Public air quality monitoring relies on expensive monitoring stations which are highly reliable and accurate but require significant maintenance and cannot be used to form a high spatial resolution measurement grid. Recent technological advances have enabled air quality monitoring that uses low-cost sensors. Being inexpensive and mobile, with wireless transfer support, such devices represent a very promising solution for hybrid sensor networks comprising public monitoring stations supported by many low-cost devices for complementary measurements. However, low-cost sensors can be influenced by weather and degradation, and considering that a spatially dense network would include them in large numbers, logistically adept solutions for low-cost device calibration are essential. In this paper, we investigate the possibility of a data-driven machine learning calibration propagation in a hybrid sensor network consisting of One public monitoring station and ten low-cost devices equipped with NO₂, PM₁₀, relative humidity, and temperature sensors. Our proposed solution relies on calibration propagation through a network of low-cost devices where a calibrated low-cost device is used to calibrate an uncalibrated device. This method has shown an improvement of up to 0.35/0.14 for the Pearson correlation coefficient and a reduction of 6.82 µg/m³/20.56 µg/m³ for the RMSE, for NO₂ and PM₁₀, respectively, showing promise for efficient and inexpensive hybrid sensor air quality monitoring deployments.

In this paper, we explore the impact of the COVID-19 lockdown in Serbia on the air pollution levels of CO, NO₂ and PM₁₀ alongside the possibility for low-cost sensor usage during this period. In the study, a device with low-cost sensors collocated with a reference public monitoring station in the city of Belgrade is used for the same period of 52 days in 2019 (pre-COVID-19 period), 2020 (COVID-19 lockdown) and 2021 (post-COVID-19 period). Low-cost sensors’ measurements are improved by using a convolutional neural network that applies corrections of the influence of temperature and relative humidity on the low-cost sensors. As a result of this study we have noticed a remarkable decrease in NO₂ (primarily related to traffic density), while on the other hand CO and PM₁₀, related to domestic heating sources and heating plants, showed constant or slightly higher levels. The obtained results are in accordance with other published work in this area. The low-cost sensors have shown a satisfactory correlation with the reference CO measurements during the lockdown, while the NO₂ and PM₁₀ measurements of 2020 were corrected using a convolutional neural network trained on meteorological and pollutant data from 2019. The results include an improvement of 0.35 for the R2 of NO₂ and an improvement of 0.13 for the R2 of PM₁₀, proving that our neural network model trained on data from 2019 can improve the performance of the sensor in the lockdown period in 2020. This means that our neural network model is very robust, as it exhibits good performance even in the case where training data from the prior year (2019) are used in the following year (2020) in very different environment circumstances—a lockdown.

This paper explores the challenges and constraints when over the horizon (OTH) maritime surveillance service utilizes an Internet of Things (IoT) as its backbone. The service is based on High Frequency Surface Wave Radars (HFSWRs) and relies on a satellite communication network as its communication infrastructure in harsh environments. The complete IoT OTH maritime surveillance network is currently deployed in the Gulf of Guinea, which due to its tropical climate represents an unfavorable environment for sensors and communications. In this paper, we have examined the service performance under various meteorological conditions specific to the Gulf of Guinea. To the best of our knowledge, this is the first analysis of IoT OTH maritime surveillance service in equatorial environment. Our analysis aims to mathematically describe the impact of harsh weather conditions on the performance of the service in order to mitigate it with careful overall system design and provide constant quality of the service. Analyses presented in the paper show that average service latency is about 90 s, but it can rise to about 120 s, which is used as a key information during the sensor data fusion algorithm design. Validity of the analyses is demonstrated through high quality of service with an outage probability of just 0.1% in the driest months up to the 0.7% in the rainiest months. The work presented here can be used as a guideline for deployment of maritime surveillance service solutions in other equatorial regions. Moreover, the gained experience presented in this paper will significantly facilitate future expansions of the existing maritime surveillance network with more HFSWRs. This will be done in such a way that it will not affect the quality of service of the entire system on a large scale.

The steady increase of the world population and economy leads to an increase in both types and amounts of goods transported over seas, which further inevitably leads to an increase of criminal activities in the maritime arena. In order to stifle criminal activities nations are forced to develop sophisticated sensor networks. The backbone of any sensor network is a communication network which connects all sensors with the command centers, most often located hundreds of kilometers away from the sensors. In developing countries, communication networks are very often poorly developed, leaving only satellite links as somewhat reliable means of communication. Henceforth, in this paper, a laboratory for the Internet of Things (IoT) communication infrastructure environment designed to facilitate maritime sensor network design process in areas where communication network is dependent on data transfer over satellite links is presented. In order to successfully describe and develop a laboratory for IoT communication infrastructure environment, necessary data are collected during the design and deployment of a maritime surveillance network in the Gulf of Guinea. The main advantage of the proposed laboratory environment is the inclusion of satellite link simulation in the IoT laboratory environment. This feature provides an opportunity to cover a much broader scope of IoT solutions compared to other IoT laboratories.

Pseudo-random number generators (PRNGs) are used in many applications (cryptography, traffic simulations, etc.) and have significant influence on their performances. In this paper, the PRNG based on irrational numbers will be introduced. It will be proved that under the given set of conditions the maximum period of the generated sequence can be guaranteed. Then, the proposed PRNG implementation will be presented and discussed. It will be shown that the generated numbers are uncorrelated. Also, the influence that the finite number of digits used to represent the irrational numbers in computer systems has on the period of the generated sequence will be analyzed.

Large telecom-service-provider networks are typically based on complex communications infrastructures comprising millions of network devices. The performance monitoring of such networks is a very demanding and challenging task. A large amount of data is collected and processed during performance monitoring to obtain information that gives insights into the current network performance. Using the obtained information, providers can efficiently detect, locate, and troubleshoot weak spots in the network and improve the overall network performance. Furthermore, the extracted information can be used for planning future network expansions and to support the determination of business-strategy decisions. However, traditional methods for processing and storing data are not applicable because of the enormous amount of collected data. Thus, big-data technologies must be used. In this paper, a big-data platform for the performance monitoring of telecom-service-provider networks is proposed. The proposed platform is capable of collecting, storing, and processing data from millions of devices. Typical challenges and problems in the development and deployment process of the platform, as well as the solutions to overcome them, are presented. The proposed platform is adjusted to HFC (Hybrid Fiber-Coaxial) network and currently operates in the real HFC network, comprising millions of users and devices.