Mr. Praveen Pal

Optical Network Traffic Alert System with Machine Learning

Abstract:

The exponential growth in global data traffic has placed immense pressure on optical networks—the backbone of modern digital communication. Fueled by cloud computing, artificial intelligence, streaming applications, and the proliferation of connected devices, the demand for bandwidth continues to rise dramatically. This surge is further accelerated by the expansion of hyperscale data centers and the deployment of high-capacity long-haul and subsea optical links that connect continents and sustain the global Internet infrastructure.

As network scales and technologies evolve, the operational complexity of optical systems has increased significantly. Today’s optical networks span multiple layers, vendors, and domains, generating massive amounts of telemetry data that are challenging to interpret in real time. Conventional rule-based monitoring methods often fail to detect subtle degradations or early warning signs of impending failures, leading to unplanned outages and service disruptions. These incidents directly impact customer service-level agreements (SLAs), where even brief traffic interruptions can translate into significant operational and financial losses.

This talk presents a novel Optical Network Traffic Alert System powered by Machine Learning (ML)—an intelligent approach that moves from reactive to predictive network management. By leveraging advanced ML models to analyze network telemetry, identify anomalies, and predict traffic disruptions before they occur, the proposed system enhances reliability, minimizes downtime, and ensures SLA compliance. The discussion will explore the evolution of optical networking, the growing data demands driving its transformation, the complexity of managing large-scale networks, and how ML-based predictive analytics is shaping the future of autonomous, self-healing optical networks.

Profile:

Mr. Praveen Kumar Pal is a seasoned professional at Nokia of America Corp., with over 20 years of experience in embedded systems, optical communications, and AI/ML-based network optimization. His expertise spans DWDM systems, transponders, Raman and EDFA amplifiers, and advanced coherent pluggables up to 1600 G per wavelength. He has contributed extensively to next-generation optical technologies and the integration of machine learning for intelligent network automation and traffic prediction.
Mr. Pal’s research cover AI/ML, Generative AI, network protocols (GMPLS, BGP, OSPF), and high-speed optical standards including IEEE 802.3 400 GbE/800 GbE and ZR/ZR+ (ITU-T). His interests also extend to wireless systems, IoT, and embedded real-time applications. He is keen on exploring and supporting high-quality research that advances optical, wireless, and AI-driven networking innovations for the next decade of intelligent communications.