Machine Learning Enabled Detection for QPSK-PD-NOMA System Employing Single Mode Fiber

Abstract

In this paper, we propose a Successive Interference Cancellation (SIC) technique which uses k-means clustering for a standard single mode fiber based non return to zero (NRZ)-quadrature phase shift keying (QPSK) transmission system that employs power division non-orthogonal multiple access (PD-NOMA) scheme to enhance the capacity. Results obtained from the numerical simulation confirm that for an aggregated transmission of 20 Gbps signal for two users, each transmitting a data rate of 10 Gbps over standard single mode fiber (SSMF), a BER value of 10−5 is achievable with an OSNR penalty of 2 dB for a far user, which is situated 100 km away from the transmitter and utilizes NOMA. The proposed system doesn't include any compensation technique for mitigating fiber induced impairments. Simulation results also reveal that, a 1:7 power division ratio is required between the two users for optimum performance.