Abstract The rapid evolution of electronic media in recent decades has exponentially amplified the propagation of fake news, resulting in widespread confusion and misunderstanding among the masses, especially concerning critical topics like the COVID‐19 pandemic. Consequently, detecting fake news on social media has emerged as a prominent area of research, attracting significant attention. This article introduces a novel cascaded group multi‐head attention (CGMHA) model for COVID‐19 fake news detection. Our research collected Twitter datasets with accurate and fake tweets in Urdu. The novel CGMHA model and depth‐wise convolution capture local and global contextual information by employing multiple attention heads in a cascaded fashion, enabling a comprehensive understanding of fake news. While achieving state‐of‐the‐art performance, we also highlight challenges such as language variations and misinformation nuances in the detection process, contributing to a more comprehensive understanding of the complexities involved in combatting fake news. Our proposed model surpasses the performance of state‐of‐the‐art models in classifying fake news and achieves accuracy, F1 score, precision, and recall of 0.98, 0.96, 0.95, and 0.95, respectively.

Abstract The growing ubiquity of Internet of Things (IoT) devices within smart homes demands the use of advanced strategies in IoT implementation, with an emphasis on energy efficiency and security. The incorporation of Artificial Intelligence (AI) within the IoT framework improves the overall efficiency of the network. An inefficient mechanism of parent selection at the network layer of IoT causes energy drain in the nodes, particularly near the sink node. As a result, nodes die earlier, causing network holes that further increase the control message overhead as well as the energy consumption of the network, compromising network security. This research introduces an AI‐based approach to parent selection of the Routing Protocol for Low Power and Lossy networks (RPL) at the network layer of IoT to enhance security and energy efficiency. A novel objective function, named Energy and Parent Load Objective Function (EA‐EPL), is also proposed that considers the composite metrics, including energy and parent load. Extensive experiments are conducted to assess EA‐EPL against OF0 and MRHOF algorithms. Experimental results show that EA‐EPL outperformed these algorithms in improving energy efficiency, network stability, and packet delivery ratio. The results also demonstrate a significant enhancement in the overall efficiency of IoT networks and increased security in smart home environments.

DeepWalk is used to convert nodes in an original graph into equivalent vectors in a latent space for performing various predictive tasks. To ensure second-order structural similarity between nodes in the original graph and their vectors in the latent space, dot products are applied to each pair of nodes explored on the random walk (RW) in the latent space. However. dot products for graphs with millions of nodes and billions of edges are computationally expensive. To minimize the computation time required for calculating the second-order structural similarity, DeepWalk with reinforcement learning (DWRL) is proposed herein. In DWRL, a level pointer for each node in the original graph is prepared. By identifying common nodes between each pair of nodes in the original graph, the number of computations in the dot product in the latent space is reduced, thereby ensuring second-order structural similarity. Additionally, repeated selection of the same node during RWs produces redundant samples for training. Therefore, the subsampling technique is used to choose the next node based on its degree, which improves the generalization of node representations in the latent space and increases accuracy. The proposed techniques are applied to popular datasets to perform multilabel classification and link prediction tasks, and their efficiency in reducing the computation time is verified. The proposed DWRL minimizes the computation time 47% for large graphs to build latent vectors and improves the average micro and macro F1 scores up to 12%. The link prediction performance also increases up to 20%.

Due to the rise in the use of crypto-currencies, such as Bitcoin and Ethereum, the fraud activities in the financial sector are increasing at the same pace. It becomes very challenging to detect frauds in crypto-currencies, because of it's distributed and anonymized nature and not having central control. In this paper, we exploited the decision-tree based machine learning model using a boosting approach, particularly XGBoost, to identify the fraudulent addresses in the Ethereum crypto-currency. To select a best decision tree and learning approach on our fraud dataset, initially, we chose four highly performed decision tree learning approaches including CART, random forest, LGBM, XGBoost, and applied a cross validation mechanism to select the top one based on accuracy. Among CART, random forest, gradient boosting tree, we selected the XGBoost model as the final model and tuned it for the best hyper parameters. Finally, we built the model on 80% of the training data, which has produced accuracy of more than 96% on test data. Further, the model is highly efficient to work in a real environment, as proved by running extensive experiments.

In An era where technology advances at an unprecedented pace, the healthcare sector stands at the cusp of a transformative revolution. The confluence of distributed Big Data intelligence with instantaneous e-healthcare services heralds a new paradigm, where the boundaries between medicine, artificial intelligence, and data science are blurred, giving rise to innovative solutions that redefine patient care. The emergence of personalized medicine, bolstered by the power of machine learning, graph-based techniques, and real-time analysis, is not merely a technological triumph but a testament to human ingenuity. It's a response to a world grappling with complex diseases, burgeoning healthcare costs, and an ever-increasing demand for precision and efficiency.

This is the editorial of the SS entitled ‘`Digital Twinning: Integrating AI-ML and Big Data Analytics for Virtual Representation’'.

Abstract Secret sharing schemes are being widely used to distribute a secret between various participants so that an authorized subset of participants belonging to appropriate access structures can reconstruct this secret. However, a dealer might get corrupted by adversaries and may influence this secret sharing or the reconstruction process. Verifiable secret sharing (VSS) overcomes this issue by adding a verifiability protocol to the original secret sharing scheme. This article proposes a computationally secure publicly verifiable secret sharing scheme based on the three‐dimensional cellular automata (3D‐CA). Unlike the more widely used linear secret sharing schemes or secret sharing scheme based on the Chinese remainder theorem, our proposed scheme performs the secret sharing using 3D‐CA. The secret is considered one of the initial configurations of the 3D‐CA, and the following configurations are devised to be the shares distributed among the participants. Update mechanisms and various rules make it hard for an adversary to corrupt or duplicate a share. To make it even more efficient, we have added a verifiability layer such that a dealer posts a public share and private share to each shareholder. The verifiability layer reduces the interaction between dealer and participants and hence increases the security. The randomness of the shares has been calculated using the National Institute of Standards and Technology statistical test suite.

Abstract Convolutional Neural Networks (CNNs) have made a great impact on attaining state‐of‐the‐art results in image task classification. Weight initialization is one of the fundamental steps in formulating a CNN model. It determines the failure or success of the CNN model. In this paper, we conduct a research based on the mathematical background of different weight initialization strategies to determine the one with better performance. To have smooth training, we expect the activation of each layer of the CNN model follow the standard normal distribution with mean 0 and SD 1. It prevents gradients from vanishing and leads to more smooth training. However, it was obtained that even with the appropriate weight initialization technique, a regular Rectified Linear Unit (ReLU) activation function increases the activation mean value. In this paper, we address this issue by proposing weight initialization based (WIB)‐ReLU activation function. The proposed method resulted in more smooth training. Moreover, the experiments showed that WIB‐ReLU outperforms ReLU, Leaky ReLU, parametric ReLU, and exponential linear unit activation functions and results in up to 20% decrease in loss value and 5% increase in accuracy score on both Fashion‐MNIST and CIFAR‐10 databases.

Abstract A successful implementation of demand response (DR) always depends on proper policy and their empower technologies. This article proposed an intelligent multiagent system to idealize the residential DR in distributed network. In our model, the primary stakeholders (smart homes and retailers) are demonstrated as a multifunctional intelligent agent. Home agents (HAs) are able to predict and schedule the energy load and retailer agents (RAs) predicts wholesale market price, sells energy to HAs. Both HAs and RAs are modeled to predict the real‐time pricing. Deep neural networks, that is, long short‐term memory network and hybrid CNN‐LSTM are used to predict the electricity load and energy price. Simulation results present good accuracy. Proposed work is compared with existing model w.r.t RMSE, MSE, and MAE. Comparison shows our model outperformed the existing models.

Summary In the last decade, artificial intelligent systems based on neural networks have gradually become primary source for clinical decision support systems (CDSS) and are being used in diverse areas of medical diagnosis, classification, and prediction. An artificial neural network (ANN) consists of a large number of processing units which performs the computation in a parallel and distributed environment. They learn the pattern from the examples provided to it and then generalize based on the concepts they have learned while training. This paper presents a review of the current status of ANN and its variants as CDSS in various medical disciplines. The work focuses and describes the methods making use of simple ANN and use of real‐time approaches based on big data using ANN in cloud computing environment for various medical applications. Critical analysis of various methods based on smart approaches indicates that feed‐forward back propagation ANN performs sufficiently better in the domain of medicines with a high degree of accuracy.

Summary Big data is largely influencing business entities and research sectors to be more data‐driven. Hadoop MapReduce is one of the cost‐effective ways to process large scale datasets and offered as a service over the Internet. Even though cloud service providers promise an infinite amount of resources available on‐demand, it is inevitable that some of the hired virtual resources for MapReduce are left unutilized and makespan is limited due to various heterogeneities that exist while offering MapReduce as a service. As MapReduce v2 allows users to define the size of containers for the map and reduce tasks, jobs in a batch become heterogeneous and behave differently. Also, the different capacity of virtual machines in the MapReduce virtual cluster accommodate a varying number of map/reduce tasks. These factors highly affect resource utilization in the virtual cluster and the makespan for a batch of MapReduce jobs. Default MapReduce job schedulers do not consider these heterogeneities that exist in a cloud environment. Moreover, virtual machines in MapReduce virtual cluster process an equal number of blocks regardless of their capacity, which affects the makespan. Therefore, we devised a heuristic‐based MapReduce job scheduler that exploits virtual machine and MapReduce workload level heterogeneities to improve resource utilization and makespan. We proposed two methods to achieve this: (i) roulette wheel scheme based data block placement in heterogeneous virtual machines, and (ii) a constrained 2‐dimensional bin packing to place heterogeneous map/reduce tasks. We compared heuristic‐based MapReduce job scheduler against the classical fair scheduler in MapReduce v2. Experimental results showed that our proposed scheduler improved makespan and resource utilization by 45.6% and 47.9% over classical fair scheduler.

Hadoop MapReduce is one of the cost-effective ways for processing huge data in this decade. Despite it is opensource, setting up Hadoop on-premise is not affordable for small-scale businesses and research entities. Therefore, consuming Hadoop MapReduce as a service from cloud is on increasing pace as it is scalable on-demand and based on pay-per-use model. In such multi-tenant environment, virtual bandwidth is an expensive commodity and co-located virtual machines race each other to make use of the bandwidth. A study shows that 26%-70% of MapReduce job latency is due to shuffle phase in MapReduce execution sequence. Primary expectation of a typical cloud user is to minimize the service usage cost. Allocating less bandwidth to the service costs less but increases job latency, consequently increases makespan. This trade-off is compromised by minimizing the amount of intermediate data generated in shuffle phase at application level. To achieve this, we proposed Time Sharing MapReduce Job Scheduler to minimize the amount of intermediate data; thus, service cost is cut down. As a by-product, MapReduce job latency and makespan also are improved. Result shows that our proposed model minimized the size of intermediate data upto 62.1%, when compared to the classical schedulers with combiners.

A rapid development has been seen in the Vehicular ad hoc networks (VANETs) because of their applicability and significance in the fields of traffic management, road monitoring and safety, infotainment, and on-demand services. Route planning in vehicular networks based on efficient collection of real-time data can effectively mitigate traffic congestion problems in urban areas. Furthermore, real-time data is shared by using an effective sharing mechanism to avoid redundancy of the collected information. However, dynamic route replanning and effective sharing mechanisms based on real-time data are still challenging problems. Therefore, based on the aforementioned constraints, this paper describes a route discovery technique that uses real time data collected from various vehicles using the Internet of Things. The proposed scheme is based on the novel data dissemination technique for information sharing among the roadside units. RTRD is comprised of VANETs, vehicular traffic servers, and a 5G-based cellular system of public transportation. By considering the traffic congestion in urban areas, the optimal path is calculated to re-plan routes based on the k shortest path algorithm, and a load balancing technique is adopted to avoid further congestion.

In parallel to the recent technological advancements and developments, there is a consistent development in the field of Intelligent Transportation Systems (ITS). Though advancements are considered in one side, there is also a negative side of increased road-accidents, traffic jams, and traffic congestions etc. In this paper, to address the above-mentioned problems, dense urban traffic surveillance data analysis and principal component analysis on the vehicular data are performed. This study and analysis provide some findings, which are discussed to come up with various perspectives on the data patterns and analysis to develop a healthy, user-friendly and highly sophisticated environment to the daily commuters in the roadway sector.

Sleep position is directly related to sleep quality especially in patients with sleep disorders such as sleep apnea or snoring. We propose SleeP-Manager, a wearable embedded system, that is designed to aid Sleep Positional Therapy (SPT). SleeP-Manager with two wristbands monitors the sleep position of the user and gives a vibration feedback when a poor position is detected. We experimentally evaluate the effectiveness of SleeP-Manager. In order to accomplish this, an additional device of chestband is designed. The chestband provides the true sleep position and also measures the response of users to the vibration feedback. The results indicate that the accuracy of sleep position detection higher than 80%, and the ratio of desired sleep position per night increased significantly by the use of SleeP-Manager. Our questionnaire survey shows the wristband-typed device is most preferred for SPT due to the cost-effectiveness, easy-to-wear, and practicality.

Internet of Things (IoT) network contains heterogeneous resource‐constrained computing devices which has its unique reputation in IoT environments. In spite of its distinctiveness, the network performance deteriorates by the distributed contention of the nodes for the shared wireless medium in IoT. In IoT network, the Medium Access Control (MAC) layer contention impacts the level of congestion at the transport layer. Further, the increasing node contention at the MAC layer increases link layer frame drops resulting in timeouts at the transport layer segments and the performance of TCP degrades. In addition to that, the expiration of maximum retransmission attempts and the high contentions drive the MAC retransmissions and the associated overheads to reduce the link level throughput and the packet delivery ratio. In order to deal with aforementioned problems, the Adaptive Contention Window (ACW) is proposed, which aims to reduce the MAC overhead and retransmissions by determining active queue size at the contending nodes and the energy level of the nodes to improve TCP performance. Further, the MAC contention window is adjusted according to the node’s active queue size and the residual energy and TCP congestion window is dynamically adjusted based on the MAC contention window. Hence, by adjusting the MAC Adaptive Contention Window, the proposed model effectively distributes the access to medium and assures improved network throughput. Finally, the simulation study implemented through ns‐2 is compared with an existing methodology such as Cross‐Layer Congestion Control and dynamic window adaptation (CC‐BADWA); the proposed model enhances the network throughput with the minimal collisions.

For quite some time, the analysis of leukocyte images has drawn significant attention from the fields of medicine and computer vision alike where various techniques have been used to automate the manual analysis and classification of such images. Analysing such samples manually for detecting leukocytes is time-consuming and prone to error as the cells have different morphological features. Therefore, in order to automate and optimize the process, the nature-inspired Plant Growth Simulation Algorithm (PGSA) has been applied in this paper. An automated detection technique of white blood cells embedded in obscured, stained and smeared images of blood samples has been presented in this paper which is based on a random bionic algorithm and makes use of a fitness function that measures the similarity of the generated candidate solution to an actual leukocyte. As the proposed algorithm proceeds the set of candidate solutions evolves, guaranteeing their fit with the actual leukocytes outlined in the edge map of the image. The experimental results of the stained images and the empirical results reported validate the higher precision and sensitivity of the proposed method than the existing methods. Further, the proposed method reduces the feasible sets of candidate points in each iteration, thereby decreasing the required run time of load flow, objective function evaluation, thus reaching the goal state in minimum time and within the desired constraints.

Internet of Things (IoT) plays a key role in connecting the e-health system with the cyber world through new services and seamless interconnection between heterogeneous devices. Therefore, it becomes computationally inefficient to analyze and select features from such massive volume of data. Therefore, keeping in view the needs above, this paper presents a system architecture that selects features by using Artificial Bee Colony (ABC). Moreover, a Kalman filter is used in Hadoop ecosystem that is used for removal of noise. Furthermore, traditional MapReduce with ABC is used that enhance the processing efficiency. Moreover, a complete four-tier architecture is also proposed that efficiently aggregate the data, eliminate unnecessary data, and analyze the data by the proposed Hadoop-based ABC algorithm. To check the efficiency of the proposed algorithms exploited in the proposed system architecture, we have implemented our proposed system using Hadoop and MapReduce with the ABC algorithm. ABC algorithm is used to select features, whereas, MapReduce is supported by a parallel algorithm that efficiently processes a huge volume of data sets. The system is implemented using MapReduce tool at the top of the Hadoop parallel nodes with near real-time. Moreover, the proposed system is compared with Swarm approaches and is evaluated regarding efficiency, accuracy, and throughput by using ten different data sets. The results show that the proposed system is more scalable and efficient in selecting features.

Load balancing and energy conservation techniques are one of the important constraints in the design of in wireless sensor network (WSN). Usually, clustering technique helps the network in the minimum utilization of energy that results in enhancing network lifetime. Moreover, various nodes in the multihop network that are near to the base station drain their battery very quickly thus result in creating hot spot problem in a network. To overcome such constraints, this paper proposes a multi-layer clustering architecture for selection of forwarding node, rotation of cluster head, and inter and intra-cluster routing communication. The proposed scheme efficiently tackle the rotation of forwarder node by incorporating routing table (table list) at each node. Moreover, the rotation is performed by the consideration of two threshold levels of the residual energy of a node. Also, the exploitation of decision maker node, forwarder node, backup forwarder node, and non-forwarder node enhancing the routing strategy in a network. The performance of the proposed scheme is tested and evaluated by C programming language. The results show that the proposed scheme successful achieve better results than TLPER and EADUC in energy consumption per node, end-to-end communication, hop count in cluster formation.

Geosocial Network data can be served as an asset for the authorities to make real-time decisions and future planning by analyzing geosocial media posts. However, there are millions of Geosocial Network users who are producing overwhelming of data, called “Big Data” that is challenging to be analyzed and make real-time decisions. Therefore, in this paper, we proposed an efficient system for exploring Geosocial Networks while harvesting data as well as user's location information. A system architecture is proposed that processes an abundant amount of various social networks' data to monitor Earth events, incidents, medical diseases, user trends, and views to make future real-time decisions and facilitate future planning. The proposed system consists of five layers, i.e., data collection, data processing, application, communication, and data storage. The system deploys Spark at the top of the Hadoop ecosystem in order to run real-time analyses. Twitter and Flickr are analyzed using the proposed architecture in order to identify current events or disasters, such as earthquakes, fires, Ebola virus, and snow. The system is evaluated with respect to efficiency while considering system throughput. We proved that the system has higher throughput and is capable of analyzing massive Geosocial Network data at real-time.

Sleep plays an important role in recovering physical and mental functions. Sleep position is known to affect sleep quality, hence, managing sleep position is beneficial for patients suffering from sleep disorders. For a long-term sleep management, we propose a sleep position tracking system using two wristbands. From the data collected from the wristbands, the system detects sleep positions and their changes. We define a sleep position motion model that consists of seven transitions between three sleep positions. Then, we propose pre-processing methods to overcome difficulties in analyzing sleep motion data, i.e., discontinuity, uncertainty, and time-variability. We tested experimental data in state-of-art pre-trained convolution neural networks by transfer learning. The accuracy of our proposed system was 96.03% and 88.02% in pilot experiment and on-site sleep experiment, respectively. Our experimental results demonstrate that the proposed system effectively and accurately keeps track of sleep positions without causing any inconvenience to users, and hence, serves as a key building block for cost-effective 24/7 sleep monitoring solutions

Recently, a rapid growth in the population in urban regions demands the provision of services and infrastructure. These needs can be come up wit the use of Internet of Things (IoT) devices, such as sensors, actuators, smartphones and smart systems. This leans to building Smart City towards the next generation Super City planning. However, as thousands of IoT devices are interconnecting and communicating with each other over the Internet to establish smart systems, a huge amount of data, termed as Big Data, is being generated. It is a challenging task to integrate IoT services and to process Big Data in an efficient way when aimed at decision making for future Super City. Therefore, to meet such requirements, this paper presents an IoT-based system for next generation Super City planning using Big Data Analytics. Authors have proposed a complete system that includes various types of IoT-based smart systems like smart home, vehicular networking, weather and water system, smart parking, and surveillance objects, etc., for dada generation. An architecture is proposed that includes four tiers/layers i.e., 1) Bottom Tier-1, 2) Intermediate Tier-1, 3) Intermediate Tier 2, and 4) Top Tier that handle data generation and collections, communication, data administration and processing, and data interpretation, respectively. The system implementation model is presented from the generation and collection of data to the decision making. The proposed system is implemented using Hadoop ecosystem with MapReduce programming. The throughput and processing time results show that the proposed Super City planning system is more efficient and scalable.

A multi-representative re-fusion (MRRF) approximate data collection approach is proposed in which multiple nodes with similar readings form a data coverage set (DCS). The reading value of the DCS is represented by an R-node. The set near the Sink is smaller, while the set far from the Sink is larger, which can reduce the energy consumption in hotspot areas. Then, a distributed data-aggregation strategy is proposed that can re-fuse the value of R-nodes that are far from each other but have similar readings. Both comprehensive theoretical and experimental results indicate that the MRRF approach increases lifetime and energy efficiency.

The Internet of Things (IoT) is increasingly becoming a worldwide network of interconnected things that are uniquely addressable, via standard communication protocols. The use of IoT for continuous monitoring of public health is being rapidly adopted by various countries while generating a massive volume of heterogeneous, multisource, dynamic, and sparse high-velocity data. Handling such an enormous amount of high-speed medical data while integrating, collecting, processing, analyzing, and extracting knowledge constitutes a challenging task. On the other hand, most of the existing IoT devices do not cooperate with one another by using the same medium of communication. For this reason, it is a challenging task to develop healthcare applications for IoT that fulfill all user needs through real-time monitoring of health parameters. Therefore, to address such issues, this article proposed a Hadoop-based intelligent care system (HICS) that demonstrates IoT-based collaborative contextual Big Data sharing among all of the devices in a healthcare system. In particular, the proposed system involves a network architecture with enhanced processing features for data collection generated by millions of connected devices. In the proposed system, various sensors, such as wearable devices, are attached to the human body and measure health parameters and transmit them to a primary mobile device (PMD). The collected data are then forwarded to intelligent building (IB) using the Internet where the data are thoroughly analyzed to identify abnormal and serious health conditions. Intelligent building consists of (1) a Big Data collection unit (used for data collection, filtration, and load balancing); (2) a Hadoop processing unit (HPU) (composed of Hadoop distributed file system (HDFS) and MapReduce); and (3) an analysis and decision unit. The HPU, analysis, and decision unit are equipped with a medical expert system, which reads the sensor data and performs actions in the case of an emergency situation. To demonstrate the feasibility and efficiency of the proposed system, we use publicly available medical sensory datasets and real-time sensor traffic while identifying the serious health conditions of patients by using thresholds, statistical methods, and machine-learning techniques. The results show that the proposed system is very efficient and able to process high-speed WBAN sensory data in real time.

Mobile Cloud Computing (MCC) is a developing technology that assists in improving the quality of the mobile services. Since the increase in mobile resources, the researchers have taken the initiative to take into contemplation resource sharing among heterogeneous mobile devices. Therefore, to design a system architecture for mobility models and resource sharing are key issues that require utmost efforts to be solved to achieve anticipated objectives. Therefore, keeping in view the desired goals, in this paper, we present a system architecture based on the hierarchical resource sharing mechanism for MCC. The proposed system architecture is divided into three domains, such as Global Cloud Server (GCS), Local ISP Server (LIS), and Gateway Server (GWS). Also, the novel paradigm for minimizing the delay in the network based on deploying Foglets at each proposed algorithm of clustering mechanism is also present. Moreover, the fuzzy rule-based scheme is proposed to eliminate the inappropriate foglets before deciding an optimal foglet for handover. A foglet selection scheme is developed based on the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) decision mechanism. Various parameters such as delay, jitter, Bit Error Rate (BER), packet loss, communication cost, response time, and network load are considered for selecting an optimal network. To check the feasibility and performance of the proposed system architecture, the mobility scenario is considered with a different speed of a mobile node ranging from very high to very low. The simulation results and an analytical model are compared with existing scheme for foglet selection by a mobile node. From the analysis and discussion, it is shown that the proposed system architecture helps in minimizing handover delay, packet loss, average queuing delay, and device lifetime. in a network.

As we delve into the Internet of Things (IoT), we are witnessing the intensive interaction and heterogeneous communication among different devices over the Internet. Consequently, these devices generate a massive volume of Big Data. The potential of these data has been analyzed by the complex network theory, describing a specialized branch, known as 'Human Dynamics.' The potential of these data has been analyzed by the complex network theory, describing a specialized branch, known as 'Human Dynamics.' In this extension, the goal is to describe human behavior in the social area at real-time. These objectives are starting to be practicable through the quantity of data provided by smartphones, social network, and smart cities. These make the environment more intelligent and offer an intelligent space to sense our activities or actions, and the evolution of the ecosystem. To address the aforementioned needs, this paper presents the concept of 'defining human behavior' using Big Data in SIoT by proposing system architecture that processes and analyzes big data in real-time. The proposed architecture consists of three operational domains, i.e., object, SIoT server, application domain. Data from object domain is aggregated at SIoT server domain, where the data is efficiently store and process and intelligently respond to the outer stimuli. The proposed system architecture focuses on the analysis the ecosystem provided by Smart Cities, wearable devices (e.g., body area network) and Big Data to determine the human behaviors as well as human dynamics. Furthermore, the feasibility and efficiency of the proposed system are implemented on Hadoop single node setup on UBUNTU 14.04 LTS coreTMi5 machine with 3.2 GHz processor and 4 GB memory.

The rate of data generation is enormously growing due to the number of internet users and its speed. This increases the possibility of intrusions causing serious financial damage. Detecting the intruders in such high-speed data networks is a challenging task. Therefore, in this paper, we present a high-speed Intrusion Detection System (IDS), capable of working in Big Data environment. The system design contains four layers, consisting of capturing layer, filtration and load balancing layer, processing layer, and the decision-making layer. Nine best parameters are selected for intruder flows classification using FSR and BER, as well as by analyzing the DARPA datasets. Among various machine learning approaches, the proposed system performs well on REPTree and J48 using the proposed features. The system evaluation and comparison results show that the system has better efficiency and accuracy as compare to existing systems with the overall 99.9 % true positive and less than 0.001 % false positive using REPTree.

Internet service providers (ISPs) and telecommunication authorities are interested in detecting VoIP calls either to block illegal commercial VoIP or prioritize the paid users VoIP calls. Signature-based, port-based, and pattern-based VoIP detection techniques are not more accurate and not efficient due to complex security and tunneling mechanisms used by VoIP. Therefore, in this paper, we propose a rule-based generic, robust, and efficient statistical analysis-based solution to identify encrypted, non-encrypted, or tunneled VoIP media (voice) flows using threshold approach. In addition, a system is proposed to efficiently process high-speed real-time network traffic. The accuracy and efficiency evaluation results and the comparative study show that the proposed system outperforms the existing systems with the ability to work in real-time and high-speed Big Data environment.

Machine-to-Machine communication (M2M) is nowadays increasingly becoming a world-wide network of interconnected devices uniquely addressable, via standard communication protocols. The prevalence of M2M is bound to generate a massive volume of heterogeneous, multisource, dynamic, and sparse data, which leads a system towards major computational challenges, such as, analysis, aggregation, and storage. Moreover, a critical problem arises to extract the useful information in an efficient manner from the massive volume of data. Hence, to govern an adequate quality of the analysis, diverse and capacious data needs to be aggregated and fused. Therefore, it is imperative to enhance the computational efficiency for fusing and analyzing the massive volume of data. Therefore, to address these issues, this article proposes an efficient, multidimensional, big data analytical architecture based on the fusion model. The basic concept implicates the division of magnitudes (attributes), i.e., big datasets with complex magnitudes can be altered into smaller data subsets using five levels of the fusion model that can be easily processed by the Hadoop Processing Server, resulting in formalizing the problem of feature extraction applications using earth observatory system, social networking, or networking applications. Moreover, a four-layered network architecture is also proposed that fulfills the basic requirements of the analytical architecture. The feasibility and efficiency of the proposed algorithms used in the fusion model are implemented on Hadoop single-node setup on UBUNTU 14.04 LTS core i5 machine with 3.2GHz processor and 4GB memory. The results show that the proposed system architecture efficiently extracts various features (such as land and sea) from the massive volume of satellite data.

The Machine-to-Machine (M2M) communication has the potential to connect millions of devices in the near future. Since they agree on this potential, several standard organizations need to focus on improved general architecture for M2M communications. Currently, there is a lack of consensus to improve the general feasibility of M2M communication. Heterogeneous Mobile Ad hoc Networks (HetMANETs) can normally be considered appropriate for M2M challenges. When a mobile node (MN) moves inside a HetMANET, various challenges including a selection of the target network and energy efficient scanning take place, which need to be addressed for efficient handover. To cope with these issues, we propose a handover management scheme that efficiently initiates a handover process and selects an optimal network. Our proposed scheme is composed of two phases, i.e., i) the MN performs handover triggering based on the optimization of the Receive Signal Strength (RSS) from an access point/base station (AP/BS), and, ii) the network selection process is carried out by considering different parameters such as delay, jitter, velocity, network load, and energy consumption by the network interface. Moreover, if there are more networks available, then the MN selects the one that can provide the highest quality-of- service (QoS) using the Elimination and Choice Expressing Reality (ELECTRE) decision model. The performance of the proposed scheme is compared in the context of the number of handovers, average stay-time of an MN in the network, and energy consumption against periodic and adaptive scanning. Similarly, a two- state Markov model is defined that efficiently distribute the number nodes on the available access points and base stations. The proposed scheme efficiently optimizes the handoff related parameters and outperforms existing schemes.

A reasonable growth has been noticed in the Web of Things (WoT), in which different embedded devices are inter-connected with each other. These devices are capable of sharing and communicating over the web based application. On the other hand, integration of such devices (capillaries) need a comprehensive architecture, which is still missing. Therefore, this paper proposes the concept of Smart Society; propelling the notion of smart home. In the proposed smart society, we present an architecture for smart society. The proposed smart society consists of three domains, i.e., smart home and smart community, with supportive techniques and related challenges, and visualize of value added smart community. We then describe how to realize a robust networking among individual society. The feasibility and efficiency of the proposed system are implemented on Hadoop single node setup by testing the sample medical, sensory data sets and fire detection datasets. Finally, the results show that the proposed system architecture efficiently process, analyze, and integrates different datasets efficiently and triggers actions to provide safety measurements for elderly age people in smart home, vehicles in smart transportation system, and others.

In this work, an embedded system working model is designed with one server that receives requests by a requester by a service queue that is monitored by a Power Manager (PM). A novel approach is presented based on reinforcement learning to predict the best policy amidst existing DPM policies and deterministic markovian nonstationary policies (DMNSP). We apply reinforcement learning, namely a computational approach to understanding and automating goal-directed learning that supports different devices according to their DPM. Reinforcement learning uses a formal framework defining the interaction between agent and environment in terms of states, response action, and reward points. The capability of this approach is demonstrated by an event-driven simulator designed using Java with a power-manageable machine-to-machine device. Our experiment result shows that the proposed dynamic power management with timeout policy gives average power saving from 4% to 21% and the novel dynamic power management with DMNSP gives average power saving from 10% to 28% more than already proposed DPM policies.

The concept of smart homes (SH) has been extensively popularized, and there are a lot of technologies that need to be continuously utilized and integrated in such a concept. In this article, some applied problems of camera array (CA) in the SH are discussed and solved. Determining how to build an effective management method for CA in order to ensure that user privacy is not encroached upon is an important issue. In SH, the applications of CA are very diversified. We suggest that a satisfactory management method of CA should be based on the open service gateway initiative (OSGi) that includes resource management and monitoring (RMM) and UPnP security for the problems of resources and privacy, respectively. Finally, an applied example of CA is addressed in an elder-care center (EC). Simulation results show that the management strategy and application of CA based on an OSGi is satisfactory.

In this study, we present a video search and indexing system based on the state support vector (SVM) network, video graph, and reinforcement agent for recognizing and organizing video events. In order to enhance the recognition performance of the state SVM network, two innovative techniques are presented: state transition correction and transition quality estimation. The classification results are also merged into the video indexing graph, which facilitates the search speed. A reinforcement algorithm with an efficient scheduling scheme significantly reduces both the power consumption and time. The experimental results show the proposed state SVM network was able to achieve a precision rate as high as 83.83% and the query results of the indexing graph reached 80% accuracy. The experiments also demonstrate the performance and feasibility of our system.

Computational load of motion estimation in advanced video coding (AVC) standard is significantly high and even worse for HDTV and super-resolution sequences. In this article, a video processing algorithm is dynamically mapped onto a new parallel reconfigurable computing (PRC) architecture which consists of multiple dynamic reconfigurable computing (DRC) units. First, we construct a directed acyclic graph (DAG) to represent video coding algorithms in which motion estimation is the focus. A novel parallel partition approach is then proposed to map motion estimation DAG onto the multiple DRC units in a PRC system. This partitioning algorithm is capable of design optimization of parallel processing reconfigurable systems for a given number of processing elements in different search ranges. This speeds up the video processing with minimum sacrifice.

This study presents an intensity gradient approach for intra-prediction in H.264 encoding system, which enhances the performance and efficiency of previous fast algorithms. We propose a preprocessing stage in which eight orientation features are extracted from a macro block by the intensity gradient filters. The orientation features are utilized to select a subset of prediction modes to be involved in the rate-distortion calculation so that the encoding time can be reduced. The simulation results indicate that the intensity gradient based algorithm for intra-prediction contributes better tradeoff between rate-distorion performance and encoding complexity than the previous algorithms. Compared to H.264 reference software, the proposed algorithm introduces slight PSNR degradation and bit rate increase but saves around 76% of the total encoding time with all intra-frame coding.

This study presents an intensity gradient approach to intra prediction in H.264 encoding system, which enhances the performance and efficiency by means of edge orientation of the gradient filter. We propose a pre-processing stage in which eight-orientation feature are extracted from a macro block that selects four modes to be applied to the block among a set of predefined modes. It is shown that by choosing number of modes used in rate-distortion calculation lead to significant enhancement in the performance of intra prediction. The validity of this algorithm is confirmed experimentally. The simulation results indicate that the intensity gradient-based algorithm for intra prediction contributes to the bit-rate reduction compared to that of previous algorithms and saves around 53% of the total encoding time in H.264.

This paper presents a fast mode decision algorithm for H.264 intra prediction based on dominant edge strength (DES). In H.264 intra prediction, the computation-extensive rate distortion optimization (RDO) technique with full intra mode search is used to select the best mode for each macroblock. To reduce the computational load in mode decision, the DES which is corresponding to a decision mode is detected first. In accordance with the detected dominant edge, a subset of the prediction modes is then chosen for RDO calculation. The proposed algorithm only searches 4 modes instead of 9 for the 4/spl times/4 luma blocks. As for the 16/spl times/16 or 8/spl times/8 chroma blocks, instead of 4 modes, only 2 modes are required to be searched. Experimental results revealed that the computation time of the proposed fast intra prediction algorithm is averagely reduced to 40% of the full search method with slight PSNR degradation.