To achieve more high‐speed communication systems, multiple‐input multiple‐output orthogonal frequency division multiplexing (MIMO/OFDM) is adopted in several standardisations. On the other hand, the conventional MIMO/OFDM requires many pilot signals to estimate the channel state information (CSI). To achieve an accurate CSI with the small number of pilot signals, high time resolution carrier interferometry (HTRCI) for MIMO/OFDM has been proposed. Moreover, QR decomposition with M‐algorithm maximum likelihood detection (QRM‐MLD) and channel ranking has been proposed to achieve both the reducing complexity and the improving system performance. However, the conventional HTRCI‐MIMO/OFDM with the QRM‐MLD and channel ranking did not consider the problem of the fast fading channel. In the fast fading environment, many errors are caused in the last part of the packet. To overcome this problem, the decision‐direct method has been proposed. However, it requires large complexity because of the iterative processing. Therefore, in this paper, we propose the non‐linear prediction method for an HTRCI‐MIMO/OFDM with the QRM‐MLD and channel ranking. Copyright © 2014 John Wiley & Sons, Ltd.