arXiv subjects
Publications and source records attributed to Agastya Raj.
Challenges associated with densification of radio access networks are motivating exploration of more efficient and scalable architectures. We examine recent progress in one direction that involves moving beyond radio and optical convergence to full confluence.
We present a DLM-anchored hybrid physics/ML framework for brownfield optical links that accurately predicts per-channel power, OSNR, and GSNR. Calibrating span/ILA boundaries via DLM yields OSNR/GSNR errors of no more than 0.39/0.43 dB across single-channel and OSaaS provisioning.
We present a fully unsupervised Fast-Slow DSVDD detector for continuous State-of-Polarization monitoring on a deployed subsea cable. Trained without event labels, it ranks all five confirmed trawler contacts within the top 13 of 122,174 recordings and surfaces additional corroborated cable-contact events.
We present an ML-based vessel detection and localization system, trained with weak supervision from imperfect AIS labels, that achieves a 97.8% detection rate at 1.98% false-trigger rate, successfully identifies dark-vessel events from unlabeled data.
We present a case study on the Emerald Fibre Bridge Link, an operational subsea telecom cable connecting Dublin and North Wales, examining DAS vessel-related signatures jointly with concurrent AIS data. The observations show that vessel-related DAS responses depend on local cable sensitivity and background conditions, while their interpretation is complicated by imperfect AIS reporting. Examining vessel-crossing events jointly, we identify representative DAS-AIS association patterns, ranging from clear vessel matches to offset, ambiguous, AIS-incomplete, AIS-silent-candidate, and non-vessel confounders. These observations reveal the gap between physical measurements at the cable and cooperative vessel reporting, providing practical insights for designing future DAS-assisted cable-protection workflows.
We develop an adaptive OFDM framework for underwater acoustic communications based on PatchCSI-T, a Transformer-based multistep channel prediction model with feature-independent modeling and parameter sharing. Combined with a greedy adaptive modulation and power allocation scheme, the proposed approach enables accurate, low-latency CSI forecasting and improves end-to-end BER and spectral efficiency on real-world UWA channel datasets.
We propose a spectrum-configuration framework for open and disaggregated optical systems that maximizes throughput while guaranteeing the quality of transmission (QoT) margins. The framework jointly optimizes transceiver parameters, including modulation format, symbol rate, pulse-shaping roll-off factor, and wavelength-selective switch (WSS) bandwidth, under fixed spectral allocation constraints. The impact of roll-off factor optimization is first experimentally evaluated in the presence of cascaded WSS filtering, demonstrating measurable QoT gains for both single- and multi-channel transmission. Building on these observations, a knapsack-based optimization is applied in the context of Optical Spectrum as a Service (OSaaS) to select service configurations that maximize aggregate throughput within a fixed spectrum width and limited transceiver resources. Experimental validation on a metro-scale open testbed confirms the effectiveness of the proposed approach in achieving efficient spectrum utilization and adaptive throughput-margin trade-offs.
Accurate modeling of the gain spectrum in Erbium-Doped Fiber Amplifiers (EDFAs) is essential for optimizing optical network performance, particularly as networks evolve toward multi-vendor solutions. In this work, we propose a generalized few-shot transfer learning architecture based on a Semi-Supervised Self-Normalizing Neural Network (SS-NN) that leverages internal EDFA features - such as VOA input or output power and attenuation, to improve gain spectrum prediction. Our SS-NN model employs a two-phase training strategy comprising unsupervised pre-training with noise-augmented measurements and supervised fine-tuning with a custom weighted MSE loss. Furthermore, we extend the framework with transfer learning (TL) techniques that enable both homogeneous (same-feature space) and heterogeneous (different-feature sets) model adaptation across booster, preamplifier, and ILA EDFAs. To address feature mismatches in heterogeneous TL, we incorporate a covariance matching loss to align second-order feature statistics between source and target domains. Extensive experiments conducted across 26 EDFAs in the COSMOS and Open Ireland testbeds demonstrate that the proposed approach significantly reduces the number of measurements requirements on the system while achieving lower mean absolute errors and improved error distributions compared to benchmark methods.
With the growing demand for high-bandwidth, low-latency applications, Optical Spectrum as a Service (OSaaS) is of interest for flexible bandwidth allocation within Elastic Optical Networks (EONs) and Open Line Systems (OLS). While OSaaS facilitates transparent connectivity and resource sharing among users, it raises concerns over potential network vulnerabilities due to shared fiber access and inter-channel interference, such as fiber non-linearity and amplifier based crosstalk. These challenges are exacerbated in multi-user environments, complicating the identification and localization of service interferences. To reduce system disruptions and system repair costs, it is beneficial to detect and identify such interferences timely. Addressing these challenges, this paper introduces a Machine Learning (ML) based architecture for network operators to detect and attribute interferences to specific OSaaS users while blind to the users' internal spectrum details. Our methodology leverages available coarse power measurements and operator channel performance data, bypassing the need for internal user information of wide-band shared spectra. Experimental studies conducted on a 190 km optical line system in the Open Ireland testbed, with three OSaaS users demonstrate the model's capability to accurately classify the source of interferences, achieving a classification accuracy of 90.3%.
We demonstrate the end-to-end transmission of digital coherent and analog radio-over-fiber signals, at mmWave and sub-THz frequencies, over the HEAnet live production metro network using Optical Spectrum-as-a-Service (OSaaS), transparently connected to a passive optical network.
We present a framework to identify and mitigate rogue OOK signals and user-generated power interference in a multi-user Optical-Spectrum-as-a-Service network. Experimental tests on the OpenIreland-testbed achieve up to 89% detection rate within 10 seconds of an interference event.
We implement a ML-based attention framework with component-specific decoders, improving optical power spectrum prediction in multi-span networks. By reducing the need for in-depth training on each component, the framework can be scaled to multi-span topologies with minimal data collection, making it suitable for brown-field scenarios.
We introduce a ML-based architecture for network operators to detect impairments from specific OSaaS users while blind to the users' internal spectrum details. Experimental studies with three OSaaS users demonstrate the model's capability to accurately classify the source of impairments, achieving classification accuracy of 94.2%.
The gain spectrum of an Erbium-Doped Fiber Amplifier (EDFA) has a complex dependence on channel loading, pump power, and operating mode, making accurate modeling difficult to achieve. Machine Learning (ML) based modeling methods can achieve high accuracy, but they require comprehensive data collection. We present a novel ML-based Semi-Supervised, Self-Normalizing Neural Network (SS-NN) framework to model the wavelength dependent gain of EDFAs using minimal data, which achieve a Mean Absolute Error (MAE) of 0.07/0.08 dB for booster/pre-amplifier gain prediction. We further perform Transfer Learning (TL) using a single additional measurement per target-gain setting to transfer this model among 22 EDFAs in Open Ireland and COSMOS testbeds, which achieves a MAE of less than 0.19 dB even when operated across different amplifier types. We show that the SS-NN model achieves high accuracy for gain spectrum prediction with minimal data requirement when compared with current benchmark methods.
We present a novel ML framework for modeling the wavelength-dependent gain of multiple EDFAs, based on semi-supervised, self-normalizing neural networks, enabling one-shot transfer learning. Our experiments on 22 EDFAs in Open Ireland and COSMOS testbeds show high-accuracy transfer-learning even when operated across different amplifier types.
Accurate estimation of the Underwater acoustic (UWA) is a key part of underwater communications, especially for coherent systems. The severe multipath effects and large delay spreads make the estimation problem large-scale. The non-stationary, non-Gaussian, and impulsive nature of ocean ambient noise poses further obstacles to the design of estimation algorithms. Under the framework of compressed sensing (CS), this work addresses the issue of robust channel estimation when measurements are contaminated by impulsive noise. A first-order algorithm based on alternating direction method of multipliers (ADMM) is proposed. Numerical simulations of time-varying channel estimation are performed to show its improved performance in highly impulsive noise environments.