Arrays designed for multiple-input multiple-output (MIMO) communication systems must provide good performance in terms of information theoretic and microwave theoretic performance measures for a large number of channel scenarios. Consequently, the general MIMO array design problem is challenging. This paper proposes a novel optimization methodology for designing circular patch arrays (CPAs), under a clustered MIMO channel model assumption. Based on the ergodic capacity, it is found that the single-cluster scenario can approximate CPA performance in clustered channels, reducing the number of channel scenarios that need to be considered. Using a new definition of channel spatial correlation that allows for pattern diversity, metrics are derived for evaluating average and outage capacity performance of CPAs, which can be expressed as a function of only the antenna parameters. With this expression, an optimization is formulated and solved numerically for optimizing the parameters of a 2-CPA on a capacity based objective function and microwave theory performance constraints. An example design is given corresponding to a 2-CPA in the 2.4 GHz unlicensed frequency band.

Previous research has shown that adaptive switching between multiple-input multiple-output (MIMO) transmission strategies like spatial multiplexing and beamforming increases link reliability and capacity gains, as compared to fixed transmission strategies. To get the full benefit of adaptive switching it is necessary to obtain accurate estimates of the SNR values when we switch between the transmission strategies. In this paper, it is shown that (relatively more) accurate switching point estimates can be obtained by taking into account real-life effects like mutual coupling and antenna efficiencies, for switching between statistical beamforming and spatial multiplexing. Using simulations, it is shown that accounting for these effects can make the switching point estimate more accurate by as much as 12 dB, compared to the case when the practical effects are not considered.

In this paper, a novel transmission technique for the multiple-input multiple-output (MIMO) broadcast channel is proposed that allows simultaneous transmission to multiple users with limited feedback from each user. During a training phase, the base station modulates a training sequence on multiple sets of randomly chosen orthogonal beamforming vectors. Each user sends the index of the best beamforming vector and the corresponding signal-to-interference-plus-noise ratio for that set of orthogonal vectors back to the base station. The base station opportunistically determines the users and corresponding orthogonal vectors that maximize the sum capacity. Based on the capacity expressions, the optimal amount of training to maximize the sum capacity is derived as a function of the system parameters. The main advantage of the proposed system is that it provides throughput gains for the MIMO broadcast channel with a small feedback overhead, and provides these gains even with a small number of active users. Numerical simulations show that a 20% gain in sum capacity is achieved (for a small number of users) over conventional opportunistic space division multiple access, and a 100% gain (for a large number of users) over conventional opportunistic beamforming when the number of transmit antennas is four.

In this paper, a novel transmission technique for the multiple-input multiple-output (MIMO) broadcast channel is proposed that allows simultaneous transmission to multiple users under a limited feedback requirement. During a training phase, the base station modulates a training sequence on multiple sets of randomly generated orthogonal beamforming vectors. Then, based on the users' feedback, the base station opportunistically selects the users and corresponding orthogonal vectors that maximize the sum capacity. From theoretical analysis, the optimal amount of training to maximize the sum capacity is derived as a function of the system parameters. The main advantage of the proposed system is that it provides throughput gains for the MIMO broadcast channel with a small feedback overhead, and provides these gains even with a small number of active users. Numerical simulations show that a 20% gain in sum capacity is achieved (for a small number of users) over conventional opportunistic space division multiple access, and a 100% gain (for a large number of users) over conventional opportunistic beamforming.

This paper considers low complexity transmission for MIMO bit-interleaved coded modulation (BICM) in spatially-correlated Rayleigh channels. We consider both statistical beam-forming (SB) and spatial-multiplexing with a zero-forcing (ZF) receiver. We derive tight closed-form bit error rate (BER) expressions based on a saddlepoint approximation. We then propose a practical adaptive algorithm which selects, based on the analytical results, the combination of code-rate, modulation format, and MIMO transmission scheme (SB or ZF) that maximizes throughput whilst maintaining a pre-defined BER.

We present a low complexity adaptive transmission approach for spatially correlated MIMO channels. The proposed scheme adaptively switches between orthogonal space-time block codes (OSTBC) and spatial multiplexing (SM), depending of the channel correlation and SNR. We derive an exact closed-form expression and tight upper bound on the OSTBC capacity in double-correlated channels, and examine the relative capacity of OSTBC and SM in terms of the spatial correlation. We then derive efficient closed-form BER expressions for practical OSTBC transmission employing bit-interleaved coded modulation (BICM). Based on these results, we propose a practical adaptive algorithm which selects the combination of MIMO transmission scheme (OSTBC or SM), and BICM mode, which achieves the highest spectral efficiency whilst satisfying a pre-defined BER

In this paper, we analyze a multiple-input multiple-output (MIMO) array consisting of two circular microstrip antennas, designed to exploit pattern diversity. We analytically derive the spatial correlation coefficients of this array as a function of the mode excited, for realistic clustered MIMO channel models. We compare the performance of the circular patch array (CPA) against an array of two spaced dipoles. In particular, we compute a theoretical tradeoff to predict when the pattern diversity provided by the CPA is more effective than space diversity from the uniform linear array (ULA), based on the eigenvalues of the spatial correlation matrix. Through simulations, we show that CPAs yield better performance or satisfy more restrictive size constraints than ULAs in clustered MIMO channels, depending on the element spacing of the ULA. These results make the CPA an attractive solution for miniaturized MIMO arrays for portable devices or access points.

We consider a low-complexity adaptive MIMO transmission approach for spatially correlated channels. The proposed scheme adaptively switches between different transmission modes depending on the changing channel conditions, as a means to enhance system capacity. Each mode is a combination of a transmission technique (i.e. statistical beamforming, double space-time transmit diversity and spatial multiplexing) and a modulation/coding scheme. We first motivate our adaptive algorithm by deriving new closed-form capacity expressions, and demonstrating significant information theoretic improvements over non-adaptive transmission. We then present a practical method to switch between different modes, based on the channel statistics. Our approach is shown to yield significant improvements in spectral efficiency for typical channel scenarios.

For a multiple antenna broadcast channel, the sum-rate capacity achieving transmit strategy requires the centralized transmitter to simultaneously communicate with multiple receivers. The objective of this paper is to design an implementable sum-rate capacity achieving transmit strategy that uses a combination of beamforming and coding for known interference. For a Gaussian broadcast channel with two transmit antennas and two receivers with one antenna each, results indicate that with QAM constellations there is significant gain in sum-rate capacity over an approach that uses only beamforming.