Closed-Form Enhanced Detection of Clipped OFDM Symbol
Adriana Lipovac,Ante Mihaljević,Vlatko Lipovac
DOI: https://doi.org/10.1109/access.2024.3353614
IF: 3.9
2024-01-01
IEEE Access
Abstract:Large peak-to-average power ratio (PAPR) and carrier frequency offset (CFO) are dominant impairments of the orthogonal frequency-division multiplexing (OFDM) symbol transmission that is applied within the state-of-the-art wireless operator networks. In this work, we deal with consequences of the amplitude peak clipping that is commonly used at the transmitter to reduce the PAPR of the OFDM symbol, and thus prevent its non-linear distortion which would otherwise be imposed by the output high-power amplifier (HPA). Accordingly, regardless of the clipping generating mechanism at the transmitter being either inherent (related to the HPA) or deliberate (due to PAPR reduction), the clipped incoming OFDM symbol at the receiver may lead to degraded detection accuracy and transmission performance. However, the methods that have been applied so far at the receiver for compensating non-linear distortion due to clipping, are quite complex and computationally demanding. On the contrary, we propose effective mitigation of the problem to be performed at the receiver, by deriving the closed-form enhanced detection criterion, which requires common measurements of the mean and the rms values, as well as the autocorrelation of the received OFDM symbol comprising both un-clipped and clipped sections. Such improved detection was shown to significantly reduce the side effects of clipping, and restore satisfactory transmission performance – the bit error rate (BER) in particular. The proposed analytical model was preliminarily verified by versatile Monte-Carlo simulations and professional industry-standard vector signal analysis (VSA) test system, as well as by BER testing. The evident convergence of the three methods’ test results leads to the conclusion that the proposed clipped OFDM symbol detection method provides clear improvement with respect to the conventional one.
computer science, information systems,telecommunications,engineering, electrical & electronic