Source author record

Francesca Cuomo

Francesca Cuomo appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Networking and Internet Architecture Artificial Intelligence Cryptography and Security Distributed, Parallel, and Cluster Computing

Catalog footprint

What is connected

4works

4topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Enabling Deep Reinforcement Learning Research for Energy Saving in Open RAN

The growing performance demands and higher deployment densities of next-generation wireless systems emphasize the importance of adopting strategies to manage the energy efficiency of mobile networks. In this demo, we showcase a framework that enables research on Deep Reinforcement Learning (DRL) techniques for improving the energy efficiency of intelligent and programmable Open Radio Access Network (RAN) systems. Using the open-source simulator ns-O-RAN and the reinforcement learning environment Gymnasium, the framework enables to train and evaluate DRL agents that dynamically control the activation and deactivation of cells in a 5G network. We show how to collect data for training and evaluate the impact of DRL on energy efficiency in a realistic 5G network scenario, including users' mobility and handovers, a full protocol stack, and 3rd Generation Partnership Project (3GPP)-compliant channel models. The tool will be open-sourced and a tutorial for energy efficiency testing in ns-O-RAN.

preprint2026arXiv

Towards Multi-Model LLM Schedulers: Empirical Insights into Offloading and Preemption

Modern deployments of Large Language Models (LLMs) increasingly require serving multiple models with diverse architectures, sizes, and specialization on shared, heterogeneous hardware. This setting introduces new challenges for resource allocation, dispatching, and scheduling, particularly under GPU memory constraints where partial CPU-GPU offloading and preemption become necessary. While existing systems primarily optimize throughput for a single model, comparatively little work addresses multi-model scheduling under these conditions. In this paper, we present an empirical study of how different LLMs behave across hardware platforms, focusing on the performance implications of layer offloading and preemption. We show that offloading leads to strongly non-linear and model-dependent degradation in decode throughput, with smaller models exhibiting sharper sensitivity to reduced GPU residency. We further demonstrate that preemption incurs substantial overhead, largely dominated by model state reload rather than key-value cache transfer, and that this cost varies significantly across models and hardware platforms. Additionally, we highlight the role of sequence length and interconnect bandwidth in amplifying data movement and execution inefficiencies. Based on these findings, we identify a set of key features that future schedulers must consider, including model-specific offloading sensitivity, workload characteristics, and the cost structure of preemption and data transfer. These insights provide guidance for the design of next-generation LLM serving systems capable of efficiently managing heterogeneous, multi-model workloads with hybrid CPU-GPU execution.

preprint2022arXiv

Intrusion Detection Systems for IoT: opportunities and challenges offered by Edge Computing and Machine Learning

Key components of current cybersecurity methods are the Intrusion Detection Systems (IDSs) were different techniques and architectures are applied to detect intrusions. IDSs can be based either on cross-checking monitored events with a database of known intrusion experiences, known as signature-based, or on learning the normal behavior of the system and reporting whether some anomalous events occur, named anomaly-based. This work is dedicated to survey the application of IDS to the Internet of Things (IoT) networks, where also the edge computing is used to support the IDS implementation. New challenges that arise when deploying an IDS in an edge scenario are identified and remedies are proposed. We focus on anomaly-based IDSs, showing the main techniques that can be leveraged to detect anomalies and we present machine learning techniques and their application in the context of an IDS, describing the expected advantages and disadvantages that a specific technique could cause.

preprint2013arXiv

Analysis of urban traffic data sets for VANETs simulations

Vehicular Ad-hoc Networks (VANET) are self-organized, distributed communication networks built up from moving vehicles where each node is characterized by variable speed, strict limits of freedom in movement patterns and a variety of traffic dynamics. In the last few years vehicular traffic is attracting a growing attention from both industry and research, due to the importance of the related applications, ranging from traffic control to road safety. However, less attention has been paid to the modeling of realistic user mobility through real empirical data that would allow to develop much more effective communication and networking schemes. The aim of this study is then twofold: on one hand we derived real mobility patters to be used in a VANET simulator and on the other hand we simulated the VANET data dissemination achieved with different broadcast protocols in our real traffic setting.