Investigation on frequency diversity effects for various transmission schemes Using frequency domain equalizer in DFT-precoded OFDMA

This paper presents frequency diversity effects of localized transmission, clustered transmission, and intra-subframe frequency hopping (FH) for Discrete Fourier Transform (DFT)-precoded Orthogonal Frequency Division Multiple Access (OFDMA) using a frequency domain equalizer (FDE). In the evaluations, we employ the normalized frequency mean square covariance (NFMSV) as a measure of the frequency diversity effect, i.e., randomization level of the frequency domain interleaving associated with turbo coding. Link-level computer simulation results show that frequency diversity is very effective in decreasing the required average received signal-to-noise power ratio (SNR) at the target average block error rate (BLER) according to the increase in the entire frequency region in DFT-precoded OFDMA using a linear minimum mean square error (LMMSE) based FDE. Moreover, we show that the NFMSV is an accurate measure of the frequency diversity effect for localized transmission in terms of the average BLER and that the NFMSV for the entire frequency region well approximates the frequency diversity effect for clustered transmission. Finally, we show that the effect of intra-subframe FH is saturated at a relatively narrow FH separation and that there is no close relationship between the FH separation and the NFMSV for the intra-subframe FH.