Authors: Salman A. Khan
Addresses: Department of Electrical and Computer Engineering, McGill University, 3480 University Street, Montreal, QC H3A 2A7, Canada
Abstract: This paper extends the existing class of autocorrelation function (ACF) based velocity estimation schemes, well-studied for single-antenna channels, to Multiple-Input Multiple-Output channels. The complexity of an intuitive extension approach, i.e. averaging individual velocity estimates from all MIMO sub-channels, increases exponentially with increasing MIMO antennas. This paper proposes two low-complexity schemes for the symmetric correlated MIMO channel with 'N' antennas. The first scheme, for a MIMO channel correlated on strictly either transmitter or receiver side, leads to complexity reduction by a factor of 'N/2'. The second scheme, for MIMO channels exhibiting correlation on both sides of the channel is a more intelligent velocity estimation scheme which allows reduction of complexity by a factor of 'N²/4'. In addition to the reduced complexity, velocity estimates obtained using both proposed schemes are more accurate than the intuitive extension approach. With 'massive MIMO' being envisioned as a reality in 5G, these schemes offer a low-complexity solution to estimate the important velocity parameter, eventually aiding diverse dynamic resource allocation schemes.
Keywords: mobile velocity estimation; correlated MIMO channels; multiple input multiple output channels; autocorrelation function; dynamic resource allocation; complexity reduction.
International Journal of Wireless and Mobile Computing, 2015 Vol.8 No.1, pp.1 - 8
Received: 18 Jun 2014
Accepted: 30 Jun 2014
Published online: 02 Jan 2015 *