ENHANCED PARTIAL TRANSMIT SEQUENCE TECHNIQUE WITH IMPROVED PARTITIONING AND PHASE FACTOR METHODS FOR ORTHOGONAL FREQUENCY DIVISION MULTIPLEXING SYSTEMS

Abstract

Orthogonal Frequency-Division Multiplexing (OFDM) is an important technology that enables transmission at high data rates. However, a major problem of OFDM systems is the high peak-to-average power ratio (PAPR) of OFDM signals. High PAPR drives the transmitter power amplifier (PA) into saturation, and causes nonlinear distortions. In order to improve the performance of the OFDM systems, PAPR should be significantly reduced at low computational complexities. Motivated by the requirement of efficient PAPR reduction techniques associated with multi-carrier systems, such as OFDM, this research explores the state-of-the-art PAPR reduction techniques and proposes new signal processing techniques that can achieve a minimum PAPR at lower computational complexities for given system parameters that are compatible with the appropriate standards. This thesis proposes a novel flexible and practical scheme for PAPR reduction, based on partial transmit sequence (PTS) technique, which is a well known probabilistic approach to the PAPR problem. This approach decreases the PAPR without any significant performance loss, adverse impact, or changes to the system. The proposed approach eliminates the known drawback of current PTS techniques, namely the high implementation complexity to achieve a significant PAPR reduction. In this thesis, three methods have been proposed to realize minimal PAPR with low computational complexity in an OFDM system. These approaches enhance both partitioning and phase rotation schemes within the PTS framework in order to minimize PAPR at lower computational complexities. The performances of the approaches were evaluated via computation of PAPR achieving these approaches and comparing them with the PAPR realized by traditional PTS schemes. The following are the principal contributions of this thesis. First, a new PTS partitioning method, namely the Blocked Interleaved Partitioning (BIP), which leads to better PAPR reduction performance, is proposed. Second, the Smallest Correlation Phase Vector (SCPV) and Blocked Random Phase Selection (BRPS) are proposed to determination the optimal phase rotations. Combination of proposed partitioning and phase rotation schemes resulted in two proposed PAPR reduction schemes namely BIPSC-PTS and BIPBR-PTS respectively. These approaches significantly reduce the PAPR of the system and simultaneously lower computational complexities. The performance of the proposed algorithms was verified via computer simulations. The results show that the proposed methods did improve the performance of the OFDM system over the current ones. The proposed schemes significantly lowered PAPR at lower computational complexities. Benchmarked comparative analysis against the latest published work revealed that the proposed methods BIPSC-PTS and BIPBR-PTS outperformed the published work with 29.7% and 22.9% improvements respectively. Also, the benchmarking of computational time showed that these two proposed methods reduced the requirement of CPU time for phase rotation searching by factors of 78.5 and 15.7, respectively compared to traditional PTS.