2nd MC and 1st Technical Meeting - Bologna, Italy, February 8-11

TD Number
TD Author
TD Title
TD Abstract
TD(22)01002Esteban Egea-Lopez, Jose Maria Molina-Garcia-Pard1, Martine Lienard, Pierre Degauque, Leandro Juan-LlacerOpal: An open source ray-tracing propagation simulator for electromagnetic characterizationAccurate characterization and simulation of electromagnetic propagation can be obtained by ray-tracing methods, which are based on a high frequency approximation to the Maxwell equations and describe the propagating field as a set of propagating rays, reflecting, diffracting and scattering over environment elements. However, this approach has been usually too computationally costly to be used in large and dynamic scenarios, but this situation is changing thanks the increasing availability of efficient ray-tracing libraries for graphical processing units. In this paper we present Opal, an electromagnetic propagation simulation tool implemented with ray-tracing on graphical processing units, which is part of the Veneris framework. Opal can be used as a stand-alone ray-tracing simulator, but its main strength lies in its integration with the game engine, which allows to generate customized 3D environments quickly and intuitively. We describe its most relevant features and provide implementation details, highlighting the different simulation types it supports and its extension possibilites. We provide application examples and validate the simulation on demanding scenarios, such as tunnels, where we compare the results with theoretical solutions and further discuss the tradeoffs between the simulation types and its performance.WG1
TD(22)01003Luoyan Zhu, Yinsheng Liu, Danping H, Ke Guan, Bo Ai, Zhangdui Zhong,Xi LiaoAn efficient target detection algorithm via Karhunen-Loève Transform for FMCW radar applicationsThis paper investigates an advanced effective signal processing technique to suppress noise, addressing a modern high-performance detection in the field of radar sensing. To achieve a higher accuracy, the frequency modulated continuous wave (FMCW) radar is taken as a case study to derive the algorithm based on Karhunen – Loeve transform (KLT) before detection. KLT defines a linear projection of the signal statistics on the eigenfunctions domain, which makes the input-dependent signals orthogonal to each other under new eigen-basis and eigenvalues. The highest energy along slow time dimension of each range bin are concentrated in the transformed domain corresponding to the largest N eigenvalues. The performance of the algorithm is evaluated by different eigenvalue selection strategies. Numerical experiments are employed to obtain the relationship between signal-to-noise ratio (SNR) and different eigenvalue selection strategies. Pertaining to the detection performance, constant false alarm ratio (CFAR) detector is applied to demonstrate the detection ability as a result of the processor by use of probability of detectionWG2
TD(22)01004Dong Yan, Ke Guan, Danping He, Hao Qiu, Junhyeong Kim, Heesang Chung, Bo Ai, and Zhangdui ZhongChannel Characterization for Millimeter-Wave Vehicular Communications with the Consideration of Beam Switching TechnologyVehicle-to-Vehicle (V2V) communication with high data rates and low latency is crucial for realizing autonomous driving. In particular, the shadowing effect of large vehicles (such as delivery vans) affects communication quality between vehicles seriously. Therefore, vehicles need to be equipped with various sensors to perceive the complex environment. Recently, millimeter-wave (mmWave) technology is utilized to meet this high data rate requirement. This paper uses a measurement-validated ray-tracing (RT) simulator to explore the mmWave V2V link channel characteristics in a typical urban scenario. First, the vehicular communication measurement campaigns are conducted in the target scenario. The measurement results can be used to calibrate the RT simulator. Then, through extensive simulations, the significance of beam switching technology is verified. Moreover, the impacts of different distances between communication vehicles and different traffic flows on the V2V link are discussed as well. Finally, channel parameters such as the Rician K-factor are extracted. The conclusions obtained can help researchers understand the significance of beam switching and the influence of the traffic flow on V2V communication, which can contribute to the link-level design of vehicular communication.WG1
TD(22)01005Rooderson Andrade, Rui R. Paulo, Salomão Francisco, Emanuel Bordalo Teixeira and Fernando J. VelezCharacterization of Indoor Small Cells PropagationThe characterization of the wireless medium in indoor small cell networks is essential to obtain appropriate modeling of the propagation environment. Software defined radios like Universal Software Radio Peripherals (USRPs) and simple dipole antennas can emulate LTE-Advanced networks and may be used as base stations (BSs) and user equipments (UEs) in field tests. In this work, we verify WINNER II propagation modeling for the indoor femtocell environment by considering different classrooms of 7.32 x 7.32 square meters near a common corridor in our Department while measuring the power received in UEs (radiated by the small eNodeB of the own cell, in room 1, and from the interferer cells, in rooms 2 or 3). These measurements consisted of tens of repetitions in each of the 49 points of the room and have been carried out either by using the Software Radio Systems LTE that emulates the LTE-Advanced network eNodeBs and UEs themselves, or by measuring the received power in the UES with a Rohde & Schwarz FSH8 spectrum analyzer. Analytical results and measurements are for room 1 where, by varying the UEs positions, the highest values of the received power have occurred close to the central BS. Nevertheless, it was verified that the received power does not decrease suddenly because of the effect of the radiation pattern of the BS and UE antennas for large angles of apertures as well as due to the non-omnidirectional horizontal antenna pattern. In addition, it was demonstrated that there is an effect of “wall loss” proven by the fact that path loss increases between room 2 and room 1 (or between room 3 and 2). If we consider an attenuation for each wall of circa 7-9 dB the behavior of the WINNER II model at 2.625 GHz for the interference coming across different walls is verified.WG1,WG3
TD(22)01007Diego Dupleich, Sebastian Semper, Mohanad Dawood Al-Dabbagh, Alexander Ebert, Thomas Kleine-Ostmann, and Reiner ThomäVerification of THz Channel Sounder and Delay Estimation with Over-The-Air Multipath ArtifactIn the present paper we introduce the empirical results of measurements with an over-the-air based propagation artifact for verification and validation of sub-THz and THz channel sounders and parameter estimation algorithms. This experiment produces a fixed number of multipath components with traceable propagation properties in the different domains that can be used to test resolution and performance. Because of the inherent characteristics of the measurement hardware, we have introduced an adaptation on a parametric high resolution estimation algorithm to account the imperfections of the channel sounder. The results have shown to account for a relative good performance of the sounder and the tested parametric and non-parametric estimation algorithms.WG1
TD(22)01008Ladislav Polak, Stanislav Rozum, Martin Slanina, Tomas Bravenec, Tomas Fryza, Aggelos PikrakisReceived Signal Strength Fingerprinting-Based Indoor Location Estimation Employing Machine LearningThe fingerprinting technique is a popular approach to reveal location of persons, instruments or devices in an indoor environment. Typically based on signal strength measurement, a power level map is created first in the learning phase to align with measured values in the inference. Second, the location is determined by taking the point for which the recorded received power level is closest to the power level actually measured. The biggest limit of this technique is the reliability of power measurements, which may lack accuracy in many wireless systems. To this end, this work extends the power level measurement by using multiple anchors and multiple radio channels and, consequently, considers different approaches to aligning the actual measurements with the recorded values. The dataset is available online. This article focuses on the very popular radio technology Bluetooth Low Energy to explore the possible improvement of the system accuracy through different machine learning approaches. It shows how the accuracy–complexity trade-off influences the possible candidate algorithms on an example of three-channel Bluetooth received signal strength based fingerprinting in a one dimensional environment with four static anchors and in a two dimensional environment with the same set of anchors. We provide a literature survey to identify the machine learning algorithms applied in the literature to show that the studies available can not be compared directly. Then, we implement and analyze the performance of four most popular supervised learning techniques, namely k Nearest Neighbors, Support Vector Machines, Random Forest, and Artificial Neural Network. In our scenario, the most promising machine learning technique being the Random Forest with classification accuracy over 99%WG2,HA1,HA3
TD(22)01009YI CHU, HAMED AHMADI, DAVID GRACE, DAVID BURNSDeep Learning Assisted Fixed Wireless Access Network Coverage PlanningWireless network coverage planning is crucial for mobile network operators and fixed wireless
network providers to estimate the performance of their networks and plan future antenna mast deployments.
To generate accurate coverage maps for target buildings, traditional wireless coverage planning tools
either require manual input of Customer-Premises Equipment (CPE) antenna locations or need to compute
received signal strength from nearby Access Points (APs) to all geolocations in the area of interest which
consumes computational resource unnecessarily. In this paper we propose a Deep Learning (DL) based
universal enhancement to wireless coverage planning tools which automatically extracts potential CPE
antenna locations from aerial images of the target buildings. We evaluate the performance of the pixel level
object detection provided by Mask Region-based Convolutional Neural Network (Mask R-CNN) trained
on an image dataset with suburban and rural residential properties across North Yorkshire, UK. We also
demonstrate a complete task flow to generate informative building coverage reports while combining the
DL based building detection with the WISDM industrial wireless coverage planning system
TD(22)01010Mohammad Furqan Ali and Dushantha Nalin K. JayakodyOcean-Ocean UVLC Communication SystemRecently, Underwater Visible Light Communication (UVLC) has become a potential wireless carrier candidate in the acrimonious mingled ocean straits. The North and Baltic ocean’s combined strait is a harsh and strongly turbid aqueous zone that contributes to signal to fade at a large scale. Due to this, we are proposing a UVLC system within the Baltic-North ocean mingled water under strong turbulence channel conditions. This study uses the Gamma-Gamma distribution to model UVLC link under the OOK modulation scheme. Subsequently, the reason of unavailability the latest North-Baltic oceanographic data within this bayou, we investigate the BER and outage probability performance of the proposed system within the mingled strait for the whole year during the 1996s. It is noteworthy that the analytical work has been considered of the following distinct physio-chemical properties and the data provided for each ocean. Additionally, the simulation results are verified the analytical work of the proposed system model.WG1,WG2
TD(22)01011Muhammed Sohaib J. Solaija,Hanadi Salman, Abuu B. Kihero, Mehmet İzzet Sağlam, Hüseyin ArslanGeneralized Coordinated Multipoint Framework for 5G and BeyondThe characteristic feature of 5G and beyond networks is the diversity of services, which is required to support different user needs. However, the requirements for these services are often competing in nature, which impresses the necessity of a coordinated and flexible network architecture. Although coordinated multipoint (CoMP) systems were primarily proposed to improve the cell edge performance in 4G, their collaborative nature can be leveraged to support the diverse requirements and enabling technologies of 5G and beyond networks. To this end, we propose the generalization of CoMP to a proactive and efficient resource management framework capable of supporting different user requirements such as reliability, latency, throughput, and security while considering network constraints. This article elaborates on the multiple aspects, inputs, and outputs of the generalized CoMP (GCoMP) framework. Apart from user requirements, the GCoMP decision mechanism also considers the CoMP scenario and network architecture to decide upon outputs such as CoMP scheme or appropriate coordinating clusters. To enable easier understanding of the concept, a case study illustrating the effect of different combinations of GCoMP framework’s outputs on varying user requirements is presented.WG1,WG2
TD(22)01012Saied El-Faitori and Sana SalousReflection and penetration loss measurements of buildings materialThis technical document presents, the results of wideband measurements of several common building material that were conducted at 28, and 39 GHz in the anechoic chamber at Durham University in co-polar and cross-polar alignments to investigate the reflection and penetration loss at different incident angles using a wideband channel sounder developed at Durham University. The results show that, the lowest value of penetration loss was when both antennas of the transmitter and receiver are perpendicular on the material, and it was more than 30 dB (co-polar) and 20 dB (cross-polar) at angles larger than 10 degrees on either side from the zero angle for most of the tested materials, where narrow-beam lens antennas were used (4.5 degrees).WG1
TD(22)01013Olga Błaszkiewicz, Robert Burczyk, Agnieszka Czapiewska, Małgorzata Gajewska, Sławomir Gajewski, Jarosław SadowskiEstimation of transmission quality in cellular systems based on indicators measured by the radio moduleThe article presents a measurement stand for testing transmission quality in cellular systems such as LTE, NB-IoT, UMTS and GSM. The results of BLER and throughput measurements for various propagation conditions as well as results of quality indicators appropriate for mentioned technologies measured by commercial radio module are presented. Those measurements were utilized to analyze the possibility of estimating the transmission quality in LTE, NB-IoT, UMTS and GSM only based on the results of measurements carried out by a commercial radio module.WG3
TD(22)01014De Beelde Brecht, Tanghe Emmeric, Plets David, and Joseph WoutVegetation Loss at D-Band FrequenciesWith the potential of fixed wireless access networks as an alternative to optical fiber, it is necessary to determine vegetation loss at mmWave and sub-THz frequencies for network design, and to perform accurate link budget calculations.
In this paper, we present vegetation loss measurement results at D-band frequencies for different vegetation types.
The average loss per meter vegetation width is derived by subtracting free space path loss from measured path loss, and averaging over different vegetation widths.
The resulting vegetation loss ranges from 0.1 dB/m for light foliage, up to 10 dB/m for dense vegetation.
TD(22)01015Charles Wiame; Luc Vandendorpe, Claude OestgesJoint rate and power density analysis in user-centric cell-free massive MIMO networksThe objective of this treatise is to analyze the statistics of the rate and of the incident power density (IPD) in user-centric cell-free networks (UCCFNs). In addition to the coverage probability, this study proposes a number of new performance metrics derived using stochastic geometry (SG). On the one hand, the mean, the variance and the marginal distribution of the IPD are derived. On the other hand, lower bounds on the joint distributions of the rate and IPD are provided for two scenarios: when it is relevant to obtain IPD values above a threshold (for energy harvesting purposes), and when these values should instead remain below the threshold (for public health reasons). In addition to deriving these metrics, this work incorporates features related to UCCFNs which are new in SG models. First, the power allocation is non-uniformly performed. This allocation is based on the collective channel gains of the cluster of radio heads serving the user equipment (UE), and on an equity criterion between all UEs. Second, the analytical model takes into account potential overlaps between the cooperative clusters. Last, the mathematical expressions also capture the cluster size distribution, as well as the UE statistics. The results derived from the SG framework are compared to Monte Carlo simulations.WG1,WG2,WG3
TD(22)01016Simona Valbonesi and Paolo GraziosoAnalysis of Covid-19 restrictions impact on EMF levels in residential areasThe sanitary emergency that ensued after the outbreak of Covid-19 pandemic in Italy led to massive closures of non-essential manufacturing and commercial activities in March and April, 2020 with an outstanding reduction in citizens mobility; analogous, though less severe, restrictions were put into effect also in the same period of the year 2021. The need to perform nearly every activity from home brought to a sharp increment in the use of digital communication tools, such as platforms for web conferences and distance learning. Many research groups carried out analyses aimed at verifying the impact of the closure of manufacturing activities and of reduced citizens mobility on the electrical field levels detected on air.
The work presented here is aimed at verifying how the closure periods, characterised by a massive use of online services and tools for work, gaming, distance learning, social relationships, impacted on the measured electrical field levels in residential areas. This analysis is carried out by comparing measurements taken in the closure periods with the levels detected when people movement and work activities didn’t suffer any limitation. We carried out two different studies, the first one comparing the overall cumulative distribution functions of measured field levels in years 2019, 2020, and 2021, and the second one consisting in a more specific analysis focused on the total closure periods and on the areas where the most severe restrictions were enforced (dubbed “red zones”).
TD(22)01017Carlos S. Alvarez-Merino, Hao Qiang Luo-Chen, Joel Llanes Michel, Emil J. Khatib, Raquel BarcoWiFi and UWB characterization for localisation in construction sitesHigh-precision localisation is becoming a necessity in the future location-based services that will come up in this decade. However, the construction sector remains particularly obsolete in the technology services. In this work we study the accuracy and penetration capacity of two technologies that are expected to deal with the future high-precision localisation services such as Ultra Wide Band (UWB) and WiFi Fine Time Measurement (FTM). Moreover, the fusion of technologies is presented at the Weighted Least-Square algorithm to see the performance of fusing different ranging technologies in construction sites. The Maximum Likelihood Estimator (MLE) evaluates the ranging information quality and introduces a weighting matrix for the fusion algorithm. An experimental setup is used to analyse the presented technologies and the fusion algorithmWG2
TD(22)01018D.P. Gaillot, P. Laly, E.P. Simon, N. Dahmouni, G. Delbarre, M. Liénard, J. Molins-Benlliure, M. Cabedo-Fabrés, E. Antonino-Daviu, M. Ferrando-Bataller, J-M. Garcia Molina-PardoExperimental characterization of propagation in vehicular scenarios using the real-time Massive MIMO channel sounder MaMIMOSAThis TD presents the results of a V2I channel sounding campaign with the real-time massive MIMO radio channel sounder named MaMIMOSA. This equipment was jointly developed by the University of Lille (FR) and the University of Ghent (BE) for 5G V2X applications. The system is equipped with a massive 64 antenna array for Tx, while up to 16 individual antennas can be deployed for Rx. MaMIMOSA’s hardware and software capabilities allow the probing parameters to be freely adapted to the scenario under investigation, demonstrating its versatility and flexibility. Radio channels were measured at 5.89 GHz with a bandwidth of 80 MHz on the University of Lille campus, with an average speed of 30 km/h.  A Doppler analysis illustrates the measurement results.WG1,VT2
TD(22)01019R.F. RuddStatistical modelling of short-range interference paths

There is a frequent requirement in spectrum sharing studies (e.g. in relation to frequency assignment for 5G systems) to assess the basic transmission loss on short (typically less than 1 km) outdoor paths in a variety of environments.
In most sharing studies, it is required that the results of the modelling are generic (i.e. no site specific inputs to the model are necessary) and are presented in statistical terms (typically path loss not exceeded for a given probability).
A useful model must also be suitable for computer implementation, often within a Monte Carlo framework. Many existing models are deterministic in the sense that they apply to specific geometries or environments which makes it impossible to apply such models in the general case where statistics covering all path geometries are required.
This paper proposes a new, unambiguous, method for the prediction of the statistics of basic transmission loss in such cases.

TD(22)01020Manuel M. Ferreira, Filipe D. Cardoso, Sławomir J. Ambroziak, Kenan Turbic, Luís M. CorreiaUser Mobility’s Influence on System Loss  in Off-body BAN ScenariosIn this paper, a measurement campaign for off-body communication in an indoor environment was investigated for a set of on-body antennas. The channel impulse response was measured with the user approaching and departing from an off-body antenna using two user dynamics: the user standing at fixed positions and the user walking. The processing of the measurement data was used to evaluate system loss statistics. Different antenna configurations were classified in terms of mobility and visibility depending on the on-body antenna placements. A dependence on distance was found for the antennas with the lowest mobility (chest and head), while no significant dependence was found for the antennas with highest mobility (arm and leg). For the standard deviation of the system loss, higher values were found in the walking scenarios (above 2.6 dB) than in the standing scenarios (below 0.6 dB). The values of the standard deviation also show a clear dependence on the mobility of the scenario: 1.6 dB for walking scenarios and 0.5 dB for standing scenarios.WG1
TD(22)01021Stefan Zelenbaba, Benjamin Rainer, Markus Hofer, Thomas ZemenWireless Digital Twin for Assessing the Reliability of Vehicular LinksDigital twins are becoming a key technology in evaluating hardware and system performance in wireless communications. The explosion of new use cases in time-sensitive, safety critical scenarios and the stringent requirements for new communication standards are creating a demand for repeatable and cost-effective verification methodologies. In this paper we use a digital twin for reliable wireless communications to obtain time-variant packet error rates which allow us to analyze the benefit of using a relay link for vehicular communications in a safety-critical overtaking scenario. The digital twin utilizes a real-time geometry-based stochastic channel model to simulate the frequency responses, and a hardware-in-the-loop setup to obtain packet error rates. We then use the obtained error rates to estimate the 90% tail probability of the link level latency and show that using a relay link to complement an obstructed link decreases the average latency from 16.4ms to 7.9ms and reduces the average error rate from 1.2 x 1e-2 to 1.4 x 1e-6.WG1
TD(22)01022Simone Del Prete, Franco Fuschini, Marina Barbiroli, Marco Zoli, André Noll BarretoA Study on Physical Layer Security Through Ray Tracing SimulationsDue to its broadcast nature, the wireless channel cannot be really private, i.e. restricted to the intended users only. Rather, it can always be available also to some unauthorized outsider, and, thus, security in wireless communications is crucial. Protection of communications from possible eavesdroppers is usually achieved through standard cryptography schemes, which nonetheless may not always represent the best solution due to their complexity and power consumption. Physical Layer Security, in particular Secret Key Generation, leverages the fading/dispersive properties of the wireless channel to distill encryption keys with limited computational effort. In this study, ray tracing propagation modeling is proposed as a tool to investigate the performance of the physical layer approach to communications security, thanks to its capabilities to model the channel statistics and the more flexibility compared to a measurement campaign.WG1
TD(22)01023Adrian Agustin, Adriano Pastore, Monica NavarroRandom Access Networks with Spatial ReuseProviding ultra-reliable, low-latency and massive access is a technical challenge that demands a redesign of current Media Access Control (MAC) layer in wireless cellular networks. This work focuses on studying the conventional slotted ALOHA protocol and ways to improve its efficiency. In particular, we concentrate on the case where multiple neighboring cells use the same resources, so that the system operates under inter-cell interference. While in conventional slotted ALOHA, terminals transmit with a fixed probability, in our scenario we propose instead to exploit channel-state information at terminals so as to define (minimum) signal-to-noise and (maximum) interference conditions, under which a terminal is allowed to transmit. We conduct a theoretical analysis for a simple scenario with Rayleigh fading where the system throughput can scale linearly with the number of coexisting cells and in logarithmic scale with the number of terminals per cell.WG2,VT4
TD(22)01024Krzysztof CICHOŃ, Łukasz KADAJ, Damian PiotrowskiSub-6 GHz Electromagnetic Field Prediction for Urban ScenarioIn the work the prediction of EMF voltage (Electromagnetic Field Voltage) in urban scenario is considered. To this aim, the narrowband EMF voltage measurements were conducted in five various urban locations. Then at 100 measurement points EMF values for 35 subbands were acquired. With such measurements a database is formed. Finally additional attributes were defined, such as distance, azimuth or elevation angle, cellular network operator. After data preparation, the prediction of EMF value was performed.WG1,VT1,HA1
TD(22)01026Kun Chen Hu, Ana Garcia ArmadaOn the feasibility of a structured configuration search for Reconfigurable Intelligent Surface-assisted communicationsReconfigurable Intelligent Surfaces (RIS) are being lately envisioned as an energy efficient solution capable of enhancing the signal coverage in cases where obstacles block the direct communication from Base Stations (BSs), especially at high frequency bands due to attenuation loss increase. In the current literature, wireless communications aided by necessitate the estimation of accurate Channel State Information (CSI) and the optimization of the RIS phase configuration. However, this requirement results in an increased overhead, complexity and delay. Aiming at a low-complexity and low-overhead solution for CSI estimation and RIS optimization, we get inspiration from the procedure of beam-searching that is currently performed in 5G NR. In this paper, we explore the feasibility of a structured search over a reduced number of pre-defined phase configurations of the RIS. We show by numerical evaluation the performance of the prosed scheme with respect to other solutions available in the literature.WG2
TD(22)01028Allan Wainaina Mbugua, Yun Chen, Wei FanRadio Channel Emulation for Virtual Drive Testing with Site-Specific ChannelsVirtual drive testing (VDT) with measured or ray tracing (RT) simulated channels requires the simplification of the channel to match the hardware specification of radio channel emulators. That is, reduction of the number of multipath components (MPC)s and adjustment of the arbitrary delays of the MPCs to the sampling grid of the radio channel emulator. In this paper, a fractional delay (FD) filter approximated by a general least square (GLS) finite impulse response (FIR) filter is used for band-limited interpolation to align the delay of the MPCs to the sampling grid of the radio channel emulator. A dominant power selection strategy is then employed to reduce the taps to match a given hardware specification. The GLS FIR method is shown to have superior performance in preservation of the channel frequency response (CFR) compared to rounding delays to the nearest integer multiple of the sampling time.WG1
TD(22)01029Konstantinos Katzis, Lazar Berbakov, Gordana Gardašević, Olivera Šveljo, and Dragana BajićBreaking Barriers in Emerging Biomedical ApplicationsThe recent global pandemic COVID-19 revealed that the existing health system in modern society can hardly cope with the increasing number of patients. Part of the burden can be mitigated by installing smart health infrastructure in the current system to allow patients remote monitoring and personalized treatment. Technological advances in communications and sensor devices have enabled the development of new, portable, and energy-efficient biomedical sensors, as well as innovative healthcare applications. However, such applications require reliable, resilient, and secure networks. This paper aims to identify communication requirements for the mass application of such smart health sensors by providing an overview of basic Internet of Things (IoT) technologies. Moreover, it emphasizes the importance of information theory in understanding the boundaries and barriers in this emerging field. With this motivation, the paper points out how data compression and entropy used in security algorithms can pave the way for the mass use of such IoT health devices. Future medical practices and paradigms are also discussed.VT1
TD(22)01030Zohreh Ebadi, Heidi Steendam, Francois Horlin and Francois QuitinAn over-the-air CFO-assisted synchronization algorithm for TDOA-based localization and velocity estimationThis paper investigates the problem of time difference of arrival (TDOA)-based localization in a wireless network with unsynchronized anchor nodes. We propose a new synchronization method called the carrier frequency offset (CFO)-assisted synchronization method. We use the observation that the clock offset that affects TDOA estimation is also observable in the CFO of the messages received by the anchors since the same local oscillator is used to generate the clock and carrier frequencies. By combining TOA and CFO information, our algorithm allows the anchor nodes to implicitly synchronize their clocks when performing TDOA estimation. The method is implemented and validated experimentally on an ultra-wideband (UWB) localization system. This CFO-assisted TDOA estimation algorithm can also be used to estimate the velocity of a moving target node, by adopting some small changes to include the doppler shift in the estimated CFO. Therefore, we propose a velocity estimation method called the three-step method. The proposed three-step method first synchronizes the anchors using CFO-assisted synchronization. Then, it estimates the location of the target node using the estimated TDOAs. Finally, it estimates the velocity of the target node. The significance of this work lies in the fact that we need only a single message from the target node as well as the master node to successfully synchronize the anchors and find the location and velocity of a moving target.WG2
TD(22)01031Chiara Buratti, Giampaolo Cuozzo, Roberto VerdoneOCDMA: A MAC Protocol for Industrial Intra-Machine TeraHertz Networks

We consider an industrial machine, where wireless sensor nodes (denoted as tags or nodes)  support control applications.
This scenario poses very challenging communication requirements: hundreds of tags per cubic meter can provide an overall offered throughput of tens of Gbit/s; at the same time, control applications require a latency of less than 0.1 ms.
To fulfill them, we consider using the TeraHertz (THz) frequency band, and we propose an Orthogonal Chirp Division Multiple Access scheme.
With THz communications, even at short distances, propagation delays can be of the same order of magnitude as the packet transmission time. This requires proper consideration of such delays in the protocol design and performance evaluation. This paper mathematically derives network throughput and latency of the proposed protocol, comparing it to benchmarks; two scenarios are considered, where tags are in fixed positions or move.

TD(22)01032Silvia Mignardi, Danila Ferretti, Riccardo Marini, Francesca Conserva, Stefania Bartoletti, Roberto Verdone, Chiara BurattiOptimizing Beam Selection and Resource Allocation in UAV-Aided Vehicular NetworksFuture mobile radio networks require a degree of flexibility that technologies like Unmanned Aerial Vehicles (UAVs) carrying Base Stations (BSs) can provide.
It is expected that the lower space above cities will be populated by many different types of UAVs, such as taxis and smaller drones used for logistics or patrolling, which can be equipped with BS to serve users on the ground while flying for a different mission. This paper investigates an urban scenario with terrestrial macro BSs (MBSs) deployed, where multiple UAVs fly on a predefined path.
Vehicles in the area are moving while relying on network services, and MBSs alone might not serve them adequately. We propose a network model where UAVs can be activated to work as BSs if needed and help the MBSs to fulfill the vehicular demand. To this purpose, a radio resource management optimization algorithm is designed to jointly consider: i) interference between MBSs and UAVs; ii) a limited backhaul capacity; iii) beamforming at UAVs, where a limited number of beams can be activated at the same time.
Numerical results show the notable improvement of the network performance when the flying BSs are activated and report the impact of beamforming and backhaul capacity.
TD(22)01033Concepcion Garcia-Pardo, Alejandro Fornes-Leal, Matteo Frasson, Vicente Pons-Beltrán, Narcis CardonaEffect of Breathing on UWB Propagation Characteristics for Ingestible and Implantable DevicesWireless in-body devices are those in which a medical sensor is introduced –implanted or ingested– inside the human body and communicates with a remote node. Some in-body applications demand high data rates are necessary, so Ultra-Wideband (UWB) spectrum has been proposed as a good candidate because of its large bandwidth available. Besides, breathing can lead to internal movement of the torso and consequently, of devices installed in this area. Thus, the radio channel performance can be affected by such movement leading to a malfunction of the radio interface of the medical device. This work aims at analyzing the effect of breathing on the propagation channel by means of in vivo measurements in living animal models. Continuous wave (CW) measurements have been carried out for five different frequencies in the lower part of the UWB band, and the effects of breathing on the relative received power (module and phase), are analyzed and discussed.WG1,VT1
TD(22)01034Sofia Patricio, Luis M. Correia, Monica GomesInfluence of Active Antennas on EMF Restrictions in  5G Base Stations DeploymentIn this paper, a model to analyze the influence of active antennas on electromagnetic field restrictions in 5G base stations deployment is presented. The model allows for the computation of electromagnetic exposure in the vicinity of base station antennas with active 5G in order to compute the respective exclusion zone that guarantees the safety of the population, taking realistic maximum power levels. Representative scenarios with co-location of antennas are analysed. For an urban scenario, the exclusion zone is reduced from 25.2 m to 16.4 m. Results are given for a large antenna simulated using CST Studio Suite.VT1
TD(22)01035Riccardo Marini, Leonardo Spampinato, Silvia Mignardi, Roberto Verdone, Chiara BurattiReinforcement Learning-based Trajectory Planning for UAV-aided Vehicular CommunicationsIt is widely expected that future 6G networks will rely on autonomous Unmanned Aerial Vehicles (UAVs) acting as flying Base Stations (BSs) to provide a wide range of services that current networks are not able to handle. One of the major trends deals with Vehicle-To-Everything (V2X) communications, where vehicles must be connected to the network in order to offer different applications, such as vehicle platooning, advanced driving, cooperative awareness and extended sensing. In this context, vehicles could deeply count on flying BS, in order to increase the throughput or reduce the experienced latency, thus satisfying the constraints such services require. As a consequence, path planning must be carefully designed so that UAVs can follow and keep stable links with moving vehicles. In this sense, Reinforcement Learning (RL) techniques are becoming the main enabler for solving such problem, since, besides the inability to solve it with classical methods, they offer the possibility to learn how to act in an environment with little prior information, given that full knowledge of the scenario is usually not available. In this paper, we present a RL-based approach to solve the path planning problem in a 6G vehicular scenario, where, differently from typical Internet of Things (IoT) applications, UAVs are required to follow cars rather than providing service to different cluster of users distributed in an area. Different RL architectures are compared, in order to analyze their performance and select the best strategy maximising the sum throughput. The model account for beamforming techniques used at the UAVs.WG3
TD(22)01036Mate Boban, Ziming Yu, Jian Li, Tommaso ZugnoTHz Channel Measurements and Modeling in Indoor Scenarios: Summary of Use Cases and Initial Pathloss ResultsCommunication systems in the THz range (between 0.1 and 10 THz) are expected to be an integral part of future 6G networks. However, due to their specific properties (large pathloss, atmospheric attenuation on the one hand, and the large bandwidths available and inherently limited interference on the other hand), they will be suitable for a specific subset of the envisioned 6G use cases. To that end, we provide an overview of relevant use cases for THz communication, with a focus on the specific aspects relevant for channel modeling scenarios. These relate not only to specific environmental setting (e.g., office, classroom, industrial), but also to mode of operation (e.g., communication only, static/mobile nodes, integrated communication and sensing, etc.). Based on the existing literature, we identify relevant indoor scenarios for THz channel measurements. Finally, for three relevant scenarios identified (meeting room, open office area, and office hallway), we perform channel measurements at 140 GHz and 220 GHz and show initial pathloss results.WG1
TD(22)01037Mengting Li, Fengchun Zhang, Yilin Ji and Wei FanVirtual Antenna Array with Directional Antennas for Millimeter-Wave Channel CharacterizationMillimeter-wave (mm-Wave) band channel modeling and characterization are essential for system design and analysis in the fifth generation (5G) and future communication systems. Reliable channel sounding in the deployment scenarios is required for accurate and realistic channel modeling and characterization. In the state-of-the-art, directional scanning sounding (DSS) and virtual antenna array (VAA) sounding are two popular methods due to their simplicity and cost-effectiveness. The DSS and the VAA methods are typically based on mechanically rotatable directional antennas and mechanically movable omnidirectional antennas, respectively. However, the spatial resolution is limited by the directivity and high side lobes of the directional antennas used in the DSS method. The conventional VAA method also has limitations in terms of the low signal-to-noise ratio (SNR) and the unavailability of suitable omnidirectional antennas that support mmWave (above 60 GHz) band measurements and wideband horizontal polarization measurements. In this paper, a novel directional antenna based VAA framework in combination with the associated beamforming algorithm is proposed. Compared to the state-of-the-art methods, the proposed framework can achieve high angular resolution and high SNR for mm-Wave channel measurements without introducing additional cost and measurement time. Furthermore, it is a generic solution which can be applicable for arbitrary frequency bands and polarizations, unlike the conventional VAA method. To validate the effectiveness and robustness of the proposed method, experiments in two scenarios (a clean anechoic chamber and a realistic indoor meeting room) were conducted over 28 – 30 GHz with two types of directional antennas (i.e., a horn antenna and a corrugated antenna). Besides, the advantages of the proposed method are highlighted with a comparison to the conventional VAA and DSS methods.WG1
TD(22)01038Yejian Lyu, Allan Wainaina Mbugua, Kim Olesen, Pekka Kyösti, and Wei FanDesign and Validation of the Phase-Compensated Long-Range Sub-THz VNA-based Channel SounderThis paper presents the first vector network analyzer (VNA)-based sub-Terahertz (sub-THz) phase-compensated channel sounder at 220-330 GHz using radio-over-fiber (RoF) techniques that could enable long-range phase-coherent measurements. The optical cable solution enables long-range channel measurements at sub-THz bands since it can effectively minimize cable loss. This paper also proposes a novel phase compensation scheme to stabilize the phase variations introduced by the optical fiber of the channel sounder to enable its application in multichannel/antenna measurements. This proposed channel sounder is validated in back-to-back measurements under two optical cable conditions, i.e., with the presence of thermal changes and mechanical stress. The phase variation introduced by the cable effects in the system is shown to be over 400 degree in 220-330 GHz, compared to 15 degree at 220-288 GHz and 37 degree in 288-330 GHz after compensation, respectively, demonstrating the robustness and effectiveness of the developed channel sounder in practice. The developed system, which has a dynamic range of 106.7 dB, can support measurement range up to 300m (limited by the optical cable length in our system and subject to over-the-air signal transmission loss in the practical environment).WG1
TD(22)01039Ahmed Boujnoui, Abdellah Zaaloul, Luis Orozco-Barbosa and Abdelkrim HaqiqEnhanced Pricing Strategy for Slotted ALOHA with ZigZag Decoding: A Stochastic Game ApproachIn this paper, we propose a novel pricing mechanism to enhance the performance of the Slotted ALOHA mechanism combined with ZigZag Decoding (SA-ZD). We model the system using a stochastic game approach where the game state is given as a Markov process. We assume a cooperative game framework where users seek to optimize the same utility function. In our previous pricing mechanism, we associated a cost C ∈ [0, 1] for every transmission and retransmission attempt. Thus, if the transmission succeeds, the user receives a reward equal to 1 − C. Otherwise, in the case of collision, he pays a penalty equal to C. Following this approach, users prefer not to take the risk of paying the penalty cost, which means they choose to wait rather than transmit, especially in heavy traffic conditions. Even though it seems optimal not to transmit in such conditions, our results show that this behavior dramatically decreases the system performance. Besides, it leads to an inherent tradeoff between the backlogged and newly arrived traffic. Toward this end, we propose in this paper a novel pricing strategy where we associate a cost not only to transmission attempts but also to the idle event (i.e., when no one is transmitting).
Moreover, we address the tradeoff problem by associating different Cb and Cs costs to backlogged and newly arrived packets. Therefore, users pay a cost price denoted by Cs or Cb when a successful transmission goes through the channel. On the other hand, if a collision occurs, they pay a cost price Cc, and when no one is transmitting, they pay the idling cost price Cidle. Our results show that the proposed approach achieves the best performance and maintains a fairness level between backlogged and newly arrived packets compared to the old pricing mechanism.
TD(22)01040Margot Deruyck, German Castellanos, Luc Martens, Wout Joseph5G Network Exposure: a case study

Recently, human exposure related to 5G technology has been receiving much attention. The question also arises whether it is possible to implement 5G while complying with the current exposure limits. In this study, we investigate a 5G roll-out for the city center of Brussels, Belgium. The study provides answers to four main questions:
1) Can we provide 5G coverage with 6 V/m limit by upgrading the 4G base stations to 5G?
2) If not, would it be possible with a 14.5 V/m limit?
3) If not, how many small cells would be required to provide 5G coverage with 6 V/m limit?
4) How many small cells would be required if we increase the limit to 14.5 V/m?
To solve these questions, a capacity-based wireless outdoor network planner is used, allowing to optimize the network not only towards coverage but also towards human exposure.

TD(22)01041Vittorio Todisco, Stefania Bartoletti, Claudia Campolo, Antonella Molinaro, Antoine O. Berthet, Alessandro BazziPerformance Analysis of Sidelink 5G-V2X Mode 2 Through an Open-Source SimulatorThe Third Generation Partnership Project (3GPP) has recently published a new set of specifications to enable advanced driving applications in fifth generation (5G) vehicle-to-everything (V2X) scenarios, with particular effort dedicated to the sidelink resource allocation in the autonomous mode, named Mode 2. In this paper, we conduct a comprehensive analysis of Mode 2 performance via an open-source system-level simulator, which implements the 5G New Radio (NR) flexible numerology and physical layer aspects together with the newly specified sidelink resource allocation modes for V2X communications and different data traffic patterns. Results collected through extensive simulation campaigns, under a wide variety of vehicle density, data transmission settings and traffic patterns, showcase the effects of the new 5G-V2X features on the sidelink resource allocation performance and provide some insights into possible ways to further improve Mode 2 performance.VT3
TD(22)01042Hamid Taramit, Luis Orozco Barbosa, Abdelkrim HaqiqEnergy Efficiency Framework for Time-limited Contention in the IEEE 802.11ah StandardThe Restricted Access Window (RAW) mechanism introduced in the IEEE 802.11ah standard aims to address the high channel contention of Internet of Things (IoT) networks. It allows the Access Point (AP) to periodically limit channel access to only one group of stations during a short time interval called RAW slot. Such a time limit prevents the channel access contention from reaching the steady state. Henceforth, we present a renewal theory based analytical model for the time-limited contention within the RAW slot. Furthermore, we construct a counting process to track transmissions within the contention interval and derive closed-form analytical results. As IoT networks are typically composed of battery-powered devices, it is compulsory to manage the energy consumption of the network in order to extend the lifetime of devices. Hence, we analyze and evaluate the energy consumption and energy efficiency during the RAW slot period. Extensive simulations using Matlab software validate the analytical model and prove the effectiveness of our proposals.WG3,VT3,VT4
TD(22)01043Sara Cavallero, Chiara Buratti, Alexey Tsarev, Giampaolo Cuozzo, Emil Khayrov, Yuliya Gaidamaka, Roberto VerdoneApplying Carrier Sense Multiple Access to Industrial IoT at Terahertz FrequenciesThis paper considers an Industrial Internet of Things (IIoT) scenario, where tags, deployed over an industrial machine, transmit the measured data to a final Gateway (GW) using Terahertz (THz) frequencies. The GW is located in the vicinity to the machine. To overcome the large path loss caused by such high frequencies, the GW is equipped with multiple antennas and sweeps the entire machine to gather data from tags, while tags are equipped with a single antenna due to miniaturisation issues. The feasibility of applying a Carrier Sense Multiple Access protocol to such scenario is addressed in this paper. A mathematical model incorporating physical-layer aspects that also considers both, hidden node problem caused by the asymmetric nature of the link, and propagation delays that cannot be neglected at such frequencies, is presented. The model is validated via comparison with simulations, and the impact of the different simplifying assumptions is shown. Numerical results also illustrate the impact of propagation delays on the performance and provide a comparison with an Aloha-based protocol.WG1,WG3,VT3
TD(22)01044Nicol Sarcone Grande, Marco Skocaj, Roberto VerdoneUsing MDT Data for 5G Initial Access OptimisationThis work proposes an algorithm for latency reduction during the initial access phase in 5G networks. Beamforming is a key technology to contrast the propagation losses in the mmWave band, but it introduces non-negligible delays since the base station (gNB) and the UEs (users equipment) have to find a suitable beam pair to communicate, hence the purpose of this algorithm is to implement an intelligent sequence of beams to be used at the gNB. This latter arises from the analysis of the propagation environment and, also, from the information contained in a historical database generated by exploiting MDT (Minimization of Drive Test) data.WG3
TD(22)01045Conor Brennan, Imtiaz Islam, Jason Basquill and Kirk SoodhalterMethod of moments approach to computing EM scattering from random rough surfacesThe problem of computing 2D EM wave scattering from rough surfaces is addressed using an integral equation formulation discretised using the method of moments. Successive Symmetric Over-Relaxation (SSOR) is applied to the governing matrix equations along with eigenvalue deflation which is designed to separately account for the effect of those eigenvectors of the iteration matrix that have eigenvalues greater than 1. Numerical results are presented applying the method to a variety of scattering profiles and examining the resultant convergence performanceWG1
TD(22)01046Petr Hron, Jozef Lukac, Jan SykoraNoncoherent WPNC 2-source MAC channel demodulation methodsAbstract— This paper addresses the problem of differential demodulation in a 2-source Hierarchical MAC (H-MAC) in the context of a Wireless Physical Layer Network Coded system with Hierarchical Decode and Forward relaying strategy. We consider a parametric H-MAC with known attenuation values but unknown phases. Component phases have generally a strong impact on the resulting superposed constellation and need to be accurately estimated to facilitate a coherent demodulation. The use of a differentially encoded modulation eliminates the need of precise channel phase tracking, which is a nontrivial problem in the WPNC scenario. At the same time it can operate with the same 3dB penalty as in a differential single user case. In this paper we summarize several differential demodulation procedures and additionally introduce new ad hoc methods, which can substantially reduce the computation complexity of the demodulation. We compare the various demodulation algorithms by means of evaluating the hierarchical bit error rate performance for different channel states. Additionally we analyze the robustness of differential demodulation w.r.t. channel phase noise.WG2
TD(22)01047Francesco Linsalata, Silvia Mura, Marouan Mizmizi, Maurizio Magarini, Umberto SpagnoliniProactive Link Selection in High Frequencies IoV networkRecent advances in Vehicle-to-Everything (V2X) technology and the upcoming 6G sub-THz, will dawn a new era for vehicular services with unprecedented communication capabilities. This will further boost the development of the so-called Internet Of Vehicles (IoV)s network. Vehicles are expected to communicate with each other and infrastructures to deliver a new transportation experience and increased safety in road networks. Communication at high frequencies and high mobility scenarios, however, introduce several challenges that need to be addressed. The radio propagation experiences severe path-loss attenuation and requires beamforming techniques. Additionally, high mobility produces frequent beam-blockage, which impairs communication reliability. Several solutions are being investigated, and the most promising is using relaying techniques. In our work we propose a novel predictive relaying strategy that exploits the cooperation in the IoV network by predicting the dynamic condition of the radio links. The proposed framework allows to compare different relay selection techniques.WG2,WG3,VT2
TD(22)01048Thomas Wilding, Erik Leitinger, Klaus WitrisalMultipath-based Localization and Tracking considering Off-Body Channel EffectsThis paper deals with multipath-based positioning and tracking in off-body channels.
An analysis of the effects introduced by the human body and the implications on positioning and tracking is presented based on channel measurements obtained in an indoor scenario.
It shows the influence of the radio signal bandwidth on the human body induced field of view (FOV) and the number of multipath components (MPCs) detected and estimated by a deterministic maximum likelihood (ML) algorithm.
A multipath-based positioning and tracking algorithm is proposed that associates these estimated MPC parameters with floor plan features and exploits a human body-dependent FOV function. The proposed algorithm is able to provide accurate position estimates even for an off-body radio channel in a multipath-prone environment with the signal bandwidth found to be a limiting factor.
TD(22)01049Youssef Agram, Jianqiao Cheng and François QuitinDirection-of-arrival estimation with virtual multi-antenna arrays using dual-antenna receivers: algorithms and controlled experiments.

Localization of radio-frequency (RF) transmitters can be achieved using virtual multi-antenna arrays. The method relies on a mobile, single-antenna receiver that captures successive messages from a transmitter, thereby emulating a multi-antenna array. However, two main challenges emerge from this technique: 1) Successive positions and orientations of antennas have to be determined, meeting spatial Nyquist criterion; 2) the local oscillator (LO) frequency offset between transmitter and receiver adds a drifting phase component to the received signal on each antenna of the array. In this paper, we extend the algorithm to dual-antenna receivers, which allows to recover the azimuth (without ambiguity) and elevation. A linear model of the LO frequency drift is also used, allowing for low-quality, drifting LOs. Controlled experimental results from a software-defined radio testbench are presented. Increased performance is noted when moving from a single antenna to a multi-antenna system.

Keywords : Direction of arrival, MUSIC, Virtual Multi-antenna array, RF transmitter localization.

TD(22)01050Mohammed Mallik, Sofiane Kharbech, Taghrid Mazloum, Shanshan Wang, Joe Wiart, Davy P. Gaillot and Laurent ClavierEME-Net: A U-net-based Indoor EMF Exposure Map Reconstruction MethodIn wireless communication systems, in order to respond to the perception of risks related to electromagnetic field exposure and allocate radio resources, the estimation of the received power and exposure map is an essential task and a challenge. This paper proposes an algorithm for estimating electromagnetic field exposure maps using U-net architecture based on convolutional neural networks. The power map estimation is transformed into an image reconstruction task by image color mapping, where every pixel value of the image represents received power intensity. The designed model learns wireless signal propagation characteristics in a realistic indoor environment while considering various positions of the Wi-Fi access points. Results show that indoor propagation phenomena and environment models can be learned from data producing an accurate power map to measure the electromagnetic field.WG1,VT1
TD(22)01051Antonio Tarrias, Sergio Fortes & Raquel Barco5G failure modeling and insights in ns-3 network simulation scenariosFailure management has been one of the most researched fields in cellular networks paradigm.
Networks operators has experienced many problems on their deployments with each of the past generations.
5G networks aim high to encompass a wide variety of services, which means a large amount of resources on network management and failure resolution.
The objective of the present work is to use the previous generation as base and provide, together with the updates on 3GPP specification, insights about what would be the problems that networks will handle.
For this, they were identified and categorised some of these failures at the same time their effect on system performance was evaluated.
TD(22)01052Sami Chajii, Philippe De Doncker, François QuitinEnvironment mapping with 28 GHz beamsteering transceivers: hardware architecture and preliminary resultsThe evolution of radio communications systems towards millimeter-wave (mm-wave) frequencies allows for new opportunities in terms of localization and environment mapping. The increase in bandwidth and size of antenna arrays creates great similarities between 28 GHz communication systems and radar systems. There are several challenges to use 28 GHz communication systems for environment mapping: 1) it is unclear whether multipath components can be identified with arrays that rely on beamsteering, 2) the local oscillator (LO) is not shared between the transmitter and the receiver, causing considerable LO frequency offsets, and 3) the high power of the line-of-sight component tends to “blind” the receiver to all other multipath. In this paper, we realize (relatively) low-cost 28 GHz software-defined radio-based transceivers that use 4×4 antenna arrays. The transceiver architecture is presented, and the setup is used to perform some preliminary measurements. These first results indicate that some amount of multipath components can be recovered, even with limited angular resolution of a 4×4 beamsteering array.WG2
TD(22)01053Srdjan Sobot, Milan Lukic, Marko Beko, Dejan VukobratovicUAV-assisted 3GPP NB-IoT Network Extension for Deep Rural DeploymentsIn this work, we develop and demonstrate a two-tier Low Power Wide Area Network (LP-WAN) system based on Unmanned Aerial Vehicle (UAV) base stations suitable for dynamic deployment in deep rural environments. The proposed UAV-based LP-WAN network augments the existing macro-cellular 3GPP NB-IoT network (Tier 1) with an additional layer of mobile base stations (Tier 2) which, in our setup, are based either on LoRa or NB-IoT technology. Mobile Tier 2 LoRa/NB-IoT base stations provide connectivity to static or mobile NB-IoT/LoRa user equipment deployed in the areas without direct Tier 1 NB-IoT network coverage. The proposed two-tier LP-WAN network scenario is suitable for various agricultural, forestry and environmental applications such as livestock or wild animal monitoring. In this experimental work, we report the prototype that was successfully deployed and used in a real-world deep rural environment without Tier 1 LP-WAN network coverage.WG3,VT2
TD(22)01054Carolina Gijón, Matías Toril, Salvador LunaModelling Performance in Sliced Radio Access Networks with Supervised LearningIn 5G, Network Slicing (NS) feature allows to deploy several logical networks customized per specific business models over a common physical infrastructure. In the Radio Access Network (RAN),  slice-level performance models are required for spectrum sharing and network re-dimensioning purposes, among other tasks. Throughput is often regarded as a key performance metric due its strong impact on satisfaction of users demanding enhanced Mobility Broadband (eMBB) services. In this work, we present the first comprehensive analysis to assess well-known supervised learning algorithms for estimating slice throughput in the down link from information collected in the Operation Support System (OSS) in RAN-sliced networks. The considered approaches include support vector regression, k-nearest neighbors, ensemble methods based on decision trees and neural networks. All these algorithms are tested in two NS scenarios considering single-service and multi-service slices, respectively. To this end, 5 datasets with performance indicators and connection traces are created with a system-level simulator emulating the activity of a live cellular network. Results show that the best SL algorithm and input information required to estimate throughput at different level (i.e. cell/slice) and NS scenario (single-service/multi-service) may vary. Nonetheless, the best model in each scenario have shown adequate accuracy (i.e., error under 10%).WG3
TD(22)01055Dennis BeckerMeasurement Based Identification of MPCs in an Urban Drone-to-Drone Propagation ScenarioThe risk of mid-air collisions between flying drones has to be minimized to the greatest possible extent as it endangers people in the air and on the ground especially when being integrated into dense urban airspace. For a safe and efficient operation drones will need to exchange information in a robust and reliable manner and one essential part will be direct Droneto-Drone (D2D) communications. Especially, for dense drone scenarios in urban environments a communication system must cope with the specific underlying channel propagation conditions. In order to characterize the propagation effects between two moving drones in urban environments, we performed a channel sounding measurement campaign and presented an approach in previous work to localize the origin of the measured multipath components (MPCs) for a three-dimensional layout. In this work, we apply it on our measured flight scenarios in three different environments in order to identify the MPCs by assigning them to real-world objects. Furthermore, we describe first characteristics for them and show that the measured urban scenario consists of different kinds of components that must be considered in a future D2D channel model.WG1,WG2
TD(22)01056J. M. Sanchez-Martin, M. Toril, C. Gijon, S. Luna-Ramirez, M. Fernandez-NavarroCoordination and Capacity Analysis in C-RAN for HetNetsIn this work, a capacity analysis method is proposed for a centralized radio access network composed of macrocells and small cells. Radio remote head (RRH) are grouped in base band unit (BBU) pools using graph-partitioning algorithms. For this purpose, Inter-Cell Interference Coordination (ICIC) and Coordinated Multi-Point Transmission/Reception (CoMP) techniques are used to evaluate their influence on the network under different load levels and coordination restrictions. The testing phase is carried out using a radio network planning tool. This tool is used to evaluate the spectral efficiency metrics and allocation of shared resources per cell obtained over a real Long-Term Evolution (LTE) heterogeneous network. The results present the load and coordination conditions between cells necessary to achieve capacity improvements in the system.WG3
TD(22)01057Paul Unterhuber, Michael Walter, Thomas KürnerGeometry-Based Stochastic Channel Model for Train-to-Train Communication in Open Field EnvironmentFuture railway applications, e.g., wireless train bus or virtual coupled trains, will rely on train-to-train (T2T) communication. Those future applications require an exchange of safety critical data between trains within one train-set based on wireless communication. Hence, the investigation of the propagation effects and the influence on the wireless communication is tremendously important. Future developments of communication standards and systems demand a detailed characterization and modeling. We investigated the propagation mechanisms based on channel sounding measurements and derived statistics for the propagation effects. Based on the environment, geometry and the propagation statistics we derive channel models for T2T communication. To cope with the movement of the trains, the changing environment and resulting temporal correlation effects we present a geometry-based stochastic channel model (GSCM) for T2T communication in open field environment.WG1,VT2
TD(22)01059Botond Tamás Csathó, Bálint Péter Horváth, János BitóFinite Element Method-based Analysis and Control of Reconfigurable Intelligent SurfacesReconfigurable intelligent surface (RIS) is a promising paradigm for future wireless networks. The core idea is to utilize a nearly-passive, real-time configurable metasurface between the transmitter and the receiver, thus gaining partial control over the wireless channel. It is a relevant yet unsolved question whether this paradigm can enable significant performance improvement. The design, performance analysis, and optimal control of RISs are a gold mine of research topics. A physically consistent model of the metasurface which captures the essential phenomenons is vital for both analysis and control of RISs. In this technical document, we analyze the reflection from a RIS with the finite element method (FEM). The physical structure of the RIS is based on a prototype described in the literature. We numerically optimize the parameters of the RIS to tune the reflected electromagnetic waves into the desired direction.WG1
TD(22)01060Pekka Kyösti, Pasi Koivumäki, Katsuyuki Haneda, Wei FanOn usability of lower radio frequency channel information for high frequency beam searchHow similar is the angular power profile of propagation channel at different frequency bands between 4 and 86 GHz? How well can low band channel information be used for beam search procedures at a higher frequency, considering practical antenna size limitations? In this TD we aim to address these questions by analysing point cloud ray tracing data of an airport check-in area at 4, 15, 28, 60, and 86 GHz frequencies. Prior to the analysis we develop and propose an analysis method that considers both antenna beam patterns and the multi path propagation channel.WG1
TD(22)01061Raheeb Muzaffar, Mahin K. Atiq, Emiliano Sisinni, Thilo Sauter, and Hans-Peter Bernhard5G Deployment Models and Configuration Choices for Industrial Networks – a State of Art OverviewThe Industry 4.0 digital transformation is led by the automation of manufacturing processes. Communication being a backbone can be enabled through the much promising 5G technology to provide connectivity to industrial internet of things. The 5G new radio support for ultra-reliable low latency communication and integration with time-sensitive networking can enable automated operation for latency-critical and deterministic industrial applications. In this paper, we provide an overview on 5G enabling technologies and non-public network deployment variants including details on multi-access edge computing infrastructure that can be selected according to the needs of an industrial setup. Moreover, we discuss 5G integration with time-sensitive networking and open platform communications unified architecture for ubiquitous industrial communication requirements. 5G QoS support and security aspects for industrial networks are also discussed.WG3,VT3
TD(22)01062Danila Ferretti, Paolo Lanci, Buket Torun, Davide Amato, Roberto VerdoneLoRa-Based Railway Signalling System for Secondary LinesItalian railway signalling systems are slowly evolving from national solutions, such as SCMT-SSC systems, towards European solutions, which are the ERTMS-ETCS systems. The rationale behind this work is that on the national railway network, only high-speed lines implement European-type signalling, while this project proposes a technological and cost-effective solution for the implementation of ERTMS-ETCS for secondary regional lines, which are about 60 per cent in Italy. This research activity investigates the possibility to implement a LoRa-based wireless train network for a feasible onboard train integrity solution following a European signalling system. In addition, it is proposed a train-to-ground network, which is based on the same technology, to develop a communication system with train localization/detection capability. This paper investigates a railway scenario with electronic devices placed on each vehicle of the train and on some catenary masts at a fixed distance. Both are equipped with LoRa modules for wireless communication and localisation purposes. To this aim, a LoRa protocol is designed to jointly consider: i) train-to-ground network; ii) multi-hop ground network.VT2
TD(22)01063Simon Demey, Charles Wiame, Claude Oestges and Philippe De DonckerInvestigation on the trade-offs between deterministic ray-tracing and stochastic methods in a Manhattan-like environmentThe proposed work tries to quantify how relevant it is to use a deterministic channel model like ray tracing at a network deployment level in a Manhattan-like environment. More precisely, the goal is to compare it with simplier and more computationally efficient models to see if similar results can be obtained at a lower cost. In the framework of the analysis, ray tracing is thus compared to a stochastic geometry-based model in terms of coverage probability and exposure at a centric user for which we were able to derive semi-closed form expressions. In addition, a good lower bound for the joint coverage-exposure distribution was found. In order to easily mimic the propagation mechanisms implemented in the UCLouvain RT software (i.e. LOS, reflection and edge diffraction) and enable fair comparison, the Berg recursive model was considered as channel model. Finally, the analytical expressions were further enhanced to take into account the antenna height, the street width as well as the line-of-sight probability in the typical street and to distinguish a typical user located at a crossroad from a typical user located inside a street.WG1
TD(22)01064Network Attack Classification in IoT Using Support Vector MachinesChristiana Ioannou and Vasos VassiliouMachine learning (ML) techniques learn a system by observing it. Events and occurrences in the network define what is expected of the network’s operation. It is for this reason that ML techniques are used in the computer network security field to detect unauthorized intervention. In the event of suspicious activity, the result of the ML analysis deviates from the definition of expected normal network activity and the suspicious activity becomes apparent. Support vector machines (SVM) are ML techniques that have been used to profile normal network activity and classify it as normal or abnormal. They are trained to configure an optimal hyperplane that classifies unknown input vectors’ values based on their positioning on the plane. We propose to use SVM models to detect malicious behavior within low-power, low-rate and short range networks, such as those used in the Internet of Things (IoT). We evaluated two SVM approaches, the C-SVM and the OC-SVM, where the former requires two classes of vector values (one for the normal and one for the abnormal activity) and the latter observes only normal behavior activity. Both approaches were used as part of an intrusion detection system (IDS) that monitors and detects abnormal activity within the smart node device. Actual network traffic with specific network-layer attacks implemented by us was used to create and evaluate the SVM detection models. It is shown that the C-SVM achieves up to 100% classification accuracy when evaluated with unknown data taken from the same network topology it was trained with and 81% accuracy when operating in an unknown topology. The OC-SVM that is created using benign activity achieves at most 58% accuracy.WG3,VT3
TD(22)01065Lennart Thielecke, Thomas KürnerConsideration of Latencies in an SDR-based V2X Channel EmulatorIn this paper, we analyze possible sources of latency in an SDR-based channel emulator. A tapped delay line (TDL) is used to model the channel impulse response. The impact of the TDL parameterization on the overall latency of the channel emulator is investigated. The performance difference between a CPU-based and an FPGA-based implementation is measured and compared. The test environment is set up using open-source frameworks.WG1,WG2
TD(22)01066Rafael Asorey-Cacheda,  Luis M. Correia, Concepcion Garcia-Pardo, Krzysztof Wojcik, Kenan Turbic, and Pawel KulakowskiWhen Body Area Network Reaches the Nano Level: An Architecture for Cardiovascular Health ApplicationsCardiovascular events occurring in the bloodstream are responsible for about 40% of human deaths in developed countries. Motivated by this fact, we present a new global network architecture for a system for the diagnosis and treatment of cardiovascular events, focusing on problems related to pulmonary artery occlusion, i.e., situations of artery blockage by a blood clot. The proposed system is based on bio-sensors for detection of artery blockage and bio-actuators for releasing appropriate medicines, both types of devices being implanted in pulmonary arteries. The system can be used by a person leading an active life and provides bidirectional communication with medical personnel via nano-nodes circulating in the bloodstream constituting an in-body area network. We derive an analytical model for calculating the required number of nano-nodes to detect artery blockage and the probability of activating a bio-actuator. We also analyze the performance of the body area component of the system in terms of path loss and of wireless links budget. Results show that the system can diagnose a blocked artery in about 3 hours and that after around 3 hours medicines can be released in the exact spot of the artery occlusion, while with current medical practices the average time for diagnosis varies between 5 to 9 days.VT1
TD(22)01067Christia Charilaou, Spyros Lavdas, Ala Khalifeh, Vasos Vassiliou and Zinon ZinonosFirmware update using multiple GWs in LoraWan networksThe remarkable evolution of the IoT raised the need for an efficient way to update the device’s firmware. Recently, a new process was released summarizing the steps for firmware updates over the air (FUOTA) on top of the LoRaWAN protocol. The FUOTA process needs to be completed quickly to reduce the systems’ interruption and, at the same time, to update the maximum number of devices with the lowest power consumption. However, as the literature showed, a single gateway cannot optimize the FUOTA procedure and offer the above mentioned goals since various trade-offs arise. In this paper, we conducted extensive experiments via simulation to investigate the impact of multiple gateways during the firmware update process. To achieve that, we extended the FUOTAsim simulation tool to support multiple gateways. The results revealed that several gateways could eliminate the trade-offs that appeared using a single gateway.WG3,VT4
TD(22)01068Julian Karoliny, Thomas Blazek, Hans-Peter Bernhard, Andreas SpringerMulti Hypothesis Interference Tracking in Wireless NetworksWith the rise of the Internet-of-Things and Industry 4.0, the number of wireless sensor networks (WSN) and devices with wireless transceivers is steadily increasing. Most of these devices rely on the use of unlicensed frequency bands. These are however limited in number and each additional device may act as new source of interference for the others. If a WSN has strict energy limitations or real-time constraints, interference detection and coexistence becomes very important. Especially in industrial  environments, unintended or in the worst case intended interference is a high security risk and counter measures have to be applied as soon as possible. This paper proposes an approach that allows to detect, identify and track interference in WSNs. Using deeper understanding of the source of interference counter measures for the WSN are applied and for example future transmission collisions prevented. For this we adapted the Multi Hypothesis Tracking algorithm and used it with timeslot-based interference measurements on low-cost sensor nodes. With this we are able to detected multiple sources of periodic interference and track them over time. We show the applicability of the algorithm with extensive simulations and demonstrate the performance with measurements of an implemented WSN. With our measurements we were able to estimate the timing of the interference with a root mean squared error (RMSE) of 0.146ms, and the corresponding transmit period with a RMSE of 0.024ms.WG2,VT3
TD(22)01069O. S. Penaherrera-Pulla, Carlos Baena, Sergio Fortes, Eduardo Baena, Raquel BarcoKQI evaluation for 360-Video services over mobile networksNew generation services have become the pillars and basis for the development of cutting-edge mobile networks as 5G and 6G. These technologies are intended to provide high-quality services which currently are not available, for instance, VR (Virtual reality). This paper presents a framework for LTE and 5G that aims to assess the performance of the 360-video service through Key Quality Indicators (KQI). This proposal integrates the immersive experience of omnidirectional video using an HMD (Head Mounted Device) while the performance measurement is done along the user session. The video content uses DASH (Dynamic Adaptive Streaming over HTTP) from a video server located in the cloud. Finally, a performance comparison is provided making use of the data collected through an iterative experiment. The results obtained show the potential of the mobile networks and encourage their use for a high-quality VR service deployment.WG3,VT1,VT4,HA1,HA2
TD(22)01070Angesom Ataklity Tesfay, Sofiane Kharbech, Eric Pierre Simon, and Laurent ClavierSignal Denoising and Detection for Uplink in LoRa Networks based on Bayesian-optimized Deep Neural NetworksLong-range and low-power communications are one solution to address the IoT challenge. The long-range implies a very low signal-to-noise ratio at the receiver. In addition, low power consumption requires reduced signaling, hence the use of minimum complexity protocols, such as ALOHA, and the lack of communication coordination. Therefore, the increase of objects using this technology will automatically lead to an increase in interference.
This paper combines autoencoder and convolutional neural networks for signal denoising and detection in the uplink of a LoRa network. An autoencoder is used to denoise and cope with interference, and a convolutional neural network (CNN) is used to detect the symbols. Simulation results demonstrate that the proposed approach outperforms both the convolutional neural network-based receiver and the classical LoRa receiver in the presence of interference from other LoRa users. The proposed receiver shows around 2 dB gain for a target SER of 0.0001.
TD(22)01071Marco Skocaj, Lorenzo Mario Amorosa, Giorgio Ghinamo, Davide Micheli, Giuliano Muratore, Flavio Zabini, Roberto VerdoneCapacity and Coverage Optimization with MDT Data and Deep Reinforcement LearningRecent years witnessed a remarkable increase in the availability of data and computing resources in communication networks. This contributed to the rise of data-driven over model-driven algorithms for network automation. This paper investigates a Minimization of Drive Tests (MDT)-driven Deep Reinforcement Learning (DRL) algorithm to optimize coverage and capacity by tuning antennas tilts on a cluster of cells from TIM’s network. We jointly utilize MDT data, electromagnetic simulations, and network Key Performance indicators (KPIs) to define a simulated network environment for the training of a Deep Q-Networks (DQN) agent. Some tweaks have been introduced to the classical DQN formulation to improve the agent’s sample efficiency and performance. In particular, a custom collection policy, namely depth-wise \(\epsilon – \eta\)  greedy policy, is designed to introduce constraints at training time. Results show that the proposed algorithm outperforms baseline approaches like DQN and best-first search in terms of long-term reward and sample efficiency. Our results indicate that MDT-driven approaches constitute a valuable tool for autonomous coverage and capacity optimization of mobile radio networks.WG3
TD(22)01072Golsa Ghiaasi, Thomas Blazek, Fjolla Ademaj, Julian Karoliny, Stefan Marksteiner, Markus Wolf, Peter Priller, Hans-Peter BernhardSDR-Based Communication Sniffing for Determining the Proliferation of ITS-G5This  paper  describes  the  measurement  setup  for the  proliferation  of  the vehicular  communication  systems  in operating  cars.  The  setup  builds  on  a  communication  sniffer which  detects  cooperative  awareness  messages broadcast  from ETSI  ITS-G5  systems,  decodes  them  and  records  the  public information in these messages. The experimental characterization of the setup in an outdoor scenario is presented in order to verify the  sniffer’s  performance  with  respect  to  transmit  power  and communication  range.  The  setup  is  deployed  for  measurements on a busy road with slow moving traffic in the center of the city of Linz,  Austria.  The  number  of  active  users  as  well  as  their transmit  power, duration  of  the  transmission  and  packet  length have  been  recorded.WG3
TD(22)01073Juan Cantizani-Estepa, Juan Bravo-Arrabal, J.J. Fernández-Lozano, Sergio Fortes, Raquel Barco, Alfonso García-CerezoClose detection robotic platform for Search And Rescue missions based on Bluetooth Low EnergyImprovements in telecommunications and digitalization directly improve the efficacy of a wide variety of processes. Recently, detection systems have received considerable attention because of the importance of tracking infected people contacts during SARS-CoV-2 pandemic. Such implementations can be useful in the task of finding potential victims in the context of emergency response, especially in situations where GPS is not available or inspection by imaging is not practical. Radio signals come into play, and specifically from devices that transmit periodically and with low power consumption. With the rise of Internet of Things over the last years, the number of wearable devices that support BLE, such as smartbands, smartwatches or smartphones, has been increasing constantly, as well as the number of users that carry them. Those devices can provide considerable assistance in locating injured or unconscious people. This work presents a system for detecting victims by means of a terrestrial search and rescue (SAR) robot. A real implementation of a close detection robotic platform based on BLE for SAR interventions is laid out. To estimate the distance between a robotic agent and potential victims within an experimental area, a Log-distance path loss model is presented. The proposed scheme has been tested in realistic scenarios during SAR exercises.WG2
TD(22)01074Quentin Gontier, Charles Wiame, Shanshan Wang, Christo Tsigros, François Horlin, Joe Wiart, Claude Oestges, Philippe De DonckerModeling the Spatial Distributions of Macro Base Stations with Homogeneous Density: Theory and Application to Real NetworksStochastic geometry is a highly studied field in telecommunications as in many other scientific fields. In the last ten years in particular, theoretical knowledge has evolved a lot, whether for the calculation of metrics to characterize interference, coverage, energy or spectral efficiency, or exposure to electromagnetic fields. Many spatial point process models have been developed but are often left aside in favor of the Poisson point process because of their unfamiliarity, their lack of tractability or the ease of use of the Poisson point process. This article is intended to be a short guide presenting a complete and simple methodology to follow to infer a real stationary macro antenna network using tractable spatial models. The focus is mainly on repulsive point processes and in particular on determinantal point processes which are among the most tractable repulsive point processes. This methodology is applied to Belgian and French cell towers. The results show that for all stationary distributions in France and Belgium, the best inference model is the $\beta$-Ginibre point process.WG2,WG3,VT1,HA1
TD(22)01075Benjamin Rainer, Stefan Zelenbaba, Anja Dakić, Markus Hofer, David Löschenbrand, Thomas Zemen, Xiaochun Ye, Guo Nan, Stefan Teschl, Peter PrillerV2V and V2I Wireless Channel Measurement Dataset including Vehicle Sensor DataThis paper discusses a freely available and open dataset containing vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) OFDM-based wireless channel measurement data including synchronised sensor data such as RADAR, LiDAR and high precision GPS. The wireless channel measurement is conducted at the carrier frequencies of 3.2 GHz and 5.9 GHz which are the most promising frequency bands in which future V2X communication systems will operate. The dataset contains the wireless channel measurement data of various V2X scenarios along with synchronized sensor information from a vehicle. In addition to the wireless channel measurement data, the dataset also includes frame error rate measurements from a IEEE 802.11p based communication system, synchronized to the other measurement data. The dataset includes V2V, V2I and V2P measurements with different scenarios.WG1,HA1
TD(22)01076Franco Fuschini, Marina Barbiroli, Enrico Maria Vitucci, Vittorio Degli-EspostiMulti-Band Outdoor-to-Indoor Propagation Measurements Using a DroneOutdoor-to-indoor propagation path-loss measurements have been carried out at 27 and 38 GHz, two of the frequencies allocated for 5G networks, for two different buildings: an office glass, steel and concrete building and a brick-wall residential house. The outdoor station has been mounted on a drone in order to have more placement flexibility and reach the desired height above ground without the use of a crane truck. Overall outdoor-to-indoor loss seems to depend primarily on the window surface and on the construction material. While loss is higher at 38GHz with respect to 27 GHz for the office building, brick-walls of the residential house appear to be less frequency selective.WG1
TD(22)01077Silvia Mignardi, Danila Ferretti, Riccardo Marini, Francesca Conserva, Stefania Bartoletti, Roberto Verdone, Chiara Buratti.Beam and Resource Assignment Optimization for UAV-Aided Vehicular Networks: Centralized and Distributed ArchitecturesFuture mobile radio networks require a degree of flexibility that technologies like Unmanned Aerial Vehicles (UAVs) carrying Base Stations (BSs) can provide. It is expected that the lower space above cities will be populated by many different types of UAVs, flying on a predefined path for a given mission, such as patrolling and logistics. This paper investigates an urban scenario with terrestrial macro BSs (MBSs) deployed and UAVs travelling on a predefined path carrying BS equipment with beamforming to be activated on demand. In the same area, several vehicles are moving while relying on network services,
and MBSs alone might not serve them adequately. We propose a network model where UAVs can be activated to work as BSs if needed and help the MBSs to fulfill the vehicular demand. To this purpose, we propose an optimization algorithm able to select the best set of beams each UAV has to activate, together with the number of resource units (in terms of time and frequency), maximizing the number of served vehicles. The optimization algorithm accounts for i) a joint resource allocation for both MBSs and UAVs that avoids interference, ii) a limited backhaul capacity, and iii) the architectural choice. Indeed, we analyse two different system architectures: i) a distributed model in which MBSs run the Radio Resource Management algorithm (RRM) separately, sharing environment information with each other; ii) a fully centralized model in which MBSs send their information to the network core, which is in charge of running the RRM algorithm. Numerical results compare the two architectures and show the impact of backhaul capacity on the performance.
TD(22)01078N. González, M. Solera, F. Ruiz, S. Luna, M.TorilA quality of experience evaluation method for an UAV first person view systemUnmanned Aerial Vehicles (UAV) communication systems are an increasingly widespread and emerging technology due to their flexibility, low cost and usability properties. Hence, the demand for Beyond Visual Line of Sight (BVLOS) cases that require large data transmission and low latency in cellullar networks are increasingly. In this work, the assembly, integration and networking of a UAV quadrotor for First Person View (FPV) system connected by LTE is presented. Different configurations of the link between the UAV and the Ground Control Station (GCS) are proposed, such as connection by LTE cloud-based server, direct LTE connection and direct WiFi connection. With these configurations, experiments are carried out to characterise the network metrics that model this service according to the telemetry, control and video traffic. The main contribution is the definition of a closed mathematical expression provided to define the Quality of Experience (QoE) for FPV use cases considering the video quality in terms of Video Multimethod Assessment Fusion (VMAF), network latency and video resolution as inputs. This expression will be applied to real experiments considering link performance, in which network changes based in packet loss and latency alterations will be introduced to measure the QoE of the UAV system.WG3
TD(22)01079Rolando Guerra-Gómez, Sílvia Ruiz-Boqué, Mario Garcia-Lozano and Umar SaeedFlexible Radio Access Network Optimization with Cell CoordinationThis paper focuses on Beyond fifth generation (B5G) non-linear data modeling and decision-making tools to optimize cost reduction versus coverage-QoS trade-off. Especially, the distribution of active Remote Radio Heads or Units (RRHs) needed, according to traffic demands, is improved. The proposed optimization platform is based on a multi-objective optimization model, which is designed to reduce the network cost while maintaining the coverage-QoS. Capacity constraints, User Equipments (UEs), and different slices are considered to test the results under realistic conditions. Results at 3.6 and 28 GHz are presented by analyzing and comparing several Cloud Radio Access Network (C-RAN) split options in a heterogeneous deployment with Macro-RRHs (MRRHs) and Small-RRHs (SRRHs). Results show cost reductions from 30% to 70% depending on the scenario. Moreover, the proposed algorithm aggregates the possibility to consider the coordination between cells in order to improve the cost reduction. The results considering cooperation has been presented at both frequency bands with a fully centralized C-RAN (split option 8).WG3
TD(22)01082Jerzy Mizeraczyk, Ryszard Studanski, Andrzej Zak and Agnieszka Czapiewska A Method for Underwater Wireless Data Transmission in a Hydroacoustic Channel under NLOS Conditions Wireless data transmission in the hydroacoustic channel under non-line-of-sight (NLOS) propaga-tion conditions, for example, during a wreck penetration, is difficult to implement reliably. This is mostly due to the multipath propagation, which causes a reduction in the quality of data recep-tion. Therefore, in this work an attempt has been made to develop a reliable method of wireless underwater communication test it under the NLOS conditions. In our method, we used multiple frequency-shift keying (MFSK) modulation, sending a single bit on two carriers, and diversity combining. The method was tested in laboratory conditions which simulated underwater signal propagation during the penetration of the wreck. The propagation conditions were investigated by determining the impulse responses at selected measurement points using the correlation meth-od. Additionally, for comparison, the data transmission quality was determined by the bit error rate (BER) under the same conditions using direct sequence spread spectrum (DSSS) and binary phase shift keying (BPSK) modulation. The obtained results confirmed the usefulness of the appli-cation of the developed method for wireless data transmission in a hydroacoustic channel under NLOS conditions. WG1, WG2 
TD(22)01083Simona Valbonesi and Paolo GraziosoPredictive evaluation of the increase in on-air E-field levels as consequence of 5G AAS systems deploymentsIn order to fully exploit their capabilities, 5G networks need a high availability of electromagnetic space. Overly precautionary exposure limits may hinder the correct and complete development of 5G, and in some cases even the full penetration of 4G networks. The context is extremely complicated, because on the one side the implementation of new generations network requires deploying more antennas on the territory, while on the other side population, and in some case also politicians, keep expressing concerns about potential risks associated to radio-frequency electromagnetic emissions.
This works presents a numerical evaluation, achieved through extrapolation, of electrical field levels that will be present in a given urban scenario when adding the contribution of 5G emissions (considering only broadcast channel, or adding variable traffic or, finally, including traffic that uses the whole capacity of 5G) to the pre-existing field levels due to 2G, 3G, and 4G sites. This evaluation has been carried out through the analysis of cumulative distribution functions of real measurements data.
Beyond the evaluation of expected field levels due to the new deployments, we present an estimate of the impact of new technologies on the saturation of electromagnetic space in regions and countries where field exposure limits are particularly restrictive, such as Italy.
WG1, VT4
TD(22)01084Simona Valbonesi and Paolo Grazioso Predictive evaluation of the impact of legacy technologies dismantling on E-field levels in urban areasThe commercial launch of 5G services spurs reflections on the exploitation of spectral resources and of electromagnetic field levels present in residential and working environments. Analyses carried out at European level showed that in a near future a remarkable fraction of mobile data traffic requests might not be satisfied in such areas or countries, like Italy, where the regulatory framework foresees more stringent limits than those suggested by the ICNIRP.
A possible solution to face this issue could be the (total or even partial) dismantling of legacy technologies (2G and/or 3G). Such dismantling could allow to free valuable resources that would be useful for the 5G development, both in terms of “electromagnetic space” in the transmission sites, and in terms of spectrum, i.e. of bandwidth that could be used to maximise 5G capacity. However, prior to make this choice, one should consider the actual use of legacy technologies by some crucial vertical services such as smart metering and e-calls. In this contribution we don’t take into account the regulatory, logistical and economic issues related to the possible replacement of 2G and/or 3G technologies with 5G; rather, we perform an evaluation of electrical field levels that may be expected if this replacement is carried out, considering different cases: with only the broadcast signal or with traffic (from light traffic to full cell occupation).
The analysis was carried out by comparing cumulative distribution functions for the current configuration with cases when one or both legacy technologies are completely replaced by 5G.
WG1, VT5
TD(22)01085H Ahmadi, A Nag, Z Khan, K Sayrafian, S RahadrjaNetworked Twins and Twins of Networks: an Overview on the Relationship Between Digital Twins and 6GDigital Twin (DT) is a promising technology for the new immersive digital life with a variety of applications in areas such as Industry 4.0, aviation, and healthcare. Proliferation of this technology requires higher data rates, reliability, resilience, and lower latency beyond what is currently offered by 5G. Thus, DT can become a major driver for 6G research and development. Alternatively, 6G network development can benefit from Digital Twin technology and its powerful features such as modularity and remote intelligence. Using DT, a 6G network (or some of its components) will have the opportunity to use Artificial Intelligence more proactively in order to enhance its resilience. DT’s application in telecommunications is still in its infancy. In this article we highlight some of the most promising research and development directions for this technology.WG3 
TD(22)01086Afonso Carvalho, Luís M. Correia, António Grilo, Ricardo DinisAnalysis of Strategies for Minimising End-to-End Latency in 4G and 5G NetworksThe main purpose of this paper is to identify and study a variety of strategies that effectively reduce the end-to-end latency in both 4G and 5G networks. This latency reduction will allow the operators to provide URLLC services to the users, such as: remote surgeries, the Intelligent Transport Systems and factory automation services. To verify if those services can be implemented using the 4G and the 5G systems, the developed model considers several variables: the MEC node deployment option, the functionality splitting options, the radio techniques, and the network architectures. The MEC technology appears in the paper as the solution that allows the end-to-end latency values to reach values below 1 ms, which are required for some of the URLLC services. The results obtained show that the 4G system does not have enough capacity to allow the existence of the upcoming services. Even with the MEC node deployment that minimizes the latency, the LTE network is not able to provide the URLLC services under study. The simulations show that using the adequate latency reduction strategies and radio techniques, the 5G system has enough capacity and sufficiently low latencies to provide the upcoming services.WG3
TD(22)01087Sibren De Bast, Evgenii Vinogradov, Sofie PollinUser localisation in distributed Massive MIMO networksThe recent interest in using Massive MIMO (MaMIMO) networks for user positioning and the trend towards cell-free MaMIMO with a large number of APs, result in the question of how these two will be combined efficiently. Current algorithms process all signals in a central processing unit, resulting in a high fronthaul requirement when deploying them in a distributed system. In this paper, we compare state-of-the-art localisation methods which process the signal in a central or distributed way. We propose novel distributed localisation methods and use state-of-the-art algorithms as a benchmark. Next to comparing the localisation accuracy, we evaluate the sample efficiency, the performance in a dynamic setting, the scalability and fronthaul requirement of the methods. The proposed Distributed ML-MUSIC method, which combines machine learning and analytical signal processing, reaches a median error of 34.2 mm while only using 500 training samples. Furthermore, due to its distributed design, it lowers the fronthaul requirement 1200-fold in comparison to centralised methods. Furthermore, this method has the lowest computational complexity of all analysed methods, making it an ideal method for upcoming cell-free networks.WG2
TD(22)01088Yann Maret, Franck Legendre, Jean-Frédéric WagenPreliminary investigations of a Graph Neural Networks for routing in MANETsGraph Neural Networks (GNNs) are popular algorithms to learn from graph-based applications. MANETs are networks of radio mobiles and can be presented as a graph evolving with the dynamic of the topology. A graph contains nodes, the radio mobiles; and links, the channel of communication between radios. The proposed approach based GNNs are expected to improve the performance of MANETs by adapting the route to the topology scenario. The contribution is based on the GNN-based Data Driven Routing (GDDR) framework developed by Oliver Hope. The DDR framework use GNNs to learn multi-commodity routing in Software Defined Networks. The DDR framework is investigated to develop an agent capable of learning a policy that balances the route according to a QoS in MANETs. The central agent observes a number of traversals on each node and should minimize the node utilization. The performance of the agent is compared to our best-known solution based on a genie approach: called Omniscient Dijkstra Routing. Both algorithms are evaluated on the tactical Anglova Company 1 MANET scenario and are assessed using real-time emulation based on the open source EMANE framework. Work in progress is reported. 
TD(22)01089Xiping Wang, Zhao Zhang, Danping He, Ke Guan, Zhangdui Zhong, Dongliang LiuRay-tracing simulation and Analysis of 5G Channel Characteristics in Dense Urban AreasRay-tracing(RT)technology has received much attention with the fast advances of the Fifth Generation (5G) wireless communication. Based on our self-developed ray-tracing platform CloudRT, 5G band (3.5 GHz-3.6 GHz, 4.8 GHz-4.9 GHz) wireless channels are simulated in 8 different dense urban areas in 4 Chinese major cities (Beijing, Shanghai, Hangzhou, Xi’an) and the second-order channel characteristics including delay spread (DS), azimuth angle spread of arrival (AASA), elevation angle spread of arrival (EASA), rays power ratio, etc, are visualized and analyzed. This is of great significance for analyzing the channel characteristics under the 5G frequency band, for better signal coverage and for other engineering issues. In addition, a large database covering multi-path and multi-dimensional channel characteristics has also been constructed. At the end, this research is concluded by highlighting several potential research directions and open issues which should be further addressed in the futureWG1, WG2, WG3
TD(22)01090Wenbin Li, Danping He, Ke Guan, Xiangyu Shi, Zhangdui ZhongMillimeter-Wave Radar Measurement and Ray-Tracing Simulation for Urban Street EnvironmentAutonomous driving technology greatly contributes to the evolution of intelligent transportation systems (ITSs). As an important sensor, the millimeter-wave (mmWave) radar plays an irreplaceable role in the overall ITSs. In this paper, we use a mmWave automotive radar system to perform measurement in an urban street environment. The radar cross section of key targets is estimated from the measurements, and it is used to calibrate the electromagnetic parameters for the ray-tracing (RT) simulation. The calibrated simulation results are compared with simulation results in distance, angle, and velocity to verify the accuracy and reliability of the RT simulator. The presented method, process, and analysis in this work can be used for the understanding of mmWave radar channel and the design of autonomous driving technology.WG1, WG2, WG4
TD(22)01091Pengxiang Xie, Ke Guan, Danping He, Haofan Yi, Jianwu Dou, Bo Ai, Zhangdui ZhongTerahertz Wave Propagation Characteristics on Rough Surfaces Based on Full-wave SimulationsTerahertz (THz) communications are considered as a critical technology for the future wireless communication. The THz band (0.1-10 THz) offers ultra-wide bandwidths and promises to satisfy the need for ultra-high-speed wireless communication. Although propagation characterization is necessary for the THz system design and validation, the response mechanism between the surface roughness and the THz wave is rarely studied. According to the Rayleigh Criterion, most surfaces in a physical environment may be considered to be rough. Scattering on rough surfaces will play a significant role in the THz propagation. Due to the difficulty of practical measurements at the THz band, we use a full-wave simulation method to study the propagation characteristics of scattering on rough surfaces. Firstly, the Monte Carlo method is used to model rough surfaces with different root-mean-squared (RMS) heights and correlation lengths. Then, with full-wave simulations, frequency dependency, material dependency, and surface size dependency are discussed. By changing the roughness of surfaces, scattering on different rough surfaces at 300 GHz are simulated. Afterwards, the cross-polarization discrimination (XPD) of each case is analyzed and counted to evaluate the depolarization effect. This study shows that scattering can become a prominent propagation mechanism with the growth of roughness. Besides, the depolarization effect becomes much more severe for a rougher surface. Studying THz propagation characteristics on rough surfaces is critical to accurate THz channel modeling, which further supports the design and deployment of THz communication systems.WG1, WG2, WG5
TD(22)01092Xiyu Wang, Pengxiang Xie, Ke Guan, Haofan Yi, Danping He, Ben Chen, Jianwu DouTCharacterization for Scattering of Rough Surfaces in the Terahertz Band Terahertz (THz) communication has been recognized by the global communication industry as a key technology for realizing the vision of the sixth-generation mobile communication (6G). The THz band (0.1-10 THz) offers ultra-wide bandwidths and promises to support diverse application scenarios in the future. According to the Rayleigh Criterion, when the wavelength is close to the surface roughness, a surface that is considered smooth at low frequencies will become rough for THz waves. Hence, the scattering on rough surfaces will play a much more important role in THz wave propagation. However, due to the great difficulty in carrying out extensive measurements in the THz band, few studies on scattering of THz waves have been conducted. In this paper, we use the full-wave simulation method to study the scattering of rough surfaces in the THz Band. By changing the roughness of surfaces, scattering on different rough surfaces at 300 GHz is simulated in Feko. Then, the directional scattering (DS) model is used to characterize the scattering on rough surfaces. The equivalent roughness alpha_R and scattering coefficient S are key parameters of the DS model. Based on the full-wave simulation results, we investigate the impact of the roughness on alpha_R and S to explore the applicability of the DS model in the THz band. Characterization for scattering of rough surfaces is crucial to the THz channel modeling towards standardization in the future, which further supports the design and deployment of THz communication systems. WG1, WG2, WG6
TD(22)01093Jean-Frédéric Wagen, Yann Maret, Nicolas Ramosaj, Franck LegendreProfiling routing and message completion rate using real time emulations and simulated fadingIn the context of beyond 5G, V2V, D2D, WiFi Mesh, etc., wireless system simulations are often use to compare new proposals to enhance the performances. When the performance metric is the link goodput or Packet Error Rate versus SINR, bit level simulations of the radio system, e.g., 802.11, and of fading via standard radio propagation models, e.g., 802.11 TG, COST IRACON Geometry-based Stochastic Channel Model or Winner, are usually favored. When multi-user system performance metrics are considered, e.g., message completion rate, the detailed radio propagation simulation approach is computationally expensive. Another approach is to pre-compute the Packet Error Rate (PER)-vs-SINR for a given Tx-Rx standard including the modulation and coding scheme used, a specific radio channel model including fading and a given packet size. Previous COST IRACON TDs contributed to show the benefit of the so-called “LOS+G” approach: using link level simulation, for 802.11n, the PER-vs-SINR curves in fading conditions, e.g., for 802.11 TGn B, can be reproduced by keeping the PER-vs-SINR for the non-fading Line-Of-Sight (LOS) case but adding, to each packet, a log-normal fading, i.e., a Gaussian random variable to the pathloss (PL) in dB: LOS+G(m=2.6, s=5). This contribution shows some results comparing the “no fading”, “LOS+G every second” and “PER-VS-SINR” approaches on the performance, here the message completion rate, for a 24 nodes MANET using real time emulations playing the realistic anglova.net pathloss scenario. The fading effects on three routing algorithms for MANET are presented. The three routing algorithms are: 1) OLSRd2 from olsr.org, 2) a so-called Omniscient Dijkstra Routing provided with non-faded pathloss values, and 3) P3ODR provided with the “LOS+G every second” values. 
TD(22)01094C. Ben Issaid (U. Oulu), C. Antón-Haro (CTTC), X. Mestre, M.S. Alouini (KAUST)Coupling Gradient-Boosting Decision Trees and Classifier Chains for Efficient User Grouping in MIMO-NOMAIn this paper, we propose a data-driven approach to group users in a Non-Orthogonal Multiple Access (NOMA) MIMO setting. Specifically, we formulate user clustering as a multi-label classification problem and solve it by coupling a Classifier Chain (CC) with a Gradient Boosting Decision Tree (GBDT), namely, the LightGBM algorithm. The performance of the proposed CC-LightGBM scheme is assessed via numerical simulations. For benchmarking, we consider two classical adaptation learning schemes: Multi-Label k-Nearest Neighbours (ML-KNN) and Multi-Label Twin Support Vector Machines (ML-TSVM); as well as other naive approaches. Besides, we also compare the computational complexity of the proposed scheme with those of the aforementioned benchmarks.WG2
TD(22)01095Paweł Sroka (presenter), Adrian KliksApplication of Edge Intelligence Tools for Autonomous Platooning – Channel Selection Use Case AnalysisEdge intelligence is one of the key paradigms related to the efficient implementation of future wireless networks. In this context, the surrounding telecommunication infrastructure is intended to support the functioning of the wireless system. Thus, it is necessary to offload some computational efforts to the system edge, allowing the infrastructure to learn and make prospective decisions. However, for the reliable realization of the edge intelligence concept, the wireless system architecture should be designed properly to minimize both storage cost and induced latency. In this paper, we compare three architecture proposals (centralized, distributed, and hybrid) tailored to the autonomous driving use case, being one of the key vertical scenarios in contemporary and future wireless networks. We evaluate the performance of an autonomous platooning system, where the operating frequency is selected dynamically with the support of infrastructure to minimize the overall interference level in the whole band. Extensive computer simulations have been carried out to analyze the impact of the induced delay in signal processing and storage efficiency.    WG1, VT2
TD(22)01096Adrian Kliks (presenter), Paweł Sroka, Paweł Kryszkiewicz, Michał Sybis Results of the channel measurement campaigns for autonomous drivingPlatooning is considered to be one of the possible prospective implementations of the autonomous driving concept, where the train-of-cars moves together following the platoon leader’s commands. However, the practical realization of this scheme assumes the use of reliable communications between platoon members. In this paper, the results of the measurement experiment have been presented showing the impact of the blocking cars on the signal attenuation. The tests have been carried out for the high-frequency band, i.e. for 26.555 GHz. It has been observed that on one hand side, the attenuation can reach even tens of dB for 2 or 3 blocking cars, but in some locations, the impact of a two-ray propagation mitigates the presence of obstructing vehicles. WG1, VT3