Selected mapping (SLM) works effectively in peak‐to‐average power ratio reduction for the single‐carrier frequency‐division multiple access (SC‐FDMA) system. However, the need for side information (SI) transmission results in data rate loss and increase of the system complexity. To avoid explicit SI transmission for SLM‐based SC‐FDMA systems, an SI embedding and detection scheme is proposed by exploiting the block pilot symbols . Specifically, to enable SI embedding, the authors propose to represent the SI index for each data symbol in the subframe by a location set of some selected subcarriers in the block pilot symbol. For SI detection, an optimal pilot‐aided maximum‐likelihood SI detector is presented. As a low‐complexity alternative, a suboptimal log‐likelihood ratio SI detector is also proposed. In addition, the authors provide theoretical analysis of SI detection error rate and discuss the effect of the proposed scheme on channel estimation. Simulation results show that, in SLM‐based SC‐FDMA systems, the proposed scheme significantly outperforms existing methods in terms of both SI detection and bit error rate (BER) and achieves the same BER performance as its counterpart with perfect SI.

In the LTE system, uplink synchronization can be established through the random access channel, by which timing and frequency offsets between transceivers can be estimated and adjusted. As the single carrier frequency division multiple access (SC-FDMA) structure is utilized in LTE uplink, preambles for the random access channel should be generated and detected in the transform-domain at user equipments and the base station, respectively. However, in available literature, an explicit transform-domain signal model with respect to timing and frequency offsets has not been provided, and the impact of frequency offsets on the transform-domain signal is not analyzed either. Therefore, suffering from unknown interference among multiple user equipments, existing multiuser detection algorithms provide unsatisfied multiuser detection and estimation performance. In this paper, an explicit transform-domain signal model is first derived with respect to both timing and frequency offsets. The characteristic of transform-domain frequency offsets is then analyzed, which indicates that transform-domain frequency offsets behave differently with time-domain frequency offsets. After that, a multi-steps hybrid multiuser detection algorithm is proposed to improve the multiuser detection and estimation performance, and the Cramër-Rao bound to the accuracy of parameters estimation for the random access procedure is finally obtained. Simulation results show that the proposed algorithm is able to significantly improve the multiuser detection and estimation performance compared to existing ones.