A non-data-aided error vector magnitude adaptive algorithm in fading channels

Abstract

In order to meet the ever-increasing requirements of spectrum efficiency in mobile communications, an adaptive modulation algorithm (NDA-EVM-AM) using non-data-aided error vector magnitude to evaluate channel quality is proposed in this paper. The algorithm dynamically allocates the statistical length of NDA-EVM according to the coherence time of the fading channel. Under the limitation of constant transmit power, the maximum modulation order suitable for the current channel is obtained from the calculated EVM value. The NDA-EVM statistical length of the channel in the algorithm of this paper can be dynamically changed according to the channel fading characteristics, and the EVM value can reflect the channel change in real time; in addition, according to the derived EVM analytical formula in the fading channel, it can predict the transmission of data with any modulation order in the current channel environment High performance, accurately select the maximum modulation order suitable for the current channel, and reduce the number of adjustments of the modulation order. Both theoretical analysis and simulation show that: compared with the traditional adaptive modulation algorithm based on signal-to-noise ratio (SNR-AM), the algorithm in this paper has better performance in terms of modulation order selection, channel fading speed and channel coherence time change speed. So there is also a higher spectrum efficiency.