We introduce two novel joint radio-frequency (RF)-baseband designs for receivers in a MIMO system with N t transmit antennas, N r receive antennas, but only L < N r RF chains at the receiver. The joint design introduces an RF preprocessing matrix that processes the signals from the different antennas, and is followed by selection (if necessary), downconversion, and further processing in the baseband. The schemes are similar to conventional antenna selection in that they use fewer RF chains than antenna elements, but achieve superior performance by exploiting the spatial correlation of the received signals. The first of our proposed designs uses an L × N r RF pre-processing matrix that outputs only L streams followed by baseband signal processing, and, thus, eliminates the need for a selection switch. The second one uses an N r × N r RF preprocessing matrix that outputs N r streams and is followed by a switch that selects L streams for baseband signal processing. Both spatial diversity and spatial multiplexing systems are considered and the optimum pre-processing matrices are derived for all cases. To accommodate practical RF design constraints, which prefer a variable phase-shifter-based implementation, a suboptimal phase approximation is also introduced. Performance better than conventional antenna selection and close to the full complexity receiver is observed in both single cluster and multi-cluster wireless channels. A beam-pattern-based geometric intuition is also developed to illustrate the effectiveness of the optimal solutions.

Multiple antenna transmitter and receiver architectures that combine antenna selection with RF pre-processing have been shown to significantly outperform conventional antenna selection with the same number of RF chains. Often, performance close to a full complexity architecture (with more RF chains) is also achieved. This paper studies the effect of hardware and signal processing non-idealities on such architectures. We show that they are robust to quantization, phase, and calibration errors introduced by RF phase-shifters, and also to the channel estimation errors. While insertion loss does lead to performance degradation, performance better than conventional antenna selection is observed for typical insertion loss values. RAWCON 2004This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. W1B.1Abstract-Multiple antenna transmitter and receiver architectures that combine antenna selection with RF pre-processing have been shown to significantly outperform conventional antenna selection with the same number of RF chains. Often, performance close to a full complexity architecture (with more RF chains) is also achieved. This paper studies the effect of hardware and signal processing non-idealities on such architectures. We show that they are robust to quantization, phase, and calibration errors introduced by RF phase-shifters, and also to the channel estimation errors. While insertion loss does lead to performance degradation, performance better than conventional antenna selection is observed for typical insertion loss values.

No abstract

For a multiple input multiple output system, antenna selection reduces complexity at the expense of performance. In this paper, we propose two novel RF pre-processing architectures that significantly improve the performance of antenna selection, which marginally increasing the complexity. These architectures introduce a RF pre-processing matrix, M, that multiplies the vector of incoming signals prior to down-conversion. The elements of M use only the knowledge of the channel statistics. In the first architecture, M outputs a reduced number of streamsan explicit selection algorithm is therefore not required. In the second architecture, the number of output streams equals the number of input streams, and a selection switch that uses instantaneous channel state information achieves the reduction of the number of streams. We show that the optimal pre-processing receiver projects the received signal along the eigenvectors of the correlation matrix. In a correlated channel, both these architectures significantly outperform conventional antenna selection. We also develop a beam-pattern based intuition and compare the performance of our scheme to other RF pre-processing schemes previously proposed in the literature. Globecom 2004This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. Abstract-For a multiple input multiple output system, antenna selection reduces complexity at the expense of performance. In this paper, we propose two novel RF pre-processing architectures that significantly improve the performance of antenna selection, which marginally increasing the complexity. These architectures introduce a RF pre-processing matrix, M, that multiplies the vector of incoming signals prior to downconversion. The elements of M use only the knowledge of the channel statistics. In the first architecture, M outputs a reduced number of streams -an explicit selection algorithm is therefore not required. In the second architecture, the number of output streams equals the number of input streams, and the reduction of the number of streams is achieved by a selection switch that uses instantaneous channel state information. We show that the optimal pre-processing receiver projects the received signal along the eigenvectors of the correlation matrix. In a correlated channel, both these architectures significantly outperform conventional antenna selection. We also develop a beam-pattern based intuition and comp...

No abstract