Equation-Method for Reducing the Bit-Errors Caused by Clipping Distortion in OFDM Signals

Orthogonal Frequency Division Multiplexing (OFDM) is the digital modulation technique used by 4G and many other communication systems. OFDM waveforms have high amplitude fluctuations resulting in high Peak to Average Power Ratios (PAPR). This can make an OFDM transmitter susceptible to non-linear distortion produced by its high power amplifiers (HPA). Limiting the peaks before an OFDM signal is applied to the HPA causes in-band distortion which introduces bit-errors at the receiver additional to those due to additive white Gaussian noise (AWGN). There have been many ideas for reducing the resulting bit-error rate. An Equation-Method is now proposed, based directly on the fast Fourier transform (FFT). Clipped time-domain samples become the unknowns of a set of equations derived from the received constellation. The use of dither and recursion at the transmitter allows the receiver to identify reliable frequency-domain symbols which may be used to formulate the equations. This approach is receiver oriented and the receiver is assumed to know the clipping levels. Simulation results are presented to show that the Equation-Method performs well with severe clipping as compared to other receiver-oriented techniques such as the Buss gang noise cancellation (BNC) iterative receiver.