,
Tikrit University , Salahaddin , Iraq
,
Tikrit University , Salahaddin , Iraq
Tikrit University , Salahaddin , Iraq
The growing need to improve the speed and quality of communication networks currently discussed in this paper is motivated by the uncontrolled increase in the world data traffic. Although fiber optic is essentially well placed to satisfy this requirement, it has long term problems like scattering, distortion nonlinearity and oscillating noise. The current Machine Learning (ML)-enabled optical communication designs mainly maximize the Bit Error Rate (BER) and the throughput in the separation, but they do not take into account the computational burden and network latency of the ML models themselves. To address these shortcomings, we offer a new hybrid intelligent optical communication system which integrates Convolutional Neural Networks (CNNs) and an evolution-based adaptive modulation selection scheme (Genetic Algorithm (GA)). The main innovation is that three key metrics are optimized jointly and multi-objectively: BER, latency, and evaluation overhead which is a key distinguishing factor compared to the previous single object-optimized modulation adaptation frameworks. The suggested solution is a dynamical control of the modulation scheme, i.e., the choice of QPSK, 16-QAM, and OFDM, according to the real-time Signal-to-Noise Ratio (SNR) and dispersion patterns. It was simulated and verified with the help of the MATLAB R2023a and Opti System 17 using the multi-wavelength Dense Wavelength Division Multiplexing (DWDM) platform. The system recorded the highest throughput of 96 Gbps and a 45 percent reduction in BER over traditional systems, which validated a BER improvement over the older system models. Attenuation was set at 1.5 dB km over a 50km fiber connection with the average latency of less than 10ms. The effectiveness of this hybrid method is better established and confirmed through comparative analysis with six benchmark studies that prove the superiority and scalability of this hybrid method in next-generation and ultra-high-speed fiber systems.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
0
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.