Forthcoming Networks and Sustainability in the IoT Era

Online house price appraisal involved complex and quite challenging task. Several researches have been proposed in the literature, providing various techniques and tools for finding and pricing houses to make the process more efficient, comfortable, and reliable for realtors and clients. Traditional house appraisal approaches focused on the economic and demographic variables and mainly used statistical methods to estimate the houses’ values. Even though those estimates provide valuable information, they are extremely unreliable in certain situations. The interior and exterior appearance, which is not considered in the estimation using these techniques, is one of the crucial variables influencing the valuation of a house. Recent advances in digital cameras, machine learning, deep learning, computer vision, and the Internet of Things (IoT) have led to the development of sophisticated house appraisal techniques, taking into account the houses’ economic, demographic, and pictorial information. This survey article investigates the current state of the art and future trends in house price appraisal methods.

A key application for IoT based technologies in the field of healthcare is wireless medical sensors that can be used to monitor patients’ physiological information such as heartbeat, bowel activity and lung sounds. Real-time detection of bowel motility after major abdominal surgery has significant importance for the patients’ healing process. Due to temporal cessation of intestinal motility after the surgery, a period of fasting is commonly practiced, and patients are fed with fluids following the recovery of bowel motility. Many studies have been conducted to monitor intestinal motility and automatically detect bowel activity. Detection and identification are challenging because of the ambient noise in clinical environments. Active noise cancellation methods remove unwanted signals by using adaptive filters. In this paper, active noise cancellation simulations were performed in order to remove ambient noise from gastrointestinal auscultation recordings. The simulation setup was created based on a previously developed IoT-driven electronic stethoscope by our group. Five widely used adaptive filter algorithms: Least Mean Squares, Normalized Least Mean Squares, Affine Projection, Recursive Least Squares, and Adaptive Lattice were tested, and performance evaluations are reported.

Since individuals with needs in the general public increased, the work introduced is a navigation system that will give a solid and durable obstacle detection and environmental imager and navigation for the user. It provides minimal cost system to permit navigation. The obstacle detection is to distinguish the deterrent and guide the visually impaired (VIP) about a suitable pathway. The framework utilises sensor based obstacle detection, and sends back buzzer or audio sound as a reaction that warns the VIP about position. The primary technique utilised by each blind or visually impaired is the strolling stick for identifying deterrent in which its functionality is restricted, it doesn’t secure territories close to the head let alone all obstacle. This framework acquires data about impediments close to the head and provides the right pathway for the VIP. When utilised with a mobile stick, the VIP is completely ensured against a snag, and the route is made simple. The environmental imager and navigation mode is the sound and visual guide for the VIP which permits users to just touch a button and proposed destination to the caregiver. This includes GPS and live video feed direction. The general system is versatile and can be conveyed by a VIP. The accuracy achieved for the system differs from 94.15\(\%\) to 99.72\(\%\). The percentage rate of the snag discovery for either indoor or outside varies from 95.40\(\%\) to 99.67\(\%\). This examination will Increase the VIP mobility significantly.

Reliable data access is essential to an increasingly smart automated and pervasive digital environment. Mobile networks are very important in a fully connected smart digital world, everything needs to be linked, from people to vehicles, sensors, things, cloud services and even robotic agents. 5 G wireless networks currently being deployed offer significant enhancements beyond LTE, but may not satisfy the full networking requirements of the growing digital society. This paper outlines technology which are intended to convert the sixth-generation 6 G wireless network and which we consider to be an enabler for several potential cases of 6 G use. We offer a detailed system-level perspective on 6 G scenarios and specifications, frameworks, standards, research activities and 6 G technology that can either be addressed by enhancing the 5 G architecture or implementing entirely new communication paradigms.

With the increasing demand of electric power and pressure of mitigating GHG emissions, electric utilities are inclined towards increasing the renewable capacity in their electricity mix. Solar photovoltaic systems, being one of the major contributors in sustainable energy production, cover a vital portion of global cumulative installed renewable capacity. To ensure the optimal efficiency and avoid any forthcoming outage, monitoring of photovoltaic plants is an essential element of integrating renewable into current generation systems. Authors review the types of photovoltaic plants based on configuration and the parameters that are optimal for energy monitoring. It also includes the measuring techniques for the different parameters of monitoring. Familiarity with these parameters and their measuring techniques is essential in development of an efficient photovoltaic energy monitoring system. Various components of these monitoring systems are exposed to extreme weather conditions which reduce their life span. In addition, the efficiency of the photovoltaic modules degrade over time and the cost and complexity of energy monitoring systems limits their usage at a larger scale.

The work proposed in this paper is the application of machine learning techniques in recognizing patterns and predicting student success rate on the bases of their performance on their previous grades in this IoT era. With this, using machine learning algorithms improves predicting student grade efficiently. This method is implemented with their previous academic data for students present in the tertiary institution. However, the education system of students in Portugal have enhanced during the past decades. Precisely, the inadequate achievement of success in critical courses like the Portuguese language and also Mathematics is a grave issue. In this paper, we intend to analyze student’s success in tertiary institutions using ML techniques. Real-world raw data were received by using existing data from the school. The two core courses were modeled, also four ML techniques were tested. The results gotten shows that student success rates can greatly be instigated by their previous performance. With the direct outcome of the research, a more adequate predicting tool can also be developed, which improves education quality and enhances resource management for schools. This study is said to increase student performance greatly if taken into consideration.

The exponential growth in the number of Internet of Things (IoT) devices, the sensitive nature of data they produce, and the simple nature of these devices makes IoT systems vulnerable to a wide range cyber-threats. Physical attacks are one of the major concerns for IoT device security. Security solutions for the IoT have to be accurate and quick since many real time applications depend on the data generated by these devices. In this article, we undertake the IoT authentication problem by proposing a fast protocol RapidAuth, which also restricts physical attacks. The proposed protocol uses Physical Unclonable Functions to achieve the security goals and requires the exchange of only two messages between the server and an IoT device. The analysis of RapidAuth proves its’ robustness against various types of attacks as well as its’ efficiency in terms of computation, communication, memory overheads and energy consumption.

As botnet style distributed denial of service (DDoS) attacks continue to proliferate the Internet of Things (IoT) landscape, researchers have struggled to provide a definitive way of addressing concerns related to the IoT’s security. In this paper, we work from the axiom that DDoS attacks are easiest to detect at the target of the attack but are best mitigated closer to the attacker by implementing four machine learning models that detect botnet-infected DDoS attackers on their access network. These models operate on network packet counts, which can easily be gathered by an access router, and run in real-time or near real-time, even on a low power device, namely a Raspberry Pi. We introduce a novel method for visualizing network activity as graphical heatmaps and use convolutional neural network (CNN) models designed for embedded devices and mobile platforms to classify network traffic as benign or malicious. We compare this approach using a support vector machine (SVM) and a long short-term memory recurrent neural network (LSTM). Based on our results, we conclude that the use of lightweight CNNs to analyze network traffic through graphical heatmaps provides highly accurate botnet-based DDoS attack detection for IoT access networks, with an average accuracy of 99.8%, despite our training dataset being between 73×–2170× smaller than those seen in related works, and runtimes ranging from 334 ms to 2 s on a Raspberry Pi.

Optical Wireless Communication (OWC) is regarded as an auspicious communication approach that can outperform the existing wireless technology. It utilizes LED lights, whose subtle variation in radiant intensity generate a binary data stream. This is perceived by a photodiode, that converts it to electric signals for further interpretation. This article aims at exploring the use of this emerging technology in order to control wirelessly industrial robots, overcoming the need for wires, especially in environments where radio waves are not working due to environmental factors or not allowed for safety reasons. We performed experiments to ensure the suitability and efficiency of OWC based technology for the aforementioned scope and “in vitro" tests in various Line-of-Sight (LoS) and Non-Line-of-Sight (NLoS) configurations to observe the system throughput and reliability. The technology performance in the “clear LoS" and in the presence of a transparent barrier, were also analyzed.

The Internet of Things (IoT) also called the Internet of Everything is a system of smart interconnected devices. The smart devices are uniquely identifiable over the network and perform autonomous data communication over the network with or without human-to-computer interaction. These devices have a high level of diversity, heterogeneity, and operates with various computational capabilities. It is highly necessary to develop a framework that allows to classify the devices into different categories from effective management, security, and privacy perspectives. Various solutions such as network traffic analysis, network protocols analysis, etc. have been developed to solve the problem of device classification. The signal of a device is an important feature that could be utilized to classify various network devices. We propose a framework to identify network devices based on their signal analysis. We have developed a training data set, by collecting signals from various Wi-Fi and Bluetooth devices in a specific geographic area. A machine learning-based model is proposed for the prediction of network device classification (e.g., a Wi-Fi or Bluetooth device) with 100% accuracy. Furthermore, clustering techniques are applied to the acquired signals to predict the total number of active Wi-Fi devices in a given region.

The Internet-of-Things (IoT) is one of the modern technological revolutions that enables communication amongst a plethora of different devices. To date 30 billion devices are connected to the internet more than 75 billion devices are foreseen to be connected worldwide by 2025, a five fold increase in ten years. Devices can have different brands and developers and can be designed to function on a proprietary ecosystem, with separate applications, gateways and tools to support them. This fragmentation can be disastrous in certain industries, such as the medical ones, and limit integration between different systems. In this paper, we envision a solution to overcome this interaction problems. We propose Share a novel programming standard through a design pattern. This allows on the fly service composition of resource constrained IoT devices. To this ending, IoT devices exchange integration codes which specify the data format and the interaction protocol. The design by contract scheme (DCS) is used to make sure that the matching services verify the constraints dictated by the composition. Unlike other on the fly approaches, Share can run on very small and resource constrained devices. Share has been implemented by using LUA programming language and has been validated on the ESP30 embedded device.

There were an increasing number of innovative applications of Wireless Sensor Networks (WSNs) in health care domain. It has never been such clearer to appreciate the advantages and benefits of applying the WSNs to improve the quality of health care in a wide variety of areas. Thanks to the sensing and communications technology of today, it has also reached a point where these WSNs applications can be readily implemented and deployed to function although there are some limits and hinderance from the viewpoint of security concerns.

In this paper, we provide a protocol stack applicable to the WSNs for health care systems, and to outline a framework to implement the WSNs in two different health care settings. Following the proposed framework, we have simulated a WSNs based health care application for the settings of hospitals and/or nursing homes for the performance study.

Advancements in the network infrastructure have caused a positive influence in our day to day life. Many reform initiatives have been taken all over the world which are related to the digitization of the countries methodologies of handling information. The usage of modern techniques also has a drawback, which allows data theft. Hence, a secure system is required, which can detect any kind of fraudulent activity and alert the administrator. Such a system is called an Intrusion Detection System (IDS). There are many types of IDSs available at our disposal, and a lot of research has also been done on their various types. This paper presents the implementation of IDS based on CatBoost technique which is a part of the ensemble machine learning strategy. The results of the implementation have been evaluated on the evaluation metrics like accuracy, precision, recall, and F1-score. The programming environment used is Python. The implementation has experimented on the NSL-KDD dataset, and the results have been analyzed on the detection accuracy, which shows the proposed scheme has reached an accuracy of 99.46% on the NSL-KDD dataset.