A novel SLM-based approach for reducing PAPR in LFDMA systems

Abstract

Localized Frequency Division Multiple Access (LFDMA) systems are gaining attention because of its low Peak to Average Power Ratio (PAPR). Selective Mapping (SLM) is one of the most widely used methods for achieving large PAPR reductions with minimal signal distortion. Traditional SLM optimization approaches such as Particle Swarm Optimization (PSO) and Simulated Annealing (SA) have constraints such as the demand for defined performance criteria, the reliance of solution quality on computation and a loss in efficiency for complex situations. In this paper, an SLM system based on Teaching–Learning Based Optimization (TLBO) is developed which substantially enhances PAPR performance without requiring computational preset parameters. The solutions produced by the teacher and learner phases of TLBO algorithm lack diversity, which may hamper the outcomes. In the proposed Modified Teaching–Learning Based Optimization (MTLBO), diversity learning is adopted in which students are sorted into groups prior to knowledge updating and the criteria for grouping in the two phases are fundamentally different. The MTLBO strategy fosters solution heterogeneity throughout the teaching and learning stages while decreasing local trapping and premature solution by applying diverse learning strategies in both phases along with differential learning and neighborhood learning for high-performing and low-performing students. Weights are also employed in some steps to enhance teaching and learning process. MTLBO-SLM-LFDMA systems utilizing Quadrature Phase Shift Keying (QPSK) and Quadrature Amplitude Modulation (QAM) schemes are compared to existing optimization strategies and the simulation results show that MTLBO-SLM-LFDMA outperforms existing approaches while ensuring good solution quality, better convergence and less computational effort.