Search

Federated learning (FL) has been promoted as a popular technique for training machine learning (ML) models over edge/fog networks. Traditional implementations of FL have largely neglected the potential for inter-network cooperation, treating edge/fog devices and other infrastructure participating in ML as separate processing elements. Consequently, FL has been vulnerable to several dimensions of network heterogeneity, such as varying computation capabilities, communication resources, data qualities, and privacy demands. We advocate for cooperative federated learning (CFL), a cooperative edge/fog ML paradigm built on device-to-device (D2D) and device-to-server (D2S) interactions. Through D2D and D2S cooperation, CFL counteracts network heterogeneity in edge/fog networks through enabling a model/data/resource pooling mechanism, which will yield substantial improvements in ML model training quality and network resource consumption. We propose a set of core methodologies that form the foundation of D2D and D2S cooperation and present preliminary experiments that demonstrate their benefits. We also discuss new FL functionalities enabled by this cooperative framework such as the integration of unlabeled data and heterogeneous device privacy into ML model training. Finally, we describe some open research directions at the intersection of cooperative edge/fog and FL., This paper has been accepted for publication in IEEE Communications Magazine

International audience; Internet of Things (IoT) technologies are increasingly used in our daily life. Among these technologies, LoRa is widely used for low power IoT data collection in wide areas. LoRa typically operates in sub-GHz bands which are region-specific, and thus has limited worldwide applications. The version of LoRa developed for the 2.4 GHz ISM band has spurred recent interest in the scientific community, as it enables the use of worldwide frequency channels. However, it requires the coexistence with numerous concurrent technologies within this 2.4 GHz band, and an evolution of the MAC and networking protocols for utilizing LoRaWAN connectivity. In this paper, we make the first experimental comparison between LoRa sub-GHz and LoRa 2.4 GHz in both indoor and outdoor scenarios. We show that LoRa 2.4 GHz can have a communication range of more than 2 km outdoor, and it can achieve a packet delivery rate of at least 0.98 for a distance of up to 39 meters indoor with the presence of obstacles. We also make an extensive discussion on the researches challenges that need to be addressed in order to fully take advantage of this new technology.

Wireless networks are inherently graph-structured, which can be utilized in graph representation learning to solve complex wireless network optimization problems. In graph representation learning, feature vectors for each entity in the network are calculated such that they capture spatial and temporal dependencies in their local and global neighbourhoods. Graph neural networks (GNNs) are powerful tools to solve these complex problems because of their expressive representation and reasoning power. In this paper, the potential of graph representation learning and GNNs in wireless networks is presented. An overview of graph learning is provided which covers the fundamentals and concepts such as feature design over graphs, GNNs, and their design principles. Potential of graph representation learning in wireless networks is presented via few exemplary use cases and some initial results on the GNN-based access point selection for cell-free massive MIMO systems., Comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible

There is a widely held belief that future Radio Access Network (RAN) architectures will be characterized by increased levels of virtualization, whereby base station functionalities, traditionally residing at a single location, will be scattered across different logical entities while being interfaced via high-speed fronthaul (FH) links. For the deployment of such FH links, operators are faced with the challenge of maintaining acceptable radio access performance while at the same time keeping deployment costs low. A common practice is to exploit statistical multiplexing by allowing several cells to utilize the same FH link. As a result, in order to cope with the resulting aggregated traffic, different techniques can be used to reduce the required FH data rates. Herein, we focus on FH compression control strategies for multiple-cell/multiple-user scenarios sharing a common FH link. We propose various methods for sounding reference signal (SRS) handling and analyze different FH-aware modulation data compression and scheduling strategies. Considering a full system setup, including the radio and FH access networks, numerical evaluation is conducted using a 5G NR system-level simulator implemented in ns-3. Simulation results show that, under stringent FH capacity constraints, optimized modulation compression strategies provide significant user-perceived throughput gains over baseline strategies (between 5.2x and 6.9x). On top of them, SRS handling methods achieve additional 2% to 41% gains., paper to appear in IEEE Communications Magazine

International audience; Under the Network Function Virtualization multidomain orchestration approach, many service providers jointly manage the lifecycle of network services composed by Virtual Network Functions (VNFs). Many orchestration algorithms have been published focusing on different tasks for the lifecycle of network services. Nowadays, some general-purpose and architectural orchestration surveys have been published. However, currently, no survey classifies and evaluates distributed multidomain orchestration algorithms. In this paper, we focus on multi-domain orchestration works from an algorithm perspective by considering the cooperative and competitive approaches. Algorithms are classified and evaluated. We propose a taxonomy for works that consider different lifecycle tasks of network services. Classification and evaluation are done by identifying key requirements present in competitive and cooperative scenarios. This allows us to have a perspective on the current state of multi-domain orchestration algorithms. We also identify future research directions based on such a current perspective.

Envisioned use cases of unmanned aerial vehicles (UAVs) impose new service requirements in terms of data rate, latency, and sensing accuracy, to name a few. If such requirements are satisfactorily met, it can create novel applications and enable highly reliable and harmonized integration of UAVs in the 6G network ecosystem. Towards this, terahertz (THz) bands are perceived as a prospective technological enabler for various improved functionalities such as ultra-high throughput and enhanced sensing capabilities. This paper focuses on THz-empowered UAVs with the following capabilities: communication, sensing, localization, imaging, and control. We review the potential opportunities and use cases of THz-empowered UAVs, corresponding novel design challenges, and resulting trade-offs. Furthermore, we overview recent advances in UAV deployments regulations, THz standardization, and health aspects related to THz bands. Finally, we take UAV to UAV (U2U) communication as a case-study to provide numerical insights into the impact of various system design parameters and environment factors.

In this paper we describe the recent 3GPP Release 16 specification for positioning in 5G networks. It specifies positioning signals, measurements, procedures, and architecture to meet requirements from a plethora of regulatory, commercial and industrial use cases. 5G thereby significantly extends positioning capabilities compared to what was possible with LTE. The indicative positioning performance is evaluated in agreed representative 3GPP simulation scenarios, showing a 90 percentile accuracy of a few meters down to a few decimeters depending on scenarios and assumptions.

Free-space optical (FSO) systems are able to offer the high data-rate, secure, and cost-efficient communication links required for applications such as wireless front- and backhauling for 5G and 6G communication networks. Despite the substantial advancement of FSO systems over the past decades, the requirement of a line-of-sight connection between transmitter and receiver remains a key limiting factor for their deployment. In this paper, we discuss the potential role of intelligent reflecting surfaces (IRSs) as a solution to relax this requirement. We present an overview of existing optical IRS technologies; compare optical IRSs with radio-frequency IRSs and optical relays; and identify various open problems for future research on IRS-assisted FSO communications.

Digital contact tracing has emerged as a viable tool supplementing manual contact tracing. To date, more than 100 contact tracing applications have been published to slow down the spread of highly contagious Covid-19. Despite subtle variabilities among these applications, all of them achieve contact tracing by manipulating the following three components: a) use a personal device to identify the user while designing a secure protocol to anonymize the user's identity; b) leverage networking technologies to analyze and store the data; c) exploit rich sensing features on the user device to detect the interaction among users and thus estimate the exposure risk. This paper reviews the current digital contact tracing based on these three components. We focus on two personal devices that are intimate to the user: smartphones and wearables. We discuss the centralized and decentralized networking approaches that use to facilitate the data flow. Lastly, we investigate the sensing feature available on smartphones and wearables to detect the proximity between any two users and present experiments comparing the proximity sensing performance between these two personal devices., Accepted at the IEEE Communications Magazine