Massive MIMO Systems with Hardware Imperfections
Doctoral thesis, 2019
However, massive MIMO systems, i.e. employing hundreds or even thousands of antennas, will be a viable solution in the future only if low-cost and energy-efficient hardware is deployed. Unfortunately, low-cost, low-quality hardware is prone to hardware impairments such as in-phase and quadrature imbalance (IQI) and phase noise.
Moreover, one of the dominant sources of power consumption in massive MIMO systems are the data converters at the BS. The baseband signal at each radio-frequency (RF) chain is generated by a pair of analog-to-digital converters (ADCs). The power consumption of these ADCs increases exponentially with the resolution (in bits) and linearly with the bandwidth. In conventional multi-antenna systems, each RF port is connected to a pair of high-resolution ADCs (e.g., 10-bit or more). For massive MIMO systems this would lead to prohibitively high-power consumption due to the large number of required ADCs. Hence, the ADC resolution must be limited to keep the power budget within tolerable levels.
In this thesis, we investigate the performance of massive MIMO systems in non-ideal hardware. We begin with by studying the impact of IQI on massive MIMO systems. Important insights are gained through the analysis of system performance indicators, such as channel estimation and achievable rates by deriving tractable approximations of the ergodic spectral efficiency.
First, a novel pilot-based joint estimator of the uplink augmented MIMO channel matrix and receiver IQI coefficients is described and then, a low-complexity IQI compensation scheme is proposed which is based on the receiver IQI coefficients' estimation and it is independent of the channel gain.
Second, we investigate the impact of the transceiver IQI in massive MIMO considering a time division duplexing (TDD) system where we assume uplink/downlink channel reciprocity in the downlink precoding design. The uplink channel estimation accuracy and the achievable downlink rate of the regularized zero-forcing (RZF) and maximum ratio transmission (MRT) is studied when there is mismatch between the uplink and downlink channels.
Finally, we analyze the quantization distortion in limited-precision ADCs in uplink massive MIMO systems whose channel state information (CSI) is not known a priori to transmitter and receiver. We show that even a small percentage of clipped samples at the receiver can downgrade considerably the systems performance and propose near-optimal low-complexity solutions to reconstruct the clipped signal.
in-phase and quadrature imbalance
random matrix theory
Chalmers, Electrical Engineering, Signalbehandling och medicinsk teknik, Signal Processing
Massive MIMO Systems with IQ Imbalance: Channel Estimation and Sum Rate Limits
IEEE Transactions on Communications,; Vol. 65(2017)p. 2382 - 2396
IQ Imbalance in Multiuser Systems: Channel Estimation and Compensation
IEEE Transactions on Communications,; Vol. 64(2016)p. 3039-3051
Kolomvakis, N, Ericsson, T, Coldrey, M, Viberg, M. Recontruction of Clipped Signals in Massive MIMO Systems
Doktorsavhandlingar vid Chalmers tekniska högskola. Ny serie: 4590
Chalmers University of Technology
EF, Hörsalsvägen 11, Chalmers
Opponent: Prof. Nikolaos Sidiropoulos, Department of Electrical and Computer Engineering, University of Virginia, USA