Search

Abstract The Internet of Things (IoT) is the recent technology intended to facilitate the daily life of humans by providing the power to connect, control and automate objects in the physical world. In this logic, the IoT helps to improve our way of producing and working in various areas (e.g. agriculture, industry, healthcare, transportation etc). Basically, an IoT network comprises physical devices, equipped with sensors and transmitters, that are interconnected with each other and/or connected to the Internet. Its main objective is to gather and transmit data to a storage system such as a server or cloud to enable processing and analysis, ultimately facilitating rapid decision‐making or enhancements to the user experience. In the realm of Connected Objects, an effective IoT data collection system plays a vital role by providing several benefits, such as real‐time data monitoring, enhanced decision‐making, increased operational efficiency etc. However, because of the resource limitations linked to connected objects, such as low memory and battery, or even single‐use devices etc. IoT data collecting presents several challenges including scalability, security, interoperability, flexibility etc. for both researchers and companies. The authors categorise current IoT data collection techniques and perform a comparative evaluation of these methods based on the topics analysed and elaborated by the authors. In addition, a comprehensive analysis of recent advances in IoT data collection is provided, highlighting different data types and sources, transmission protocols from connected sensors to a storage platform (server or cloud), the IoT data collection framework, and principles for streamlining the collection process. Finally, the most important research questions and future prospects for the effective collection of IoT data are summarised.

This study aims to detect and count people using impulse radio ultra-wideband radar and machine learning algorithms. However, the data quality, difficulty distinguishing human signals from noise and clutter, and instances where human presence is not detected make it challenging to count multiple humans. To overcome these challenges, we apply wavelet transformation to reduce signal size and use simple moving averages to eliminate noise. Next, we create features based on statistical and entropic properties of the signal and apply several classification algorithms, including ANN, Random Forest, KNN, XGBOOST, and multiple linear regression, to predict the number of people present. Our findings reveal that using the ANN classifier with the Daubechies 4 (db4) wavelet provides better results than other classifiers, with an accuracy rate of 99%. Additionally, filtering the data improves accuracy, and labeling the data after extracting essential characteristics significantly improves the model’s accuracy.

The Cloud of Things (IoT) that refers to the integration of the Cloud Computing (CC) and the Internet of Things (IoT), has dramatically changed the way treatments are done in the ubiquitous computing world. This integration has become imperative because the important amount of data generated by IoT devices needs the CC as a storage and processing infrastructure. Unfortunately, security issues in CoT remain more critical since users and IoT devices continue to share computing as well as networking resources remotely. Moreover, preserving data privacy in such an environment is also a critical concern. Therefore, the CoT is continuously growing up security and privacy issues. This paper focused on security and privacy considerations by analyzing some potential challenges and risks that need to be resolved. To achieve that, the CoT architecture and existing applications have been investigated. Furthermore, a number of security as well as privacy concerns and issues as well as open challenges, are discussed in this work.

Flood crises are the consequence of climate change and global warming, which lead to an increase in the frequency and intensity of heavy rainfall. Floods are, and remain, natural disasters that result in huge loss of lives and material damage. Flood risks threaten all countries of the globe in general. The Far-North region of Cameroon has suffered of flood crises on several occasions, resulting in significant loss of human lives, infrastructural and socio-economic damage, with the destruction of homes, crops and grazing areas, and the halting of economic activities. The models used for flood forecasting in this region are generally physical-based, and produce unsatisfactory results. The use of artificial intelligence based methods for flood forecasting in order to limit its consequences is a way to be explored in the Far-North region of Cameroon. The aims of the present research work is to design and compare the performance of Machine Learning and Deep Learning based models such as one dimensional Convolutional Neural Network, Long and Short Term Memory and Multi Layer Perceptron for short-term and long-term flood forecasting in the Far-North region of Cameroon. The models designed take as input the temperature and rainfall time series recorded in this region. Performance criteria used for evaluating models are Nash–Sutcliffe Efficiency, Percent Bias, Coefficient of Determination and Root Mean Squared Error. As the results of the design and performance comparison of the models, the best model for short-term flood forecasting is the LSTM model , and the best model for long-term flood forecasting is still the LSTM model. The best models obtained from the comparisons have satisfactory performance and good generalization capabilities, as reflected by the performance criteria. The results of our research work can be used for implementation of floods warning systems and for definition of an effective and efficient flood risk management policies in order to make the Far-North region of Cameroon more resilient to flood crises.

In this paper, we introduce a Caputo-sense fractional derivative to an existing model for the Chikungunya transmission dynamics, replacing integer derivative. After reviewing previous work on the integer order model, we suggest a change whereby the integer derivative is replaced by the Caputo derivative operator, and then we obtain results about the asymptotic stability of equilibrium points in this new fractional model. We prove the existence and uniqueness of the solution as well as the global stability of the fractional system. We then construct a numerical scheme for the fractional model, and use parameter values from the Chikungunya epidemic in Chad to perform numerical simulations. This allows us to validate our theoretical results and compare the dynamics behaviour of both models. We find from numerical simulations, that for the fractional-order parameter η in the range (0.85;1), daily detected cases are closer to those the model predict. Thus in the case of Chikungunya epidemic in Chad, the model with fractional derivative produces better results than which with integer derivative.

One of the main concerns of cybersecurity is the detection of hoaxes across social media. Hoaxers propagate such messages to mislead users and to promote violence. Several approaches exist in literature to address this issue. They are mainly limited to detect hoax activities by characterizing the message nature and detecting provenance of messages. However, unless hoaxes are detected, they continue to propagate across social media nodes. This work aims at reducing the dissemination of hoaxes across group of users. Relying on social graph structure, this research develops a mechanism based on smart contract logics to prevent a group to consume a fake post. To achieve this objective, we used a smart contract to exploit a trust index computed based on message characteristics and group features such as graph density, group status, group degree, group acceptability. Based on the value of trust index, the message is forwarded or blocked. Experiments realized on groups of different characteristics revealed that the proposed smart contract is even able to reactively block a fake post of the same nature than the group type. Results indicate that the proportion of targeted groups could be reduced even if their interests match with the message subject. This research is an important step forward to anti-promote hoaxes with the novelty of exploiting smart contract approach to contain their propagation.

The constant evolution of transportation systems and traffic in developing countries is nowadays confronted with a problem of road safety and therefore of a high accident rate, especially in the context of inter-urban road transport. In this work, we propose a communication architecture for an Intelligent Transport System (ITS) to provide surveillance in an inter-urban transport network in the context of developing countries. We introduce two types of control points: Relay Control Points (RCP) and Treatment Control Points (TrCP). We also designed two multi-objective models for the deployment of these points. In order to ensure good coverage, to minimize the cost of installation and to put more emphasis in the areas having a high number of accidents. To ensure optimal deployment of these points, we have used the Non-Dominated Sorted Genetic Algorithm II (NSGA-II) that will allow to realize the Pareto front of non-dominated solutions. The results of the simulations show the effectiveness of our proposal.

Wireless sensor networks are typically characterized by large network size and sensor nodes with low energy capacity and a limited bandwidth for data transmission. Over-activity of the sensor nodes will therefore cause many issues to the network, such as an increase in the network depletion rate and poor data transmission. Data prediction methods that exploit the inter-relationship between sensor nodes can be used to reduce data traffic across wireless networks. Several related work in data prediction do not consider the time-series distribution of the sensing data. However, exploiting sequential features of the historical observations can improve prediction accuracy, and increase the number of sensing data predicted per sequence. This work propose a sequence to sequence data prediction model, using a one dimensional layer convolutional neural network to extract spatial features from the pre-processed sensing data, and an encoder-decoder model to predict the next two outputs in the sequence by exploiting the temporal distribution of the data. The aforementioned approach has the capacity of generating more accurate information, which can reduce network traffic and energy expenditure in WSNs. Furthermore, the experimental results reveals that, based on a suitable choice of nodes, our proposed model perform accurate predictions, with reduced root mean squared error as compared to related work. We also propose an approach to regulate and control the data traffic toward the base station.

In recent years, the interest in using wireless communication technologies and mobile devices in the healthcare environment has increased. However, despite increased attention to the security of electronic health records, patient privacy is still at risk for data breaches. Thus, it is quite a challenge to involve an access control system especially if the patient’s medical data are accessible by users who have diverse privileges in different situations. Blockchain is a new technology that can be adopted for decentralized access control management issues. Nevertheless, different scalability, security, and privacy challenges affect this technology. To address these issues, we suggest a novel Decentralized Self-Management of data Access Control (DSMAC) system using a blockchain-based Self-Sovereign Identity (SSI) model for privacy-preserving medical data, empowering patients with mechanisms to preserve control over their personal information and allowing them to self-grant access rights to their medical data. DSMAC leverages smart contracts to conduct Role-based Access Control policies and adopts the implementation of decentralized identifiers and verifiable credentials to describe advanced access control techniques for emergency cases. Finally, by evaluating performance and comparing analyses with other schemes, DSMAC can satisfy the privacy requirements of medical systems in terms of privacy, scalability, and sustainability, and offers a new approach for emergency cases.

Spread spectrum techniques, such as the Chirp Spread Spectrum (CSS) used by LoRa technology, are important for machine-to-machine communication in the context of the Internet of Things. They offer high processing gain, reliable communication over long ranges, robustness to interference and noise in harsh environments, etc. However, these features are compromised by their poor spectral efficiency, resulting in a very low data transmission rate. This paper deals with a spectrally efficient variant of CSS. The system uses M-ary phase keying to modulate the data and exploits CSS’s properties to transmit the modulated symbols as overlapping chirps. The overlapping of chirp signals may affect the system performance due to inter-symbol interference. Therefore, we analyse the relationship between the number of overlaps and the effect of inter-symbol interference (ISI), and we also determine the BER expression as a function of the number of overlaps. Finally, we derive the optimal number of overlapping symbols that corresponds to the minimum error probability.

Telecommunication network infrastructures in Africa and the Middle East regions, are deployed and operated in challenging environments that are highly scattered particularly in rural areas. Moreover, considerable number of cell sites are located in areas difficult to access. Furthermore, low income in rural areas does not allow a fast return on investment since the cost of deployment and operation of a cell site is considerable. These issues lead to a difficult human resource management, particularly, in the assignment of technicians to cell site for maintenance purpose. In this paper, an optimized scheme for costs of maintenance operations on cell sites is proposed. We used the k-means clustering algorithm for allocating field technician to a pool of cell sites. Moreover, to alleviate the k-means sensitivity to initialization, we proposed an initialization method that is based on the geometry of a sphere. We conducted series of experiments with sample of thousands of cell towers from OpenCellID and the results demonstrate the effectiveness of the proposal.

The Internet of Things (IoT) invention has taken the growth of sensors technology to a completely high step. New challenges in terms of data delivery have emerged due to strict QoS conditions. Among the solutions proposed in the literature is the subdivision of the large-scale network into several clusters. Except that most of these solutions are conventional. However, prior research generally confirms that bio-inspired paradigms are more flexible and effective compared to traditional methods. When it comes to a heterogeneous network, additional constraints appear. Nodes have different buffer sizes. Then, data captured must be sent before their buffers are full, otherwise, some data will be lost. This is not suitable for a real-time application where time and information are crucial elements. In this study, a comprehensive overview of the use of sensors in IoT contexts is performed. Two algorithms as Grey Wolf Optimizer (GWO) and Whale Optimization Algorithm (WOA), combined with the Imperialist Competitive Algorithm (ICA) based Cluster Head (CH) selection with a novel approach for heterogeneous networks are proposed. These algorithms can support data exchange over a heterogeneous Wireless Sensor Network (WSN) infrastructure with taking into consideration the buffer overflow problem. Simulation results are presented and discussed in different network designs. The research demonstrated that knowing well how to manage buffers using bio-inspired techniques, leads to a significant reduction in data loss.

Internet of Vehicles (IoV) has the potential to enhance road-safety with environment sensing features provided by embedded devices and sensors. This benignant feature also raises privacy issues as vehicles announce their fine-grained whereabouts mainly for safety requirements, adversaries can leverage this to track and identify users. Various privacy-preserving schemes have been designed and evaluated, for example, mix-zone, encryption, group forming, and silent-period-based techniques. However, they all suffer inherent limitations. In this paper, we review these limitations and propose WHISPER, a safety-aware location privacy-preserving scheme that adjusts the transmission range of vehicles in order to prevent continuous location monitoring. We detail the set of protocols used by WHISPER, then we compare it against other privacy-preserving schemes. The results show that WHISPER outperformed the other schemes by providing better location privacy levels while still fulfilling road-safety requirements.

In recent years, technological advances and the ever-increasing power of embedded systems have seen the emergence of so-called smart cities. In these cities, application needs are increasingly calling for Large-Scale Wireless Sensor Networks (LS-WSN). However, the design and implementation of such networks pose several important and interesting challenges. These low-cost, low-power devices are characterized by limited computing, memory storage, communication, and battery power capabilities. Moreover, sensors are often required to cooperate in order to route the collected data to a single central node (or sink). The many-to-one communication model that governs dense and widely deployed Wireless Sensor Networks (WSNs) most often leads to problems of network overload and congestion. Indeed, it is easy to show that the closer a node is geographical to the sink, the more data sources it has to relay. This leads to several problems including overloading of nodes close to the sink, high loss rate in the area close to the sink, and poor distribution of power consumption that directly affects the lives of these networks. In this context, we propose a contribution to the problem of LS-WSN energy consumption. We designed a hierarchical 3-tier architecture of LS-WSNs coupled with a modeling of the activities of the different sensors in the network. This architecture that is based on clustering also includes a redeployment function to maintain the topology in case of coverage gaps. The results of the performed simulations show that our architecture maximizes the lifetime than compared solutions.

Large Scale Wireless Sensor Networks (LS-WSNs) are Wireless Sensor Networks (WSNs) composed of an impressive number of sensors, with inherent detection and processing capabilities, to be deployed over large areas of interest. The deployment of a very large number of diverse or similar sensors is certainly a common practice that aims to overcome frequent sensor failures and avoid any human intervention to replace them or recharge their batteries, to ensure the reliability of the network. However, in practice, the complexity of LS-WSNs pose significant challenges to ensuring quality communications in terms of symmetry of radio links and maximizing network life. In recent years, most of the proposed LS-WSN deployment techniques aim either to maximize network connectivity, increase coverage of the area of interest or, of course, extend network life. Few studies have considered the choice of a good LS-WSN deployment strategy as a solution for both connectivity and energy consumption efficiency. In this paper, we designed a LS-WSN as a tool for collecting big data generated by smart cities. The intrinsic characteristics of big data require the use of heterogeneous sensors. Furthermore, in order to build a heterogeneous LS-WSN, our scientific contributions include a model of quantifying the kinds of sensors in the network and the multi-level architecture for LS-WSN deployment, which relies on clustering for the big data collection. The results simulations show that our proposed LS-WSN architecture is better than some well known WSN protocols in the literature including Low Energy Adaptive Clustering Hierarchy (LEACH), E-LEACH, SEP, DEEC, EECDA, DSCHE and BEENISH.

Data collection is one of the main operations performed in Wireless Sensor Networks (WSNs). Even if several interesting approaches on data collection have been proposed during the last decade, it remains a research focus in full swing with a number of important challenges. Indeed, the continuous reduction in sensor size and cost, the variety of sensors available on the market, and the tremendous advances in wireless communication technology have potentially broadened the impact of WSNs. The range of application of WSNs now extends from health to the military field through home automation, environmental monitoring and tracking, as well as other areas of human activity. Moreover, the expansion of the Internet of Things (IoT) has resulted in an important amount of heterogeneous data that are produced at an exponential rate. Furthermore, these data are of interest to both industry and in research. This fact makes their collection and analysis imperative for many purposes. In view of the characteristics of these data, we believe that very large-scale and heterogeneous WSNs can be very useful for collecting and processing these Big Data. However, the scaling up of WSNs presents several challenges that are of interest in both network architecture to be proposed, and the design of data-routing protocols. This paper reviews the background and state of the art of Big Data collection in Large-Scale WSNs (LS-WSNs), compares and discusses on challenges of Big Data collection in LS-WSNs, and proposes possible directions for the future.

MEC and IoT are rapidly expanding technologies that offer numerous opportunities to enhance efficiency and application performance. However, the huge volume of data generated by IoT devices, coupled with computational and latency constraints, poses data processing challenges. To address this within the MEC architecture, deploying computing servers at the network edge near IoT devices is a promising approach. This reduces latency and traffic load on the core network while improving the user experience. However, offloading computations task from IoT devices to MEC servers and efficiently allocating computing resources is a complex problem. IoT tasks may have specific requirements in terms of latency, bandwidth and energy efficiency, while computing resources and capacities maybe limited or shared between several users. We propose an approach called FedMeta2Ag, which we evaluate using the MNIST database. With 20 epochs, the training accuracy reached 91.5%, while the test accuracy achieved 92.0%. Performance consistently improved during the initial 20 iterations and gradually stabilized thereafter. Additionally, we compared the performance of our proposed model with existing methods, finding that our approach outperforms existing models in predicting performance more accurately. Thus, this approach effectively meets the demanding performance requirements of wireless communication systems. [ABSTRACT FROM AUTHOR]

Predictive maintenance contributes to Industry 4.0, as it enables a decrease in maintenance costs and downtime while aiming to increase production and return on investment. Despite the increasing utilization of machine learning techniques in predictive maintenance in industrial systems over the past few years, several challenges remain to be addressed in the implementation of ML, including the quality of the data collected, resource constraints, and equipment heterogeneity. This study proposes an adaptive framework for predictive maintenance in the context of Industry 4.0, specifically in internet of things (IoT) systems, using machine learning (ML) models. In particular, this study introduces PdM-FSA, a new framework based on an ensemble classifier that takes advantage of four widely adopted ML models in the predictive maintenance literature: random forest (RF), support vector machine (SVM), extreme gradient boosting (XGBoost), and k-nearest neighbors (KNN). The performance evaluation results showed that the PdM-FSA framework can perform well for predictive maintenance according to the severity of equipment malfunctions in a smart factory. The results of this study provide significant knowledge to researchers and practitioners on predictive maintenance in the context of Industry 4.0. and enables the optimization of processes and improves productivity. [ABSTRACT FROM AUTHOR]