Search

Intrusive Load Monitoring (ILM) is a method to measure and collect the energy consumption data of individual appliances via smart plugs or smart sockets. A major challenge of ILM is automatic appliance identification, in which the system is able to determine automatically a label of the active appliance connected to the smart device. Existing ILM techniques depend on labels input by end-users and are usually under the supervised learning scheme. However, in reality, end-users labeling is laboriously rendering insufficient training data to fit the supervised learning models. In this work, we propose a semi-supervised learning (SSL) method that leverages rich signals from the unlabeled dataset and jointly learns the classification loss for the labeled dataset and the consistency training loss for unlabeled dataset. The samples fit into consistency learning are generated by a transformation that is built upon weighted versions of DTW Barycenter Averaging algorithm. The work is inspired by two recent advanced works in SSL in computer vision and combines the advantages of the two. We evaluate our method on the dataset collected from our developed Internet-of-Things based energy monitoring system in a smart home environment. We also examine the method’s performances on 10 benchmark datasets. As a result, the proposed method outperforms other methods on our smart appliance datasets and most of the benchmarks datasets, while it shows competitive results on the rest datasets.

As a representative model for privacy preserving data publishing, K-anonymity has raised a considerable number of questions for researchers over the past few decades. Among them, how to achieve data release without sacrificing the users’ privacy and how to maximize the availability of published data is the ultimate goal of privacy preserving data publishing. In order to enhance the clustering effect and reduce the unnecessary computation, this paper proposes a weighted K-member clustering algorithm. A series of weight indicators are designed to evaluate the outlyingness of records, distance between records, and information loss of the published data. The proposed algorithm can reduce the influence of outliers on the clustering effect and maintain the availability of data to the best possible extent during the clustering process. Experimental analysis suggests that the proposed method generates lower information loss, improves the clustering effect, and is less sensitive to outliers as compared with some existing methods.

The novel coronavirus (Covid-19) infection outbreak has posed a major threat to the international health system and economy. This study is aimed at investigating the biological and epidemiological trends in the prevalence and mortality due to outbreaks of novel coronavirus (COVID-19) infections. The data on the global outbreak of COVID-19, were obtained from World Health Organization (WHO), Worldometer, Centers for Disease Control and Prevention (CDC), and research institutes. The information was also recorded from research documents published in global scientific journals indexed in Pub Med and Institute of Scientific Information (ISI) Web of Science on the trends in the prevalence and mortality due to COVID-19 infection outbreaks. The results show rising trends in the transmission, prevalence and mortality rate due to coronavirus COVID-19. During the period of December 29, 2019 through March 31, 2020, it has infected 750,890 people worldwide, resulting in 36,405 deaths with a mortality rate of 4.84%. The infections were more frequent among male gender over 60 years of age. The mean growth rate index for total number of cases from January 23 to March 31, 2020 was 1.20 and growth rate index for mortality rate was 1.12. There was a positive association between the prevalence and mortality rate (R2 = 0.996). The novel coronavirus COVID-19 is highly contagious and has affected a large number of people worldwide. It is still spreading with mutable prevalence and mortality outbreak trends. The global health officials have taken priority measures to prevent further outbreaks of this emerging pathogen across the globe. However, the rising number of cases and mortality risk estimates are demonstrating that enhanced public health mediations, good hygienic conditions, social distancing, and movement limitations may control the COVID-19 epidemics.

Flight delay is a significant problem that negatively impacts the aviation industry and costs billion of dollars each year. Most existing studies investigated this issue using various methods based on historical data. However, due to the highly dynamic environments of the aviation industry, relying only on historical datasets of flight delays may not be sufficient and applicable to forecast the future of flights. The purpose of this research is to study the flight delays from a new angle by utilising data generated from the emerging Internet of Things (IoT) paradigm. Our primary goal is to improve the understanding of the roots and signs of flight delays as well as discovering related factors. In this paper, we present a framework that aims at improving the flight delay problem. We consider the IoT data generated from distributed sensors that have not been considered in existing works in the analysis of flight delays, and for that purpose, an automatic tool is developed to collect IoT data from various data sources including flight, weather, and air quality index. Based on the heterogeneous data, an algorithm is developed to merge different features from diverse data sources. We adopt predictive modelling to study the factors that contribute to flight delays and to predict the flight delays in the future. The results of our work show a high correlation among the developed features. In particular, the results clearly demonstrate the association between the flight delays and the air quality index factor. In particular, our current prediction model achieves 85.74% in accuracy.

Level of Trust can determine which source of information is reliable and with whom we should share or from whom we should accept information. There are several applications for measuring trust in Online Social Networks (OSNs), including social spammer detection, fake news detection, retweet behaviour detection and recommender systems. Trust prediction is the process of predicting a new trust relation between two users who are not currently connected. In applications of trust, trust relations among users need to be predicted. This process faces many challenges, such as the sparsity of user-specified trust relations, the context-awareness of trust and changes in trust values over time. In this paper, we analyse the state-of-the-art in pair-wise trust prediction models in OSNs, classify them based on different factors, and propose some future directions for researchers interested in this field.

Centralized publishing of big location data provides great convenience for various location-based interactive queries and services. Privacy protection of users' location information is an indispensable issue in the security of big data applications. Partition publishing is an effective way to release statistical information of two-dimensional big location data. By combining with the differential privacy model, it can provide more accurate range counting query service on the premise of ensuring location privacy. In order to further improve the availability of location data subsequent to centralized publishing, this paper analyzes the primary noise sources of partition publishing and discusses the constraints among publishing errors, the spatial partition structure, and privacy budget allocation. An unbalanced quadtree partition algorithm based on regional uniformity is proposed. Accordingly, the gradient privacy budget allocation scheme and adjustment method are designed to ensure the effectiveness of the differential privacy model. Experimental comparison of the real-world datasets proves the advantages of the proposed algorithm in improving the querying accuracy of the published data.

As a promising next-generation network architecture, named data networking (NDN) supports name-based routing and in-network caching to retrieve content in an efficient, fast, and reliable manner. Most of the studies on NDN have proposed innovative and efficient caching mechanisms and retrieval of content via efficient routing. However, very few studies have targeted addressing the vulnerabilities in NDN architecture, which a malicious node can exploit to perform a content poisoning attack (CPA). This potentially results in polluting the in-network caches, the routing of content, and consequently isolates the legitimate content in the network. In the past, several efforts have been made to propose the mitigation strategies for the content poisoning attack, but to the best of our knowledge, no specific work has been done to address an emerging attack-surface in NDN, which we call an interest flooding attack. Handling this attack-surface can potentially make content poisoning attack mitigation schemes more effective, secure, and robust. Hence, in this article, we propose the addition of a security mechanism in the CPA mitigation scheme that is, Name-Key Based Forwarding and Multipath Forwarding Based Inband Probe, in which we block the malicious face of compromised consumers by monitoring the Cache-Miss Ratio values and the Queue Capacity at the Edge Routers. The malicious face is blocked when the cache-miss ratio hits the threshold value, which is adjusted dynamically through monitoring the cache-miss ratio and queue capacity values. The experimental results show that we are successful in mitigating the vulnerability of the CPA mitigation scheme by detecting and blocking the flooding interface, at the cost of very little verification overhead at the NDN Routers.

As one promising way to inquire about any particular information through a dialog with the bot, question answering dialog systems have gained increasing research interests recently. Designing interactive QA systems has always been a challenging task in natural language processing and used as a benchmark to evaluate machine’s ability of natural language understanding. However, such systems often struggle when the question answering is carried out in multiple turns by the users to seek more information based on what they have already learned, thus, giving rise to another complicated form called Conversational Question Answering (CQA). CQA systems are often criticized for not understanding or utilizing the previous context of the conversation when answering the questions. To address the research gap, in this paper, we explore how to integrate the conversational history into the neural machine comprehension system. On one hand, we introduce a framework based on publicly available pre-trained language model called BERT for incorporating history turns into the system. On the other hand, we propose a history selection mechanism that selects the turns that are relevant and contributes the most to answer the current question. Experimentation results revealed that our framework is comparable in performance with the state-of-the-art models on the QuAC (http://quac.ai/) leader board. We also conduct a number of experiments to show the side effects of using entire context information which brings unnecessary information and noise signals resulting in a decline in the model’s performance.

The rapid development of mobile Internet services and the wide application of intelligent terminals has accelerated the advent of the promising era of big data. A number of big data services based on location information bring convenience to users, however, it also results in serious leakage of personal privacy. The partitioning and publishing method combined with the differential privacy model can provide better range counting query results under the premise of ensuring the privacy of users' location. Nevertheless, most of the existing research studies only focus on the structural design during the partitioning process of location big data and ignore the impact of differential privacy budget allocation methods on the published results. This paper, therefore, proposes an efficient arithmetic privacy budget allocation strategy for the tree-based partitioning and publishing of location big data which satisfies the e-differential privacy. Experimental results over a large number of real-world datasets prove that the proposed privacy budget allocation method is superior in contrast to the existing methods for improving the usability of the published data.

Since the term first coined in 1999 by Kevin Ashton, the Internet of Things (IoT) has gained significant momentum as a technology to connect physical objects to the Internet and to facilitate machine-to-human and machine-to-machine communications. Over the past two decades, IoT has been an active area of research and development endeavours by many technical and commercial communities. Yet, IoT technology is still not mature and many issues need to be addressed. In this paper, we identify 10 key research topics and discuss the research problems and opportunities within these topics., 10 pages. IEEE CIC 2020 vision paper

Recently, the convergence of the paradigm of the Internet of Things (IoT) with social networking has been widely recognized as an emerging interdisciplinary area, lately, referred to as the Social Internet of Things (SIoT). SIoT enables the object-to-object interaction and provides the platform for these objects to autonomously socialize with the other objects in the network in a bid to overcome the key challenges of IoT, i.e., data discovery and composition, network navigability, trust management, etc. Trust plays a significant role while establishing these inter-object social relationships and it is also essential to observe the trustworthiness of an object before relying on information provided by them. A number of trust evaluation models for SIoT have been proposed in the literature, nevertheless, most of these models suffer in validating and testing their respective models due to lack of appropriate datasets. To address this issue, this paper proposes a scalable plug and play trust platform referred to as SCalable and Robust Trust platform for SIoT (SCaRT-SIoT) to provide the dataset to test and analyze various SIoT-based trust models of the research community.

The promising paradigm of the Internet of Things (IoT) has rapidly proliferated in a number of domains, e.g., energy monitoring, transportation, smart cities, healthcare, and the like. Leveraging on the same, in this paper, we proposed an Internet of Plants system to remotely monitor soil moisture level and automate watering control for the plants. The envisaged system not only facilitates in over watering or under watering of the plants but also provides data collection for the gardeners to understand the key features pertinent to healthy growth of plants. Such kind of system also encourages more people to grow their small-scale organic gardens in urban areas, thereby, increasing greenery in modern cities.

Centralized publishing of big location data can provide accurate and timely information to assist in traffic management and for facilitating people to decide travel time and route, mitigate traffic congestion, and reduce unnecessary waste. However, the spatio-temporal correlation, non-linearity, randomness, and uncertainty of big location data make it impossible to decide an optimal data publishing instance through traditional methods. This paper, accordingly, proposes a publishing interval predicting method for centralized publication of big location data based on the promising paradigm of deep learning. First, the adaptive adjusted sampling method is designed to address the challenge of finding a reasonable release time via a prediction mechanism. Second, the Maximal Overlap Discrete Wavelet Transform (MODWT) is introduced for the decomposition of time series in order to separate different features of big location data. Finally, different deep learning models are selected to construct the entire framework according to various time-domain features. Experimental analysis suggests that the proposed prediction scheme is not only feasible, but also improves the prediction accuracy in contrast to the traditional deep learning mechanisms.

Energy optimization for cloud computing services has gained a considerable momentum over the recent years. Unfortunately, minimizing energy consumption of cloud services has its own unique research problems and challenges. More specifically, it is difficult to select suitable servers for cloud service systems to minimize energy consumption due to the heterogeneity of servers in cloud centers. In this paper, the energy minimization problem is considered for cloud systems with stochastic service requests and system availability constraints where the stochastic cloud service requests are constrained by deadlines. An energy minimization algorithm is proposed to select the most suitable servers to achieve the energy efficiency of cloud services. Our intensive experimental studies based on both simulated and real cloud instances show the proposed algorithm is much more effective with acceptable CPU utilization, saving up to 61.95% energy consumption, than the existing algorithms.

Vertical Handover Decision (VHD) algorithms are indispensable components of forthcoming 4G heterogeneous wireless networks architecture – so as to provide requisite Quality of Service to an assortment of applications anywhere at any time, while allowing seamless roaming in highly dynamic scenarios (i.e. multitude of access network technologies that vary in bandwidth, latency, monetary cost, etc.) using Mobile Terminals (MTs) enabled with multiple access interfaces. In this article, a critical review of the existing VHD algorithms has been carried out as an effort to update the previous studies. To offer a methodical contrast, recently published VHD algorithms have been classified into four major classes depending on the key handover decision criterion used, i.e. RSS based algorithms, bandwidth based algorithms, cost function based algorithms, and the combination algorithms. Moreover, operational fundamentals, advantages, and disadvantages of exemplary VHD algorithms for each class have been presented to assess the tradeoffs between their intricacy of implementation and the efficacy. Copyright © 2018, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Considerable innovative advancements in vehicular communication have led to the emerging yet promising paradigm of Internet-of-Vehicles, wherein vehicles exchange safety-critical information with minimal delay for ensuring road safety as well as efficacious traffic flows. It is, therefore, indispensable that these safety messages are authentic and reliable, and have originated from a legitimate vehicle. This demands establishing trust among vehicles such that the malicious and dishonest vehicles (and their malicious content) could be flagged and subsequently eliminated from the network. The risk manifolds if a malicious vehicle gets elected as the cluster head of other vehicles thereby compromising the safety of vehicular passengers and pedestrians on the road. Hence, intelligent algorithms should be in place to opt for the trusted and resource efficient cluster heads which could enhance the overall security and efficiency of their respective clusters. To this end, in this paper, we have proposed a scalable hybrid trust model which takes into account a composite metric encompassing the weighted trust score and available resources of each vehicle for identification of multiple malicious vehicles in real-time, and for meeting the stringent performance requirements of vehicular safety applications. Moreover, an optimal role assignment scheme based on the Hungarian algorithm has been proposed for electing the optimal cluster head, proxy cluster head, and followers among the members of a vehicular cluster so as to maximize its overall efficiency. Preliminary simulations have been carried out and are also presented in this paper.

The promising advancements in the telecommunications and automotive sectors over the years have empowered drivers with highly innovative communication and sensing capabilities, in turn paving the way for the next-generation connected and autonomous vehicles. Today, vehicles communicate wirelessly with other vehicles and vulnerable pedestrians in their immediate vicinity to share timely safety-critical information primarily for collision mitigation. Furthermore, vehicles connect with the traffic management entities via their supporting network infrastructure to become more aware of any potential hazards on the roads and for guidance pertinent to their current and anticipated speeds and travelling course to ensure more efficient traffic flows. Therefore, a secure and low-latency communication is highly indispensable in order to meet the stringent performance requirements of such safety-critical vehicular applications. However, the heterogeneity of diverse radio access technologies and inflexibility in their deployment results in network fragmentation and inefficient resource utilization, and these, therefore, act as bottlenecks in realizing the aims for a highly efficient vehicular networking architecture. In order to overcome such sorts of bottlenecks, this article brings forth the current state-of-the-art in the context of intelligent transportation systems (ITS) and subsequently proposes a software-defined heterogeneous vehicular networking (SDHVNet) architecture for ensuring a highly agile networking infrastructure to ensure rapid network innovation on-demand. Finally, a number of potential architectural challenges and their probable solutions are discussed.

Recent advancements in the telecommunications and automotive sectors, combining in the form of Intelligent Transportation Systems (ITS), are leading to intelligent connected and autonomous vehicles. At present, IEEE 802.11p Dedicated Short Range Communication (DSRC) and the next-generation cellular networks are considered as the key wireless access technologies for the vehicular networking industry. Nevertheless, none of them are capable of meeting the stringent latency requirements of diverse vehicular safety-critical applications. This paper, therefore, presents an ongoing research work pertinent to software defined heterogeneous vehicular networking architecture for ensuring an extremely reliable and low-latency communication for the safety-critical vehicular cooperative applications and services.

State-of-the-art advances in vehicular communication have fostered the development of smart mobility platforms, where moving vehicles exchange safety-critical information with minimal latency, to improve passenger and pedestrian safety. It is, therefore, of paramount importance that the information that is exchanged is not only authentic, but its sender should be trustworthy. Consequently, incorrect information needs to be detected as such, and its source needs to be flagged as untrustworthy if it consistently sends incorrect information. In the case of highly dynamic vehicular networks, ignoring trustworthiness poses a serious threat to vehicular communications, especially if there is a chance that the malicious vehicle could be elected as the cluster head of other vehicles within a group. We propose a hybrid trust management scheme to identify such malicious vehicles and to inhibit them from being elected as the cluster head. The scheme encompasses a composite metric (i.e., trust values assigned to the vehicles coupled with their resource availability) for cluster head and proxy cluster head selection via intermittent elections. This approach helps to form trustworthy and resource efficient vehicular networks. Simulations of the proposed scheme have been conducted using MATLAB and are also presented in this manuscript.

The recent state-of-the-art advancements in vehicular ad hoc networks (VANETs) have led to the emergence and rapid proliferation of the promising notion of the Internet-of-Vehicles (IoV), wherein vehicles exchange safety-critical messages with one another to ensure safe, convenient, and highly efficient traffic flows. Nevertheless, such inter-vehicular communication could not be realized until the network is completely secured as the dissemination of even a single malicious message may jeopardize the entire network. Accordingly, numerous trust models have been proposed in the research literature to ensure the identification and elimination of malicious vehicles from a network. These trust models primarily depend on the aggregation of both direct and indirect observations, and which themselves are computed depending on the diverse influential parameters pertinent to dynamic and distributed networking environments. Still, optimum weights need to be allocated to these parameters for generating accurate and intuitive trust values. Furthermore, once the trust for a target vehicle has been computed, a specific threshold value equal to the minimum acceptable trust score has been selected for identifying the malicious vehicles. Quantification of these weights and selecting of an optimal threshold poses a significant challenge in VANETs. Accordingly, this paper focuses on employing machine learning techniques as to cope with the said problems in VANETs. It thus utilizes a real IoT data set by transforming it into an IoV format and computes the feature matrix for three parameters, i.e., similarity, familiarity, and packet delivery ratio, in two different ways, (a) all of the stated parameters computed by each trustor for a trustee are treated as individual features, and (b) the mean of each single parameter computed by all of the trustors for a trustee is regarded as a collective feature. Different machine learning algorithms were employed for classifying vehicles as trustworthy and untrustworthy. Simulation results revealed that the classification via the mean parametric scores yielded much more accurate results in contrast to the one which takes into account the parametric score of each trustor for a trustee on an individual basis.

The automotive industry is shifting its focus from performance and features to safety, entertainment, and driver comfort. In this regard, driver assistance and autonomous driving technology are gaining more attention. Such technology has the potential to reduce road accidents, traffic congestion, and fuel usage. However, vehicles cannot become fully autonomous, until they are able to sense their context efficiently (context sensing), and to use ambient learning to respond appropriately and within short timescales to the data they have sensed. Context sharing will also become essential, because a single vehicle will not be able to gain a holistic view of its context without cooperation from other nearby vehicles and from the roadside infrastructure. Indeed, there are further advantages when a group of vehicles make intelligent decisions based on a common understanding of their context. This chapter highlights the significance of ambient intelligence for next-generation connected and autonomous vehicles, describes its current state of the art, and also shows how its potential might be achieved. One of the main challenges refers to how to provision and coordinate cloud-based services to meet the needs of real-time (low latency) data-intensive (high data rate) ambient intelligence, particularly for safety-critical vehicular safety applications. It indicates how autonomous or semi-autonomous vehicles are likely to make seamless use of any available wireless networking technologies to improve both coverage and reliability and, where feasible, to cache critical content near the network edge so as to minimize the number of network hops and hence service latencies. Both of these approaches should improve the network quality of service afforded to driving applications.

Internet of Vehicles (IoV) is an integral component of the Intelligent Transportation System (ITS) with the vision to acquire and process data for improving traffic safety. In IoV, vehicles communicate by exchanging messages with other vehicles and roadside infrastructure subsequently generating a large volume of data often referred to as big data, and which demands a massive storage and high computational capability with minimal possible delay. The aim of this paper is to take a step forward in outlining the current state-of-the-art in IoV and analyzing the improved data processing and storage features by utilizing big data analytics and fog computing paradigm. We further presented a preliminary simulation result to demonstrate that the introduction of fog improves the overall system performance by ensuring low-latency critical for especially vehicular safety applications. Moreover, this paper briefly discusses a fog computing based IoV crime assistant and highlight some of its challenges which needs to be addressed in order to enhance the performance of such systems.

Managing network resources in highly dynamic vehicular environments is difficult. It is made more challenging when vehicular safety applications demand high bandwidth, reliability and extremely low latency. Standalone wireless access technologies cannot meet these requirements simultaneously and so heterogeneous wireless networking has been proposed as a solution. However heterogeneity introduces other problems, particularly network fragmentation and inefficient use of network resources. Software-defined Networking (SDN) offers logically centralized control and re-programmability of the control plane, with the potential to address these problems of heterogeneity. However, SDN has difficulty dealing with the high levels of mobility associated with highly dynamic vehicular networks. This paper presents ongoing work towards an SDN-based heterogeneous vehicular networking, employing edge-based caching to ensure extremely reliable and low-latency communication for safety-critical cooperative vehicular services.

Rapid advances in wireless communication technologies coupled with ongoing massive development in vehicular networking standards and innovations in computing, sensing, and analytics have paved the way for intelligent transportation systems (ITS) to develop rapidly in the near future. ITS provides a complete solution for the efficient and intelligent management of real-time traffic, wherein sensory data is collected from within the vehicles (i.e., via their onboard units) as well as data exchanged between the vehicles, between the vehicles and their supporting roadside infrastructure/network, among the vehicles and vulnerable pedestrians, subsequently paving the way for the realization of the futuristic Internet of Vehicles. The traditional intent of an ITS system is to detect, monitor, control, and subsequently reduce traffic congestion based on a real-time analysis of the data pertinent to certain patterns of the road traffic, including traffic density at a geographical area of interest, precise velocity of vehicles, current and predicted travelling trajectories and times, etc. However, merely relying on an ITS framework is not an optimal solution. In case of dense traffic environments, where communication broadcasts from hundreds of thousands of vehicles could potentially choke the entire network (and so could lead to fatal accidents in the case of autonomous vehicles that depend on reliable communications for their operational safety), a fall back to the traditional decentralized vehicular ad hoc network (VANET) approach becomes necessary. It is therefore of critical importance to enhance the situational awareness of vehicular drivers so as to enable them to make quick but well-founded manual decisions in such safety-critical situations.

The evolution of Big Data in large-scale Internet-of-Vehicles has brought forward unprecedented opportunities for a unified management of the transportation sector, and for devising smart Intelligent Transportation Systems. Nevertheless, such form of frequent heterogeneous data collection between the vehicles and numerous applications platforms via diverse radio access technologies has led to a number of security and privacy attacks, and accordingly demands for a secure data collection in such architectures. In this respect, this chapter is primarily an effort to highlight the said challenge to the readers, and to subsequently propose some security requirements and a basic system model for secure Big Data collection in Internet-of-Vehicles. Open research challenges and future directions have also been deliberated.