In 2013, the International Agency for Research on Cancer classified particulate matter (PM) as carcinogenic to humans. It is therefore essential to measure PM concentrations to minimize the exposure of individuals. Our objective was to investigate personal exposure to PM2.5 (PM with diameter <= 2.5 mu m) in Grenoble (France) during commuting in different transportation modes: bike, walk, bus and tramway. PM2.5 measurements were found to be the highest for bikes, followed by walk, bus, and tramway. In this study, conducted in spring during low pollution levels of PM, exposure levels are greatly influenced by the time of day. Pedestrian and cyclists' exposure generally stayed under background reference values. Exposure in public transportation was usually below reference values, but when background PM2.5 levels went lower (evening), levels registered in the tramway or bus reached those of the reference. Therefore, public transport users could be less exposed than active commuters, except when ambient pollutant levels are low. Environmental parameters like wind might be important in Grenoble, and it would be worthwhile to reproduce this study at a time when wind speed is lower.
This paper proposes two different approaches to be applied in gateway placement problem in LoRaWAN sensor networks. The first approach is based on finding the minimal set to contain all the coverage intersections of the sensors and the second approach is based on optimization via integer programming over the distance between the gateways and sensors. Our results show that using automated gateway placement provides significantly less number of gateways to be used.
Considering the ongoing growth of Wireless Sensor Networks (WSNs) and the challenges they pose due to their hardware limitations as well as the intrinsic complexity of their interactions, specialized algorithms have the potential to help solving these challenges. We present a survey on recent developments regarding algorithmic problems which have applications in wireless systems and WSNs in particular. Focusing on the intersection between WSNs and algorithms, we give an overview of recent results inside this intersection, concerning topics such as routing, interference minimization, latency reduction, localization among others. Progress on solving these problems could be potentially beneficial for the industry as a whole by increasing network throughput, reducing latency or making systems more energy-efficient. We summarize and structure these recent developments and list interesting open problems to be investigated in future works.
The emergence of the Internet of things is highly related to the development of wireless sensor networks (WSNs) and their evolving protocols, such as Internet Protocol version 6 (IPv6) over Low-Power Wireless Personal Area Networks (6LoWPAN). Providing security within a sensor network, including achieving authentication between WSN nodes, is critical. The node and the server create an encryption session key for future communications. Therefore, developing a lightweight and efficient authentication and key establishment (AKE) scheme is imperative. Symmetric cryptographic and public key-based AKE methods have been developed to address these issues. Nevertheless, some known attacks and large communication and computational overheads remain as problems for the developed solutions. This study proposes a secure and lightweight authenticated encryption scheme for 6LoWPAN (SLAE6) that uses a lightweight hash function and an authenticated encryption primitive, known as ACE, to enable the AKE process to occur securely. SLAE6 is effective in dealing with computing and communication complexities while simultaneously withstanding well-known attacks. First, SLAE6 validates the authenticity of information from sensor networks (SNs) and then establishes a secret key between an SN and the server to guarantee security. The proposed system is proven reliable on the basis of the Canetti-Krawczyk and Dolev-Yao threat models. In addition, SLAE6 is logically demonstrated to be exact through Burrows-Abadi-Needham logic. Compared with other schemes, SLAE6 is lightweight, efficient, and requires less bandwidth and shorter execution time.
The development of Highly Automated Driving (HAD) systems is necessary for automated vehicles in terms of various complex functionalities. HAD systems consist of complex structures containing different types of sensors. The functionality of HAD systems needs be tested to ensure the overall safety of automated vehicles. Methods such as real-world testing require a large number of driving miles and are enormously expensive and time-consuming. Therefore, simulation-based testing is widely accepted and applicable in the development of HAD systems, including sensor performance improvement. In order to identify the functional insufficiency of such sensors that affect the safety of HAD systems, it is critical to test these sensors extensively under a variety of conditions such as, road types, environment and traffic situations. Based on this motivation, the main contributions of this paper are as follows: First, a simulation-based test concept of radar sensors with methods for the Safety Of the Intended Functionality (SOTIF) use case is presented. Second, a specific radar effect is evaluated through simulation-based testing of two different radar models to support and realize the sensor's functional insufficiency. Finally, the development of a filter is proposed to improve the sensor performance considering the radar specific multipath propagation effects.
The demand for sensor-based information has seen a rapidly increasing demand due to the massive deployment of Structural Health Monitoring (SHM) systems. SHM allows for the analysis aimed to the prediction of forthcoming incidents and enables the evaluation of a structure's status. The advances in the Internet of Things (IoT) structures to retrieve data anytime, everywhere through the internet represents a promising paradigm for SHM. Among the various technologies and topologies that are now evolving, Wireless Sensor Networks (WSNs) have become well suited for the implementation of monitoring systems, especially in low power wide area network (LPWAN) structures. LoRa modulation technology is a suitable technical solution for sensor node communication. In this study, two LoRa-based systems for Structural Health Monitoring (SHM) are presented, located in Sicily and Calabria, Italy. Accelerometric sensors are encapsulated into solar harvesting powered sensor nodes and are used to monitor the variation of inclinations of the mounting location. LoRaWAN gateways interface the nodes towards the internet, enabling the Internet of Things (IoT) paradigm for the monitoring solution. In this article, an overview of the system structure is given, with nodes and gateways' hardware features provided. Inclination monitoring using accelerometric data is explained, and real scenario recorded data are given. Brief power analysis for the sensor nodes is also reported.
Interactions among employees at a workplace facilitate spread of infectious diseases. This paper proposes to integrate traditional IoT sensor data, wastewater analysis and data from organizational information systems for timely identification of threats and adjustment of work activities. The overall approach combining predictive, preventive and prescriptive capabilities is described as well as the overall technical solution is presented. The proposed approach allows tailoring of work activities depending on macro and micro monitoring results in a non-intrusive manner.
Forensics investigations need to be conducted efficiently and accurately especially in situations where time is a scarce resource. Novel technologies, like forensic sensors, can aid investigators in trace detection, visualisation, identification and interpretation on site. Arising from the need to connect different sensors to a remote digital management software, a network-enabled Sensor API is proposed to enable any compliant CBRNe Sensor to connect and exchange information in a harmonised and interoperable way. As a result, a Standardisation Workshop agreement, on CBRNe SENSOR API - Network Protocols, Data Formats and Interfaces, was initiated to promote standardisation of the Sensor API. The new proposed standard will allow sensor manufacturers to focus on sensor development work, benefitting from already defined interfaces and data models. Moreover, forensic investigators, acting as end-users, can better understand and analyse (well defined) sensor outputs, thus improving their work efficiency and facilitating technology acceptance.
Efficiently querying data on embedded sensor and IoT devices is challenging given the very limited memory and CPU resources. With the increasing volumes of collected data, it is critical to process, filter, and manipulate data on the edge devices where it is collected to improve efficiency and reduce network transmissions. Existing embedded index structures do not adapt to the data distribution and characteristics. This paper demonstrates how applying learned indexes that develop space efficient summaries of the data can dramatically improve the query performance and predictability. Learned indexes based on linear approximations can reduce the query I/O by 50 to 90% and improve query throughput by a factor of 2 to 5, while only requiring a few kilobytes of RAM. Experimental results on a variety of time series data sets demonstrate the advantages of learned indexes that considerably improve over the state-of-the-art index algorithms.
The Time-Slotted Channel Hopping (TSCH) from the IEEE 802.15.4-2015 standard has been proposed as a MAC protocol for industrial sensor networks as it provides a reliable media access control protocol under harsh conditions. This paper presents an algorithm for the optimization of a TSCH schedule based on a combined determination of interference free transmissions and Differential Evolution (DE) algorithm. Because of the inconsistent number of transmissions that a TSCH schedule can support, the standard mutation and cross-over steps of the DE algorithm had to be modified and a unique approach is presented that accommodates for this inconsistency. The feasibility of the optimization algorithm is demonstrated by showing an improvement in the throughput of a sensor network with heterogeneous sensors rates as the DE optimizer iterates through the population of schedules.
Access control systems consist in managing access to buildings or any secure area where access is restricted. This paper presents a model that helps build access control systems along with its internal architecture. This system is modeled according to the behavior of the access control system. The OMNeT++ network simulator, in addition to the INET framework, is used to model the behavior of a studied system as well as its energy consumption. The paper aims to compare the energy consumption of the studied system and its simulated model with the same working scenario. The challenge is to create a simulation model with a set of configurable parameters, where users will be able to modify the value of the latter, based on the intended application. By this way, the simulated model calculates promptly the energy consumption.
Increasing field productivity is not just a financial need, but also a social issue. Several technologies converge to promote food production and, in this context, the fog computing paradigm can support the development of solutions for precision agriculture. This paper proposes an application of the Jetson Nano device, embedded in an agricultural spraying implement. This device supports the decision on irrigation activity, based on data collected by sensors distributed in the field. The sensors read information about the plant’s stress level from electrical signals and the Jetson Nano enables real-time analysis, through machine learning algorithms, to manage the product spray rate, according to the condition of the crop. Initial studies validated the proposed solution on an experimental basis, showing that the device can be an alternative for this purpose, since it can be used efficiently in machine learning tasks from data collected by the sensors. The experiment also highlighted some limitations of the proposed solution, such as the importance of observing the conditions of the system as a whole, its context and environment, in order to improve performance in spraying process.
Wireless Sensor Networks (WSNs) consist of a number of resource-constrained sensor nodes and a designated node called a sink, which collects data from the sensor nodes. A WSN can be used in numerous applications such as subject tracking and monitoring, where it is often desirable to keep the location of the subject private. In these types of applications, an adversary can locate the monitored subject, if a location privacy protection scheme is not applied. In this paper, we propose an adaptive energy and delay efficient scheme, called Snowflake, that conceals the location of subjects from a global adversary. Snowflake can be adapted to make the delivery delay smaller, or to make the packet overhead low. The simulation results show that Snowflake performs better than an existing algorithm.
Alan Turing introduced a novel Reaction-Diffusion (RD) model in 1952 to explain biological pattern formation found in animals. Since then, studies based on the RD model have long proved the feasibility of adapting it to spatial patern formation in distributed systems, especially in networking systems. In the past two decades, RD mechanism started being applied to Wireless Sensor Networks, and the possiblity of expanding to new applications is promising. In this paper, we first review the original RD model and further show its variants, known as activator-inhibitor models. Several research efforts on applying them to model tasks in wireless sensor networks will be presented and summarized.
In this paper, we present a new study to improve the automated segmentation of blood vessels in diabetic retinopathy images. Pre-processing is necessary due to the contrast between the blood vessels and the background, as well as the uneven illumination of the retinal images, in order to produce better quality data to be used in further processing. We use data augmentation techniques to increase the amount of accessible data in the dataset to overcome the data sparsity problem that deep learning requires. We then use the CNN VGG16 architecture to extract the feature from the preprocessed background images. The Random Forest method will then use the extracted attributes as input parameters. We used part of the augmented dataset to train the model (1764 images, representing the training set); the rest of the dataset will be used to test the model (196 images, representing the test set). Regarding the model validation phase, we used the dedicated part for testing the DRIVE dataset. Promising results compared to the state of the art were obtained. The method achieved an accuracy of 98.7%, a sensitivity of 97.4% and specificity of 99.5%. A comparison with some recent previous work in the literature has shown a significant advancement in our proposal.
IoT systems have taken on an essential role in our life. IoT devices are strongly integrated into several sectors such as Smart Healthcare, Smart Cities, Smart Energy, Smart Industry, etc. and deliver important data. Designing, building, and implementing IoT systems are significant challenges because of IoT requirements such as mobility, energy consumption, and limited device memory. To mitigate such challenges, opportunities to test and evaluate IoT systems early in the first development phases are important to reduce cost and effort. Different systems have been proposed to aid such development, aiming at different key challenges. One of these systems is FIWARE, an open source IoT middleware, designed to ease data transportation and big data tasks. It has been established as an ecosystem technology used for optimizing the development of several applications and services in IoT. Key feature is the standardized architecture for gathering context information and managing these contexts in cloud based IoT and big data applications. In this paper, we discuss the integration of FIWARE software and IoT-NDN. IoT-NDN is an IoT system based on the Named Data Networking (NDN) communication paradigm. NDN is a communication protocol developed for the Internet and uses hierarchical names instead of IP addresses to deliver data on the Internet. IoT-NDN is an advanced architecture of NDN, conceding the requirements and limitations of IoT devices. In this paper we present an approach and architecture to integrate FIWARE and IoT-NDN. This integration eases implementation of IoT-NDN in existing applications, since a transparent compatibility between both systems can be achieved.
Networking security continue to be a serious challenge for all domains because of the increasing number of attacks launched every day due to the advent of connected devices and the emergence of the Internet. Hence, Intrusion detection system comes into focus, especially with the inception of big data challenges. In this paper, we propose a distributed and parallel intrusion detection system suitable for big data environments using machine learning-based multi agent system and big data analysis.
