IoT and Analytics for Sensor Networks

With the large-scale improvements in communication technology and easy accessibility of the same has led to the advent of the IoT technology, which in turn has resulted in more, and more devices being IoT enabled. This research paper embarks a next generation smart mirror with the aim of connecting the conventional mirror into the IoT network along with a certain set of features, which were never introduced before. The whole range of features focuses on increasing the utility of daily mirrors. This mirror along with acting as a piece of glass will also be acting as a huge glass notepad upon which the user can leave any handwritten messages which are to be conveyed to others and can never go unnoticed. Keeping in mind the smudging effect and the fingerprint mark that are left behind when the mirror surface is touched, this new feature allows one to write the message on the mirror without touching the mirror surface itself. Additionally, the mirror will also support some of common features as well namely date, weather forecast, news headlines and daily remainders.

Wireless Sensor Networks (WSNs) are considered as one of the most prevailing technologies in today’s world due to the diversified applications. These applications are huge in the range such as environmental monitoring, health care, civil and military, disaster management to other surveillance systems. Minimization of energy is one of the most exciting tasks in WSNs as small sensor nodes are battery powered and deploy in remote environments. Clustering is one such imperative technique that can conserve energy more broadly, and evenly which makes the network operational for a longer period. This research paper aims at developing an energy-aware cluster-based routing protocol using Artificial Neural Network (ANN) which finds an optimal number of clusters and rotates the cluster head periodically to balance the energy consumption throughout the network. Typically, the proposed algorithm is developed based on Self-Organizing Map (SOM) to form the clusters, and the K-means algorithm is used to form different sizes of clusters. Finally, an optimal number of clusters is found that impacts the network load and balances the energy consumption. MATLAB is used as a simulation tool for experimental analysis. Simulation results prove that the proposed cluster-based routing protocol “LEACH-SOM” dominates the traditional LEACH routing protocol in terms of minimal energy consumption and makes the network active for long period.

Image forgery has been increased enormously due to the development of software and introduction of new cheaper digital devices like cameras and cellular phones. The introduction of new devices leads to inadvertent capturing of images. A lot of misuse of these devices led to the need for their identification. Source camera identification deals with the issue of identification of the mobile device using their camera through which image has been taken. This empowers the scientific agent to spot mobile device that has been used for capturing the specific image during the study. This is significant as this alphanumeric content is considered as a inaudible observer. In this proposed work, we are reviewing the various approaches for SCI which depend on classical machine learning algorithms feature extraction. Then, a CNN model is proposed for identification of mobile device using vision dataset. High accuracy of 92% is achieved for vision dataset.

Fault tolerance provisioning is extremely important in cloud. Literature suggests that the existing frameworks are service provider centric. Being a pay-per-use model, fault tolerance in the cloud should be flexible with respect to the users’ requirement. Our previously proposed framework FFTF (Flexible Fault Tolerance Framework) is further extended to E-FFTF in the present paper to provide user transparency in the selection of service category based on the task completion deadline and slack time. E-FFTF successfully establishes task execution price savings through flexible fault tolerance. E-FFTF is proved to be beneficial for both the cloud service providers and service users as per the obtained experimental results.

Today, almost all the places are observed by surveillance cameras. The aim is to monitor the activities in and around, especially for abnormal activities. But it requires manual assistance to watch/monitor. However, manual inspection is a monotonous job, that reflects on information lost. It shows the importance of an automatic abnormal activity detection system. This is considered a difficult task because of object overlapping, light variation, clutter background, camera angle, presence of various activity, and posture variations. It manipulates the human actions in videos. Hence, it is challgening to recognize abnormal human actions. This work examines the concert of the classical SVM and ResNet50 model among four datasets: The ‘SAIAZ’ (S tudents A ctivities I n A cademic Z one)-Corridor, ‘SAIAZ’-Open space, Classroom Violence from YouTube (cc), and Mixed Background Dataset (MBD). The ‘SAIAZ’ was created by student volunteers of our Institution, and MBD is a collection of selected frames from various videos. Abnormal actions like slapping, punching, kicking, running, and fighting is commonly extant in these datasets. Here, MBD is assumed to adopt various real-world situations. The SVM achieved the classification accuracy of 85%, 92%, 60%, and 44% on SAIAZ-Corridor, SAIAZ-Open space, Classroom Violence, MBD respectively. The ResNet50 achieved significant improvements in all the given datasets.

Quantum-dot Cellular Automata is an evaluation paradigm in which transistors are not used and viable candidate for replacing the CMOS based technology. QCA is one of the boosting nanotechnology devices with the aim to replace the CMOS technology. QCA is implemented by utilizing the tunneling of the electrons with the given potential within the quantum cell. We made an attempt to suggest a multiplexer architecture in QCA using micro-level polarization. The proposed multiplexer design saves 16.67% of effective area compared to the best designs reported till date. In this paper, new designs of universal gates are proposed. The proposed NOR and NAND gates requires less number of quantum cells which results in less effective area compared to the conventional majority gate based designs. By using these micro-level polarized gates, the multiplexer which is proposed in this paper is implemented. The proposed multiplexer is designed and simulated by QCA Designer.

Fiber optic systems are used for the prolonged reach transmission systems, but by increasing the bit rate which is the main requirement of the current time, dispersion gets arisen which results in intersymbol interference. Compensation of dispersion to improve the transmission capability of the fiber optic system provides a vast field for research. From the literature survey done, use of Dispersion compensation fiber has been found as the most reliable method for compensating the dispersion, but it becomes expensive as the length of Dispersion compensation fiber is increased for long distance transmission. The Fiber Bragg Grating is also used as a dispersion compensation module as reported in previous works but has been found inefficient method. However, the Performance of the Fiber Bragg Grating can be enhanced by adapting optimum Chirping technique and Apodization profile. From the previous reported works, Tanh-Apodized Fiber Bragg grating and Gaussian-Apodized Fiber Bragg grating are found to have optimum performance characteristics in terms of side lobe suppression and maximum reflectivity, which motivates us to analyze the respective Fiber Bragg Gratings for compensating the dispersion at various chirping techniques and variable grating lengths. In this work, Tanh-Apodized Fiber Bragg grating and Gaussian-Apodized Fiber Bragg grating are analyzed and simulated in various chirping techniques individually, as well as along with the Dispersion compensation fiber, in the hybrid model of dispersion compensation for a 100 km long optical fiber link at the data rate of 10Gbps. The simulation software used is optisystem. Also, the grating length has been varied and the different performance characteristics like Q-factor, BER, and Eye diagram are analyzed and compared. It has been observed that the Gaussian-Apodized quadratic-chirped Fiber Bragg Grating at the grating length of 26.6 mm along with the 11 km long Dispersion compensation fiber makes the cheaper dispersion compensation module with the finest performance.

Wireless Mesh Network (WMN) is one type of mobile ad hoc network. So, if there is a link breakage, it can heal and reorganize the network by itself. Portals, Mesh Gates and Access Points, Stations (STAs) and external station (STA) are main components of WMN. Portals are used to connect the network to the internet. STAs are the nodes through which routing of packets take place. External STAs are the users of WMN. Mesh Gates and Access Points are the stations which route the packets and also act as Access Points for the external STAs. In case of WMN, IEEE 802.11 s is the standard. When the external STA moves in the WMN, it goes out of coverage of a Mesh Gate and Access Point and enters into the coverage of another one. Hybrid Wireless Mesh Protocol (HWMP), routing protocol of IEEE 802.11s, does not support external Station’s (STA’s) mobility. To integrate mobility, Adaptive mobility management technique had been proposed. The scheme considers session and mobility activities of the external STA by taking into account its session-to-mobility ratio (SMR). Based on SMR value, it makes the decision about transmission of route management packets. This paper presents numerical analysis of adaptive mobility management scheme and HWMP. Simulation of both techniques are carried out using NS-2. Performance comparison shows that adaptive mobility management scheme outperforms HWMP. Moreover, a comparison between the numerical analysis and simulation results have also been performed. It has been observed that results in case of HWMP are very close.

Attendance plays a major role in every education system. Taking attendance of students manually can be a great burden for teachers. It may cause many problems like loss of time, repetition, incorrect markings, and difficulties in marking them. To avoid this, there is a need to design an automatic system that overcomes the issues with the traditional attendance system. There are many automatic methods available for this purpose like fingerprint systems, RFID systems, face recognition systems and iris recognition systems, etc. But among these Face Recognition proved to be more efficient. The main objective of this paper is to propose a model that captures images from videos, detect and recognize the faces, predict the recognized face, and then marks attendance. In this work, a basic step has been performed which uses fifteen classes from LFW Dataset and faces are detected, recognized, and then prediction is done on a randomly selected image from the used dataset. This system uses a combination of Multi Task Cascaded Neural Network (MTCNN) algorithm along with FaceNet that can be used to detect faces and extract facial features from images. SVM is used to predict the face of the person from the image. The proposed system obtains a classification accuracy of 99.177% for the training set and 100% on the test set. The accuracy, precision, recall, and F1-score are computed.

With the increasing use of sensors and intelligent devices, Internet of Things (IoT) becomes an important area of research to establish connectivity among connected devices. Traditional algorithms used for encryption are found to be highly complex and with higher number of rounds for encryption which is computationally expensive. In an IoT network, communicating nodes adapt fewer complex algorithms for maintaining security. Hence, lightweight security algorithms are used for this purpose. In this paper, we have proposed a method for encryption and decryption process called Enhanced Secure IoT (ESIT) of an image using shift (<< or > >) and bitwise binary modulo 2 (+2) operation for data transmission. It is a block cipher that accepts a 64-bit key which performs Left shift (<<) and (+2) function for encryption, and right shift (>>) and (+2) decryption. It is the normal bitwise left and right shift by removing the sequence of q-bits from the left and right side, respectively. The performance of the proposed method is established by correlation, entropy, and image histogram. Experimental evaluation on four available images clearly shows the advantages of the given approach.

Diversity in MIMO applications tend to ameliorate the architecture of system for compensation with upgraded hardware and software requirements. Single-user MIMO systems allow one user to be serviced per transmission interval. This maximizes the throughput of a single user, but its disadvantage is that it does not take advantage of multi-user diversity. Multi-user MIMO systems (MU-MIMO) have become the main technique for meeting the requirements. In MU-MIMO, one of the biggest issues to deal with is eliminating co-channel interference. The block diagonalization (BD) is a linear precoding method used in broadcast channels of MU-MIMO that has been effective in removing multi-user interference (MUI), but is not computationally efficient. To counter this, we have developed a novel optimization algorithm (Bacterial Foraging Optimization), and implemented it with variation in the number of users and modulation orders. This paper exhibits the improvement in the diversity, and proves the efficiency of the proposed algorithm.

Big Data is used in a spacious, distinct set of information that is continuously evolving estimates. Excessively vast array of tools can be computed to reveal models, patterns, and relationships, exclusively linked to human attitudes and communication. Its applications are applied in a range of real-time applications such as agriculture, weather forecasting, healthcare, and water resource management. Each Big Data research-oriented process can provide the outcome of the prediction or forecasting process. Analyzing the quality of water is still so important to a human being, because human beings cannot live without water. Smart Cities have been growing in the recent period, but many rural development areas are not yet emerging and do not have a lot of facilities like a smart city, and these areas are affected by the chemical contaminations like Arsenic, Cadmium, and some heavy metals. In this research work, Prediction of Chemical contamination for Water Quality Assurance using ML-based Techniques is proposed. This article offers a brief overview of Machine Learning Algorithms such as Decision Tree, Support Vector Machine, Random Forest, Naïve bayes, K-Nearest Neighbor. These algorithms are used for classification and prediction using datasets to analyze which Algorithm provides the best Predicting Accuracy as seen in this paper.

It proposes and simulates a new Ultra-wide (UWB) Multiple Input Multiple Output (MIMO) circular polarization (CP) antenna. The developed antenna’s configuration is uncomplicated and incorporates dual T-shaped feed ports at left and right edges to achieve both LHCP and RHCP at same frequency band. A wide circular slot with a hexagonal stub is created in the ground plane in order to achieve better bandwidth for the realization of UWB. The evolution measures of the antenna put forward are presented ahead in this research paper, and its parameters are designed to obtain the optimal return loss, isolation, polarization, and ARBW. Also, a parametric study is done on various types of materials chosen as substrate. The parametric study results, portrays that substrate with higher dielectric tends to give poor ARBW. Operating bandwidth (S11 <  = –10db) of the planned antenna is obtained to be 6.7 GHz (3.4 to 10.1 GHz). An isolation of more than 15 dB is also achieved. 3 dB ARBW is obtained in the extent of 3.5–5.5 GHz. Furthermore, the circular polarization features of the antenna proposed are favorable. The surface current and radiation patterns obtained illustrates the circular polarization. Both RHCP and LHCP are obtained at the same frequency in this proposed antenna design. This proposed antenna is a solution to orientation problems of transmitter and receiver, complicated designing, multipath fading, and weather penetration. The simulated antenna is a low profile, compact sized antenna simulated for UWB circular polarization and next generation applications.

Traffic Control System has faced many issues, such as high waiting time, resulting in the emission of CO2 (carbon) gas, which results in environment degradation and the loss of crucial time for the emergency vehicles, which increases impatience in drivers, which could be one of the primary causes of increase in accidents. There is also a considerable delay in vehicles’ arrival time, particularly emergency vehicles that get stuck in the long queue. It is a matter of concern, as these vehicles are being categorized as priority vehicles, so their waiting time should be minimum. The current Traffic Control System does not use any specific algorithm to control the traffic for priority vehicles such as fire brigade, ambulance. These drawbacks can be resolved by replacing the existing Traffic Control System with a more intelligent Traffic Control System. In this work, we proposed an Intelligent Traffic Control System for Emergency Vehicles model (ITCSEV), which can identify the vehicles and separate them based on their types like “emergency,” “passenger.” It can change the duration and traffic signal accordingly. This model has been tested by us through a practical simulation using an open-source simulator, i.e., SUMO (Simulation of Urban MObility), on the map of India’s Gwalior city, which demonstrates the efficiency and reliability of the model.

The main purpose of this paper is to replace the manual work in sewage cleaning and monitoring with automation. In this era, there is a lack of utilization of technology for sewage disposal tasks. At present, manual effort has been used for drainage cleaning, which leads to loss of human life while cleaning the blockage in pipes. In order to overcome these types of problems, “Digital Controller based automated drainage water monitoring and controlling” will be helpful. This project includes PLC (Programmable Logic Controller), which plays a vital role in solving real-time problems. Using the resource in this paper, it is possible to monitor and control the sewage system continuously without using manual work.

The evolution of Intelligent Technology Systems (ITS) has made possible the designing of vehicles with more sophisticated computing, communicating, and sensing capabilities. Vehicular Cloud (VC) magnifies these capabilities by cost-effectively sharing their resources. Location information about a node plays a prime role in many VC services which are based on the location of the user node. In this paper, we present a cooperative agent-based location verification framework for VC. Central Authority (CA), a trusted entity that coordinates the activities in VC, takes care of the location verification process with the help of special nodes known as verifier nodes. Verifier nodes are distributed in the network randomly and secretly. The presence of such nodes is known only to the CA, which makes it stronger in terms of security.

Traffic grooming is the optimization of resources in a network. An efficient traffic grooming provides better resource utilization, enhanced performance at a lower cost. Traffic grooming has become very significant for all types of computer networks. As WDM (wavelength division multiplexing) networks are providing very high speed for a huge amount of data transfer, so traffic grooming is even more important for these networks. There are so many parameters that may be considered for traffic grooming. The waiting probability is one of the key parameters for traffic grooming. Waiting probability is a measurement of time spent by a call to get the required resources for further communication. We have proposed an efficient traffic grooming technique. This technique is based on waiting probability calculations. Waiting probability is calculated, and then traffic grooming problems are addressed based on the waiting probability calculations. Some of the other key network performance parameters such as the number of servers required, ideal path length for a source–destination pair, number of free wavelengths required, etc., are also analyzed. It is a low complexity technique for handling traffic grooming problems efficiently in telecommunication and call center management.

Software Defined Networks (SDN) have become more efficient and popular by having effective controller. Centralized controller makes the decision to handle traffic in data plane by analyzing the entire network. In addition, handling network attack is equivalently complex for controllers. Low-rate Distributed Denial of Service (LR-DDoS) attack restricts legitimate users to access resources by sending unwanted and half-opened request towards the devices in data plane. Hence, it is vital to detect and mitigate LR-DDoS attack in its early stage because nature of attack is very similar to original request. The nature of this attack exhausts the network resources and leads to resource unavailability or delay while processing the legitimate requests. In this paper, we propose a flow-based detection and mitigation framework using machine learning models like Support Vector Machine (SVM), C4.5 Decision tree and Naïve Bayes as classifiers to detect LR-DDoS attack. From every traffic flow samples, we extract the essential features to detect attack. In mitigation phase, we handle the attack flow information and install the mitigation rules to avoid LR-DDoS attack from same source. Our experimental results show that SVM mechanism achieves better accuracy compared to C4.5 and Naïve Bayes techniques.

Accelerometer is an electromechanical device, which is used for physical measurement along the orthogonal coordinates. Micro Electro Mechanical Systems (MEMS) based capacitive accelerometers are embedded in many modern technological applications. This paper presents the comparison between two single axis MEMS based capacitive accelerometers, which have the natural frequencies of 7 and 2.2 kHz. This work includes design, simulation, analytical modelling, and finite element modelling of each MEMS comb type capacitive accelerometer with different operating frequencies. The accelerometer was designed using COMSOL Multiphysics and MATLAB simulator tool.

Nowadays, the crime has increased and the occurrence of the accidents are more in the cities. People are afraid about the current scenario. Moreover parents are worried on children when they travel to schools. In order to provide protection and safety to the children, there is a necessity for a technology based security and alert. In this paper, an idea to identify the status of the children when they are out to school so that they can identify the location of their children. Using GSM and RFID the tracking of transport is done. It has two units; one is bus unit for tracking the activity and alerting parents during the school travel. Another is school unit for tracking child inside the school. The results show that our approach will provide safety and security to school children.

Modern wireless systems require a single antenna to perform multi-band functionality considering different applications. As per the FCC report, the designed antenna works in the ultra-wide band from 0.45 to 14.37 GHz with an impedance bandwidth of 187.8%. For the designed antenna, efficiency is greater than 50% and gain is greater than 1 dBi for the entire UWB band. By tuning the feed line width, the same antenna can operate in eight narrow bands covering various mobile communication bands, military applications, global positioning system, satellite communication, and some part of Ku band that is from 0.45 to 14.37 GHz. The designed antenna provides a reflection coefficient of less than −10 dB for all eight narrow bands and has good impedance matching. Moreover, the proposed structure can also support many mobile applications like LTE, Wi-Fi, Wi-Max, and 5G.

High-performance and low power consumption are major factors that describe the significance of a design in VLSI. At low and ultra-low power applications, power consumption and logic delays are a fundamental problem. Nowadays, higher performance designs are built on the concept of computation units like ALU, where adders and multipliers are the essential components. To optimize low-power and high-performance applications, adders and multipliers need to be engineered to meet specifications such as speed and power dissipation. Using Modified Gate Diffusion Technique, this paper proposes a carry look-ahead adder (CLA). Compared with other adder models, the carry look ahead adder has much less propagation delay. The proposed adder is implemented on 90 nm technologies using the cadence environment. This paper's main objective is to use Modified Gate Diffusion Input (MGDI) Technology to develop and implement a CLA.

Scheduling is a process that manages the workflow tasks during execution on different resources. Virtual infrastructure is a dynamic mapping of system resources to applications in order to maximize its utilization. In today's technological world, cloud has taken a long stride on the success towards maximum throughput as well as highest qualitative services to its consumers. Yet, approaches for maximizing the utilization of cloud resources are at peak demand. Each cloud service provider focuses on maximum utilization with minimum consumption of cloud resources, although managing and providing computational resources to maximum number of users and to execute such huge applications is a challenging one. In this paper, a scheduling algorithm with name TBW (Tabu Bayesian Whale Optimization) has been proposed. Basically, the algorithm is used to target the improvement in scheduling of scientific workflows. The complete framework has firstly used a ranking algorithm named distributed HEFT ranking and then applied TBW algorithm on ranked tasks of input workflows. The work has been executed for five scientific workflows LIGO, MONTAGE, Epigenomics, SIPHT and Cybershake. TBW is using tabu method on workflow tasks for fast local search in cloud system, and Bayesian Optimization is used to find out best possible combinations of resources where tasks are mapped and then whale optimization maps the tasks on the resources in a smart way. In the whole process, total execution time and cost parameters are minimized under deadline constraints.

The Online Analytical Processing (OLAP) based multidimensional examination hassles for several stockpiling magnificence over huge data. For as much to recognize queries answering time companionable by OLAP framework users and understanding entire business perceive mandatory, OLAP data is structured as a data cube (a multidimensional model). The OLAP queries are responded in speedy and steady time by utilizing the cube materialization for assessments takers. But, this also involves unendurable expenses, regarding to stockpile memory and period, and as a data depot, OLAP has an average dimension and dimensionality which is to be significant on query processing. Consequently, cube assortment has got to be finished motivating to diminish inquiry management expenses, maintaining as a restraint the materializing gap. Several techniques and heuristics like deviationist and insatiable algorithms have been utilized to offer an estimated result. In this work, a Fruit Fly Optimization (FFO) approach is implemented in a lattice structure to obtain an optimal materialized data cube for reducing the query processing expenses. The results illustrate that FFO generates better performance than Particle Swarm Optimization (PSO) in terms of frequency and number of dimensions.

Extensive usage of mobile multimedia-based applications is increasing the user’s demand on the internet. Edge computing makes it possible to bring storage of data and computation in close vicinity to mobile users. To improve the experience of the users, the proposed method uses edge Internet of Things equipment-assisted Fault Tolerant Multimedia Caching Strategy (FTMCS) for information-centric networking. To accomplish this, we propose a new technique composed of optimized task distribution and fault tolerant network services. The task distribution method efficiently distributes the downloading task among the edge nodes having required data. Fault tolerant service makes it possible for the network to withstand node failures that may occur during the file downloading, and by using this service, fast and reliable transfer of content happens seamlessly according to user preferences. The experimental results show that the FTMCS provides better network performance compared to the existing solution in terms of download speed and throughput.

This research work proposes an effective solution for sizing in wireless networks supported by IEEE 802.15.4g, thus presenting the possibility of communicating networks with remote sensors in a completely transparent way for end devices, considering for this analysis restriction parameters such as the capacity, coverage, and interference knowing that the field of application of Wireless Networks is recently emerging, considerably gaining much popularity, which is increased as their features increase and new applications are discovered for them. This article has also included the application of a practical scenario, which demonstrates the communication of the devices remotely.

Parkinson disease is identified as the second most severe neurodegenerative disorder that affects the nerve system of people. This disease could mainly affect the walking, speech, and vision of patients followed by body nervousness, handwriting, harsh voice quality, depression, and sleeping problems. The proposed research study focuses on early detection and diagnosis of disease from the accelerometer sensor-based data by evaluating the deviations present in patient’s motor symptoms. A cloud-based Parkinson disease prediction system is developed for a clinical decision-making process that helps the doctor to diagnose the Parkinson-influenced patient from a remote place. Gait parameters of the patient were extracted along to provide input vectors to the classifier model for onboard Parkinson disease prediction and diagnosis. An effective expanded cat swarm optimization (ECSO)-based feature selection technique is explored to overcome the problem of dataset dimensionality. It could select the most appropriate features from the patient dataset according to a logically inspired evolutionary algorithm. Using this feature selection technique in the k-Nearest Neighbor (k-NN), classifier model could significantly improve the disease prediction accuracy, and also minimizes the disease prediction time against the existing classifier models.

Electric vehicles (EVs) are emerging as a preferred way to reduce environmental concern’s needs. Concern and energy insufficiency, and in the foreseeable term, this pattern is expected to grow. However, this inadequate charging infrastructure is now a major barrier to the adoption of EVs. Deployment of this infrastructure is expected to maximize the adoption of EVs to promote community access. Connectivity between charging substations (CS) is therefore mandatory. Entertainment real-time status of CSs can provide useful information, such as availability of charging provisions, reserves, and time to meet the SC. The purpose of this paper is to have a better solution related to EV charging mechanism leveraging the benefits of the Internet of Things (IoT) technologies. The IoT is a model that gives the present facilities a real-time worldwide communication view of the physical world employing the sensors and the transmitting networks. This article suggests a real-time server-based prediction of EV infrastructure.

A new 4D dissipative hyperchaotic system with an unstable equilibrium point is introduced. The proposed system consists of ten terms including three quadratic nonlinearities which constructed through using a nonlinear state control algorithm in the known Lorenz system, exhibits self-excited attractors and chaotic system attractors, with two positive derivations of Lyapunov. The dynamical properties of this system are analyzed using theoretical and numerical simulations based on equilibrium points, stability, dissipative, Lyapunov exponents, and phase portrait. Besides, various coexisting attractors or multistability with different initial conditions under the same parameters are investigated. Furthermore, Projective Synchronization (PS) of an identical proposed system is realized by nonlinear control strategy and Lyapunov stability theory.

Wireless Sensor Network (WSN) is an area of research that connects mutually huge subareas of communication, routing, security, and attacks. WSN is conceivably the most susceptible network to node capture attack due to its dynamic nature in huge area. A node capture attack is introduced by seizing few nodes through an intruder to capture entire WSN by extracting the useful information like keys, routing mechanism, and data from WSN. To improve the protection of WSN, we proposed a Black Hole Optimization Algorithm (BHOA) on best feasible node capture attack to discover the optimal nodes having superior possibility of attack. The BHOA is applied on a function Vertex Participation. The experiment is performed on MATLAB 2019a environment, and the results show the better quality, efficiency of BHOA against MA, OGA, MREA, GA, and FFOA based on traffic compromised ratio, power consumption cost, and attacking time.

In recent years, human activity recognition has become a prominent research area in numerous fields such as healthcare, smart home activity analysis, suspicious activity recognition, robotics, surveillance, and security. The focus of the current research work is the analysis of human activity recognition systems for video surveillance. The human activity recognition system involves the detection of normal as well as abnormal activities. The recognition of human activities is still considered a challenging issue despite the contributions of numerous researchers. The erratic human behavior and complexities of the video datasets create numerous challenges to precisely observe the human activities with significant performance. The analysis of the human activity detection systems for video surveillance is conducted on the basis of state-of-art contributions by different researchers in the field. The paper also describes the taxonomy of human activity detection. It ends with a discussion of the challenging issues in the field along with the concluding remarks.

With the advent of IoT, Internet dominance has been extended far and wide, resulting in management of billions of smart devices online. However, all the management frameworks in IoT developed so far are based on centralized models, which have their own set of issues like single point of failure and security constraints. To encounter major issues like this, Blockchain provides an effective alternative to encounter the issues of security and privacy. Blockchain is being a distributed and decentralized ledger framework, when used with IoT helps to encounter a lot of security related constraints. In this paper, a detailed study related to Blockchain has been conducted and how well it works in collaboration with IoT to overcome all the major security and privacy issues in an IoT ecosystem. Many other applications of Blockchain in an IoT environment were also studied, with the intention of exploring all the potential benefits it can provide while working with IoT.

Future communications are developed with new communication standards for multi access technologies, where spectrum sharing and Non-orthogonal multiple access (NOMA) are the two latest approaches. To transfer the data from one end to another end through this wireless communication medium interference take place, which affects the total system performance. To minimize the interference in these communications the efficient spectrum sharing technique was developed. Cooperative relaying in NOMA system with spectrum sharing using threshold modeling were proposed in past. To improve the resource allocation in NOMA with cooperative relaying, instantaneous signal to noise ratio were used in obtaining higher outage throughput in a spectrum sharing cognitive radio-NOMA (CR-NOMA) system. In the previous methods constant thresholding is considered in the energy detection model and the quality of signaling is not observed, which results in lower throughput. In this work, a Quality Assisted Spectrum Allocation (QASA) based on loss probability is proposed and a dynamic threshold modeling is suggested for spectrum sensing under dynamic channel condition. Here the Rayleigh and Rician fading models are considered and used the Lagrange mathematical concept to calculate the bit error and to minimize the distortion. The throughput of the proposed work performs well when compared with the conventional method of Simultaneous wireless information and power transfer (SWIPT).

Recent developments in cloud services have increased number of data owners to store their encrypted data in the cloud whereas equal or more data users participate to retrieve data. Secure retrieval of relevant data has become a challenging issue. In this paper, a secure ranking based multi-keyword search using semantic index is being developed. Initially, owner builds an index file by semantic representation of keyword. Security key is provided by Trusted Authority (TA) for decrypting the obtained results at the user side. TA manages dual security processes such as managing secret keys and issuing security devices to the data users. User query reaches proxy server, and it checks whether any frequent keyword matches with given query using Boolean Search. If it does not match, query enters into the main server which stores all document and index files to obtain relevant result using Deep Learning Neural Network. In deep learning neural network, the query is processed with vector space model to retrieve the relevant documents. Finally, user decrypts the relevant results obtained from deep neural network. The experimental result shows that our proposed model provides better performance in terms of recall, ranking privacy, precision, and searching time.

The global pandemic caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) virus has had historical impact on the world. The virus causes severe respiratory problems and with an R0 of 5.7, spreads at a rapid rate. At the time of writing, there were over 85 million cases and 1.8 million deaths caused by COVID-19. In the proposed methodology, Deep Convolutional Neural Networks (DCNNs) have been trained, with the help of transfer learning, to learn to identify whether a suspected patient is suffering from this disease using their chest CT scan image. Transfer learning technique enables the transfer of knowledge from pre-trained models which have been previously trained on extremely large datasets. Various DCNN models have been applied such as AlexNet, ResNet-18, ResNet-34, ResNet-50, VGG-16, and VGG-19. The DCNNs were evaluated on a set of 2,481 chest CT scan images. Various performance metrics (Accuracy, MCC, Kappa, F1 score, etc.) were calculated for all DCNN models to enable their comparative evaluation. After extensive testing, ResNet50 was found to give the best results in this binary classification task. The highest accuracy achieved was 97.37% and highest kappa was 0.947. Identification of presence of COVID-19 using this method would provide great benefit to society and mankind.

Cloud computing offers several profitable services as and when required to the customer, and so it is growing up as a forthcoming drift in the IT sector. Software-as-a-service (SaaS) is one of the outstanding and fastest-growing fields in cloud computing history. It is a license to acquire cloud applications via the Internet. Out of all this curiosity, security is found as one of the key issues that delay the growth of SaaS. The motivation of this research work is to provide security for SaaS by handling massive amounts of data. At first, feature selection is accomplished by Oppositional Crow Search Algorithm (OCSA). These nominated features are sent for detecting the attacks via Deep Belief Network (DBN). The main objective is to introduce an innovative, secure framework for SaaS by performing attack detection when there is enormous traffic. The proposed and conventional models are evaluated using a benchmark dataset. Results prove that the proposed OCSA + DBN outperforms the other existing methods with respect to precision, sensitivity which are positive measures and False Positive Rate (FPR) and False Detection Rate (FDR); the negative measures. Moreover, the proposed work performs better with the existing works with a performance indicator of 3% for all the metrics.

Everyone needs to express their ideas, thoughts, and emotions. For a professional or a student, they have to express or communicate their thoughts effectively. To make communication effective, it is important to understand the English language. As English is an international language, all had to have fluent English-speaking skills. Making improvements in spoken English can be done in various ways. Talking to friends or family members in English, practicing phrases and sentences, etc. are some of them. But most people feel some awkwardness talking in front of a crowd or even with friends. Here comes the helper Assistant, Miss AMMU Teacher. Miss AMMU Teacher is the advanced version of AMMU spoken English chatbot. In this work, a chatbot called AMMU, Automatic Mega-agent Managed User guide is built for helping this purpose. AMMU will communicate with the user and will help to improve the spoken English capability of users.

Recently, Convolutional Neural Network (CNN) gains more popularity over hybrid Deep Neural Network (DNN) and Hidden Markov Model (HMM) based acoustic models. CNN has the ability to deal with speech signals and it makes appropriate choice for the Automatic Speech Recognition (ASR) system. The sparse connectivity, weight sharing, and pooling allow CNN to handle a slight position shift in the frequency domain. This property helps to manage speaker and environment variations. CNN works well for speech recognition, but it was not appropriately examined for the Hindi speech recognition system. The activation functions and optimization techniques play a vital role in CNN to achieve high accuracy. In this work, we investigate the impact of various activation functions and optimization techniques in the Hindi-ASR system. All the experiments were performed on the Hindi speech dataset developed by TIFR, with the help of the Kaldi and Pytorch-Kaldi toolkit. The experiment results show that the ELU activation function with Rmsprop optimization techniques gives the best Word Error Rate (WER) 14.56%.

Now a day the application of Data mining in many areas has been tremendously increased. Data management and its processing is becoming the active research area for research community. In this paper the major concern is to study and analyse the impact and performance of various data mining techniques over sensor networks (wireless), as the characteristics of sensor nodes and its wireless nature are the primary concerns. The volume and rate for data generation is huge, which is variable in nature; therefore, it is very much essential to design and implement data mining methods for Wireless Sensor Network. This paper mainly focused on to the comprehensive survey of existing data mining techniques, wherein the various limitations and its probable solutions are highlighted in detailed. A transition of traditional mining techniques to newly introduced research is also analysed in this paper. Detailed working and description of compression algorithm has been stated in this paper.

The Internet of Things gadgets can acquire and transmit the data straightforwardly with different gadgets through the cloud, giving a gigantic measure of data to be assembled, stored and investigated for data-analytics processes. The aim of this paper is to enhance the patient’s quality of life by accessing real time visibility of the patient's condition, through measuring the physiological parameters like systolic, diastolic, pulse rate and body temperature values. The key idea is to administer care to the patients by constantly monitoring the medical parameters include blood pressure, pulse rate and body temperature without the need for the patient to move from facility to facility for constant supervision of their health. Data gathered from the blood pressure sensor and temperature sensor is analyzed and stored in the cloud, which can be monitored by the caregivers of the patient from any location and respond appropriately, based on the alert received.

Computer-aided diagnosis systems (CAD), as their name suggests, utilize computers to assist doctors to obtain a quick and correct diagnosis. They focused on several scholars as they are built upon the concept of processing and examining pictures of various parts of the individual body meant for a fast and correct outcome. CAD systems are generally area specific because they are augmented for some certain kinds of infections, various parts of the individual body, diagnosis methods, etc. They analyze dissimilar types of inputs given, for example, signs, test center, result, health pictures, etc. varying on their territory. One of the maximum common kind of diagnosis depends on medical pictures. Our approach is to develop a model to identify either a brain hemorrhage is present or not in Computed Topography (CT) scan of the brain and also identify the kind of hemorrhage. The process of detecting and identifying hemorrhage contains many steps like image pre-processing, segmentation of image, extracting the features, and classifying the images.

In modern times, every human being relies upon the internet for fulfilling their hefty needs as the internet offers a vast amount of information to users, so its availability to users is indispensable. Major objectives of security are availability, integrity, and confidentiality. DDoS (Distributed Denial of Service) is a universal cyber-attack that is a major intimidation for cyberspace. DDoS attack slows down network availability by overflowing illegal traffic over network bandwidth. Day by day, attackers improve upon their strategies by using new technologies and techniques. In this paper, a DDoS attack is proposed using python script. We focus on the volumetric DDoS attack effect on the performance of the server ultimately shutting it down. A DDoS attack is undertaken using a python script on the server in which multiple clients send multiple fake requests to the server to slow down the services/performance of the server.

Chatbots or the well-known automated conversational agents have become a raging trend among all the sectors of businesses as a result of the rapid transition happening towards automation in processes. They are already being used extensively and will spread their wings to newer horizons shortly. The basic model of Chatbots is to interact with the user to answer their questions using various modes like text messages, voice replies, or any other predefined suitable interface. This paper discusses the development of a Chatbot for the college, Prasad V Potluri Siddhartha Institute of technology, to answer various questions related to the college like the facilities, procedures, policies, etc. This is a web-based software application implemented using Flask framework. This model is designed to capture text inputs from the user through a console and outputs the response in text format using machine learning concepts. A retrieval approach is implemented to process the input and to respond with an appropriate answer using logic adapters. The performance of this model is analyzed using a questionnaire that uses various parameters like performance, humanity, effect, and accessibility. This paper presents the overall approach used to design the Chatbot and compares the web application as-is study with the to-be website when the Chatbot is incorporated. The web application along with the Chatbot showed a 20% improvement in performance and 5% increase in accessibility by analyzing the performance metrics.

The Movie Recommendation System (MRS) is part of a comprehensive class of recommendation systems, which categorizes information to predict user preferences. The sum of movies is increasing tremendously day by day, and a reliable recommender system should be developed to increase the user satisfaction. Most of the approaches are made to prevent cold-start, first-rater drawbacks, and gray sheep user problems, nevertheless, in order to recommend the related items, various methods are available in the literature. Firstly, content-based method has some drawbacks like data of similar user could not be achieved, and what category of these items the user likes or dislikes are also not known. Secondly, this paper discusses about collaborative filtering to find both user and item attributes that have been considered. Since there exist some issues pictured with collaborative filtering, so this paper further aims into hybrid collaborative filtering and deep learning with KNN algorithm of ratings of top K-nearest neighbors.

Ongoing advancements in the improvement of multilayer convolutional neural organizations have brought about upgrades in the precision of important recognition jobs, for example, huge category picture classification and cutting-edge automated recognition of speech. Custom hardware accelerators are crucial in improving their performance, given the large computational demands of Convolution Neural Networks (CNN). The Field-Programmable Gate Arrays (FPGAs) reconfigurability, computational abilities, and high energy efficacy makes it a propitious CNN hardware acceleration tool. CNN have demonstrated their value in picture identification and recognition applications; nonetheless, they require high CPU use and memory transmission capacity tasks that cause general CPUs to neglect to accomplish wanted execution levels. Consequently, to increase the throughput of CNNs, hardware accelerators using Application-Specific Integrated Circuits (ASICs), FPGAs, and Graphic Processing Units (GPUs) have been employed to improve CNN performance. To bring out their synonymity and dissimilarity, we group the works into many groups. Thus, it is anticipated that this review will lead to the upcoming development of successful hardware accelerators and be beneficial to researchers in deep learning.