arXiv subjects

Benjamin Sliwa

Publications and source records attributed to Benjamin Sliwa.

At least 19 recordsLinked to original sources

arXiv2022

System Modeling and Performance Evaluation of Predictive QoS for Future Tele-Operated Driving

Future Tele-operated Driving (ToD) applications place challenging Quality of Service (QoS) demands on existing mobile communication networks that are of highly important to comply with for safe operation. New remote control and platooning services will emerge and pose high data rate and latency requirements. One key enabler for these applications is the newly available 5G New Radio (NR) promising higher bandwidth and lower latency than its predecessors. In addition to that, public 5G networks do not consistently deliver and do not guarantee the required data rates and latency of ToD. In this paper, we discuss the communication-related requirements of tele-operated driving. ToD is regarded as a complex system consisting of multiple research areas. One key aspect of ToD is the provision and maintenance of the required data rate for teleoperation by the mobile network. An in-advance prediction method of the end-to-end data rate based on so-called Radio Environmental Maps (REMs) is discussed. Furthermore, a novel approach improving the prediction accuracy is introduced and it features individually optimized REM layers. Finally, we analyze the implementation of tele-operated driving applications on a scaled vehicular platform combined with a cyber-physical test environment consisting of real and virtual objects. This approach enables large-scale testing of remote operation and autonomous applications.

cs.NI↗

arXiv2021

DRaGon: Mining Latent Radio Channel Information from Geographical Data Leveraging Deep Learning

Radio channel modeling is one of the most fundamental aspects in the process of designing, optimizing, and simulating wireless communication networks. In this field, long-established approaches such as analytical channel models and ray tracing techniques represent the de-facto standard methodologies. However, as demonstrated by recent results, there remains an untapped potential to innovate this research field by enriching model-based approaches with machine learning techniques. In this paper, we present Deep RAdio channel modeling from GeOinformatioN (DRaGon) as a novel machine learning-enabled method for automatic generation of Radio Environmental Maps (REMs) from geographical data. For achieving accurate path loss prediction results, DRaGon combines determining features extracted from a three-dimensional model of the radio propagation environment with raw images of the receiver area within a deep learning model. In a comprehensive performance evaluation and validation campaign, we compare the accuracy of the proposed approach with real world measurements, ray tracing analyses, and well-known channel models. It is found that the combination of expert knowledge from the communications domain and the data analysis capabilities of deep learning allows to achieve a significantly higher prediction accuracy than the reference methods.

cs.NI↗

arXiv2021

Machine Learning-Enabled Data Rate Prediction for 5G NSA Vehicle-to-Cloud Communications

In order to satisfy the ever-growing Quality of Service (QoS) requirements of innovative services, cellular communication networks are constantly evolving. Recently, the 5G NonStandalone (NSA) mode has been deployed as an intermediate strategy to deliver high-speed connectivity to early adopters of 5G by incorporating Long Term Evolution (LTE) network infrastructure. In addition to the technological advancements, novel communication paradigms such as anticipatory mobile networking aim to achieve a more intelligent usage of the available network resources through exploitation of context knowledge. For this purpose, novel methods for proactive prediction of the end-to-end behavior are seen as key enablers. In this paper, we present a first empirical analysis of client-based end-to-end data rate prediction for 5G NSA vehicle-to-cloud communications. Although this operation mode is characterized by massive fluctuations of the observed data rate, the results show that conventional machine learning methods can utilize locally acquirable measurements for achieving comparably accurate estimations of the end-to-end behavior.

cs.NI↗

arXiv2021

Modeling and Simulation of Reconfigurable Intelligent Surfaces for Hybrid Aerial and Ground-based Vehicular Communications

The requirements of vehicular communications grow with increasing level of automated driving and future applications of intelligent transportation systems (ITS). Beside the ever-increasing need for high capacity radio links, reliability and latency constraints challenge the mobile network supply. While for example the millimeter-wave spectrum and THz-bands offer a vast amount of radio resources, their applicability is limited due to delicate radio channel conditions and signal propagation characteristics. Reconfigurable intelligent surfaces (RISs) as part of smart radio environments (SREs) of future ITS infrastructure promise improved radio link qualities by means of purposeful cultivation of passive reflections. With this, obstructed mmWave or THz beams can be guided around obstacles through RIS reflection paths to improve the otherwise limited coverage. In this article, application use cases of RIS-enhanced vehicular communications are proposed. Beside static deployments of RISs at exterior walls of buildings, unmanned aerial vehicles (UAV) could provide reflection capabilities on demand, while future vehicles could - in a visionary approach - consist of meta-material allowing for their opportunistic utilization within an enriched SRE. Results of a case study based on our multi-scale mobility and network simulation model clearly highlight the potential of RIS deployment for hybrid vehicular communication scenarios. Path loss and outage percentages can be reduced considerably.

cs.NI↗

arXiv2021

Towards Machine Learning-Enabled Context Adaption for Reliable Aerial Mesh Routing

In this paper, we present Context-Adaptive PARRoT (CA-PARRoT) as an extension of our previous work Predictive Ad-hoc Routing fueled by Reinforcement learning and Trajectory knowledge (PARRoT). Short-term effects, as occurring in urban surroundings, have shown to have a negative impact on the Reinforcement Learning (RL)-based routing process. Therefore, we add a timer-based compensation mechanism to the update process and introduce a hybrid Machine Learning (ML) approach to classify Radio Environment Prototypes (REPs) with a dedicated ML component and enable the protocol for autonomous context adaption. The performance of the novel protocol is evaluated in comprehensive network simulations considering different REPs and is compared to well-known established routing protocols for Mobile Ad-hoc Networks (MANETs). The results show, that CA-PARRoT is capable to compensate the challenges confronted with in different REPs and to improve its Key Performance Indicators (KPIs) up to 23% compared to PARRoT, and outperform established routing protocols by up to 50 %.

cs.NI↗

arXiv2021

Client-Based Intelligence for Resource Efficient Vehicular Big Data Transfer in Future 6G Network

Vehicular big data is anticipated to become the "new oil" of the automotive industry which fuels the development of novel crowdsensing-enabled services. However, the tremendous amount of transmitted vehicular sensor data represents a massive challenge for the cellular network. A promising method for achieving relief which allows to utilize the existing network resources in a more efficient way is the utilization of intelligence on the end-edge-cloud devices. Through machine learning-based identification and exploitation of highly resource efficient data transmission opportunities, the client devices are able to participate in overall network resource optimization process. In this work, we present a novel client-based opportunistic data transmission method for delay-tolerant applications which is based on a hybrid machine learning approach: Supervised learning is applied to forecast the currently achievable data rate which serves as the metric for the reinforcement learning-based data transfer scheduling process. In addition, unsupervised learning is applied to uncover geospatially-dependent uncertainties within the prediction model. In a comprehensive real world evaluation in the public cellular networks of three German Mobile Network Operators (MNOs), we show that the average data rate can be improved by up to 223 % while simultaneously reducing the amount of occupied network resources by up to 89 %. As a side-effect of preferring more robust network conditions for the data transfer, the transmission-related power consumption is reduced by up to 73 %. The price to pay is an increased Age of Information (AoI) of the sensor data.

cs.NI↗

arXiv2021

A Low Cost Modular Radio Tomography System for Bicycle and Vehicle Detection and Classification

The advancing deployment of ubiquitous Internet of Things (IoT)-powered vehicle detection and classification systems will successively turn the existing road infrastructure into a highly dynamical and interconnected Cyber-physical System (CPS). Though many different sensor systems have been proposed in recent years, these solutions can only meet a subset of requirements, including cost-efficiency, robustness, accuracy, and privacy preservation. This paper provides a modular system approach that exploits radio tomography in terms of attenuation patterns and highly accurate channel information for reliable and robust detection and classification of different road users. Hereto, we use Wireless Local Area Network (WLAN) and Ultra-Wideband (UWB) transceiver modules providing either Channel State Information (CSI) or Channel Impulse Response (CIR) data. Since the proposed system utilizes off-the-shelf and power-efficient embedded systems, it allows for a cost-efficient ad-hoc deployment in existing road infrastructures. We have evaluated the proposed system's performance for cyclists and other motorized vehicles with an experimental live deployment. In this concern, the primary focus has been on the accurate detection of cyclists on a bicycle path. However, we also have conducted preliminary evaluation tests measuring different motorized vehicles using a similar system configuration as for the cyclists. In summary, the system achieves up to 100% accuracy for detecting cyclists and more than 98% classifying cyclists and cars.

eess.SP↗

arXiv2021

Flying Robots for Safe and Efficient Parcel Delivery Within the COVID-19 Pandemic

The integration of small-scale Unmanned Aerial Vehicles (UAVs) into Intelligent Transportation Systems (ITSs) will empower novel smart-city applications and services. After the unforeseen outbreak of the COVID-19 pandemic, the public demand for delivery services has multiplied. Mobile robotic systems inherently offer the potential for minimizing the amount of direct human-to-human interactions with the parcel delivery process. The proposed system-of-systems consists of various complex aspects such as assigning and distributing delivery jobs, establishing and maintaining reliable communication links between the vehicles, as well as path planning and mobility control. In this paper, we apply a system-level perspective for identifying key challenges and promising solution approaches for modeling, analysis, and optimization of UAV-aided parcel delivery. We present a system-of-systems model for UAV-assisted parcel delivery to cope with higher capacity requirements induced by the COVID-19. To demonstrate the benefits of hybrid vehicular delivery, we present a case study focusing on the prioritization of time-critical deliveries such as medical goods. The results further confirm that the capacity of traditional delivery fleets can be upgraded with drone usage. Furthermore, we observe that the delay incurred by prioritizing time-critical deliveries can be compensated with drone deployment. Finally, centralized and decentralized communication approaches for data transmission inside hybrid delivery fleets are compared.

cs.NI↗

arXiv2020

PARRoT: Predictive Ad-hoc Routing Fueled by Reinforcement Learning and Trajectory Knowledge

Swarms of collaborating Unmanned Aerial Vehicles (UAVs) that utilize ad-hoc networking technologies for coordinating their actions offer the potential to catalyze emerging research fields such as autonomous exploration of disaster areas, demanddriven network provisioning, and near field packet delivery in Intelligent Transportation Systems (ITSs). As these mobile robotic networks are characterized by high grades of relative mobility, existing routing protocols often fail to adopt their decision making to the implied network topology dynamics. For addressing these challenges, we present Predictive Ad-hoc Routing fueled by Reinforcement learning and Trajectory knowledge (PARRoT) as a novel machine learning-enabled routing protocol which exploits mobility control information for integrating knowledge about the future motion of the mobile agents into the routing process. The performance of the proposed routing approach is evaluated using comprehensive network simulation. In comparison to established routing protocols, PARRoT achieves a massively higher robustness and a significantly lower end-to-end latency.

cs.NI↗

arXiv2020

Deep Learning-based Signal Strength Prediction Using Geographical Images and Expert Knowledge

Methods for accurate prediction of radio signal quality parameters are crucial for optimization of mobile networks, and a necessity for future autonomous driving solutions. The power-distance relation of current empirical models struggles with describing the specific local geo-statistics that influence signal quality parameters. The use of empirical models commonly results in an over- or under-estimation of the signal quality parameters and require additional calibration studies. In this paper, we present a novel model-aided deep learning approach for path loss prediction, which implicitly extracts radio propagation characteristics from top-view geographical images of the receiver location. In a comprehensive evaluation campaign, we apply the proposed method on an extensive real-world data set consisting of five different scenarios and more than 125.000 individual measurements. It is found that 1) the novel approach reduces the average prediction error by up to 53% in comparison to ray-tracing techniques, 2) A distance of 250-300 meters spanned by the images offer the necessary level of detail, 3) Predictions with a root-mean-squared error of approximately 6 dB is achieved across inherently different data sources.

cs.NI↗

arXiv2020

The Best of Both Worlds: Hybrid Data-Driven and Model-Based Vehicular Network Simulation

The analysis of the end-to-end behavior of novel mobile communication methods in concrete evaluation scenarios frequently results in a methodological dilemma: Real world measurement campaigns are highly time-consuming and lack of a controllable environment, the derivation of analytical models is often not possible due to the immense system complexity, system-level network simulations imply simplifications that result in significant derivations to the real world observations. In this paper, we present a hybrid simulation approach which brings together model-based mobility simulation, multi-dimensional Radio Environmental Maps (REMs) for efficient maintenance of radio propagation data, and Data-driven Network Simulation (DDNS) for fast and accurate analysis of the end-to-end behavior of mobile networks. For the validation, we analyze an opportunistic vehicular data transfer use-case and compare the proposed method to real world measurements and a corresponding simulation setup in Network Simulator 3 (ns-3). In comparison to the latter, the proposed method is not only able to better mimic the real world behavior, it also achieves a 300 times higher computational efficiency.

cs.NI↗

arXiv2020

Acting Selfish for the Good of All: Contextual Bandits for Resource-Efficient Transmission of Vehicular Sensor Data

as a novel client-based method for resource-efficient opportunistic transmission of delay-tolerant vehicular sensor data. BS-CB applies a hybrid approach which brings together all major machine learning disciplines - supervised, unsupervised, and reinforcement learning - in order to autonomously schedule vehicular sensor data transmissions with respect to the expected resource efficiency. Within a comprehensive real world performance evaluation in the public cellular networks of three Mobile Network Operators (MNOs), it is found that 1) The average uplink data rate is improved by 125%-195% 2) The apparently selfish goal of data rate optimization reduces the amount of occupied cell resources by 84%-89% 3) The average transmission-related power consumption can be reduced by 53%-75% 4) The price to pay is an additional buffering delay due to the opportunistic medium access strategy.

cs.NI↗

arXiv2020

6G White Paper on Machine Learning in Wireless Communication Networks

The focus of this white paper is on machine learning (ML) in wireless communications. 6G wireless communication networks will be the backbone of the digital transformation of societies by providing ubiquitous, reliable, and near-instant wireless connectivity for humans and machines. Recent advances in ML research has led enable a wide range of novel technologies such as self-driving vehicles and voice assistants. Such innovation is possible as a result of the availability of advanced ML models, large datasets, and high computational power. On the other hand, the ever-increasing demand for connectivity will require a lot of innovation in 6G wireless networks, and ML tools will play a major role in solving problems in the wireless domain. In this paper, we provide an overview of the vision of how ML will impact the wireless communication systems. We first give an overview of the ML methods that have the highest potential to be used in wireless networks. Then, we discuss the problems that can be solved by using ML in various layers of the network such as the physical layer, medium access layer, and application layer. Zero-touch optimization of wireless networks using ML is another interesting aspect that is discussed in this paper. Finally, at the end of each section, important research questions that the section aims to answer are presented.

cs.IT↗

arXiv2020

The Channel as a Traffic Sensor: Vehicle Detection and Classification based on Radio Fingerprinting

Ubiquitously deployed Internet of Things (IoT)- based automatic vehicle classification systems will catalyze data-driven traffic flow optimization in future smart cities and will transform the road infrastructure itself into a dynamically sensing Cyber-physical System (CPS). Although a wide range of different traffic sensing systems has been proposed, the existing solutions are not yet able to simultaneously satisfy the multitude of requirements, e.g., accuracy, robustness, cost-efficiency, and privacy preservation. In this paper, we present a novel approach, which exploits radio fingerprints - multidimensional attenuation patterns of wireless signals - for accurate and robust vehicle detection and classification. The proposed system can be deployed in a highly cost-efficient manner as it relies on off-the-shelf embedded devices which are installed into existing delineator posts. In a comprehensive field evaluation campaign, the performance of the radio fingerprinting-based approach is analyzed within an experimental live deployment on a German highway, where it is able to achieve a binary classification success ratio of more than 99% and an overall accuracy of 93.83% for a classification task with seven different classes.

cs.NI↗

arXiv2020

Simulating Hybrid Aerial- and Ground-based Vehicular Networks with ns-3 and LIMoSim

Integrating Unmanned Aerial Vehicles (UAVs) into future Intelligent Transportation Systems (ITSs) allows to exploit their unique mobility potentials for improving the performance of services such as near-field parcel delivery, dynamic network provisioning, and aerial sensing. In order to provide a controllable environment for the methodological performance analysis, simulation frameworks need to support ground- and aerial-based mobility as well as the involved communication technologies. In this paper, we present the open source Lightweight ICT-centric Mobility Simulation (LIMoSim) framework for simulating hybrid vehicular networks within Network Simulator 3 (ns-3). LIMoSim implements a shared codebase coupling approach which integrates all required components in a single simulation process. The low-level mobility behaviors rely on well-known analytical models. Different case studies discussing cutting-edge communication technologies such as Cellular Vehicle-to-Everything (C-V2X) and millimeter Wave (mmWave) are presented in order to illustrate how the proposed framework can be integrated into ns-3-based network simulation setups.

cs.NI↗

arXiv2020

Rushing Full Speed with LTE-Advanced is Economical -- A Power Consumption Analysis

Boosting data rates in LTE mobile networks is one of the key features of LTE-Advanced. This improved user experience is achieved by Carrier Aggregation (CA), in which the available spectrum of an operator is bundled out of several frequency bands. Accordingly, the user equipment has to supply multiple reception chains and therefore consumes considerably more power during a transmission. On the other hand, transmissions terminate faster, which enables a quick switchover into energy-saving mode. In order to examine these opposed facts, empirical analyses of existing devices are first carried out. Subsequently, we present a new CA enhancement of an existing context-aware power consumption model which incorporates the development density of the environment and the mobile device mobility. Based on the extended model we perform a detailed power consumption analysis and show that CA leads to power savings of 31% if the data rate doubled for large file transmissions. In addition, we show that CA can lead to power savings even from a data rate increase of 25%, regardless of mobility and urban development density. Besides, the measurement results show that CA operated in the same band leads to a lower power consumption than inter-band CA.

cs.NI↗

arXiv2020

Machine Learning Based Uplink Transmission Power Prediction for LTE and Upcoming 5G Networks using Passive Downlink Indicators

Energy-aware system design is an important optimization task for static and mobile Internet of Things (IoT)-based sensor nodes, especially for highly resource-constrained vehicles such as mobile robotic systems. For 4G/5G-based cellular communication systems, the effective transmission power of uplink data transmissions is of crucial importance for the overall system power consumption. Unfortunately, this information is usually hidden within off-the-shelf modems and mobile handsets and can therefore not be exploited for enabling green communication. Moreover, the dynamic transmission power control behavior of the mobile device is not even explicitly modeled in most of the established simulation frameworks. In this paper, we present a novel machine learning-based approach for forecasting the resulting uplink transmission power used for data transmissions based on the available passive network quality indicators and application-level information. The model is derived from comprehensive field measurements of drive tests performed in a public cellular network and can be parameterized for integrating all measurements a given target platform is able to provide into the prediction process. In a comparison of three different machine learning methods, Random-Forest models thoroughly performed best with a mean average error of 3.166 dB. As the absolute sum of errors converges towards zero and falls below 1 dB after 28 predictions in average, the approach is well-suited for long-term power estimations.

cs.NI↗

arXiv2020

LIMITS: Lightweight Machine Learning for IoT Systems with Resource Limitations

Exploiting big data knowledge on small devices will pave the way for building truly cognitive Internet of Things (IoT) systems. Although machine learning has led to great advancements for IoT-based data analytics, there remains a huge methodological gap for the deployment phase of trained machine learning models. For given resource-constrained platforms such as Microcontroller Units (MCUs), model choice and parametrization are typically performed based on heuristics or analytical models. However, these approaches are only able to provide rough estimates of the required system resources as they do not consider the interplay of hardware, compiler specific optimizations, and code dependencies. In this paper, we present the novel open source framework LIghtweight Machine learning for IoT Systems (LIMITS), which applies a platform-in-the-loop approach explicitly considering the actual compilation toolchain of the target IoT platform. LIMITS focuses on high level tasks such as experiment automation, platform-specific code generation, and sweet spot determination. The solid foundations of validated low-level model implementations are provided by the coupled well-established data analysis framework Waikato Environment for Knowledge Analysis (WEKA). We apply and validate LIMITS in two case studies focusing on cellular data rate prediction and radio-based vehicle classification, where we compare different learning models and real world IoT platforms with memory constraints from 16 kB to 4 MB and demonstrate its potential to catalyze the development of machine learning enabled IoT systems.

cs.NI↗