Network-assisted single-hop device-to-device (D2D) communication can increase the spectral and energy efficiency of cellular networks by taking advantage proximity, reuse, hop gains. In this paper we argue that D2D technology be used to further if key radio resource management algorithms are suitably extended support network assisted multi-hop communications. Specifically propose a novel, distributed utility maximizing power control (PC) scheme is able balance while into account mode...

In the resource management of wireless networks, Federated Learning has been used to predict handovers. However, non-independent and identically distributed data degrade accuracy performance such predictions. To overcome problem, can leverage clustering algorithms build a machine learning model for each cluster. traditional algorithms, when applied handover prediction, exhibit three main limitations: risk privacy breach, fixed shape clusters, non-adaptive number clusters. these limitations,...

As the standardization of network-assisted device-to-device (D2D) communications by Third Generation Partnership Project progresses, research community has started to explore technology potential new advanced features that will largely impact performance 5G networks. For 5G, D2D is becoming an integrative term emerging technologies take advantage proximity communicating entities in licensed and unlicensed spectra. The European project Mobile Wireless Communication Enablers for 2020...

Abstract This work proposes a reliable leakage detection methodology for water distribution networks (WDNs) using machine-learning strategies. Our solution aims at detecting in WDNs efficient We analyze pressure measurements from pumps district metered areas (DMAs) Stockholm, Sweden, where we consider residential DMA of the network. proposed uses learning strategies unsupervised (K-means and cluster validation techniques), supervised (learning vector quantization algorithms). The propose...

Millimeter-wave using large-antenna arrays is a key technological component for the future cellular systems, where it expected that hybrid beamforming along with quantized phase shifters will be used due to their implementation and cost efficiency. In this paper, we investigate efficacy of full-duplex mmWave communication low-resolution shifters. We assume self-interference can sufficiently cancelled by combination propagation domain digital techniques, without any analog cancellation....

Over-the-air computation (AirComp) is a known technique in which wireless devices transmit values by analog amplitude modulation so that function of these computed over the communication channel at common receiver. The physical reason superposition properties electromagnetic waves, naturally return sums values. Consequently, applications AirComp are almost entirely restricted to systems. However, use digital communications for over-the-air computations would have several benefits, such as...

A promising new transmission mode in cellular networks is the three-node full-duplex (FD) mode, which involves a base station with FD capability and two half-duplex user transmissions on same frequency channel for uplink downlink. The can increase spectral efficiency, especially low transmit power regime, without requiring at devices. However, when large set of users are scheduled this self-interference user-to-user interference substantially hinder potential gains communications. This paper...

Although in cellular networks full duplex and dynamic time-division duplexing promise increased spectrum efficiency, their potential is so far challenged by interference. While previous studies have shown that self-interference can be suppressed to a sufficient level, we show the cross-link interference for both modes, especially from base station station, remaining challenge multi-cell networks, restricting uplink performance. Using beamforming techniques of low complexity, this mitigated,...

As data generation increasingly takes place on devices without a wired connection, Machine Learning (ML) related traffic will be ubiquitous in wireless networks.Many studies have shown that traditional protocols are highly inefficient or unsustainable to support ML, which creates the need for new communication methods.In this monograph, we give comprehensive review of state-of-the-art methods specifically designed ML services over distributed datasets.Currently, there two clear themes within...

Abstract In this work, we consider a federated learning model in wireless system with multiple base stations and inter‐cell interference. We apply differentially private scheme to transmit information from users their corresponding station during the phase. show convergence behavior of process by deriving an upper bound on its optimality gap. Furthermore, define optimization problem reduce total privacy leakage. To find locally optimal solutions problem, first propose algorithm that...

Over-the-air computation (AirComp) is a well-known technique by which several wireless devices transmit analog amplitude modulation to achieve sum of their signals at common receiver. The underlying physical principle the superposition property radio waves. Since such and in amplitude, it natural that AirComp uses modulations. Unfortunately, this impractical because most today use digital It would be highly desirable communications numerous benefits, as error correction, synchronization,...

Binary power control (BPC) is known to maximize the capacity of a two-cell interference limited system and performs near optimally for larger systems. However, when device-to-device (D2D) communication underlying cellular layer supported, an objective function that considers consumption more suitable. We find BPC remains optimal D2D communications weight overall in utility bounded. Building on this insight, we propose simple near-optimal extended scheme compare its performance with recently...

This paper investigates efficient distributed training of a Federated Learning (FL) model over wireless network devices. The communication iterations the algorithm may be substantially deteriorated or even blocked by effects devices' background traffic, packet losses, congestion, latency. We abstract communication-computation impacts as an 'iteration cost' and propose cost-aware causal FL (FedCau) to tackle this problem. iteration-termination method that trade-offs performance networking...

Three-node full-duplex is a promising new transmission mode between capable wireless node and two other nodes that use half-duplex reception respectively. Although three-node transmissions can increase the spectral efficiency without requiring capability of user devices, inter-node interference - in addition to inherent self-interference severely degrade performance. Therefore, as methods provide effective mitigation evolve, management internode becoming increasingly important. This paper...

In the Internet of Things, learning is one most prominent tasks. this paper, we consider an Things scenario where federated used with simultaneous transmission model data and wireless power. We investigate trade-off between number communication rounds round time while harvesting energy to compensate expenditure. formulate solve optimization problem by considering local iterations on devices, transmit-receive updates, harvest sufficient energy. Numerical results indicate that maximum ratio...

Researches in Power Control (PC) schemes are fundamental to provide efficient algorithms and understand their behavior Device-to-Device (D2D) scenario. The aim of the current study is investigate Open Loop (OLPC), Closed (CLPC) indicate Soft Dropping (SDPC) like an alternative PC scheme D2D communications underlaying a cellular network. We briefly review schemes, suggest new point dynamic offset compensation factor σ setting issue CLPC evaluate its performance terms spectral power...

Federated Edge Learning (FEEL) is a distributed machine learning technique where each device contributes to training global inference model by independently performing local computations with their data. More recently, FEEL has been merged over-the-air computation (OAC), the calculated over air leveraging superposition of analog signals. However, when implementing OAC, there challenge on how precode signals overcome any time misalignment at receiver. In this work, we propose novel...

In cellular networks, the three-node full-duplex transmission mode has potential to increase spectral efficiency without requiring capability of users. Consequently, in networks must deal with self-interference and user-to-user interference, which can be managed by power control user-frequency assignment techniques. This paper investigates problem maximizing sum jointly determining transmit powers a distributed fashion, assigning users frequency channels. The is formulated as mixed-integer...

Full-duplex communications have the potential to almost double spectral efficiency. To realize such a potentiality, signal separation at base station's antennas plays an essential role. This article addresses fundamentals of by proposing new smart antenna architecture that allows every be either shared or separated between uplink and downlink transmissions. The benefits are investigated assignment problem optimally assign antennas, beamforming power maximize weighted sum We propose...

As data generation increasingly takes place on devices without a wired connection, machine learning (ML) related traffic will be ubiquitous in wireless networks. Many studies have shown that traditional protocols are highly inefficient or unsustainable to support ML, which creates the need for new communication methods. In this survey, we give an exhaustive review of state-of-the-art methods specifically designed ML services over distributed datasets. Currently, there two clear themes within...