1,721,047 research outputs found
A Network Coverage Algorithm for Message Broadcast in Vehicular Networks
No full textn the recent years, ICT use in vehicular technology has expanded very fast. Inter-vehicle communication (IVC) is seen as the natural next step in this evolution whose availability will give rise to the next generation of smart vehicles. As this technology is deployed, a wide range of services varying from safety to entertainment will become factually accessible to passengers. Many of these envisaged services require a one-to-many communication model which demand for intelligent solutions able to propagate the message in areas of the network where it has not been heard before. To this end, we propose an enhancement of an existent optimal message propagation scheme proposed in the context of one dimensional vehicular networks, namely Fast-Broadcast. Our proposal generalizes the former with the objective of guiding message dissemination to an entire area of interest in generic road topologies. Our proposed scheme exploits bloom filter properties to further advance message propagation into the network. We contrast our approach with the original, lightweight state-of-the-art proposal, showing that it fulfills its design objectives
Enhancing generalization in Federated Learning with heterogeneous data: A comparative literature review
Full text linkFederated Learning (FL) is a collaborative training paradigm whereby a global Machine Learning (ML) model is trained using typically private and distributed data sources without disclosing the raw data. The approach paves the way for better privacy guarantees, improved overall system scalability, and sustainability. In this context, Federated Averaging (FedAvg) is a representative FL algorithm adopting a client–server protocol that operates in synchronous rounds, where selected learners contribute to the global model via local model updates, trained using their private data, while a server entity aggregates the local contributions, producing the new-generation global model as a weighted average of the local ones. However, when clients possess (highly) dissimilar data, the FedAvg technique becomes ineffective due to divergence in client models. Consequently, FedAvg-trained models struggle to generalize when presented with unseen data from the global distribution. In this research paper, we conduct a systematic review of state-of-the-art approaches proposed to counteract global model performance degradation in the presence of heterogeneous data.
To this end, we compile an original taxonomy, highlighting the main algorithmic approaches and mechanisms behind each identified category. Advancing the current body of knowledge, we empirically evaluate the generalization performance on visual tasks of various methods under moderate and significant levels of data heterogeneity, as common practice within the surveyed literature. In addition, the paper benchmarks the performance of hybrid techniques, resulting as a combination of client- and server-side algorithmic tweaks, by shedding light on some associated performance tradeoffs. While recognizing other relevant issues in FL, such as device heterogeneity and energy consumption, which have a non-negligible impact on the learning process, these well-investigated topics are not the main focus of this article
A MECApp-aware Lifecycle Management Approach in 5G Edge-Cloud Deployments
Full text linkThe recent trend pushing towards reliance on edge computing, virtualization and programmatic 5G network control has sparked the development of a myriad of open-source resource management and orchestration projects for improved control and added flexibility, making up for a rich and complex ecosystem of frameworks and tools with varying degree of support for standardized features. In this technological panorama, the ETSI Multi-Access Edge Computing (MEC) standard proposes a conceptual reference architecture, standardizing edge integration, interoperability and application management in an extended 5G edge-core architecture. In this context, we propose an application-aware orchestration solution for 5G edge-core distributed deployments, currently lacking support in state-of-the-art frameworks and tools. The proposal is built on an experimental and distributed deployment of the OpenAirInterface minimal MEC platform implementation and relies on the Kubernetes Operator pattern for the automatic MECApp lifecycle management. To validate our approach, we conduct a series of experiments, reporting key metrics of interest
Editorial: Smart Objects and Technologies
Full text linkSmart objects are entering everyday life and are heavily modifying it. Healthcare, communication, art, entertainment, safety, environment, education, democracy, and human rights, are just a few examples of scenarios that are radically changing thanks to the use of smart objects and technologies.
In this context, the popularity of portable computing devices, such as smartphones, tablets, or smart watches combined with the emergence of many other small smart objects with computational, sensing and communication capabilities coupled with the popularity of social networks and new human-technology interaction paradigms is creating unprecedented opportunities for each of us to do something useful, ranging from a single person to the whole world. Furthermore, Internet of Things, Smart-cities, distributed sensing and Fog computing are representative examples of modern ICT paradigms that aim to describe a dynamic and globally cooperative infrastructure built upon objects intelligence and self-configuring capabilities. These connected objects are finding their way into our pockets, vehicles, urban areas and infrastructure, thus becoming the very texture of our society and providing us the possibility, but also the responsibility, to shape it
An Event and Service Mesh Architecture Supporting Service Integration in Society 5.0 enabled Smart Cities
Full text linkSociety 5.0 envisions a more resilient, sustainable, and human-centered society fostered by ever-evolving cooperation and knowledge sharing among the many digital systems already shaping our daily lives. However, the current state of smart cities often consists of siloed systems, with different actors and stakeholders managing their services and assets independently. This phenomenon is evident in both technological and operational domains, posing challenges to seamless collaboration. In this context, new cloud computing models and technologies like event and service mesh promise to reduce the burden associated with the development and integration of solutions. In the attempt to pave the way for more integrated IT environments, we propose a practical architecture that combines service and event mesh technologies, enabling the seamless exploitation of service invocation and composition based on event distribution and direct service calls. Our proposal allows applications to remain transparent of the underlying technology, facilitating various optimizations on the network and management plane, necessary to meet the diverse operational requirements of complex and heterogeneous applications. We validate our proposal in a real-use case scenario implementation, discussing the tradeoffs that emerge
Game Theoretic Analysis of AoI Efficiency for Participatory and Federated Data Ecosystems
No full text- …
