arXiv subjects

Sofie Pollin

Publications and source records attributed to Sofie Pollin.

At least 109 records · Page 6Linked to original sources

arXiv2021

Drone technology: interdisciplinary systematic assessment of knowledge gaps and potential solutions

Despite being a hot research topic for a decade, drones are still not part of our everyday life. In this article, we analyze the reasons for this state of affairs and look for ways of improving the situation. We rely on the achievements of the so-called Technology Assessment (TA), an interdisciplinary research field aiming at providing knowledge for better-informed and well-reflected decisions concerning new technologies. We demonstrate that the most critical area requiring further development is safety. Since Unmanned Aerial System Traffic Management (UTM) systems promise to address this problem in a systematic manner, we also indicate relevant solutions for UTM that have to be designed by wireless experts. Moreover, we suggest project implementation guidelines for several drone applications. The guidelines take into account the public acceptance levels estimated in state of the art literature of the correspondent field.

eess.SY↗

arXiv2021

Towards Fine-Grained Indoor Localization based on Massive MIMO-OFDM System: Experiment and Analysis

Fine-grained indoor localization has attracted attention recently because of the rapidly growing demand for indoor location-based services (ILBS). Specifically, massive (large-scale) multiple-input and multiple-output (MIMO) systems have received increasing attention due to high angular resolution. This paper presents an indoor localization testbed based on a massive MIMO orthogonal frequency-division multiplexing (OFDM) system, which supports physical-layer channel measurements. Instead of exploiting channel state information (CSI) directly for localization, we focus on positioning from the perspective of multipath components (MPCs), which are extracted from the CSI through the space-alternating generalized expectation-maximization (SAGE) algorithm. On top of the available MPCs, we propose a generalized fingerprinting system based on different single-metric and hybrid-metric schemes. We evaluate the impact of the varying antenna topologies, the size of the training set, the number of antennas, and the effective signal-to-noise ratio (SNR). The experimental results show that the proposed fingerprinting method can achieve centimeter-level positioning accuracy with a relatively small training set. Specifically, the distributed uniform linear array obtains the highest accuracy with about 1.63-2.5-cm mean absolute errors resulting from the high spatial resolution.

eess.SP↗

arXiv2021

Aerial Vehicles Tracking Using Noncoherent Crowdsourced Wireless Networks

Air traffic management (ATM) of manned and unmanned aerial vehicles (AVs) relies critically on ubiquitous location tracking. While technologies exist for AVs to broadcast their location periodically and for airports to track and detect AVs, methods to verify the broadcast locations and complement the ATM coverage are urgently needed, addressing anti-spoofing and safe coexistence concerns. In this work, we propose an ATM solution by exploiting noncoherent crowdsourced wireless networks (CWNs) and correcting the inherent clock-synchronization problems present in such non-coordinated sensor networks. While CWNs can provide a great number of measurements for ubiquitous ATM, these are normally obtained from unsynchronized sensors. This article first presents an analysis of the effects of lack of clock synchronization in ATM with CWN and provides solutions based on the presence of few trustworthy sensors in a large non-coordinated network. Secondly, autoregressive-based and long short-term memory (LSTM)-based approaches are investigated to achieve the time synchronization needed for localization of the AVs. Finally, a combination of a multilateration (MLAT) method and a Kalman filter is employed to provide an anti-spoofing tracking solution for AVs. We demonstrate the performance advantages of our framework through a dataset collected by a real-world CWN. Our results show that the proposed framework achieves localization accuracy comparable to that acquired using only GPS-synchronized sensors and outperforms the localization accuracy obtained based on state-of-the-art CWN synchronization methods.

cs.NI↗

arXiv2021

Avoiding normalization uncertainties in deep learning architectures for end-to-end communication

Recently, deep learning is considered to optimize the end-to-end performance of digital communication systems. The promise of learning a digital communication scheme from data is attractive, since this makes the scheme adaptable and precisely tunable to many scenarios and channel models. In this paper, we analyse a widely used neural network architecture and show that the training of the end-to-end architecture suffers from normalization errors introduced by an average power constraint. To solve this issue, we propose a modified architecture: shifting the batch slicing after the normalization layer. This approach meets the normalization constraints better, especially in the case of small batch sizes. Finally, we experimentally demonstrate that our modified architecture leads to significantly improved performance of trained models, even for large batch sizes where normalization constraints are more easily met.

eess.SP↗

arXiv2020

Edge Computing Assisted Autonomous Flight for UAV: Synergies between Vision and Communications

Autonomous flight for UAVs relies on visual information for avoiding obstacles and ensuring a safe collision-free flight. In addition to visual clues, safe UAVs often need connectivity with the ground station. In this paper, we study the synergies between vision and communications for edge computing-enabled UAV flight. By proposing a framework of Edge Computing Assisted Autonomous Flight (ECAAF), we illustrate that vision and communications can interact with and assist each other with the aid of edge computing and offloading, and further speed up the UAV mission completion. ECAAF consists of three functionalities that are discussed in detail: edge computing for 3D map acquisition, radio map construction from the 3D map, and online trajectory planning. During ECAAF, the interactions of communication capacity, video offloading, 3D map quality, and channel state of the trajectory form a positive feedback loop. Simulation results verify that the proposed method can improve mission performance by enhancing connectivity. Finally, we conclude with some future research directions.

cs.NI↗

arXiv2020

Massive MIMO goes Sub-GHz: Implementation and Experimental Exploration for LPWANs

Low-Power Wide-Area Networks operating in the unlicensed bands are being deployed to connect a rapidly growing number of Internet-of-Things devices. While the unlicensed sub-GHz band offers favorable propagation for long-range connections, measurements show that the energy consumption of the nodes is still mostly dominated by the wireless transmission affecting their autonomy. We investigate the potential benefits of deploying massive MIMO technology to increase system reliability and at the same time support low-energy devices with good coverage at sub-GHz frequencies. The impact of different antenna configurations and propagation conditions is analyzed. Both actual average experienced array gain and channel hardening are examined. The assessment demonstrates the effect of channel hardening as well as the potential benefits of the experienced array gain. These measurements serve as a first assessment of the channel conditions of massive MIMO at sub-GHz frequencies and are, to the best of our knowledge, the first of its kind.

eess.SP↗

arXiv2020

Drone classification from RF fingerprints using deep residual nets

Detecting UAVs is becoming more crucial for various industries such as airports and nuclear power plants for improving surveillance and security measures. Exploiting radio frequency (RF) based drone control and communication enables a passive way of drone detection for a wide range of environments and even without favourable line of sight (LOS) conditions. In this paper, we evaluate RF based drone classification performance of various state-of-the-art (SoA) models on a new realistic drone RF dataset. With the help of a newly proposed residual Convolutional Neural Network (CNN) model, we show that the drone RF frequency signatures can be used for effective classification. The robustness of the classifier is evaluated in a multipath environment considering varying Doppler frequencies that may be introduced from a flying drone. We also show that the model achieves better generalization capabilities under different wireless channel and drone speed scenarios. Furthermore, the newly proposed model's classification performance is evaluated on a simultaneous multi-drone scenario. The classifier achieves close to 99 % classification accuracy for signal-to-noise ratio (SNR) 0 dB and at -10 dB SNR it obtains 5 % better classification accuracy compared to the existing framework.

eess.SP↗

arXiv2020

3D beamforming and handover analysis for UAV networks

In future drone applications fast moving unmanned aerial vehicles (UAVs) will need to be connected via a high throughput ultra reliable wireless link. MmWave communication is assumed to be a promising technology for UAV communication, as the narrow beams cause little interference to and from the ground. A challenge for such networks is the beamforming requirement, and the fact that frequent handovers are required as the cells are small. In the UAV communication research community, mobility and especially handovers are often neglected, however when considering beamforming, antenna array sizes start to matter and the effect of azimuth and elevation should be studied, especially their impact on handover rate and outage capacity. This paper aims to fill some of this knowledge gap and to shed some light on the existing problems. This work will analyse the performance of 3D beamforming and handovers for UAV networks through a case study of a realistic 5G deployment using mmWave. We will look at the performance of a UAV flying over a city utilizing a beamformed mmWave link.

eess.SP↗

arXiv2020

Spectrum Sharing Strategies for UAV-to-UAV Cellular Communications

In this article, we consider a cellular network deployment where UAV-to-UAV (U2U) transmit-receive pairs coexist with the uplink (UL) of cellular ground users (GUEs). Our analysis focuses on comparing two spectrum sharing mechanisms: i) overlay, where the available time-frequency resources are split into orthogonal portions for U2U and GUE communications, and ii) underlay, where the same resources may be accessed by both link types, resulting in mutual interference. We evaluate the coverage probability and rate of all links and their interplay to identify the best spectrum sharing mechanism. Among other things, our results demonstrate that, in scenarios with a large number of UAV pairs, adopting overlay spectrum sharing seems the most suitable approach for maintaining a minimum guaranteed rate for UAVs and a high GUE UL performance. We also find that increasing the density of U2U links degrades their rates in the overlay-where UAVs only receive interference from other UAVs-, but not significantly so in the underlay -- where the effect of GUE-generated interference is dominant.

eess.SP↗

arXiv2020

A Blind Beam Tracking Scheme for Millimeter Wave Systems

Millimeter-wave is one of the technologies powering the new generation of wireless communication systems. To compensate the high path-loss, millimeter-wave devices need to use highly directional antennas. Consequently, beam misalignment causes strong performance degradation reducing the link throughput or even provoking a complete outage. Conventional solutions, e.g. IEEE 802.11ad, propose the usage of additional training sequences to track beam misalignment. These methods however introduce significant overhead especially in dynamic scenarios. In this paper we propose a beamforming scheme that can reduce this overhead. First, we propose an algorithm to design a codebook suitable for mobile scenarios. Secondly, we propose a blind beam tracking algorithm based on particle filter, which describes the angular position of the devices with a posterior density function constructed by particles. The proposed scheme reduces by more than 80% the overhead caused by additional training sequences.

eess.SP↗

arXiv2020

Electrosense+: Crowdsourcing Radio Spectrum Decoding using IoT Receivers

Web spectrum monitoring systems based on crowdsourcing have recently gained popularity. These systems are however limited to applications of interest for governamental organizationsor telecom providers, and only provide aggregated information about spectrum statistics. Theresult is that there is a lack of interest for layman users to participate, which limits its widespreaddeployment. We present Electrosense+ which addresses this challenge and creates a general-purpose and open platform for spectrum monitoring using low-cost, embedded, and software-defined spectrum IoT sensors. Electrosense+ allows users to remotely decode specific parts ofthe radio spectrum. It builds on the centralized architecture of its predecessor, Electrosense, forcontrolling and monitoring the spectrum IoT sensors, but implements a real-time and peer-to-peercommunication system for scalable spectrum data decoding. We propose different mechanismsto incentivize the participation of users for deploying new sensors and keep them operational inthe Electrosense network. As a reward for the user, we propose an incentive accounting systembased on virtual tokens to encourage the participants to host IoT sensors. We present the newElectrosense+ system architecture and evaluate its performance at decoding various wireless sig-nals, including FM radio, AM radio, ADS-B, AIS, LTE, and ACARS.

cs.NI↗

arXiv2020

Avoiding collisions at any (low) cost: ADS-B like position broadcast for UAVs

Unmanned Aerial Vehicles (UAVs), a.k.a. drones, are increasingly used for different tasks. With more drones in the sky, the risk of accidents rises, sparkling the need for conflict management solutions. Aircraft use a system called Automatic Dependent System-Broadcast (ADS-B) to continuously broadcast their position and speed but this system is not suitable for small drones because of its cost, complexity and capacity limitations. Broadband technologies such as Wi-Fi beacons are more suited for such dense scenarios, and they also offer the benefit of wide availability and low cost. The main challenges for Wi-Fi are (a) the multi-channel nature of the technology makes transmitter and receiver coordination difficult, and (b) standard chipsets are not designed for frequent broadcast transmission and reception. In this paper, we propose and analyze a multi-channel position broadcast solution that is robust against jamming and achieves a reliable location update within 125~ms. In addition, we implement the protocol on inexpensive embedded Wi-Fi modules and analyze the hardware limitations of such devices. Our conclusions are that even on the simplest Wi-Fi chipsets, our protocol can be implemented to achieve a realistic location broadcast solution that still perfectly mimics simulation and analytical results on the lab bench and still can achieve approximately 4~message/s throughput at a distance of 900~m on flying UAVs.

eess.SY↗

arXiv2020

Weave and Conquer: A Measurement-based Analysis of Dense Antenna Deployments

Massive MIMO is bringing significant performance improvements in the context of outdoor macrocells, such as favourable propagation conditions, spatially confined communication, high antenna gains to overcome pathloss, and good angular localisation. In this paper we explore how these benefits scale to indoor scattering-rich deployments based on a dense indoor measured Massive MIMO dataset. First, we design and implement three different and relevant topologies to position our 64 antennas in the environment: \emph{Massive MIMO, RadioStripes} and \emph{RadioWeaves} topologies. Second, we measure 252004 indoor channels for a $3x3m^2$ area for each topology, using an automated user-positioning and measurement system. Using this dense dataset, we provide a unique analysis of system level properties such as pathloss, favourable propagation, spatial focusing and localisation performance. Our measurement-based analyses verify and quantify that distributing the antennas throughout the environment results in an improved propagation fairness, better favourable propagation conditions, higher spatial confinement and finally a high localisation performance. The dataset is publicly available and can serve as a reference database for benchmarking of future indoor communication systems and communication models. We outline the implementation challenges we observed, and also list diverse R\&D challenges that can benefit from using this dataset as a benchmark.

eess.SP↗

arXiv2020

Wireless communication for safe UAVs: From Long-Range Deconfliction to short-range collision avoidance

Small drones are becoming a part of our everyday life. They are used in a wide variety of commercial applications, and the number of drones in the air is steadily growing. To ensure the safe operation of drones, traffic management rules must be designed and implemented by avionics and telecommunication experts. In this article, we propose to establish a common terminology for these two communities. We first describe the traffic management architecture and services. Next, we overview several approaches for defining the inter-drone separation distances ensuring the safe operation. Moreover, we analyze which existing technologies can be useful for each of these definitions. Finally, we present measurement results indicating that our new Wi-Fi-based messaging scheme is a potentially useful tool for the drone traffic management system.

eess.SY↗

arXiv2020

MaMIMO CSI-based positioning using CNNs: Peeking inside the black box

Massive MIMO (MaMIMO) Channel State Information (CSI) based user positioning systems using Convolutional Neural Networks (CNNs) show great potential, reaching a very high accuracy without introducing any overhead in the MaMIMO communication system. In this study, we show that both these systems can position indoor users in both Line-of-Sight and in non-Line-of-Sight conditions with an accuracy of around 20 mm. However, to further develop these positioning systems, more insight in how the CNN infers the position is needed. The used CNNs are a black box and we can only guess how they position the users. Therefore, the second focus of this paper is on opening the black box using several experiments. We explore the current limitations and promises using the open dataset gathered on a real-life 64-antenna MaMIMO testbed. In this way, extra insight in the system is gathered, guiding research on MaMIMO CSI-based positioning systems using CNNs in the right direction.

eess.SP↗

arXiv2020

Fixed mmWave Multi-User MIMO: Performance Analysis and Proof-of-Concept Architecture

In this paper, we present a fixed mmWave Multi-User Multiple-Input Multiple-Output (MIMO) system for fixed wireless access with a unique architecture. A digital MIMO system is combined with an analog multi-beam antenna array which uses a high-dimension 16x16 Butler matrix to obtain 16 orthogonal beams. A system model of this architecture is presented and used to simulate its performance comparing to the performance of common-used patch antennas. Several MIMO precoding techniques are considered and compared with basic analog beamforming. To verify these results, a prototype is built and a dedicated measurement campaign is performed. The results show that the system model is a good approximation and that the use of the multi-beam antenna array is a good alternative to patch antennas for a large number of users.

eess.SP↗

arXiv2019

UAV-to-UAV Communications in Cellular Networks

We consider a cellular network deployment where UAV-to-UAV (U2U) transmit-receive pairs share the same spectrum with the uplink (UL) of cellular ground users (GUEs). For this setup, we focus on analyzing and comparing the performance of two spectrum sharing mechanisms: (i) underlay, where the same time-frequency resources may be accessed by both UAVs and GUEs, resulting in mutual interference, and (ii)overlay, where the available resources are divided into orthogonal portions for U2U and GUE communications. We evaluate the coverage probability and rate of both link types and their interplay to identify the best spectrum sharing strategy. We do so through an analytical framework that embraces realistic height-dependent channel models, antenna patterns, and practical power control mechanisms. For the underlay, we find that although the presence of U2U direct communications may worsen the uplink performance of GUEs, such effect is limited as base stations receive the power-constrained UAV signals through their antenna sidelobes. In spite of this, our results lead us to conclude that in urban scenarios with a large number of UAV pairs, adopting an overlay spectrum sharing seems the most suitable approach for maintaining a minimum guaranteed rate for UAVs and a high GUE UL performance.

cs.IT↗

arXiv2019

CSI-based Positioning in Massive MIMO systems using Convolutional Neural Networks

This paper studies the performance of a user positioning system using Channel State Information (CSI) of a Massive MIMO (MaMIMO) system. To infer the position of the user from the CSI, a Convolutional Neural Network is designed and evaluated through a novel dataset. This dataset contains indoor MaMIMO CSI measurements using three different antenna topologies, covering a 2.5 m by 2.5 m indoor area. We show that we can train a Convolutional Neural Network (CNN) model to estimate the position of a user inside this area with a mean error of less than half a wavelength. Moreover, once the model is trained on a given scenario and antenna topology, Transfer Learning is used to repurpose the acquired knowledge towards another scenario with significantly different antenna topology and configuration. Our results show that it is possible to further train the CNN using only a small amount of extra labelled samples for the new topology. This transfer learning approach is able to reach accurate results, paving the road to a practical CSI-based positioning system powered by CNNs.

eess.SP↗