Since the emergence of wireless communication networks, the quality aspects of wireless links are among core concerns of the research community. The analysis of the existing literature on link quality estimation (LQE) using models developed from data traces indicates that the techniques used for modeling link quality estimation are becoming increasingly sophisticated. A number of recent estimators leverage machine learning (ML) techniques that require sophisticated design and development process, each of which has a great potential to significantly affect the overall model performance. In this paper, we provide a comprehensive survey on link quality estimators developed from empirical data and then focus on the subset that use ML algorithms. We analyze ML-based LQE models from two perspectives using performance data. Firstly, we focus on how they address quality requirements that are important from the perspective of the applications they serve. Secondly, we analyze how they approach the standard design steps commonly used in the ML community. Next, we review existing open source datasets suitable for LQE research. Finally, we round up the paper with the lessons learned and design guidelines for ML-based LQE development and dataset collection.
COBISS.SI-ID: 48389891
The satellite communications in Ka-band (19.7 GHz) and Q-band (39.4 GHz) provide significant capacity increase due to the availability of wider bandwidths. Thus, the experimental measurement campaigns are of crucial importance for the investigation and modeling the effects of the impairment phenomena, rain in particular, and subsequently for the development of appropriate mitigation techniques. In this paper, we presented the statistical analysis of signal rain attenuation and fade durations obtained from three years of measurements of the Alphasat satellite beacons in Ka and Q bands in the ground receiver, developed and deployed in Ljubljana at Jozef Stefan Institute. The experiment was part of the European propagation measurement campaign within the cooperation projects of the European Space Agency (ESA). We achieved very high 99.6% data availability. Complementary Cumulative Distribution Functions (CCDF) of rain attenuation are presented for each year separately and for three years combined. The analysis of CCDFs for each separate season shows that higher attenuation values are observed during summer and spring, compared to autumn and winter, which is due to higher and more intense rain precipitation. The experimental data presented in this paper are important for further analysis, development and adjusting of channel prediction models at Ka/Q bands and will be included in data banks by ITU-R Study Group 3 "Radiowave Propagation".
COBISS.SI-ID: 41917443
Radio propagation in natural caves attract little attention. Lately, modal analyses have given acceptable results in some cave passages while multiple bends and highly irregular passages remained problematic. Recently introduced rapid capture laser scanners enables creating accurate 3D models of cave passages which permits, for the first time, ray tracing to be applied to such geometrically complicated environments. The primary aim of this work was to further knowledge into cave communication and the applicability of ray tracing in this unique environment. In addition, it is also intended that the results will be of practical value to those with a requirement for communication in natural cave passages in general. In the paper an evaluation of a ray tracing approach was therefore undertaken in three selected cave passages. In particular, the issues of cave model simplification and handling of roughness have been studied and results were compared to modal analyses results at three different frequencies. Acceptable correlation was demonstrated despite the difficulty in accurately measuring or estimating average roughness. In particular ray tracing results were more accurate in eight out of the nine combinations of location and frequency compared to modal approach.
COBISS.SI-ID: 33088551
After decades of research, Internet of Things (IoT) is finally permeating real-life and helps improve the efficiency of various infrastructures and working processes as well as our health. As massive number of IoT devices are deployed, they naturally incur great operational costs to ensure intended operations. To effectively handle such intended operations in massive IoT networks, automatic detection of malfunctioning, namely anomaly detection, becomes a critical and challenging task. In this paper, motivated by a real-world experimental IoT deployment, we introduce four types of wireless network anomalies that are identified at the link layer, and we study the performance of different classifiers to automatically detect these anomalies. We examine the relative performance of three supervised and three unsupervised machine learning (ML) techniques on both non-encoded and encoded (using autoencoder) feature representations.
COBISS.SI-ID: 38799107
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-plus-noise 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.
COBISS.SI-ID: 32819495