Orthogonal or Superimposed Pilots? A Rate-Efficient Channel Estimation Strategy for Stationary MIMO Fading Channels

This paper considers channel estimation for multiple-input multiple-output (MIMO) channels and revisits two competing concepts of including training data into the transmit signal, namely, orthogonal pilot (OP) that periodically transmits alternating pilot-data symbols, and superimposed pilot (SP) that overlays pilot-data symbols over time. We investigate rates achievable by both schemes when the channel undergoes time-selective bandlimited fading and analyze their behaviors with respect to the MIMO dimension and fading speed. By incorporating the multiple-antenna factors, we demonstrate that the widely known trend in which the OP is superior to the SP in the regimes of high signal-to-noise ratio (SNR) and slow fading, and vice versa, does not hold in general. As the number of transmit antennas (n_t) increases, the range of operable fading speeds for the OP is significantly narrowed due to limited time resources for channel estimation and insufficient fading samples, which results in the SP being competitive in wider speed and SNR ranges. For a sufficiently small n_t, we demonstrate that as the fading variation becomes slower, the estimation quality for the SP can be superior to that for the OP. In this case, the SP outperforms the OP in the slow-fading regime due to full utilization of time for data transmission.