Beat Noise Compensation in OCDMA Systems using soft decoding based FEC
Résumé
In this paper we study Forward Error Correction (FEC) under beat noise assumption in the context of incoherent Optical Code Division Multiple Access (OCDMA) systems. Due to the use of unipolar codes, Multiple Access Interference (MAI) amount and thus beatings between different user contributions known as beat noise, constitute important limitations in particular for high data rates. In order to compensate for these two major limitations, we propose a solution using FEC based on a soft decoding algorithm. For this purpose, we analyze the OCDMA received signal distribution and establish a new model, based on a mixture of chi-square distributions, valuable for any OCDMA code parameters and any source coherence time values. From this model, we establish the error probability analytical expression under beat noise assumption and determine the reliability parameters required for the FEC soft decoder initialization. To illustrate the performance, we use Low Density Parity Check (LDPC) codes and show that the LDPC adapted decoder is highly efficient in reducing beat noise and MAI for incoherent OCDMA systems, even when beat noise has a severe impact. In addition, we point out that this solution permits increasing data rates or number of active users.