The bit error rate (BER) in optical communication systems is often get degraded due to various factors like launch power, dispersion, modal noise, and so on. Finding the most optimal launch power for a signal to provide acceptable BER is usually difficult on an installed link. Therefore, in this paper, an attempt has been made to use machine learning-based linear regression technique for predicting the optimal signal quality for spatial division multiplexed (SDM)-based fiber optical transmission system for a fixed distance. This technique helps in predicting the optimal value of continuous launch power. Therefore, to demonstrate the given concept in this research work, the generic setup of an SDM optical long-haul system of 20 km in length has been designed and simulated in Optisystem-14.0. The light sources in our experiments are two spatial optical transmitters with an emission wavelength of around 1550 nm. Abiding to the first step for the regression analysis, that is the data preparation, normality and correlation checks were performed. After this, a linear regression model is developed which was validated through a summary report, coefficients and diagnostic plots. Furthermore, the accuracy of the model is improved by employing Cook's distance. These tests help in dealing with the influence points that hinders the prediction ability of the proposed model. The results show that the R-squared value is 0.9366, and value of adjusted R 2 comes out to be around 0.9344; we therefore came to infer that the model explains nearly 94% variations in the dependent variables. K E Y W O R D Sbit error rate (BER), launch power (LP), linear regression, optical transmission systems, spatial division multiplex (SDM)

Summary Over the years, optical communication systems have been a significant source of fast and secure communication. However, factors like noise and mitigation error can degrade the bit error rate (BER) and quality factor (Q factor) of optical communication systems. Predicting the optimal threshold, Q factor, and BER is usually a difficult task. Therefore, in this paper, machine learning‐based linear regression, least absolute shrinkage and selection operator (LASSO) regression, and Ridge regression have been used for a dense wavelength division multiplexing (DWDM)‐based optical communication network to predict the signal quality. These techniques have been used to predict the desired BER, Q factor, threshold, and eye height of the system. To demonstrate this research concept, a DWDM‐based optical communication network of 50 km length is designed and simulated using Optisystem‐14.0. After data preparation, regression models have been developed and validated through diagnostic plots. Results show that mean square error (MSE) has a significant decline with an increase in the number of epochs for all four models. LASSO and Ridge regression have effectively resolved the issue of overfitting, which occurred in the linear regression case. Furthermore, the mean MSE plot proved the significant reduction of mean MSE in the case of LASSO regression. Results show that min BER for LASSO regression came out to be −173,627.14, providing a robust and cost‐efficient process.

No abstract

No abstract

The performance of FSO communication link is subject to numerous atmospheric factors in wireless communication like fog, rainfall, and haze which leads to deteriorate the performance of a system in term of BER and power at the receiver side. Due to numerous advantages of OCDMA over other access techniques, it allows various users to access a channel simultaneously without intervention with the other user. It has the ability to provide security, large number of users, privacy, reduce interference from multiple users and operate asynchronously. So, in this paper, Multi-Diagonal codes along with the fiber brags grating filters for Spectral amplitude coded OCDMA technique is developed for FSO system and performance analysis is performed in terms of BER and received power.