Clipping noise estimation in OFDM systems: A greedy-based approach

A major drawback of orthogonal frequency division multiplexing (OFDM) is the high peak-to-average power ratio (PAPR) which drives transmitter's power amplifier into saturation. One of the simplest methods for mitigating PAPR is to clip signal prior to the amplifier. However, this technique suffers from noise caused by clipping operation and limits its practical application. To investigate this issue, we have proposed a greedy algorithm based on orthogonal matching pursuit (OMP) utilizing a distance metric and a threshold parameter in order to define a backward condition to achieve better results in clipping noise estimation. The defined distance metric and threshold parameter are jointly used to determine whether the recovered noise indices are valid or not. Besides, a closed form equation for threshold parameter has been obtained in terms of channel statistics and has been validated by extensive numerical simulations. According to such appropriate choice of threshold values, simulations demonstrate that the proposed algorithm outperforms state-of-the-art methods such as OMP, ℓ1-minimization and classic methods, especially when SNR grows higher. Also, in terms of computational complexity, proposed algorithm due to its structure imposes a slightly further complexity to the receiver in comparison with conventional OMP algorithm.