Papers by Mihael Mohorcic
FSO in HAP-Based Communication Systems
John Wiley & Sons, Ltd eBooks, Aug 27, 2010
The Impact of Different Scheduling Policies on Traffic Class Dependent Routing in Intersatellite Link
ESA Special Publication, Jul 1, 2003
UWB Motion Detection Data Set
<strong>Introduction</strong> This data set includes a collection of measurements usi... more <strong>Introduction</strong> This data set includes a collection of measurements using DecaWave DW1000 UWB radios in two indoor environments used for motion detection functionality. Measurements include channel impulse response (CIR) samples in form of power delay profile (PDP) with corresponding timestamps for three channels for each indoor environment. Data set includes pieces of Python code and Jupyter notebooks for data loading, analysis and to reproduce the results of a paper entitled "UWB Radio Based Motion Detection System for Assisted Living" submitted to MDPI Sensors. The data set will require around 10 GB of total free space after extraction. The code included in the data set is written and tested on Linux (Ubuntu 20.04) and requires 16 GB of RAM and additional SWAP partition to run properly. The code can be modified to consume less memory but it requires unnecessary additional work. If the .npy format is compatible with your numpy version, you won't need to regenerate npy data from .csv files. <strong>Data Set Structure</strong> The resulting folder after extracting the uwb_motion_detection.zip file is organized as follows: <strong>data</strong> subfolder: contains all original .csv and intermediate .npy data files. models pdp: this folder contains 4 .csv files with raw PDP measurements (timestamp + PDP). The data format will be discussed in the following section. pdp_diff: this folder contains .npy files with PDP samples and .npy files with timestamps. Those files are generated by running the generate_pdp_diff.py script. generate_pdp_diff.py validation subfolder: contains data for motion detection validation events: contains .npy files with motion events for validation. The .npy files are generated using generate_event_x.py files or notebooks inside the /Process/validation folder. pdp: this folder contains raw PDP measurements in .csv format. pdp_diff: this folder contains .npy files with PDP samples and .npy files with timestamps. Those files are generated by running the generate_pdp_diff.py script. [...]
IEEE Access, 2018
Indoor localization is one of the key enablers for various application and service areas that rel... more Indoor localization is one of the key enablers for various application and service areas that rely on precise locations of people, goods, and assets, ranging from home automation and assisted living to increased automation of production and logistic processes and wireless network optimization. Existing solutions provide various levels of precision, which also depends on the complexity of the indoor radio environment. In this paper, we propose two methods for reducing the localization error in indoor nonline-of-sight (NLoS) conditions using raw channel impulse response (CIR) information obtained from ultra-wide band radios requiring no prior knowledge about the radio environment. The methods are based on NLoS channel classification and ranging error regression models, both using convolutional neural networks (CNNs) and implemented in the TensorFlow computational framework. We first show that NLoS channel classification using raw CIR data outperforms existing approaches that are based on derived input signal features. We further demonstrate that the predicted NLoS channel state and predicted ranging error information, used in combination with least squares (LS) and weighted LS location estimation algorithms, significantly improve indoor localization performance. We also evaluate the computational performance and suitability of the proposed CNN-based algorithms on various computing platforms with a wide range of different capabilities and show that in a distributed localization system, they can also be used on computationally restricted devices. INDEX TERMS Channel impulse response, convolutional neural network, deep learning, indoor localization, non-line-of-sight, ranging error mitigation, ultra-wide band.
Modelling of Atmospheric Impairments in Mobile Propagation Channel for Stratospheric Communications
Bluetooth-based Mobile Gateway for Wireless Sensor Network
AbstractSensor nodes are being increasingly deployed for monitoring different phenomena in a sta... more AbstractSensor nodes are being increasingly deployed for monitoring different phenomena in a stand-alone manner or as a part of larger wireless sensor networks (WSN). In addition to protocols for communication within the WSN, sensor nodes may also support ...
Combining Measurements and Simulations for Evaluation of Tracking Algorithms
Wireless device position tracking has been already thoroughly studied in literature. Most of the ... more Wireless device position tracking has been already thoroughly studied in literature. Most of the studies rely on the presumption that location information is acquired based on range measurements that are performed in a very short period of time. However, in time-division-multiple-access (TDMA) two-way-ranging (TWR) ultra-wideband (UWB) wireless localization networks, those ranging measurements are always spread in time by significant time delays. Those delays have negative impact on the tracking performance and the effects of ranging in these systems should be evaluated accordingly. In this paper we propose a time-of-flight (ToF) simulation-based approach for indoor tracking algorithm evaluation with a measurement calibration which enables changing the size of TDMA slots and thus observing the tracking performance degradation. A constant velocity movement model with a random curvature of walking path is proposed to simulate the person's random walking pattern inside the room as naturally as possible. With a proposed simulation framework the impact of individual hyperparameter changes on tracking performance can be evaluated.
Eurasip Journal on Wireless Communications and Networking, Jun 26, 2019
Precise location information will play an important role in 5G networks, their applications and s... more Precise location information will play an important role in 5G networks, their applications and services, especially in indoor environments. Ultra-wideband (UWB) technology offers exceptional temporal resolution enabling the emergence of high accuracy ranging-based indoor localization systems. In order to reduce time to market, developers need a reliable, fast and efficient method to evaluate localization and tracking system designs in the selected high-dimensional parameter space. Purely measurement-based performance evaluation of such systems is costly and cumbersome, so we propose the use of a radio frequency (RF) ray-tracing simulator augmented with a noise model based on measurements with localization equipment in real environment. We demonstrate the proposed approach by evaluating the UWB-based two-way-ranging (TWR) localization with the least squares (LS) and with the extended Kalman filter (EKF) tracking algorithms. We analyze the degradation of tracking performance due to time spreading of range measurements. Moreover, on a set of fixed locations we also show that the proposed approach is sufficiently representative for the pure measurement-based evaluation. The results obtained in a reference office environment show that EKF algorithm is twice as efficient and more resilient to the effects of time spreading of range measurements as LS algorithm.
Aeronautics and Energetics
John Wiley & Sons, Ltd eBooks, Aug 27, 2010
Performance evaluation of adaptive routing algorithms in ISL networks
Performance evaluation of adaptive routing algorithms in packet-switched intersatellite link networks
International Journal of Satellite Communications, 2002
This paper addresses the performance evaluation of adaptive routing algorithms in non‐geostationa... more This paper addresses the performance evaluation of adaptive routing algorithms in non‐geostationary packet‐switched satellite communication systems. The dynamic topology of satellite networks and variable traffic load in satellite coverage areas, due to the motion of satellites in their orbit planes, pose stringent requirements to routing algorithms. We have limited the scope of our interest to routing in the intersatellite link (ISL) segment. In order to analyse the applicability of different routing algorithms used in terrestrial networks, and to evaluate the performance of new algorithms designed for satellite networks, we have built a simulation model of a satellite communication system with intersatellite links. In the paper, we present simulation results considering a network‐uniform source/destination distribution model and a uniform source–destination traffic flow, thus showing the inherent routing characteristics of a selected Celestri‐like LEO satellite constellation. The updates of the routing tables are centrally calculated according to the Dijkstra shortest path algorithm. Copyright © 2002 John Wiley &amp; Sons, Ltd.
Cross-Polarization Discrimination Interference Analysis of Alphasat Satellite Measurements in Ka and Q Bands
2023 17th European Conference on Antennas and Propagation (EuCAP)
Future Development of HAPs and HAP-Based Applications
John Wiley & Sons, Ltd eBooks, Aug 27, 2010
arXiv (Cornell University), May 17, 2023
The so-called black-box deep learning (DL) models are increasingly used in classification tasks a... more The so-called black-box deep learning (DL) models are increasingly used in classification tasks across many scientific disciplines, including wireless communications domain. In this trend, supervised DL models appear as most commonly proposed solutions to domain-related classification problems. Although they are proven to have unmatched performance, the necessity for large labeled training data and their intractable reasoning, as two major drawbacks, are constraining their usage. The self-supervised architectures emerged as a promising solution that reduces the size of the needed labeled data, but the explainability problem remains. In this paper, we propose a methodology for explaining deep clustering, self-supervised learning architectures comprised of a representation learning part based on a Convolutional Neural Network (CNN) and a clustering part. For the state of the art representation learning part, our methodology employs Guided Backpropagation to interpret the regions of interest of the input data. For the clustering part, the methodology relies on Shallow Trees to explain the clustering result using optimized depth decision tree. Finally, a data-specific visualizations part enables connection for each of the clusters to the input data trough the relevant features. We explain on a use case of wireless spectrum activity clustering how the CNN-based, deep clustering architecture reasons.
HANNA: Human-friendly provisioning and configuration of smart devices
Engineering Applications of Artificial Intelligence, Nov 1, 2023
IEEE Access, 2019
Uplink transmissions, within coexisting distinct sub-GHz technologies operating in the same unlic... more Uplink transmissions, within coexisting distinct sub-GHz technologies operating in the same unlicensed band, can be exposed to detrimental impact of the interference. In such scenarios, transmission scheduling becomes important for mitigating interference or minimizing the impact of the interference. For this purpose, we aim to whitelist relatively better channels in terms of their yielded packet reception ratio using our proposed channel quality metric that is based on the received signal-to-interference-plusnoise ratio. In this paper, we investigate the trade-offs of the channel whitelisting in random frequency division multiple access (RFDMA) networks in the presence of the cumulative intra-and inter-technology interferences. Our main findings indicate that, although channel whitelisting reduces the degree of freedom, and thus the overall capacity, it empowers a certain amount of devices to be served at a much lower received signal power, whereas this is infeasible for non-whitelisting scenarios at larger received signal power, which signifies the energy conservation ability of our proposed whitelisting method. It is experimentally demonstrated, on Sigfox, a particular type of RFDMA network, that non-whitelisting scenarios are not capable of supporting any devices at a received signal power below −118 dBm. Even for lower received signal power, we are able to reduce the required number of retransmissions at the same reception probability, which indeed indicates that the overall reliability of the network is improved. INDEX TERMS Aloha, inter-technology interference, Internet of things, RFDMA, whitelisting.
PLOS ONE, Jun 22, 2018
Lack of unallocated spectrum and increasing demand for bandwidth in wireless networks is forcing ... more Lack of unallocated spectrum and increasing demand for bandwidth in wireless networks is forcing new devices and technologies to share frequency bands. Spectrum sensing is a key enabler for frequency sharing and there is a large body of existing work on signal detection methods. However a unified methodology that would be suitable for objective comparison of detection methods based on experimental evaluations is missing. In this paper we propose such a methodology comprised of seven steps that can be applied to evaluate methods in simulation or practical experiments. Using the proposed methodology, we perform the most comprehensive experimental evaluation of signal detection methods to date: we compare energy detection, covariance-based and eigenvalue-based detection and cyclostationary detection. We measure minimal detectable signal power, sensitivity to noise power changes and computational complexity using an experimental setup that covers typical capabilities from low-cost embedded to high-end software defined radio devices. Presented results validate our premise that a unified methodology is valuable in obtaining reliable and reproducible comparisons of signal detection methods.
IEEE Communications Magazine, Dec 1, 2018
Network testing plays an important role in the iterative process of developing new communication ... more Network testing plays an important role in the iterative process of developing new communication protocols and algorithms. However, test environments have to keep up with the evolution of technology and require continuous update and redesign. In this paper, we propose COINS, a framework that can be used by wireless technology developers to enable continuous integration (CI) practices in their testbed infrastructure. As a proof-ofconcept, we provide a reference architecture and implementation of COINS for controlled testing of multi-technology 5G Machine Type Communication (MTC) networks. The implementation upgrades an existing wireless experimentation testbed with new software and hardware functionalities. It blends web service technology and operating system virtualization technologies with emerging Internet of Things technologies enabling CI for wireless networks. Moreover, we also extend an existing qualitative methodology for comparing similar frameworks and identify and discuss open challenges for wider use of CI practices in wireless technology development.
Feature Management for Machine Learning Operation Pipelines in AI Native Networks
The Internet of Things (IoT) is being widely adopted in today's society, interconnecting smart em... more The Internet of Things (IoT) is being widely adopted in today's society, interconnecting smart embedded devices that are being deployed for indoor and outdoor environments, such as homes, factories and hospitals. Along with the growth in the development and implementation of these IoT devices, their simple and rapid deployment, initial configuration and outof-the-box operational provisioning are becoming prominent challenges to be circumvented. Considering a large number of heterogeneous devices to be deployed within next generation IoT networks, the amount of time needed for manual provisioning of these IoT devices can significantly delay the deployment and manual provisioning may introduce human-induced failures and errors. By incorporating zero-touch provisioning (ZTP), multiple heterogeneous devices can be provisioned with less effort and without human intervention. In this paper, we propose softwareenabled access point (Soft-AP)-and Bluetooth-based ZTP solutions relying only on a single mediator device and evaluate their performances using LOG-A-TEC testbed against manual provisioning in terms of the time required for provisioning (timeto-provision, TTP). We demonstrate that on average, Soft-APand Bluetooth-based ZTP solutions outperform manual provisioning with about 154% and 313% when compared to the expert provisioning, and with about 434% and 880% when compared to the non-expert provisioning in terms of TTP performances, respectively. Index Terms-Zero-touch provisioning, automated configuration, embedded device, Internet of things.
Uploads
Papers by Mihael Mohorcic