Information-Theoretic Analysis of MIMO Channel Sounding

Abstract

The large majority of commercially available multiple-input multiple-output (MIMO) radio channel measurement devices (sounders) is based on time-division multiplexed switching (TDMS) of a single transmit/receive radio frequency chain into the elements of a transmit/receive antenna array. While being cost-effective, such a solution can cause significant measurement errors due to phase noise and frequency offset in the local oscillators. In this paper, we systematically analyze the resulting errors and show that, in practice, overestimation of channel capacity by several hundred percent can occur. Overestimation is caused by phase noise (and to a lesser extent frequency offset) leading to an increase of the MIMO channel rank. Our analysis furthermore reveals that the impact of phase errors is, in general, most pronounced if the physical channel has low rank (typical for line-of-sight or poor scattering scenarios).The extreme case of a rank-1 physical channel is analyzed in detail. Finally, we present measurement results obtained from a commercially employed TDMS-based MIMO channel sounder. In the light of the findings of this paper, the results obtained through MIMO channel measurement campaigns using TDMS-based channel sounders should be interpreted with great care.

Improving MIMO detection performance in presence of phase noise using norm difference criterion

“…But the practical implementation of MIMO systems experience random independent phase perturbations termed as phase noise due to transmit and receive oscillators [2], and thus suffers performance degradation [3]. In particular the effect of phase noise is more detrimental in OFDM systems [4].…”

Capacity Bounds for MIMO Microwave Backhaul Links Affected by Phase Noise

“…The numerical evaluation of these bounds is often based on the algorithm for the computation of the information rates for finite-state channels proposed in [16] (see [17] for different approaches based, e.g., on particle filtering). Although the effect of phase noise in the measurement of MIMO channels has been extensively investigated in the literature (see, e.g., [18]), little is known about the impact of phase noise on the MIMO capacity. In [6], it was shown that different RF circuitries configurations (e.g., independent oscillators at each antenna as opposed to a single oscillator driving all antennas) yield different spatial multiplexing gains.…”

Algorithms for Joint Phase Estimation and Decoding for MIMO Systems in the Presence of Phase Noise and Quasi-Static Fading Channels

“…The phase noise manifests in a MIMO system as the random, time-varying phase differences between the oscillators connected to the antennas at the transmitter and the receiver. Practical designs of MIMO systems based on this assumption can result in significant performance losses and have to be addressed appropriately [2]. The detrimental effects of phase noise can be even more pronounced in scenarios where independent oscillators are connected to each transmit and receive antenna (or a subset of them).…”

“…One of the few works investigating this problem can be found in [16], where the VB framework is employed. In general, MIMO receiver design has focused on developing algorithms for joint channel estimation and data detection (refer to [19], [20] and the references therein)-it is perceived that the phase noise can be handled by existing channel estimation-data detection algorithms since it can be regarded to be a part of the channel [2].…”

“…The channel between the transmit-receive antennas is assumed to be a constant over the length of a frame, and known (or estimated). Each antenna is equipped with an independent free-running oscillator that is perturbed by a random phase noise process, which varies much faster than the channel [16], and conforms to the model in (1) and (2). Specifically, the phase noise process from the oscillators is assumed to be varying from symbol-to-symbol as in (2).…”