennas are divided into multiple subsets each consisting of two adjacent antennas, and the best subset is selected. Bit error rate (BER) expressions for the proposed schemes with Alamouti code are derived for independent flat Rayleigh fading channels. It is found that all the three schemes achieve a full diversity order. The relative merit of each proposed scheme is delineated based on the trade-off between the asymptotic performance loss and feedback reduction, both relative to the conventional scheme. We conclude that Schemes 1 and 3 are more favorable for practical applications, and the appropriate application scenarios are also identified. The proposed schemes enrich the choices for antenna selection system design for various feedback channel bandwidths and different requirements for quality of service.

Design of adaptive modulation in MIMO systems using outdated CSI

In this paper, we first address the effects of the feedback delay on a variable-rate variable-power adaptive modulation MIMO system that is designed under a perfect channel state information (CSI) assumption. Closed-form expressions for the average bit error rate (BER) and average spectral efficiency (ASE) are derived. Based on that, simple but effective adaptive modulation designs using the outdated instantaneous CSI and time correlation coefficient are proposed based on the signal-to-interference and noise ratio (SINK) and BER analysis. Analytical and simulation results show that these designs provide good trade-offs between the ASE and BER in an adaptive way, and make the adaptive MIMO system much more robust to the CSI imperfection

An optimized network coding scheme in two-way relay channels with multiple relay antennas

We consider two-way relay channels (TWRCs) with multiple antennas at the relay node. Two stage communications are considered, where both sources transmit during the multiple access (MA) stage, and the relay transmits during the broadcast (BC) stage. An optimized network coding scheme is proposed. For the MA stage, a maximum likelihood algorithm is proposed to decode the XOR of the signals received from the two sources. For the BC stage, an optimized beamforming algorithm is proposed to maximize the product of the effective channel gains for the two sources. Both analytical and simulation results show that the proposed scheme achieves a full diversity gain and outperforms the amplify-and-forward (AF) scheme significantly, owing to its efficient use of the relay transmit power and matched excellent performance of the two stages.

Iterative multiple beamforming algorithm for mimo broadcast channels

We consider multiuser Gaussian broadcast channels with multiple antennas at both transmitter and receivers. An iterative multiple beamforming algorithm is proposed, which is flexible in the antenna configuration and performs well in low to moderate data rates. Its capacity and bit error rate performance are compared with the ones achieved by the traditional zeroforcing method.

An iterative beamforming optimization algorithm for generalized MIMO Y channels

We consider a K-user multiple-input multiple-output (MIMO) relay channel, where each user sends K - 1 independent messages to the other K - 1 users via a common relay. All users and the relay are assumed to have multiple antennas. The communication happens in two time slots, the multiple access (MA) stage and the broadcast (BC) stage. A linear transmit beamforming and receive combining scheme is proposed based on signal subspace alignment and physical layer network coding. The optimization of the beamforming and combining vectors is addressed with an iterative suboptimal algorithm based on orthogonal projection optimization in the signal subspaces. Suboptimal power allocation is also considered to maximize the effective signal-to-noise ratios (SNRs). The bit error rate (BER) performance of the proposed scheme in various channel configurations is verified by simulation, which shows that the proposed scheme produces significant improvement over existing one.

Performance analysis of space-time trellis codes with transmit antenna selection in Rayleigh fading channels

In this paper we investigate the error performance of a multiple-input-multiple-output (MIMO) scheme combining transmit antenna selection (TAS) and space-time trellis codes (STTCs), which is referred to as the TAS/STTC scheme. In this scheme, two transmit antennas, which maximize the total received signal power, are selected to transmit the full-rank baseline STTCs designed for two transmit antennas. An upper bound on the pairwise error probability (PEP) of this scheme is derived in quasistatic flat Rayleigh fading channels. It is shown that, as long as the baseline STTC has a full rank, a full diversity order can be achieved, as if all the transmit antennas were used. This scheme has a fixed low decoding complexity as for the baseline STTC and a full diversity order can be achieved with a small memory order. Therefore, the TAS/STTC scheme provides a new approach to the design of MIMO system achieving a high diversity order based on the existing STTCs designed for a small number of transmit antennas.

Joint Semi-Blind Channel Estimation and Synchronization in Two-Way Relay Networks

In this paper, we propose a synchronization and channel estimation method for amplify-and-forward two-way relay networks (AF-TWRNs) based on a low-complexity maximum-likelihood (LCML) algorithm and a joint synchronization and channel estimation (JSCE) algorithm. For synchronous AF-TWRNs, the LCML algorithm blindly estimates general nonreciprocal flat-fading channels. We formulate the channel estimation as a convex optimization problem and obtain a closed-form channel estimator. Based on the mean square error (MSE) analysis of the LCML algorithm, we propose a generalized LCML (GLCML) algorithm to perform channel estimation in the presence of the timing offset. Based on the approximation of the LCML algorithm, the JSCE algorithm is proposed to estimate jointly the timing offset and channel parameters. The theoretical analysis shows that the closed-form LCML channel estimator is consistent and unbiased. The analytical MSE expression shows that the estimation error approaches zero in scenarios with either a high signal-to-noise ratio (SNR) or a large frame length. Monte Carlo simulations are employed to verify the theoretical MSE analysis of the LCML algorithm. In the absence of perfect timing synchronization, the GLCML algorithm selects an estimation sample, which produces the optimal channel estimation, according to the MSE analysis. Simulation results also demonstrate that the JSCE algorithm is able to achieve accurate timing offset estimation.

Adaptive Coded MIMO Systems with Near Full Multiplexing Gain Using Outdated CSI

Information theoretical study shows that multiple-input multiple-output (MIMO) systems have a much higher channel capacity than single-input single-output (SISO) systems, which can be characterized as a multiplexing gain. Adaptive modulation and coding (AMC) is proven to be an effective technique to approach the channel capacity by utilizing the channel state information (CSI) at the transmitter. In this paper, following an analysis of the multiplexing gain of a generic AMC MIMO system, we propose an adaptive coded MIMO system that achieves a near-full multiplexing gain as well as a robust BER performance in an outdated CSI environment. A comparison among various adaptive and non-adaptive MIMO systems reveals the proposed system as a good trade-off between the spectral efficiency, BER and system complexity.

A Cooperative Beamforming Scheme in MIMO Relay Broadcast Channels

We consider relay broadcast channels (RBCs) with multiple antennas at all nodes. A practical linear precoding, relaying and combining scheme is proposed. Under an overall power constraint, we derive the optimal power allocation solution in a closed form. A low complexity beamforming vector optimization algorithm is proposed to maximize the effective channel gains and improve the system performance. Simulation results are presented for various channel configurations, which show that the proposed cooperative beamforming algorithm achieves performance very close to that of the exhaustive search algorithm but with a much lower complexity, and the maximum diversity gain is always attained.