Neural network based digital pre-distorter design for DCO-OFDM visible light communications

Abstract

Direct current biased optical orthogonal frequency division multiplexing (DCO-OFDM) is an appealing modulation scheme for reliable, high-speed optical transmissions and foreseen to be used in the upcoming IEEE 802.11bb visible light communication (VLC) standard. However, non-linear characteristics of light emitting diodes (LEDs) as VLC transmitters degrade the bit error rate (BER) performance of DCO-OFDM due to its high peak-to-average-power ratio. In this paper, we propose neural network based digital pre-distorter (DPD) to mitigate non-linear LED response for DCO-OFDM transmission. The proposed scheme extends the reliable transmission range by 1.75 m and 1.8 cm for non-compensated LED response and Memory Polynomial based DPD, respectively.