Approaches for IoT-enhanced predictive process monitoring

The widespread use of mobile devices, such as smart phones, smart watches, bracelets, equipped with plenty of integrated sensors, provides collection of massive amounts of data and their processing and analysis on edge and cloud computing infrastructures. For the purpose of monitoring of health and activities of people, such Big mobility and IoT data must be properly processed and analyzed. In this paper, the RemoteHealth platform is presented, developed using contemporary mobile, IoT and Big Data technologies. The application implemented on RemoteHealth platform is based on sensor fusion and provides processing and analysis of Big mobility and IoT data originated from smart personal devices with the aim to detect and recognize human activities and behavior. The experimental evaluation shows that the sensor fusion increase the accuracy of activity recognition comparing to single sensor use.

Recent years have seen an exploration of data volumes from a myriad of IoT devices, such as various sensors and ubiquitous cameras. The deluge of IoT data creates enormous opportunities for us to explore the physical world, especially with the help of deep learning techniques. Traditionally, the Cloud is the option for deploying deep learning based applications. However, the challenges of Cloud-centric IoT systems are increasing due to significant data movement overhead, escalating energy needs, and privacy issues. Rather than constantly moving a tremendous amount of raw data to the Cloud, it would be beneficial to leverage the emerging powerful IoT devices to perform the inference task. Nevertheless, the statically trained model could not efficiently handle the dynamic data in the real in-situ environments, which leads to low accuracy. Moreover, the big raw IoT data challenges the traditional supervised training method in the Cloud. To tackle the above challenges, we propose In-situ AI, the first Autonomous and Incremental computing framework and architecture for deep learning based IoT applications. We equip deep learning based IoT system with autonomous IoT data diagnosis (minimize data movement), and incremental and unsupervised training method (tackle the big raw IoT data generated in ever-changing in-situ environments). To provide efficient architectural support for this new computing paradigm, we first characterize the two In-situ AI tasks (i.e. inference and diagnosis tasks) on two popular IoT devices (i.e. mobile GPU and FPGA) and explore the design space and tradeoffs. Based on the characterization results, we propose two working modes for the In-situ AI tasks, including Single-running and Co-running modes. Moreover, we craft analytical models for these two modes to guide the best configuration selection. We also develop a novel two-level weight shared In-situ AI architecture to efficiently deploy In-situ tasks to IoT node. Compared with traditional IoT systems, our In-situ AI can reduce data movement by 28-71%, which further yields 1.4X-3.3X speedup on model update and contributes to 30-70% energy saving.

In today’s world of technology, Data has been playing an imperative role in many different technical areas. Data confidentiality, integrity and data security over Internet from different media and applications has greater challenges tasks. Data generation from multimedia and IoT data is the biggest Big Data in Internet now. When sensitive and confidential data are counterfeit by attacks this lead to serious countermeasures to security and privacy. Data Encryption is the mechanism to forestall this issue. At present, many encryption techniques are used for multimedia and IoT but when there is massive data are developed it incur more computational challenge. This paper design and propose new coherent encryption algorithm which holds issue of IoT and multimedia big data. The proposed system can cause a strong cryptographic effect without hold much memory and easy performance analysis. Handling huge data with the help GPU is included in the proposed system to enhance the data processing more efficiently. The proposed algorithm is compared with other symmetric cryptographic algorithms like AES, DES, 3-DES, RC6 and MARS based on architecture, flexibility, scalability, security level and also based on computational Running time, Throughput for both encryption and decryption process. Avalanche effect is also calculated for proposed system with 54.2%. The proposed framework holds to secure the multimedia against real-time attacks.

Fog-based caching mechanism for IoT data in information centric network using prioritization

The Internet-of-Things is becoming an established technology, with devices being deployed in homes, workplaces, and public areas at an increasingly rapid rate. IoT devices are the core technology of smart-homes, smart-cities, intelligent transport systems, and promise to optimise travel, reduce energy usage and improve quality of life. With the IoT prevalence, the problem of how to manage the vast volumes of data, wide variety and type of data generated, and erratic generation patterns is becoming increasingly clear and challenging. This Special Issue focuses on solving this problem through the use of edge computing. Edge computing offers a solution to managing IoT data through the processing of IoT data close to the location where the data is being generated. Edge computing allows computation to be performed locally, thus reducing the volume of data that needs to be transmitted to remote data centres and Cloud storage. It also allows decisions to be made locally without having to wait for Cloud servers to respond.

Background:ENT diseases are common the world over and their occurrence often as a result of environmental factors which include air pollution and other particles causing allergens. Since the usage of IoT devices continues to rise, the current environmental details may be employed to diagnose and estimate the ENT disease's causative elements. For this reason, the role of IoT-derived environmental data and its association with the occurrence and manifestation of ENT conditions within different groups of individuals will be investigated in this research study. Objective:To explore the use of IoT data in defining the occurrence and distribution of ENT diseases and for determining the effects of environmental drives including pollution on ENT health. Methods:A cross-sectional descriptive survey research design was adopted for this study using data collected through smart devices and structured questionnaires. The participants included 250 people with differences in geographical location and age. Score of ENT symptoms, days lost due to ENT diseases, and various environmental factors like air pollution, etc, were examined. The techniques used for data analysis were the Shapiro-Wilk test to determine the level of normality and Cronbach’s Alpha regarding internal reliability or consistency of the items in the set questionnaires besides logistic regression. The multicollinearity of the independent variable was tested using the Variance Inflation Factor (VIF). Results:Descriptive analysis also showed that ENT symptoms severity, sick day, and other likes were not normally distributed ( = < 0.001). There was a poor internal consistency of all the measurement items in terms of reliability and Cronbach’s Alpha was low (-0.168). The statistical analysis said that by using logistic regression, both pollution exposure and IoT device use, it was only possible to predict sick days with an accuracy of 10.67%. The VIF analysis suggested that there was no problem with multicollinearity amongst the control variables. ENT symptoms severity distribution and RATM Sick Days were positively skewed. Conclusion:Though IoT devices are beneficial for obtaining real-time environmental data, the findings suggest that the environmental variables may not independently act as effective predictor variables for the severity of ENT disease or sick days. Because the reliability of the measurement tools used was low, and due to the poor performance of the predictive model, we recommend that future studies use more variables and better methods of assessment. More elaborate frameworks are required to capture the link between IoT data, environmental effects, and ENT health-related outcomes.

In last few years lot of work is carried out in process mining field by various researchers. Process mining deals with analysis and extraction of process related information from the event logs created by business processes. Predictive monitoring of business process is subfield of process mining which includes activities where event logs are analyzed to make various process specific predictions. The various machine learning and deep learning techniques have been proposed in predictive business process monitoring (BPM). The aim of these techniques is to predict next process event, remaining cycle time, deadline violations etc. of running process instance. The goal of this paper is to discuss the most representative deep learning approaches used for the runtime prediction of business process. The different types of deep learning approaches used in predictive BPM based on Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM) and Stacked Autoencoders have are highlighted in this paper. Also we are focusing on aspects like type of dataset, predicted values, type of data encoding and quality evaluation metrics for the categorization of these approaches. In this paper we have highlighted various research gaps in mentioned deep learning approaches which can be referred by other researchers in this field to enhance effectiveness of predictive BPM.

Cloud computing and big data: Technologies and applications

Background IoT and smart devices have become extremely popular in the last few years. With their capabilities to collect data, it is reasonable to have concerns about the protection of users’ personal information and privacy in general. Objectives Comparing existing regulations on data protection and information security rules with the new capabilities provided by IoT and smart devices. Methods/approach This paper will analyse information on data collected by IoT and smart devices and the corresponding legal framework to explore whether the legal framework also covers these new devices and their functionalities. Results Various IoT and smart devices pose a high risk to an individual's privacy. The General Data Protection Regulation, although a relatively recent law, may not adequately regulate all instances and uses of this technology. Also, due to inadequate technological protection, abuse of such devices by unauthorized persons is possible and even likely. Conclusions The number of IoT and smart devices is rapidly increasing. The number of IoT and smart home device security incidents is on the rise. The regulatory framework to ensure data controller and processor compliance needs to be improved in order to create a safer environment for new innovative IoT services and products without jeopardizing the rights and freedoms of data subjects. Also, it is important to increase awareness of homeowners about potential security threats when using IoT and smart devices and services.

The concept of Edge-to-Cloud Continuum aims to significantly reduce overall traffic to the cloud by enabling IoT data processing as close as possible to the data sources, either on near- or far-edge devices. In this highly dynamic environment, where IoT devices and edge nodes are constantly changing their state and location, services running on edge nodes have to be scheduled, deployed and managed to ensure high service availability with appropriate Quality of Service (QoS) parameters. However, once services are deployed in the edge-to-cloud continuum, the question arises how to ensure continuous data delivery from IoT devices to the appropriate services for further processing, either on edge devices or in the cloud. In this paper, we propose a general architecture for adaptive data-driven routing in the edge-to-cloud continuum and introduce an implementation of this architecture using the content-based publish/subscribe approach. We evaluate the given implementation against a real-world use case scenario for federated learning in an edge-to-cloud environment hosting digital twins. The performance evaluation of this scenario shows that our implementation efficiently adapts to service failures and reconfigures the edge-to-cloud environment with minimal latency and without data loss, while preserving data privacy and security. In addition, the experiments show that our solution is stable in an environment with IoT data sources generating data at high frequency.

Federated learning (FL) enables multiple IoT devices to collaboratively train a global model without centralizing raw data. However, achieving optimal performance for each device is challenging due to the heterogeneity of IoT data. Moreover, IoT devices, constrained in computational power and energy, necessitate innovative privacy protection solutions in FL. In this paper, we propose privacy-enhanced FL methods to address data heterogeneity in IoT. Our approach includes a dual-model framework for each device, combining a shared model to facilitate inter-client communication with a personalized model. This personalized model is developed from local aggregation, utilizing layer-wise aggregation weights generated by local hypernetworks. This integration effectively balances collective intelligence with client-specific knowledge, addressing data heterogeneity for improved personalization. Additionally, we propose a Local Gradient Shielding (LGS) method to protect training data. This method conceals gradients in locally trained personalized models, preventing data reconstruction and ensuring client privacy with minimal resource consumption. Extensive evaluations on benchmark datasets, showcasing various types and degrees of data heterogeneity, demonstrate the superiority of our approaches over current baselines.

Two of the foremost talked concerning technologies in today's world are Blockchain and IoT. They're at the initial stage of maturity and there are plenty of activities going down within development of them and finding fascinating uses of the technologies. The data transmitted by IoT devices is very important and must be secure. Based on the IoT security problems, this paper proposed a model for secure IoT data based on Blockchain; the model called service-centric networking (SCN). SCN provides communication by service names instead of addresses, and a decentralized network that allows the users of the network to communicate fast and data are reliable and secure. However, the identity of the users will be an encrypted format that is stored on their device backed by InterPlanetary File System (IPFS).

In this paper, we propose a new comprehensive realistic cyber security dataset of IoT and IIoT applications, called Edge-IIoTset, which can be used by machine learning-based intrusion detection systems in two different modes, namely, centralized and federated learning. Specifically, the dataset has been generated using a purpose-built IoT/IIoT testbed with a large representative set of devices, sensors, protocols and cloud/edge configurations. The IoT data are generated from various IoT devices (more than 10 types) such as Low-cost digital sensors for sensing temperature and humidity, Ultrasonic sensor, Water level detection sensor, pH Sensor Meter, Soil Moisture sensor, Heart Rate Sensor, Flame Sensor, etc.). Furthermore, we identify and analyze fourteen attacks related to IoT and IIoT connectivity protocols, which are categorized into five threats, including, DoS/DDoS attacks, Information gathering, Man in the middle attacks, Injection attacks, and Malware attacks. In addition, we extract features obtained from different sources, including alerts, system resources, logs, network traffic, and propose new 61 features with high correlations from 1176 found features. After processing and analyzing the proposed realistic cyber security dataset, we provide a primary exploratory data analysis and evaluate the performance of machine learning approaches (i.e., traditional machine learning as well as deep learning) in both centralized and federated learning modes. The Edge-IIoTset dataset can be publicly accessed from <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">http://ieee-dataport.org/8939</uri> .

By employing RFID tags and tiny sensors, the Internet of Things makes it possible to connect a physical object-like a book or other text typologies-with real-time communication technologies. Additionally, a significant number of libraries and colleges can be globally linked in real-time, continuously, thanks to the Internet of Things. In the context of delivering end-to-end IoT business solutions, IoT integration refers to ensuring that the combination of new IoT devices, IoT data, IoT platforms and IoT applications-as well as IT assets (business applications, legacy data, mobile and SaaS)-work well together. The collection of IoT integration capabilities required by IoT project implementers to successfully integrate end-to-end IoT business solutions is known as the IoT integration market. Monitoring the environment, occupancy rates and equipment use within library buildings is one of the most important components of IoT integration in libraries. IoT sensors give librarians the ability to monitor variables like humidity, temperature and air quality, which helps to maintain ideal conditions for both collections and users.

Information is gathered in smart agriculture from a variety of sources, such as IoT devices with There are many different kinds of sensors, such as weather sensor stations, soil moisture sensors, and sensors used by insects to check crop health. The satellites and drones are used to gather this sensor data. Farm management software tracks plant schedules, pest control methods, and irrigation techniques using the collected valuable data. Any unauthorized person manipulating agricultural data in this way leads to serious issues in the agricultural food supply chain and increases market demand for food. To avoid data manipulation in agricultural information, many strategies are utilized. This work discusses man in the middle attacks, man in the cloud attacks, SQL injection attacks, and cross-site scripting (XSS) attacks. Our strategy was to efficiently test the modified data using cyber integrated agriculture management (CIAM). This strategy focuses on tailored and effective techniques for treating plant soil health, analyzing nutrition levels and pH to guide fertilization, managing water levels, and detecting early signs of pests using image recognition and IoT data. AI-driven tools can detect anomalies and prevent attacks in real time, allowing different farming practices or weather conditions and their impacts on productivity for water usage, fertilizer application, and planting strategies to be detected and avoided with an impressive 98% accuracy, affecting both local and global food security.