High peak-to-average power ratio is one of the major drawbacks of orthogonal frequency division multiplexing (OFDM). Clipping is the simplest peak reduction scheme, however, it requires clipping mitigation at the receiver. Recently compressed sensing has been used for clipping mitigation (by exploiting the sparse nature of clipping signal). However, clipping estimation in multi-user scenario (i.e., OFDMA) is not straightforward as clipping distortions overlap in frequency domain and one cannot distinguish between distortions from different users. In this work, a collaborative clipping removal strategy is proposed based on joint estimation of the clipping distortions from all users. Further, an effective data aided channel estimation strategy for clipped OFDM is also outlined. Simulation results are presented to justify the effectiveness of the proposed schemes. © 2014 IEEE.