Nondata-Aided Error Vector Magnitude Analysis of $\eta-\mu$ Fading Channels in Device-to-Device Communications

Transmission in non-line-of-sight (NLOS) conditions has a poor throughput in device-to-device (D2D) communications. In order to achieve high throughput, adaptive modulation has been selected as a spectrally-efficient transmission technology. Moreover, quantifying the performance of a fading channel in NLOS conditions becomes a core issue. In general, the $\eta -\mu $ distribution can be employed to characterize and model the NLOS fading channel. Effective evaluating the quality of $\eta -\mu $ channels can provide an efficient theoretical reference for determining the optimum switching thresholds in adaptive modulation. However, due to the sensitivity of D2D transmission to fading channels, providing an efficient theoretical benchmark for evaluating, the quality of $\eta -\mu $ channels is still a challenge in transmission design. In this paper, the nondata-aided error vector magnitude (NDA-EVM) is considered as a novel metric to evaluate the quality of wireless fading channels. Specifically, the NDA-EVM of the multilevel quadrature amplitude modulation (MQAM) signals over $\eta -\mu $ fading channels and its lower bound is analytically derived. This can be used to further determine the optimum switching thresholds in adaptive modulation in NLOS conditions. The numerical results validate the effectiveness of the proposed formulation and also reveal the influence of the channel parameters on the lower bound of the NDA-EVM.