Search

Edge computing has revolutionized the world of mobile and wireless networks world thanks to its flexible, secure, and performing characteristics. Lately, we have witnessed the increasing use of it to make more performing the deployment of machine learning (ML) techniques such as federated learning (FL). FL was debuted to improve communication efficiency compared to conventional distributed machine learning (ML). The original FL assumes a central aggregation server to aggregate locally optimized parameters and might bring reliability and latency issues. In this paper, we conduct an in-depth study of strategies to replace this central server by a flying master that is dynamically selected based on the current participants and/or available resources at every FL round of optimization. Specifically, we compare different metrics to select this flying master and assess consensus algorithms to perform the selection. Our results demonstrate a significant reduction of runtime using our flying master FL framework compared to the original FL from measurements results conducted in our EdgeAI testbed and over real 5G networks using an operational edge testbed., 6 pages, accepted to GLOBECOM 2021

Machine Intelligence (MI) technologies have revolutionized the design and applications of computational intelligence systems, by introducing remarkable scientific and technological enhancements across domains. MI can improve Internet of Things (IoT) in several ways, such as optimizing the management of large volumes of data or improving automation and transmission in large-scale IoT deployments. When considering MI in the IoT context, MI services deployment must account for the latency demands and network bandwidth requirements. To this extent, moving the intelligence towards the IoT end-device aims to address such requirements and introduces the notion of Distributed MI (D-MI) also in the IoT context. However, current D-MI deployments are limited by the lack of MI interoperability. Currently, the intelligence is tightly bound to the application that exploits it, limiting the provisioning of that specific intelligence service to additional applications. The objective of this article is to propose a novel approach to cope with such constraints. It focuses on decoupling the intelligence from the application by revising the traditional device's stack and introducing an intelligence layer that provides services to the overlying application layer. This paradigm aims to provide final users with more control and accessibility of intelligence services by boosting providers' incentives to develop solutions that could theoretically reach any device. Based on the definition of this emerging paradigm, we explore several aspects related to the intelligence distribution and its impact in the whole MI ecosystem.

The conventional federated learning (FedL) architecture distributes machine learning (ML) across worker devices by having them train local models that are periodically aggregated by a server. FedL ignores two important characteristics of contemporary wireless networks, however: (i) the network may contain heterogeneous communication/computation resources, while (ii) there may be significant overlaps in devices' local data distributions. In this work, we develop a novel optimization methodology that jointly accounts for these factors via intelligent device sampling complemented by device-to-device (D2D) offloading. Our optimization aims to select the best combination of sampled nodes and data offloading configuration to maximize FedL training accuracy subject to realistic constraints on the network topology and device capabilities. Theoretical analysis of the D2D offloading subproblem leads to new FedL convergence bounds and an efficient sequential convex optimizer. Using this result, we develop a sampling methodology based on graph convolutional networks (GCNs) which learns the relationship between network attributes, sampled nodes, and resulting offloading that maximizes FedL accuracy. Through evaluation on real-world datasets and network measurements from our IoT testbed, we find that our methodology while sampling less than 5% of all devices outperforms conventional FedL substantially both in terms of trained model accuracy and required resource utilization., This paper is accepted for publication in the proceedings of 2021 IEEE International Conference on Computer Communications (INFOCOM)

Internet of Things (IoT) is a rapidly growing technological domain and the effort that many Standards Developing Organizations (SDOs) and alliances are dedicating to it is constantly increasing. The Internet Engineering Task Force (IETF) is certainly one of the most active SDOs operating in this broad domain. With the aim of building a comprehensive, interoperable and streamlined IoT stack, several IETF working groups are developing protocols, in multiple technological domains, which are already nowadays very relevant to the IoT. This article gives a concise but comprehensive survey of the IETF efforts in the IoT. We discuss mainstream standardization and research activities, as well as other IETF-connected supporting initiatives provided still in the context of IoT., The paper has been accepted for publication to IEEE Communications Standards Magazine

Lightweight virtualization (LV) technologies have refashioned the world of software development by introducing flexibility and new ways of managing and distributing software. Edge computing complements today's powerful centralized data centers with a large number of distributed nodes that provide virtualization close to the data source and end users. This emerging paradigm offers ubiquitous processing capabilities on a wide range of heterogeneous hardware characterized by different processing power and energy availability. The scope of this article is to present an in-depth analysis on the requirements of edge computing from the perspective of three selected use cases that are particularly interesting for harnessing the power of the Internet of Things. We discuss and compare the applicability of two LV technologies, containers and unikernels, as platforms for enabling the scalability, security, and manageability required by such pervasive applications that soon may be part of our everyday lives. To inspire further research, we identify open problems and highlight future directions to serve as a road map for both industry and academia.

Virtualization enabled by container-based technologies is a recently emerging concept in the integration of Internet of Things (IoT) and cloud computing. Due of their lightweight nature, container-based virtualization tools improve manageability of cloud-based IoT solutions by making it possible to update application software on the fly. Although different studies have demonstrated the feasibility of efficiently running container-based virtualization on low-power IoT nodes, the implication of doing so on battery-powered nodes has been overlooked. In this paper, we investigate how much energy overhead is generated by Docker-based virtualization on battery powered camera sensor nodes. In our scenario, camera nodes are most of the time in “power off”state to save energy. They are switched on for streaming video only when activity is detected by motion sensor nodes. By means of empirical measurement and subsequent analysis, we found that starting and closing of containers in the Docker platform adds-up roughly 13 percent power consumption overhead during the boot-up and shutdown of the camera nodes. Furthermore, the fixed overhead occurring from boot-up and shutdown procedures become negligible with longer video stream sessions.

The constant and fast increase in the number of heterogeneous Internet of Things (IoT) devices that populate everyday life environments brings new challenges to the full exploitation of the computation, memory, sensing, and actuation resources associated to them. In this context, device virtualization solutions and platforms may definitely play a key role in enabling the desired tradeoff between flexibility and performance. This paper focuses on lightweight virtualization technologies for IoT devices, suitably thought to effectively deploy new integrated applications and to create a novel distributed and virtualized ecosystem. Two different frameworks for container-based IoT service provisioning are compared, the one based on a direct interaction between two cooperating devices and the other based on the presence of a manager supervising the operations between cooperating devices forming a cluster. In the latter case, accounting for the growing impetus to move intelligence toward the edge of the network, management features are implemented at the network access point to provide short latency responses. We also introduce the outcomes of a thorough performance evaluation campaign conducted via a real IoT testbed. The measurements, performed by accounting for the constraints of typical IoT nodes, shed light on the actual feasibility of container-based IoT frameworks.