Phaseless pilots for OFDM

PAPR reduction and efficient channel estimation are one of the main challenges for multicarrier systems. Common approaches use pilot sequences to estimate the channel at the receiver and to compensate the PAPR at the transmitter. But these approaches need full knowledge of the pilots at the receiver. Here, we present an alternative phaseless pilot scheme such that the receiver does not need knowledge of the pilot phases. Therefore, pilot phases can be used now to compensate for each transmitted data signal the PAPR optimally. Hence, our new PAPR reduction scheme is adaptive. We establish a new method were channel estimation based on phaseless pilots is accomplished with a phase retrieval approach from suitable chosen magnitude measurements. Even more we use compressed sensing methods for estimation by exploiting the sparse nature of the channel impulse response. We can show that a significant gain in (uncoded) bit error rate can be obtained in amplitude-constrained channels as compared to classical pilot-aided schemes such as Zadoff-Chu pilot sequences.