Low Power OFDM Receiver Exploiting Data Sparseness and DFT Symmetry

Abstract

An undersampling technique appropriate for Orthogonal Frequency Division Multiplexing (OFDM) that can be implemented with low complexity hardware is presented. The OFDM receiver can operate in undersampling mode, 75% of the time. One of the advantages of the proposed scheme is the reduction down to the half of the power consumed by the Analog Digital Converter (ADC) and the Fast Fourier Transform (FFT). The FFT memory requirements for sample storage can also be reduced and its operating speed can be increased. Simulations were performed for two Quadrature Amplitude Modulation orders (16-QAM and 32-QAM), two options for the FFT size (1024 and 4096), two alternative input symbol structures for the inverse FFT (IFFT), and several sparseness levels and samples substitution options. The Symbol Error Rate (SER) and image reconstruction examples are used to show that a full reconstruction or a very low error can be achieved. Although the proposed undersampling method is evaluated for wired channel, it can also be used without any modification to wireless Single Input Single Output (SISO) systems (with an expected SER degradation). It can also be used with Multiple Input Multiple Output (MIMO) systems when an appropriate arrangement of the IFFT input symbols is adopted.