Milutin Radonjic
@ucg.ac.me
Professor at Faculty of Electrical Engineering
University of Montenegro
RESEARCH, TEACHING, or OTHER INTERESTS
Electrical and Electronic Engineering, Computer Engineering, Computer Networks and Communications, Hardware and Architecture
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Igor Stanovčić, Milutin Radonjić, Enis Kočan, Božidar Škrbić, and Božo Krstajić
IEEE
This paper presents a proposal for the approach to potential collision detection of non-SOLAS vessels using an e-navigation system based on modern IoT systems. Although there are significantly more non-SOLAS vessels than SOLAS vessels worldwide, the lack of basic navigation equipment excludes them from the navigation safety system. The implementation of the proposed innovative IoT concept and the described architecture, using available smart devices, would significantly increase the navigation safety of non-SOLAS vessels, and thus the overall safety at sea.
Suada Hadzovic, Sasa Mrdovic, and Milutin Radonjic
Institute of Electrical and Electronics Engineers (IEEE)
As technology is the driver of the economy, it is necessary to follow emerging technological trends and to create appropriate conditions for its adoption and implementation as a human-centred technology. In this regard, rules and standards for the Internet of Things (IoT) and Artificial Intelligence (AI) should be established to best use the benefits of technology and to prevent or minimize the consequences of technology misuse. The fifth industrial revolution (Industry 5.0) has already begun, although Industry 4.0 is still developing. Consequently, the original attention has shifted from IoT to AI, with the IoT debate now being a prerequisite for the AI debate. As AI is transforming our lives, a growing number of countries have considered or already adopted national AI strategies. However, in many developing countries, national AI strategies and initiatives for establishing AI and IoT regulation and legislation frameworks yet need to be discussed. The subject of this article is the research of existing initiatives related to establishing the IoT and AI regulatory and legislative framework in the EU and its applicability in developing countries.
Igor Stanovčić, Rino Bošnjak, Milutin Radonjić, and Božo Krstajić
IEEE
This paper presents the concept of increasing navigation safety, especially from the point of view of Non SOLAS vessels and in areas with limited navigation space. The concept consists of a symbiosis of the traditional Traffic Separation Scheme with the concept of E-navigation and VTS as a service provider. The idea is that Non SOLAS vessels, using smartphones with the appropriate application and IoT cloud platform (server and appropriate software), become identified by other vessels and VTS.
Milutin Radonjić, Žarko Zečević, and Božo Krstajić
MDPI AG
The excavators are heavy machines widely used in the civil engineering and surface mining industry. Recent studies show that 95% of contractors face the problem of finding skilled operators. Unskilled operators not only worsen productivity but also very often cause machine failures through unprofessional handling. Motivated by these studies and guided by the mining company’s requirements, we present a prototype of an IoT system for monitoring DC motor overload on the EKG-15 excavator. The IoT system consists of a microprocessor device mounted inside the excavator and an external cloud platform that can be accessed via the Internet. The proposed solution detects and warns the operator when the DC motor overload occurs, thus reducing the probability of its damage. In addition, overload data is sent to the cloud platform for later research, analysis and processing. The main benefit of the proposed solution is that it can be applied to existing industry machinery, thus reducing the maintenance cost and increasing productivity. After several months of use of the proposed system in real working conditions, it has been shown that the overload occurrence and its duration time are approximately reduced by 60% and 80%, respectively.
Milutin Radonjić, Sanja Vujnović, Aleksandra Krstić, and Žarko Zečević
MDPI AG
Modern predictive maintenance techniques have been significantly improved with the development of Industrial Internet of Things solutions which have enabled easier collection and analysis of various data. Artificial intelligence-based algorithms in combination with modular interconnected architecture of sensors, devices and servers, have resulted in the development of intelligent maintenance systems which outperform most traditional machine maintenance approaches. In this paper, a novel acoustic-based IoT system for condition detection of rotating machines is proposed. The IoT device designed for this purpose is mobile and inexpensive and the algorithm developed for condition detection consists of a combination of discrete wavelet transform and neural networks, while a genetic algorithm is used to tune the necessary hyperparameters. The performance of this system has been tested in a real industrial setting, on different rotating machines, in an environment with strong acoustic pollution. The results show high accuracy of the algorithm, with an average F1 score of around 0.99 with tuned hyperparameters.
Budimir Andelic, Milutin Radonjic, and Slobodan Djukanovic
IEEE
Illegal logging represents a major environmental issue which impedes the stability of forest ecosystem and supports climate change, flooding, soil erosion, weeding of habitats, extinction of animal and plant species. This paper proposes a method for mitigating the logging issue by automatically detecting the sound of logging activities. More specifically, we propose to detect the chainsaw sound using deep learning. Two deep learning approaches were considered, one based on multilayer perception (MLP) and the other based on convolutional neural network (CNN). As inputs to our models, we used time, frequency and time-frequency audio features. For this research, we collected two datasets of audio signals. First dataset, downloaded from YouTube, is used for training and validating the proposed models. Second dataset, which we recorded in a real environment, is used for testing of the proposed models. The experiments have shown that the CNN-based approach outperforms the MLP-based one, with a sound classification accuracy of 94.96% on the first dataset and 88.87% on the second dataset.
Nikola Žarić, Milutin Radonjić, Nikola Pavlićević, and Sanja Paunović Žarić
MDPI AG
Numerous researchers are working on Ambient Assisted Living systems to enable more comfortable and safer living for senior people in their homes. Due to modern lifestyles and an aging population, this has become a very important issue. According to the available literature, it is obvious that the kitchen is one of the most important and most dangerous rooms in the home. However, there is still evident lack of monitoring systems suitable for specific kitchen activities. In this paper, we propose a monitoring system capable of identifying activities related to the cooking process, and a decision-making system capable of identifying some unwanted and possibly critical conditions. The proposed systems are designed to satisfy the requirements of the modern Ambient Assisted Living systems dedicated to older adults. The proposed monitoring system consists of ultrasound, temperature, and humidity sensors. The acquired results from these sensors are the inputs for the decision-making system, which generate warnings or alarms intended for the senior users and/or formal or informal caregivers. This system is designed to improve home safety related to kitchen activities, as well as to provide information about the lifestyle and daily activities of senior users. Experimental validation of the proposed system confirms its functionality and accurate design approach.
Suada Hadzovic, Sasa Mrdovic, and Milutin Radonjic
MDPI AG
The Internet of Things (IoT) is a leading trend with numerous opportunities accompanied by advantages as well as disadvantages. Parallel with IoT development, significant privacy and personal data protection challenges are also growing. In this regard, the General Data Protection Regulation (GDPR) is often considered the world’s strongest set of data protection rules and has proven to be a catalyst for many countries around the world. The concepts and interaction of the data controller, the joint controllers, and the data processor play a key role in the implementation of the GDPR. Therefore, clarifying the blurred IoT actors’ relationships to determine corresponding responsibilities is necessary. Given the IoT transformation reflected in shifting computing power from cloud to the edge, in this research we have considered how these computing paradigms are affecting IoT actors. In this regard, we have introduced identification of IoT actors according to a new five-computing layer IoT model based on the cloud, fog, edge, mist, and dew computing. Our conclusion is that identifying IoT actors in the light of the corresponding IoT data manager roles could be useful in determining the responsibilities of IoT actors for their compliance with data protection and privacy rules.
Milutin Radonjic and Bozo Krstajic
IEEE
This paper discusses possible solution for data transfer between elements of the system for estimation of the state of rotating coal mills in thermal power plant, based on recorded sound. The system consists of the portable device for sound acquisition and state estimation and dedicated server for data storage, post processing and visualization. In the present phase of the system development, GPRS connection showed a good enough performance for necessary data transfer and public cloud platform proved suitability as a dedicated server.
Milutin Radonjic, Goran Kvascev, Milovan Radulovic, and Bozo Krstajic
IEEE
In this paper we propose one possible solution of microprocessor hardware platform for acquisition of sound signals from coal mill in thermal power plant. Proposed platform is based on Raspberry Pi 3B microcomputer, running Raspbian operating system. Sound acquisition is performed by command-line sound recorder arecord, which is part of the software package available with this operating system. Experimental results showed that this platform provides recordings comparable to recordings acquired by professional sound recorder, for particular application.
Nikola Pavlicevic, Nikola Zaric, and Milutin Radonjic
IEEE
The trend of population aging and increasing number of people with cognitive and other impairments caused more demands in Ambient Assisted Living solutions development. The kitchen space and cooking process are identified as a critical in this domain, from security and assistive aspects. In this paper, we present a short overview of technical solutions related to the cooker monitoring and corresponding sensors. In order to improve existing solutions, we invoke the ultrasound sensor utilization for cookware detection, liquid level measurement and detection of user activities regarding the cooking process. We have setup the testing prototype based on ultrasound sensor and acquired some experimental results, which will be presented.
Trifun Savic and Milutin Radonjic
IEEE
In this paper, an architecture of a ZigBee wireless sensor network for application in smart irrigation system is proposed. After a brief overview of the most important features of ZigBee standard related to Wireless Sensor Network (WSN) operation, Arduino-based sensor node for acquisition of soil moisture and air temperature is described. This node is equipped with XBee S2 communication module in order to be part of the established WSN. Furthermore, we presented the reception of the collected data on the main control unit, designed on the Waspmote microcontroller platform equipped with XBee S2 coordinator. All necessary steps required to establish the communication between Waspmote platform and XBee S2 coordinator device are described, even though Waspmote platform is designed to communicate only with Xbee routers and end devices. Proposed architecture allows flexible implementation with reliable data transfer that provides necessary information for autonomous decision-making in smart irrigation system.
Andrija Mitrovic, Slobodan Djukanovic, and Milutin Radonjic
IEEE
In this paper, we propose an algorithm for classification of three signals: sink tap, toilet and shower tap, implemented on the Raspberry Pi platform. Signals are analyzed in the frequency domain, and features used for classification are concentration measure and the amount of energy in certain frequency bands. The process consists of 1) sound acquisition from a microphone; 2) data processing and 3) signaling with led diodes. The proposed algorithm presents an efficient way to classify sink tap, toilet and shower tap into proper signal groups, and it can be used in the bathrooms, to classify the water flow signals.
Trifun Savic and Milutin Radonjic
IEEE
One possible solution for automated irrigation system is presented in this paper. Proposed solution allows flexible control of the three electromagnetic valves for releasing water to agricultural fields. It provides possibility for users to set up time schedule for irrigation, where closing of the valves can be scheduled based on duration of irrigation or amount of water that is released. System is implemented with Libelium Waspmote open source platform, supported by GPRS communication module. Configuration and control of the system can be performed manually, on the main control unit, or remotely, by commands sent over the SMS service. Information about current configuration and amount of released water are sent to user by SMS report and to remote server for further analysis and graphical presentation. Proposed solution provides easy and flexible control of the irrigation process and provides optimal water consumption.
Milutin Radonjic, Nikola Ljumovic, Danilo Misovic, Ivo Maljevic, Kenji Yoshigoe, and Igor Radusinovic
Springer Science and Business Media LLC
Gordana Gardašević, Mladen Veletić, Nebojša Maletić, Dragan Vasiljević, Igor Radusinović, Slavica Tomović, and Milutin Radonjić
Springer Science and Business Media LLC
Trifun Savic and Milutin Radonjic
IEEE
One possible weather station solution based on the Raspberry Pi platform is described in this paper. Information about current air temperature, air humidity, atmospheric pressure and relative altitude is collected from appropriate sensors. Received data are processed and transmitted from measuring point to the distant location via UDP protocol. Sensors data acquisition and UDP server are implemented on Raspberry Pi platform. On the other side, an android application which requests and presents acquired weather data at distant location is developed. Proposed solution provides real-time data acquisition and transfer of measured parameters, like much more expensive commercial systems. Moreover, it is very flexible and can be easily adapted to various applications.
Slavica Tomovic, Milutin Radonjic, and Igor Radusinovic
IEEE
In this paper we analyzed suitability of different routing algorithms for establishing bandwidth-delay constrained traffic tunnels in SDN backbone networks. Bandwidth rejection ratio was used as indicator of algorithm's ability to efficiently utilize network resources. Through set of simulations on two real backbone topologies we showed that network topology can affect algorithm's performance, and can lead to biased results. Also, we pointed out problem of high complexity of existing traffic engineering (TE) solutions, which often cannot perform calculations in real-time. Since in SDN networks the controller processing latency can turn out to be a non-negligible factor in the total round-trip latency, we explained heuristic techniques for reducing computation time of bandwidth-delay constrained algorithms and provided illustrative results showing how they impact BRR performance.
Nikola Zaric, Milutin Radonjic, Milica Pejanovic-Djurisic, and Igor Radusinovic
IEEE
Numerous Ambient Assisted Living systems have been developed in the last decade, but there is still an evident lack of monitoring systems that are suitable for some of the specific user's activities. That is certainly the case with the ones related to the bathroom environment, where the strict respect of functional and privacy requirements are imposed. The paper proposes a monitoring system that is able to identify the bathroom activities taking into account requirements that an Ambient Assisted Living systems dedicated to older adults should satisfy. Proposed monitoring system consists of presence sensor, humidity sensor, and microphone-based sensor that provide autonomous and automated classification of different signals representing possible activities. Based on the sensors' data, an efficient reasoning module has been proposed, which is able to detect critical situations for user and home safety and to provide information for lifestyle pattern reasoning and daily function monitoring module.
Gordana GARDASEVIC, Soko DIVANOVIC, Milutin RADONJIC, and Igor RADUSINOVIC
Institute of Electrical and Electronics Engineers (IEEE)
Nikola Zaric, Milutin Radonjic, Sofoklis Kyriazakos, and Milica Pejanovic Djurisic
IEEE
Characterization of a home water-flow signal, as an outcome of an analysis of the most appropriate representation in time, frequency and time-frequency domain, has not been adequately studied; mainly due to lack of such specific need. Nevertheless, being able to characterize a water-flow signal in the home-environment can support Ambient Assisted Living systems to determine the activity of a senior citizen, thus contributing to the daily pattern recognition, as well as to identify critical situation. This has triggered our research activities to develop an automated decision-making algorithm for detection of different water-flows that may be found in an ordinary bathroom. The preliminary results will be presented in this paper. The time-frequency representations are employed to analyze and characterize different water-flow signals, as well as to provide input parameters for the algorithm. These approach and algorithm can be used in numerous applications where automated decision-making scenario is required.
Igor Radusinovic, Soko Divanovic, and Milutin Radonjic
IEEE
In this paper we study the influence of Weighted Round Robin (WRR) scheduling algorithm frame size to the performance of Crosspoint Queued (CQ) crossbar switch. In order to show that throughput and cell delay can be adjusted with appropriate WRR frame size, we analyzed switch for different values of frame size and under the unbalanced bursty traffic. We show that WRR scheduling algorithm achieve throughput similar to Output Queued switch when frame size is low. Also, with an increase of the frame size, the delay decreases.
Milutin Radonjic, Ivo Maljevic, Nedjeljko Lekic, and Igor Radusinovic
IEEE
The performance analysis of the variable packet size crosspoint queued switch is presented in this paper. Packet switch throughput and average latency are evaluated under Interrupted Bernoulli Process incoming traffic pattern. It is shown that among the observed algorithms, the longest queue first algorithm has the highest throughput with short crosspoint buffers, but the highest average latency. Also, we establish that the choice of the scheduling algorithm does not play a significant role in the switch performance if the buffers are long enough. Therefore, the round robin algorithm becomes the best choice for implementation due to its simplicity.
Soko Divanovic, Milutin Radonjic, Gordana Gardasevic, and Igor Radusinovic
IEEE
Achievement of desired performance levels (quality of service - QoS) in switches and routers is one of the most important tasks in switching systems. This implies providing guaranteed throughput, cell loss probability, average and maximal latency within the required bounds. To achieve this task in crosspoint queued switch, we implemented the weighted round robin (WRR) algorithm. We showed that WRR algorithm can achieve very good performance levels regarding the throughput and latency, but had major drawback because it require knowledge of arrival traffic. To overcome this problem, we implemented and presented in this paper the dynamic weighted round robin (DWRR) algorithm that can work with unknown arrival traffic. We showed that DWRR can achieve same performance as WRR, without the need for incoming traffic information, which makes it suitable for practical implementation.
Nebojsa Maletic, Soko Divanovic, Milutin Radonjic, Igor Radusinovic, and Gordana Gardasevic
IEEE
This paper presents an analysis of new Dual Crosspoint Queued (DCQ) crossbar switch architecture. It is well-known that in order to achieve performance guarantees, the priority separation of incoming traffic is necessary. Therefore, we modified CQ crossbar switch with the separation of different traffic flows, by implementing two buffers in each crosspoint of switch fabric. Different round robin based scheduling algorithms are presented and analyzed. With the particular attention on the cell delay and cell loss probability, results of simulations show that proposed switch architecture can achieve desired QoS only for the high priority traffic flow.