9 October 2018 in IEEE Transactions on Wireless Communications
Coarsely quantized massive Multiple-Input Multiple-Output (MIMO) systems are gaining more interest due to their power efficiency. We present a new precoding technique to mitigate the Multi-User Interference (MUI) and the quantization distortions in a downlink Multi-User (MU) MIMO system with coarsely Quantized Constant Envelope (QCE) signals at the transmitter. The transmit signal vector is optimized for every desired received vector taking into account a relaxed version of the QCE constraint. The optimization is based on maximizing the safety margin to the decision thresholds of the receiver constellation modulation. Due to the linear property of the objective function and the constraints, the optimization problem is formulated as a linear programming problem. Simulation results show a significant gain in terms of the uncoded Bit Error Rate (BER) compared to the existing precoding techniques.
8 November 2018 in IEEE Transactions on Wireless Communications
In this contribution, we investigate a coarsely quantized multi-user multiple-input single-output downlink communication system, where we assume 1-bit digital-to-analog converters at the base station antennas. First, we analyze the achievable sum rate lower-bound using the Bussgang decomposition under new assumptions. In the presence of the non-linear quantization, our analysis indicates the potential merit of reconsidering traditional signal processing techniques in coarsely quantized systems, i.e., reconsidering transmit covariance matrices whose rank is equal to the rank of the channel. Furthermore, in the latter part of this paper, we propose a linear precoder design that achieves the predicted increase in performance compared with a state-of-the-art linear precoder design. Moreover, our linear signal processing algorithm allows for higher order modulation schemes to be employed.
15 June 2018 in IEEE Wireless Communications Letters
Transmit signals that have constant magnitudes and discrete phases can decrease the power consumption of future wireless communication systems. To obtain signals fulfilling that property, Constant Envelope Quantizers (CEQs) have to be deployed in the system. However, the increased power efficiency comes at the cost of degraded system performance. To mitigate this performance loss with conventional digital filters, e.g Wiener Filter (WF) precoders, the statistical signal properties at the CEQ have to be known. In this context, we study the effect of the CEQ on the statistical signal properties and derive closed-form expressions of the correlation factors at the CEQ with complex-valued Gaussian input signals.