arXiv subjects

Juliano Araujo Wickboldt

Publications and source records attributed to Juliano Araujo Wickboldt.

8 recordsLinked to original sources

arXiv2026

Conflict-Aware Autonomous Resource Orchestration for 5G Edge Slicing Using Multi-Agent Systems and LLMs

5G edge network slicing must reconcile two coupled requirements: understanding high-level operator intent and reallocating scarce computing and bandwidth resources under dynamic contention. Centralized optimization can face computational and responsiveness challenges under rapidly changing edge contention, and Multi-Agent System (MAS)-based auctions can support resource arbitration but still depend on operator-defined economic parameters and priorities. Embedding a Large Language Model (LLM) directly in a recurring control loop, by contrast, may introduce non-negligible inference latency and non-deterministic behavior. This paper proposes a two-timescale architecture that addresses this gap: an LLM-based Strategist translates natural-language network intents into economic policy parameters (budget, income, and bidding-aggressiveness) at the macro scale, while a deterministic, Vickrey-inspired second-price mechanism run by SPADE-based slice agents resolves recurring resource contention at the operational scale, with the resulting allocations enforced by patching Kubernetes CPU limits and Calico CNI bandwidth annotations directly on User Plane Function pods. We evaluate the architecture end-to-end on a real Kubernetes/Open5GS edge cluster across two complementary scenarios, each repeated 30 times: a Smart-City scenario, in which natural-language public-safety intents propagate into measurable changes in slice bids and bandwidth allocations within a few auction rounds, and a Hierarchical-slicing scenario, in which the mechanism provides priority-dependent resource and QoE differentiation, adapts to abrupt demand changes, and preserves configured minimum guarantees while maintaining near-complete allocation of the shared bandwidth pool.

cs.NI↗

arXiv2026

Dynamic Sampling for Telemetry in Microservices: A Reinforcement Learning and Entropy-Based Approach

Microservices architectures are increasingly deployed in cloud-based distributed environments, making application development and maintenance more dynamic, but also increasing the complexity of troubleshooting and observability. Distributed tracing tools are therefore essential for request analysis and debugging, despite introducing additional overhead that can be amplified by excessive and inefficient data collection. This article proposes RADAR (Reinforcement Learning Agent for Dynamic And Relevant trace sampling), an agent that combines reinforcement learning with a data entropy assessment to achieve more efficient capture of traces relevant to system monitoring, based on the OpenTelemetry standard. RADAR tests different sampling rules to discover which combination is most efficient. A test environment simulating a minimalist online store with several microservices distributed across a Kubernetes cluster served as the basis for the experiments, which evaluated the agent's convergence and the system's performance in terms of resource consumption and collected data quality. Results showed that RADAR reduced network bandwidth consumption by 97.4% and CPU usage by 99.0% compared to full data collection, also outperforming a fixed-rate sampling baseline. Beyond these resource savings, the approach preserved observability of critical scenarios, retaining approximately 85.6% of rare trace patterns and increasing the average entropy of the stored information by approximately 25%, validating the feasibility of using entropy to orchestrate telemetry autonomously and efficiently.

cs.NI↗

arXiv2025

A Study on 5G Network Slice Isolation Based on Native Cloud and Edge Computing Tools

5G networks support various advanced applications through network slicing, network function virtualization (NFV), and edge computing, ensuring low latency and service isolation. However, private 5G networks relying on open-source tools still face challenges in maturity and integration with edge/cloud platforms, compromising proper slice isolation. This study investigates resource allocation mechanisms to address this issue, conducting experiments in a hospital scenario with medical video conferencing. The results show that CPU limitations improve the performance of prioritized slices, while memory restrictions have minimal impact. The generated data and scripts have been made publicly available for future research and machine learning applications.

cs.NI↗

arXiv2025

KTWIN: A Serverless Kubernetes-based Digital Twin Platform

Digital Twins (DTs) systems are virtual representations of physical assets allowing organizations to gain insights and improve existing processes. In practice, DTs require proper modeling, coherent development and seamless deployment along cloud and edge landscapes relying on established patterns to reduce operational costs. In this work, we propose KTWIN a Kubernetes-based Serverless Platform for Digital Twins. KTWIN was developed using the state-of-the-art open-source Cloud Native tools, allowing DT operators to easily define models through open standards and configure details of the underlying services and infrastructure. The experiments carried out with the developed prototype show that KTWIN can provide a higher level of abstraction to model and deploy a Digital Twin use case without compromising the solution scalability. The tests performed also show cost savings ranging between 60% and 80% compared to overprovisioned scenarios.

cs.NI↗

arXiv2024

A Framework for testing Federated Learning algorithms using an edge-like environment

Federated Learning (FL) is a machine learning paradigm in which many clients cooperatively train a single centralized model while keeping their data private and decentralized. FL is commonly used in edge computing, which involves placing computer workloads (both hardware and software) as close as possible to the edge, where the data is being created and where actions are occurring, enabling faster response times, greater data privacy, and reduced data transfer costs. However, due to the heterogeneous data distributions/contents of clients, it is non-trivial to accurately evaluate the contributions of local models in global centralized model aggregation. This is an example of a major challenge in FL, commonly known as data imbalance or class imbalance. In general, testing and assessing FL algorithms can be a very difficult and complex task due to the distributed nature of the systems. In this work, a framework is proposed and implemented to assess FL algorithms in a more easy and scalable way. This framework is evaluated over a distributed edge-like environment managed by a container orchestration platform (i.e. Kubernetes).

cs.LG↗

arXiv2022

A Digital Twin-based Smart Home: A Proof of Concept Study

A Digital Twin is a virtual system that can fully describe a physical one. It constantly receives data from its counterpart's sensors, consults external sources, and obtains manual inputs from its stakeholders. The DT uses all this information to make various computations, such as analyses, predictions, and simulations, and then possibly sends the results back to the physical system. Domotics, or Smart Home Technologies, brings intelligence and comfort to a house by automating some of its functions. Although the research on both themes is vast, there are few implementations of Smart Homes based on Digital Twin technologies, and this work aims to prove that residences can also benefit from this concept. We implement two different use cases showing that a two-way connection between a home and its virtual counterpart can provide its owners with analyses and simulation-based automation. The first study case allows the user to visualize their home appliances' past and present state concerning their energy consumption. Based on heating simulations, the second determines the best time to turn on a heater to increase the home's thermal comfort and reduce energy usage.

cs.DC↗

arXiv2022

Context-Aware Environment Monitoring to Support LPWAN-based Battlefield Applications

The use of IoT-related technologies is growing in several areas. Applications of environmental monitoring, logistics, smart cities are examples of applications that benefit from advances in IoT. In the military context, IoT applications can support the decision-making process by delivering information collected directly from the battlefield to Command, Control, Communications, Computers, Intelligence, Surveillance and Reconnaissance (C4ISR) systems. Taking the benefit of the installed IoT network in the battlefield, the use of the data collected by the IoT nodes is a way to improve resiliency and increase the survivability of networks, as well as to optimize the use of available resources. Towards improving the communication network present on the battlefield, this work presents a context-aware environmental monitoring system that uses real-time battlefield information to increase military networks' resilience and survivability. The proposed approach is validated by a proof-of-concept experiment. The obtained results show that the implementation of this system can improve the communication process even when the network is exposed to unfavorable climatic factors.

cs.NI↗

arXiv2021

A Cloud-Fog Computing Architecture for Real-Time Digital Twins

Digital Twin systems are designed as two interconnected mirrored spaces, one real and one virtual, each reflecting the other, sharing information, and making predictions based on analysis and simulations. The correct behavior of a real-time Digital Twin depends not only on the logical results of computation but also on the timing constraints. To cope with the large amounts of data that need to be stored and analyzed, modern large scale Digital Twin deployments often rely on cloud-based architectures. A significant portion of the overall response time of a Digital Twin is spent on moving data from the edge to the cloud. Therefore, implementing Digital Twins using cloud-fog architectures emerges as an alternative to bring computing power closer to the edge, reducing latency and allowing faster response times. This paper studies how suitable the use of a cloud-fog architecture is to handle the real-time requirements of Digital Twins. Based on a realistic implementation and deployment of Digital Twin software components, it is possible to conclude that the distribution of Digital Twins in a fog computing setup can reduce response times, meeting its real-time application requirements.

cs.NI↗