On the Simulation Performance of FEC SOQPSK-TG Systems with Symbol by Symbol and SOVA Decoding Methods

Abstract

A number of encoding and decoding methods for SOQPSK-TG are evaluated with their simulation performance under additive white Gaussian noise (AWGN) channel. The two forward error correction (FEC) codes are serial concatenated convolutional code (SCCC) and low-density parity-check code (LDPC). Two decoding methods are applied: soft-output Viterbi algorithm (SOVA) and symbol by symbol (SxS) detection. Pulse amplitude modulation (PAM) and pulse truncation (PT) filters are designed for SOVA decoding, and numerically optimum (NO) and integrate and dump (I&D) filters are employed for the SxS methods. We discuss the performance of these schemes by comparing and analyzing their Matlab simulation. The LDPC and SOVA with PAM filter outperforms over all the other combinations. We also use timing and phase recovery with various loop bandwidths. The system structures and the synchronization technique are discussed in this paper.