arXiv subjects

Farnaz Moradi

Publications and source records attributed to Farnaz Moradi.

5 recordsLinked to original sources

arXiv2026

Automated Model Design using Gated Neuron Selection in Telecom

The telecommunications industry is experiencing rapid growth in adopting deep learning for critical tasks such as traffic prediction, signal strength prediction, and quality of service optimisation. However, designing neural network architectures for these applications remains challenging and time-consuming, particularly when targeting compact models suitable for resource-constrained network environments. Therefore, there is a need for automating the model design process to create high-performing models efficiently. This paper introduces TabGNS (Tabular Gated Neuron Selection), a novel gradient-based Neural Architecture Search (NAS) method specifically tailored for tabular data in telecommunications networks. We evaluate TabGNS across multiple telecommunications and generic tabular datasets, demonstrating improvements in prediction performance while reducing the architecture size by 51-82% and reducing the search time by up to 36x compared to state-of-the-art tabular NAS methods. Integrating TabGNS into the model lifecycle management enables automated design of neural networks throughout the lifecycle, accelerating deployment of ML solutions in telecommunications networks.

cs.LG↗

arXiv2025

On Measuring Available Capacity in High-speed Cloud Networks

Measurement of available path capacity with high accuracy over high-speed links deployed in cloud and transport networks is vital for performance assessment and traffic engineering. Methods for measuring the available path capacity rely on sending and receiving time stamped probe packets. A requirement for accurate estimates of the available path capacity is the ability to generate probe packets at a desired rate and also time stamping with high precision and accuracy. This is challenging especially for measurement systems deployed using general purpose hardware. To touch upon the challenge this paper describes and evaluates four approaches for sending and receiving probe packets in high-speed networks (10+ Gbps). The evaluation shows that the baseline approach, based on the native UDP socket, is suitable for available path capacity measurements over links with capacities up to 2.5 Gbps. For higher capacities we show that an implementation based on Data Plane Development Kit (DPDK) gives good results up to 10 Gbps.

cs.NI↗

arXiv2024

Towards Neural Architecture Search for Transfer Learning in 6G Networks

The future 6G network is envisioned to be AI-native, and as such, ML models will be pervasive in support of optimizing performance, reducing energy consumption, and in coping with increasing complexity and heterogeneity. A key challenge is automating the process of finding optimal model architectures satisfying stringent requirements stemming from varying tasks, dynamicity and available resources in the infrastructure and deployment positions. In this paper, we describe and review the state-of-the-art in Neural Architecture Search and Transfer Learning and their applicability in networking. Further, we identify open research challenges and set directions with a specific focus on three main requirements with elements unique to the future network, namely combining NAS and TL, multi-objective search, and tabular data. Finally, we outline and discuss both near-term and long-term work ahead.

cs.NI↗

arXiv2024

Using Large Language Models to Understand Telecom Standards

The Third Generation Partnership Project (3GPP) has successfully introduced standards for global mobility. However, the volume and complexity of these standards has increased over time, thus complicating access to relevant information for vendors and service providers. Use of Generative Artificial Intelligence (AI) and in particular Large Language Models (LLMs), may provide faster access to relevant information. In this paper, we evaluate the capability of state-of-art LLMs to be used as Question Answering (QA) assistants for 3GPP document reference. Our contribution is threefold. First, we provide a benchmark and measuring methods for evaluating performance of LLMs. Second, we do data preprocessing and fine-tuning for one of these LLMs and provide guidelines to increase accuracy of the responses that apply to all LLMs. Third, we provide a model of our own, TeleRoBERTa, that performs on-par with foundation LLMs but with an order of magnitude less number of parameters. Results show that LLMs can be used as a credible reference tool on telecom technical documents, and thus have potential for a number of different applications from troubleshooting and maintenance, to network operations and software product development.

cs.CL↗

arXiv2010

Analyzing the Social Structure and Dynamics of E-mail and Spam in Massive Backbone Internet Traffic

E-mail is probably the most popular application on the Internet, with everyday business and personal communications dependent on it. Spam or unsolicited e-mail has been estimated to cost businesses significant amounts of money. However, our understanding of the network-level behavior of legitimate e-mail traffic and how it differs from spam traffic is limited. In this study, we have passively captured SMTP packets from a 10 Gbit/s Internet backbone link to construct a social network of e-mail users based on their exchanged e-mails. The focus of this paper is on the graph metrics indicating various structural properties of e-mail networks and how they evolve over time. This study also looks into the differences in the structural and temporal characteristics of spam and non-spam networks. Our analysis on the collected data allows us to show several differences between the behavior of spam and legitimate e-mail traffic, which can help us to understand the behavior of spammers and give us the knowledge to statistically model spam traffic on the network-level in order to complement current spam detection techniques.

cs.SI↗