Research On SC-FDMA Technology For LTE/LTE-Advanced Uplink

Long Term Evolution (LTE) of the 3rd Generation Partnership Project (3GPP) is designed for improved service, higher date rate, higher spectral efficiency as well as lower latency into the next decade. LTE-Advanced is an enhancement of LTE, formally submitted as a candidate for 4G system to International Telecommunication Union (ITU) and aiming at exceeding the IMT-Advanced requirements.LTE adopts Single Carrier Frequency Division Multiple Access (SC-FDMA) for the uplink transmission scheme and LTE-Advanced uses enhanced SC-FDMA transmission scheme in the uplink. This thesis describes different types of sub-carrier mapping schemes for SC-FDMA system and analyzes the PAPR performance and link performance of Interleave FDMA (IFDMA) and Localized FDMA (LFDMA) sub-carrier mapping schemes. Simulation results show that both IFDMA system and LFDMA system have lower PAPR than OFDMA system, thus verifying the conclusion that SC-FDMA system has better PAPR performance than OFDMA system. Simulation results also demonstrate that IFDMA system has lower PAPR than LFDMA system and that IFDMA with raised cosine filter and root raised cosine filter both reduce more PAPR than LFDMA. Furthermore, IFDMA with root raised cosine filter achieves lower PAPR than IFDMA with raised cosine filter. Simulation results also show that the BER performance of LFDMA system depends highly on the location of the subband and LFDMA could achieve better BER performance than IFDMA when subband is allocated properly. So, LFDMA could employ either channel-dependent scheduling or subband hopping to achieve a better BER performance than IFDMA.This thesis also presents a new clipping-based PAPR reduction algorithm for LFDMA system, which combines clipping with time domain constellation correction and transmit power compensation processes. Simulation results show that with 0.4dB degradation in the BER performance, the new clipping algorithm can achieve 1.5dB PAPR reduction for LFDMA. Trade-off exists between PAPR performance and BER performance when using our new algorithm.