User Mobility Estimation Through Maximum Doppler Frequency Detection

This paper proposes a novel maximum Doppler frequency detection technique for user mobility estimation. The proposed detection estimates the maximum Doppler frequency by making use of the fact that user mobility such as user moving velocity is not distributed continuously. The fluctuation of the channel state information during a packet is applied for the proposed Doppler frequency detection based on likelihood estimation where the fluctuation is compared with the thresholds. The theoretical thresholds are derived on assumption that the fluctuation is distributed with an exponential function. The performance of the proposed detection is evaluated by computer simulation. The proposed detection accomplishes better detection performance as the fluctuation values are aggregated over more packets. The proposed detection achieves about 90% correct detection performance in a fading channel with the $E_{\mathrm{b}}/N_{0}=35\mathrm{dB}$, when the fluctuation values are summed over only three packets.