A Least Squares Approach for Estimating Non-linearity Parameters for OFDM Signals with Bussgang Receivers

Non-linear distortion imposed to orthogonal frequency division multiplexing (OFDM) signals from non-linear power amplifiers (PAs), has always been a challenging problem in wireless communications. This is due to the high peak-to-average power ratio (PAPR) of OFDM signals. To overcome this issue, receiver side compensation techniques based on distortion cancellation and iterative detection have been proposed. However, these techniques require knowledge of the non-linearity (NL) parameters. Therefore, in this paper we focus on NL parameter estimation and propose a least squares (LS) based estimation technique by transmitting a training signal. We study the performance of the proposed technique for the solid state power amplifier (SSPA) model and the Bussgang model. The studies are done considering both perfect channel knowledge and imperfect channel knowledge at the receiver. The results show that the proposed technique has good accuracy and performance for distortion cancellation and iterative detection based receivers such as the Bussgang receiver. The results also indicate that the estimation technique is more sensitive to channel estimation error than the additive Gaussian noise.