Emerging Research in Electronics, Computer Science and Technology

This paper presents a new approach for distance determination between two persons based on human height reference using model with single monocular 2D camera. Depth estimation from single 2D image is very difficult problem. Local and global features of the image are important for depth estimation. In this paper, human height feature is used as reference for distance estimation. First, the image is captured from the 2D camera and converted into grayscale image, and then the presence of a human is determined based on person detection. Separation of images according to the number of human and segmentation of each image is performed. Next process, the depth information estimation, is performed for each separated image (Fidan et al., IEEE Trans Ind Elect 60(12), 2013) [1]. Finally, distance is measured using human height and distortion value. For this depth estimation problem, supervised learning approach has been utilized and corresponding depth model is built.

The main challenges in wired computer networks are transferring data packets through the most reliable path and optimizing the time taken for dynamic recalculation of the backup route during the path failure. In this paper, the path with the largest number of reliable links is considered as the best path. The link reliability is decided based on several parameters like packet delivery ratio, flow capacity, bandwidth, queue capacity of the link, delay and throughput. For each link, the cost is assigned in terms of reliability using the proposed cost function. Data is sent through the calculated reliable path, thereby avoiding path failure to the maximum extent possible. If there is a failure in the path during packet transmission, the backup path is dynamically calculated by considering different scenarios. The effectiveness of the proposed protocol is proved by simulation results.

Anomaly detection in network traffic is one of the major concerns for the researches and the network administrators. Presence of anomalies in network traffic could indicate a possible intrusion on the network, increasing the need for a fast and reliable network intrusion detection system (NIDS). A novel method of using an artificial neural network (ANN) optimised with Cuckoo Search Optimizer (CSO) is developed in this research paper to act as network monitoring and anomaly detection system. Two subsets of the KDD Cup 99 dataset have been considered to train and test our model, one of 2000 instances and the other of 10,000 instances, along with the complete dataset of 61,593 instances and I have compared the result with the BCS-GA algorithm and the fuzzy K-means clustering algorithm optimised with PSO in terms of precision, recall and f1-score, and the training time for the model with the selected database instances.

A numerous advantages of wireless sensor networks have led to discover various applications incurred in the field of health systems, home, industrial, commercial, military, and environment etc. The biggest privilege of using WSN is due to its low cost, maintenance, and other desirable features. One major application involved in WSN is animal health monitoring. By incorporating this, health status of the cattle can be observed and maintained based on variety of parameters using distinct sensors. The description about unique sensors that are associated in wireless sensor network in order to monitor the health status of cattle to erupt the ill health is presented in this paper. The disadvantages incurred in the existing system are overcome by the proposed system containing android application where all the cattle’s health-related data will be stored in a database.

Machine comprehension deals with the idea of teaching machines the ability to read a passage and provide the correct answer to a question asked from it. Creation of machines with the ability to understand natural language is the prime aim of natural language processing. A machine comprehension task is an extension of question answering technique which provides the machines an ability to answer questions. This task revolutionizes the way in which humans interact with machines and retrieve information from them. Recent works in the field of natural language processing reveal the dominance of deep learning technique in handling complex tasks which suggest the use of neural network models for solving machine comprehension tasks. This paper discusses the performance of code-mixed Hindi data for handling machine comprehension using long short-term memory network and gated recurrent unit. A comparative analysis on the basis of accuracy is performed between the two sequence models to determine the best-suited model for handling this task.

Stance detection is one step ahead of sentiment analysis where author’s stance for certain topics such as an event, personality, or a government policy is considered. The author’s stance could be in “favor” or “against” the topic under consideration. A myriad amount of data is being accumulated via various social media platforms. This work considers the Kannada–English code-mixed aspect of social media text. The corpus was collected based on various trending topics such as “Bengaluru molestation,” “currency ban”, etc. using particular word phrases. The user comments on social media platform Facebook was used to collect the corpus. The collected dataset was represented using different techniques such as bag of tricks, word embedding like Word2vec and GloVe, and pre-trained embeddings. These representations were further used in combination with various deep learning architectures such as convolutional neural network (CNN), bidirectional long short-term memory (Bi-LSTM). The results for various combinations are listed.

Cloud computing is a web-based computing where in the services based on Internet such as data, storage, and computing resources are delivered to the local devices through Internet. Meanwhile, wireless sensor network deals with monitoring and gathering of the information about the physical or environmental conditions. The integration of these two domains has tremendous advantages to improve the business infrastructure and performance over worldwide. The security of data on the cloud and calculation of trust and reputation of the cloud service providers (CSP) and sensor network providers (SNP) are the issues for this new paradigm. To fulfill these issues paper presents novel techniques 1) trust and reputation calculation 2) providing data security on cloud 3) choosing desirable CSP and SNP for the service. This paper proposes data security by taking into account the service of wireless sensor network (WSN) and cloud service providers. Our experimental results help users to choose the best service providers in case of both cloud and wireless sensor network. A better security for the data in the cloud is provided using trust and reputation calculation.

In this paper, the problem of classification of imbalanced text data is addressed. Initially, imbalanceness present across the classes is reduced by converting each class into multiple smaller subclasses. Further, each document is represented in a lower-dimensional space of size equal to the number of subclasses using term-class relevance (TCR) measure-based transformation technique. Then, each subclass is represented in the form of an interval-valued feature vector to achieve the compactness and stored in a knowledgebase. A symbolic classifier has been effectively used for the classification of unlabeled text documents. Experiments are conducted on Reuters-21578 and TDT2 text datasets. The results reveal that the performance of the proposed method is better than the other existing methods.

The discrete tomography which focuses on the binary image reconstruction from two orthogonal projections, whereas the Mojette transform allows for a more general framework with any kind of values and any number of projections. The line Mojette was derived to focus on the limited number of materials scene reconstructions. Meanwhile, the scene only contained limited number of materials. So projection line might contain a unique material value, and then pixels of this line can be back-projected directly with the unique material value. The merits of the inverse line Mojette compare the Dirac Mojette that it can always solve more than one pixel at a time. In this paper, we discussed the line Mojette algorithm performances for the 2, 3, and 4 materials images. After comparison of analysis, results show the line Mojette performance ought to be paramount in the 3 materials case.

Nowadays, image segmentation performs a large-scale contribution in the medical images. Magnetic resonance (MR) image is a good substantial strategy during the fetal and neonatal periods that make earlier diagnosis of individual tumor analysis. It is used as topmost image investigation model for pattern discovery of brain tumor. The aim of our study is to discriminate the unique images related to every one of the patterns of the brain tumor on neonatal MRI. Our proposed system consists of four steps. The first step is to explore the acquisition of the neonatal MRI sequence. The second step performs segmentation of the locality of abnormality. The third step is the segregation of tumor from edema and its enhanced sequences. The fourth step is the estimation of the drawn area of tumor and calculation of the mean. Our proposed method successfully detects the tumor area in the given neonatal MRI images.

Decision tree is a well-established technique for classification in data mining and machine learning. Induction of a decision tree using high-dimensional dataset may lower the performance of the decision tree in terms of classification rate and stability. Vertical partitioning is a novel paradigm to avoid these issues; it divides the features of a dataset into subsets and creates a subset-based classifier ensemble. In our previous work, we proposed a theme-based decision tree classifier ensemble using vertical partitioning for teacher recruitment modelling. In this paper, we extend our previous work in terms of exhaustive experimental analysis to address both high-dimensionality and instability issues of decision tree using five standard datasets. The performance of the theme-based method is evaluated using classification rate, standard deviation and misclassification rate. Our experimental analysis confirms the superiority of the theme-based approach over traditional decision tree approaches.

Distributed applications are vulnerable to more unauthorized breaches in a service executing environment which runs trusted methods. The lack of security which is concerned with trust attributes decreases with many factors in upgraded trust at vendor’s end, and this leads to a lot of obstacles at vendor’s end for many of the deployments. A model which is really updated enough for executing the operations in trusted environment is required to reduce trust issues. To mitigate the risk level of security, the operational code executed in the trust wrap premises must be enabled with moderated or decreased trusted code. Processing the steps which are highly sensitive is handled in these environments only at trusted regions, whereas in untrusted part, the regions are not processed. Swarm intelligence (SI) is a technique which influences recent trends computing applications, and few of the SI techniques are particle swarm optimization (PSO) and ant colony optimization (ACO) which are revised for better optimization. Likewise, secure cloud services require a properly coordinated and decision-making system for a better-trusted transaction between cloud vendors and users. In this paper, we have implemented a decision-making system by deploying Boid-based feedback analysis that coordinates between its clusters that forward proper feedback for running the service on a trust lacking environment in mobile cloud nodes. Here we set a decision resulting layer under the distributed services to inject feedback results among the respective cluster. Therefore, an optimized-trusted approach is been introduced to manage the coordination between cloud providers.

A wireless sensor network (WSN) is an independent network comprising of huge number of sensor motes and base stations (BS). Security is a major concern in military forces and border surveillance. BS which collects information from sensors has become the main target of attack for intruders. Anonymous intruder eavesdrops into network, analyzes the radio patterns to get contextual information, and can find where the BS location is. Attacker can destroy BS or alter its behavior which results in malfunctioning or complete failure of the entire network. In this paper, a hybrid algorithm for the BS anonymity has been proposed which provides protection for the sink node against adversaries. By making the BS anonymous in the network, it will be difficult for any intruder to find the location of the BS. The proposed algorithm is developed and simulated in MATLAB. The results infer that the anonymity of BS is independent of traffic volume and the algorithm proposed is efficient with varying locations of the sensor nodes. The anonymity value of BS for each topology is below 0.04. This infers that for the topology simulated an intruder conducting the traffic pattern analysis of WSN considered have a less than 4% probability of detecting the BS on the first trail when searching for physical location for the BS.

Economy of a country is closely related to stock market. By analysing stock market performance, we can evaluate whether a country’s economic growth is increasing or decreasing. Even though country’s economic growth can be understood by predicting stock market, it is highly unpredictable. We used dynamic mode decomposition which is a spatio-temporal, equation-free, data-driven algorithm for stock market prediction Schmid (J Fluid Mech 656:5–28, [13]) by considering stock markets as a dynamical system. How the system evolves and prediction of future state is done using DMD by decomposing a spatio-temporal system to modes having predetermined temporal behaviour. We used this property of DMD to predict stock market behaviour. In Kuttichira et al. (2017 international conference on advances in computing, communications and informatics (ICACCI), IEEE, pp 55–60, [7]) DMD was used to predict Indian stock market for minutewise data. We used daywise data as our timescale. Time series data of stock price of companies listed in National Stock Exchange were used as data. Sampled daywise stock price of companies across sector was used to predict the stock price for next few days. Predicted prices were compared with original prices and mean absolute percentage error was used to calculate the deviation for every companies. We analysed the stock price prediction for both intra- and intersector companies. We used dynamic mode decomposition to predict the stock price using historical data. We also did fine tuning of sampling windows to find out the best parameters for our data set.

The task of mapping graphemes or phonemes of one language into phoneme approximations of another language is known as machine transliteration. The paper describes transliteration of Romanized Kannada words to Kannada script. Our system utilizes a bilingual corpus of around 3 lakh words, which comprise pairs of Romanized Kannada word with its corresponding word in Kannada script and employs orthographic and phonetic information. Romanized Kannada words are segmented into phonemes by predefined rules. These phonemes are transliterated to Kannada script statistically based on probabilities. The paradigm is assessed with 3000 Romanized Kannada test words and obtained an accuracy of 85.4%.

Knight’s tour has been an intriguing chess puzzle for many computer scientists and mathematicians. In this paper, we describe and analyze two of the most commonly used algorithms to solve the Knight’s tour puzzle in computer science—Warnsdorff’s rule and backtracking. Various case studies are performed to showcase the differences in time complexity of each algorithm. We also point out that the starting position and dimensions of a board play a key role in determining whether a tour is possible. By doing this, we prove that a tour does not exist when the knight starts from odd squares on boards of odd dimensions.

Fetal ECG contains possibly exact information that may support clinicians in making apt decisions during pregnancy and labor. Hence, extraction of clean fetal ECG is extremely crucial for fetal surveillance. This is done by placing electrodes on mother’s abdomen. However, it is tainted with various sources of noise. This paper compares LMS adaptive filter for FECG extraction with neural network-based adaptive filter. Twenty cases of clinical data were used. Experimental results showed superiority of later scheme in terms of SNR and MSE.

Network flow is an important topic in the field of algorithms which has more than half a century of research history and is an active area of research even now. Quiet a good number of real-life problems exists, like those involving the flow of liquids through pipes, current through wires, and delivery of goods in the field of transportation can be modeled using flow networks. Maximum flow problems appear in many practical applications. A flow network in graph theory is a directed graph involving source (S) and sink (T) and various nodes linked with edges. Each edge has an individual capacity which is the maximum limit of flow that an edge can allow. This paper proposes an algorithm to find a path which is the maximum flow rate allowed for data between the source and destination in a network. We implement the maximum flow algorithm in order to determine the optimal flow in a network.

The realization of IoT boom has motivated scientific community to enhance IoT into a more acceptable modeling of real world. Social Internet of Things (SIoT) is a step in this direction. The concept of SIoT presents social relationship aspect to network of IoT devices. It also represents real-world activities in the form of concrete model. Hence, SIoT may be termed as IoT network with social relationship affiliation. The nature of relationship among various SIoT members determines the interactions that will take place among them. Work done by researchers presents various models upon which social relationship interactions can be based. Among them the concept of trust which is the most appropriate way to quantify the reliability of interactions among members of the SIoT is found. In this paper, different trust management models for SIoT are reviewed. Trust model based on multiplicative attribute graph (MAG) model is considered as the basic concept of this survey. The different trust models are also compared using different metrics.

Healthcare or health insurance agencies have been in a continuous state of change, especially during the present technology expansion. They are beneath colossal pressure to forecast customer health issue and to generate surplus premium holders which will simultaneously reduce the cost. Examining and utilizing the vast data available are critical for healthcare companies in designing various strategies in the future. Many such healthcare organizations have already moved toward data mining and analytics with data warehouse methodologies and business intelligence with statistical analysis. However, further adaptation is required, as they must use different data from new sources in a blend with the prior sources. To consider this adaptation, predictive analysis technique is proposed. Predictive analysis comprises diverse statistical methods from predictive modeling, machine learning, and data mining that analyze present and past realities to make predictions about future. There are several advantages of using descriptive and predictive analytics in healthcare domain for concrete decision making of cost-effective solutions to their customers. This paper expands upon risk mitigation tactics to foresee high-risk patients. This is done by considering clinical data that is an electronic medical record and evaluating risk associated using stacked ensemble machine learning techniques. This technique helps achieve higher predictive accuracy of 90.17% and specificity of 94.90% in identifying high-risk patients from a lesser amount of data. It has been centered on payer analytics and customer analytics in light of novel machine learning algorithm to give full-cycle knowledge toward cost reduction and improvement in nature of care.

Over the years, data generated by the social media is extremely large in volume. This large volume of data is known as big data. Several physical resources are required to store big data; in order to avoid, these cloud resources can be used. Cloud resources are provided by Amazon Web Services, OpenStack, Rackspace, and many more. Stored big data can be used for analysis of traffic management, call center optimization, real-time fraud detection, social media, and sentimental analysis. Solutions for analyzing big data on cloud are presently not available. Users must obtain the essential cloud computing resources and install the necessary software manually. This can be a cumbersome task for complex distributed services. To address this issue, the services should be viewed as a single application consisting of virtual machines. Users should no longer be concerned about individual machines or their internal organization. To overcome this problem, an OpenStack cloud system with multi-node setup to provide Hadoop as a service to the users has been implemented. Users can easily access the Hadoop service which is provided by OpenStack cloud environment.

The societies are changing toward new and advanced directions due to new technological requirements. Flexibility, scalability, reliability, and security play a vital role in the future Internet, thus has to be achieved efficiently, so this inflexibility causes a challenge. Over the past few years, the well-liked topic in the era of networking is Software-defined networking (SDN). This gives rebirth to the programmable networks which/that open a door for new network paradigm. OpenFlow protocol is a basic building block of SDN networks. SDN operates on OpenFlow protocol through which the idea of programmable networks is accomplished. SDN converts the traditional network architecture by separating network logics into two different planes: control plane and data plane. SDN is extended for various applications; this causes imbalanced network utilization. The paper begins by introducing the SDN to this background OpenFlow architecture and SDN architecture is discussed and to examine the existing challenges in SDN. Finally, an emulation of SDN environment is presented using Mininet, which implements load balancing for the SDN with traffic segregation based on input traffic. The results show that solution provided helps to balance the performance of network in terms of Throughput, Delay, and Packet loss that fosters maximized QoS.

Agriculture is the backbone of India which indirectly contributes for an Indian economy. Farmers who are the drivers of agriculture are facing lot of problems for proper identification of the crops that can be cultivated for the specific soil conditions and to maximize the crops yield. All these problems are due to lack of technology and scientific techniques being used in agriculture. Crop yield varies as a result of variations in atmospheric and soil conditions. Data mining mainly focuses on methods to elicit useful knowledge from the dataset. There are several data mining approaches that can be used for the purpose of predicting crops yield and finding association among attributes contributing for the crops yield. This paper mainly intensifies on various association algorithms, namely Apriori, Eclat, and AprioriTid to find the association among temperature, rainfall, soil pH, soil nitrogen, and paddy yield.

The advent of blockchain technology has been set to revolutionize the real estate industry, and the potential changes are already taking its shape. The real estate industry is one among the top global sectors that is driving the economic growth of any country. The growth of this industry is well complemented by the growth of the corporate environment, demand for office space, industrial plots, urban housing accommodations, agricultural lands, etc. However, the existing world of real estate is complicated with the lack of transparency in its transactions such as leasing, purchasing, and sales and fails to attain the level of confidentiality and authenticity of operational data. Several aspects of its operations such as property sale prices, sale history, lease rental rates, market valuation, and so on expect greater demand for transparency, data integrity, and security—a trusted environment. As a result, property-related information can be made available as digitized information and hosted as decentralized database of records on distributed systems with lesser incidence of fraud and inaccuracies. The possibilities of mutual delusion and concealment during the sale, rental, and lease processes can be eliminated, thus achieving digitization of real estate ownership. Everything from making real estate investment decisions to choosing investment properties, blockchain can become a common practice by incorporating its wide benefits. This paper aims at revamping the tedious, paper-based, time-consuming system of ownership and renting into a next-generation digitized system based on blockchain technology.

Grade 316L stainless steels can be easily welded by all types of fusion welding processes. Pulsed gas metal arc welding (P-GMAW) is widely used in industries. By melting continuously fed current-carrying wire, P-GMAW achieves coalescence of metals. However, to achieve good quality of weld and attractive looking, P-GMAW needs consistent, high-quality welding procedures. This need is due to continuous control metal transfer that is necessary in P-GMAW for thin metal workpieces. This paper explores how image processing could be applied in assessment of mechanical properties in destructive testing of SS304L material weld bead. Image features like height, area, perimeter of weld bead have been extracted for different loading conditions using image processing by multivision techniques. The vision techniques play an important role in quality inspection and process monitoring. Multi Vision technology improves the edge recognition, pixel processing and reliable consistently achieved. From the study, it is found that multivision is capable of quantifying the parameters associated with soundness and performance of weld joints, and the established trend using image processing features is correlating well with traditional measurement.

Syntax analysis is the process of checking grammatical correctness of sentence by determining the relationship among words present in the given sentence. This paper proposes a syntax analyzer for Kannada sentences using (Kaaraka) framework. This proposed work is tested on stories and novels dataset from Enabling Minority Language Engineering (EMILLE) corpus and obtained an accuracy of 75%. Syntax analyzer is an essential tool in most of the natural language processing (NLP) applications like machine translation system, word sense disambiguation, and question and answering.

With the huge development in cloud computing services, cloud storage becomes an important service for individuals and organization, with the capabilities of on-demand outsourcing application. After outsourcing the data, owner has no physical control over remotely present data. To verify the data integrity, several efficient cryptographic techniques related to provable data possession (PDP) were proposed to check the remote data integrity without downloading it from the cloud storage server. In this paper, with the knowledge of interactive PDP protocol, we propose an efficient public auditing method based on probabilistic queries and periodic verification to improve the performance of audit service. The method reduces the leakage of the data by public auditing and reduces the computational and communication cost of audit services by selecting the secured parameter. Our experimental result shows it incurs less communication cost.

With the quick evaluation of Web services on the web, we find that quality of Web services plays a significant role for the selection of Web services. When a user has a list of functionally match services then service selection decisions are often made based on the non-functional attributes value such as reliability, availability, price, and rating. and the values of these non-functional attributes of the Web services are given by the user. Different decision strategies may be followed by different users for selection of Web service. In this paper, we discuss the different approaches introduced by many authors for selection of Web services based on QoS and the issues in the existing work after that we proposed a selection and ranking system based on multiple decision strategies.

The usage of vehicles has been rapidly increasing and the entry of the unauthorized vehicles in the campus has become a hectic problem. In this situation, the detection of unauthorized vehicles plays an important role nowadays. In these scenarios, the vehicle number plate recognition system has attracted many of the researchers to work with the concept of image recognition and processing. Theft of vehicles, breaking of traffic rules, entering into the restricted space, so on are increasing day by day. Thus to break this act, vehicle license registration code recognition is necessary. The recognition system can avoid the problem of vehicle theft, breaking of traffic rules, restriction of unauthorized vehicles to the secured area, and so on. The work focuses on recognizing the individual character within the registered license plate and aims to achieve high accuracy by optimizing many parameters that have higher recognition rate than the conventional techniques.

Lately, wireless sensor networks (WSNs) have been sent for a wide arrangement of utilization, including military distinguishing and following, industrial status checking, action stream checking, where tangible devices oftentimes move between various regions. Securing information and correspondences require suitable encryption key conventions. In this investigation, propose a certificate less-powerful key administration (CL-PKA) protocol for secure correspondence in one of a kind WSNs portrayed by hub versatility. The CL-PKA bolsters productive key updates when a node leaves or joins a bunch and guarantees forward and in reverse key mystery. The convention likewise bolsters effective key revocation for traded off hubs and limits the effect of a node compromise on the security of other correspondence joins. A secure examination of our plan demonstrates that our convention is effective in safeguarding against different attacks. This inquire about actualize CL-PKA in Conic OS and reproduce it utilizing Cooja test system to evaluate now is the ideal time, vitality, correspondence, and memory execution.

Recognition of handwritten character and digits is one of the major challenges in the computer vision system. One of the major application areas of character and number recognition is the financial sector, which deals with enormous amount of document data. In this paper, we have proposed to provide an automation system for bank cheque processing. The proposed system takes the cheque images as input and extracts the digits from account number, date and amount field, respectively. Initially, cheque images are preprocessed, and then digits are segmented using simple connected component analysis. Extracted digit images are normalized and given as input to convolutional neural network (CNN) classifier. Classifier is trained using large data set: MNIST and built-in MATLAB digit data set along with our data set. Total samples of 91,000 images are used for training and testing. Post-processing is done to construct the whole numbers for account number, date and amount. Recognition rate achieved for 50 cheque images is 95.59%.

Wireless adhoc networks are an infrastuctureless network of nodes with applications in many fields like agriculture, forestry, and military. Attacks like selective forwarding and denial of service occurs in network due to malicious attackers who intentionally want to take advantage of network for their own selfish needs. Many existing solutions to detect attacks have been proposed in literature, but the false positives are high in those solutions, thus it necessitates designing a solution with low false positives, high accuracy, and lower attack detection time. In this direction, a novel routing protocol for detection of attacks and avoidance of malicious nodes based on cluster-based topology is proposed in this work.

In medical field, breast cancer is the most widespread cancer among women worldwide. Effective diagnosis of breast cancer remains major challenge in research. However, breast cancer can be prevented by detecting at the early stage. Early detection is extremely important which reduces the time required for treatment. There is a scope of research on wearable technology to detect breast cancer, as the technology evolving to a point where we can wear a sensor and monitor the health of the breast at patient comfort rather than going for mammogram. This paper aims at describing the research progress and advantage on wearable devices to help in detecting breast cancer at the early stage. Study’s outcome can be applied to develop a new efficient device to detect cancer for further research and study.

The goal of any organization is to make their product to get succeed and compete with other products in the market where pricing of their products plays a vital role. To sell any product in market, the most important aspect is to determine the price. There are many traditional and new methods for estimating before pricing their products, and a method is chosen which gives more appropriate result. In this study, support vector regression analysis is used as a machine learning technique in order to predict the market price of smartphones based on their features. Many variants of features are utilized for data preprocessing or input technique for SVR model. If required factors are derived and used accordingly, it can provide a good prediction result. Different features of the smartphone are considered in this experiment in order to get more reliable outputs. Support vector regression gives more promising predictions for making better decisions in price prediction of smartphones compared to other models.

High demand of Internet by the society and digitization of most of the civil services have increased the complexity of core network. To make easy the life of administrator of network, new technology is required and software-defined networking (SDN) is better solution. SDN provides ease of configuration, flexibility, agility, and security to manage the network. The main goal of SDN is separating the network configuration from traffic forwarding. This leads to gain control over the devices to forward required amount of traffic within the network. Load balancing in the Internet is essential to keep the performance of network at the expected level. SDN balances the load in the network more efficiently than traditional network. Authors propose the distributed load balancing algorithm using weighed round-robin (WRR) mechanism. Analytical model of WRR is proposed to represent the task of load balancing algorithm. Mininet emulator is used to set up the experimental test bed. Throughput and delay parameters are used to test the performance of the proposed algorithm. Results show the reduction of delay and increase the availability of bandwidth.

Stress occurs in a human being when they are faced with exigent situations in life. Assessing stress has been always challenging. Smartphones have become a part of everyone’s day-to-day activity in the present time. Considering human–smartphone interaction, sensing of stress in an individual can be assessed as today’s youth spends most of their time with smartphones. Taking this into consideration, a study is carried out in this paper on assessing stress of an individual based on their interaction with the smartphone. In this work, human–smartphone interaction features, like ‘swipe,’ ‘scroll,’ and ‘text input,’ are examined. Text input is incorporated by disabling the autocorrection and spelling checker features of the keyboard. Moreover, sensor data is used by Google activity recognition API to analyze the physical activity of the individual to assess the stress level.

The mechanism of knowledge delivery system in a cloud-based e-learning system can be significantly improved if an advanced analytical operation can be performed over it. Although, there are various research attempts that have used big data approach on educational data by introducing various sophisticated techniques; however, the problems associated with the elementary structure of data is less reported. Therefore, this paper presents a novel concept of virtual map generation that is meant for repositing the structured data after it is transformed from unstructured educational data. The proposed system offers the advantage of independence from any existing software frameworks to introduce a simplified approach that can not only perform an effective transformation but can also perform an analytical operation using a semantic-based approach. The system also offers the benefit of faster response time with better consistency over educational big data.

Agricultural dependent countries require weather forecasting to facilitate the agricultural activities. Thus, the accuracy of weather forecasting is of utmost importance for these countries. The artificial neural network is one of the most widely used techniques for weather forecasting especially for rainfall prediction. Also, chaos theory is another widely used technique for weather forecasting. Artificial neural networks are generally classified into back-propagation neural network, recurrent neural network, and time delay neural network depending upon the architecture and type of inputs. We have considered the five regions namely Bengaluru, Chikmagalur, Dharwad, Kolar, and Mandya, which fall into Karnataka state for our rainfall prediction. The input parameters are monthly rainfall, monthly minimum temperature, and monthly maximum temperature of the above-mentioned districts from the year 1951 to 2000. This paper compares the two commonly used forecasting methodologies by building training and testing data sets and finding the number of hidden neurons and the number of layers in neural networks, the number of dimension in phase space diagram for the best performance and hence, find which method gives the most accurate prediction results.

Conventional vital signal acquisition systems use external sensors and peripheral devices for the collection of data. The mobility of a subject is limited while monitoring the vital physiological parameters of that subject. These systems tend to be bulky and also expensive. In this paper, we have designed and developed a low-cost wireless wearable vital signal acquisition system with an activity tracker to monitor and analyze physiological parameters such as electrocardiogram (ECG), photoplethysmography (PPG), galvanic skin response (GSR), body temperature based on the wearer’s activity. The data is acquired from sensors using Arduino board. The acquired signals are then processed and analyzed in real time to identify the relationship between the physiological parameters. The processing of signals includes denoising, motion artifact removal, trend analysis, parameter retrieval. The retrieved parameters are heart rate, pulse rate, HRV, and sleep cycles. Activity monitoring and eHealth monitoring are the common applications of the designed device. Conventional systems record the signals and do not perform analysis on them. Here, we perform analysis of physiological signals which provide information about early onset of chronic diseases. Thus, the system helps in maintaining a healthy lifestyle by keeping away from the chronic diseases.

Water plays an important role in flock’s performance as it is a most important nutrient. Layers can survive longer without food than they can without water. For better performance of layers, water is essential and it should be of good quality. Water quality refers to chemical, physical, biological, and radiological characteristics. A water quality testing system is developed for sensing different chemical and physical characteristics. Experimentation is carried out to test the water quality of different water sources in poultry farm. The sensed values are compared with laboratory experimental results. The system is cost effective and provides real-time monitoring.

In today’s world, a huge amount of data is stored in the form of electronic documents in the World Wide Web. Text classification algorithms have been widely used for classifying those text documents into a fixed number of predefined classes. The applicable scopes and their performances of these algorithms are different. Therefore, finding an appropriate algorithm for a dataset is becoming a significant emphasis for researchers to solve practical problems quickly. This paper puts forward an experimental evaluation of five significant text classification algorithms with each other and with TF and TF-IDF feature selection methods built using decision tree (C5.0), support vector machine, K-nearest neighbor, Naïve Bayes, and neural network on four public datasets, namely 20news-bydate, ohsumed-first-20000-docs, Reuters 21578-Apte-90 Cat, and 20 Newsgroup. The experimental results are examined from multiple perspectives and summarized to provide usefulness of different algorithms on different datasets.

In recent years, big data applications with scheduling algorithms have evolved lot due to the advancement of new technologies and techniques. We are living in digital data world where the data size is in terms of Exabyte or Pico Byte. This large volume of data is referred as big data. In today’s business environment, the performance of applications largely depends on the efficient retrieval of relevant data on time; the data analysis and retrieval of relevant data need to be done at faster rate. The traditional scheduling algorithms will not be efficient to handle such huge volume of data, considering the above facts managing big data applications and scheduling of big data on distributed architecture has become a challenging research area in the last three–four years. To process such huge volume of data, efficient scheduling algorithms need to be adopted to achieve better performance. The existing MapReduce implementation on Hadoop framework on single node cluster limits themselves to implement all the jobs on single node cluster. In this paper, we will discuss different scheduling techniques and their performance effects on a multimode clusters. The parameters considered for performance evaluation are CPU time, physical memory, and virtual memory. The main aim is to provide survey of different scheduling algorithms that can be used across distributed architecture to achieve better performance in analysis of big data considering YouTube dataset. The results interpret that capacity-based scheduling algorithm is more efficient as compared to FIFO and FAIR in terms of CPU cycles, physical and virtual memory utilization.

Recommender systems (RS) are data sieving method that recommends on the basis of liking of users and category of products from big data. In the same way, videos and movie recommender systems give some technique which helps the users to classify movie according to user’s similar interest. This mechanism process makes recommender systems very important tool for Web site and various digital marketing applications. This paper focuses on the movie recommender system by using data clustering and nature-inspired algorithm. K-means algorithm is widely used algorithm for clustering due to its simple nature, handling large amount of data and running time is low. But it falls into local optima due to its randomly generated initial centroids. This algorithm can achieve global optimum solution if it is integrated with nature-inspired algorithm. This paper integrates k-means with nature-inspired algorithms (bat, firefly, cuckoo, modified cuckoo search) on data set (movielens). Outcomes are compared on the basis of objective function, less the objective function better the results. Our proposed system (k-mean modified cuckoo search) gives improved outcome than other algorithms.

Human blood is very important to save the life of other people. During blood donation process, blood is directly collected from donors, processed and stored in the blood bank. It must be managed effectively for the need during emergency in hospitals. The area of transfusion medicine, specifically blood donation services requires an intelligent system for automation of the process. Hence, an intelligent system that can integrate major operations involved in the blood bank, make efficient decisions and improve communication is required. A system of this sort would involve machine learning algorithms for efficient donor selection. Accurate prediction of the number of blood donors can help medical professionals know the future supply of blood and plan accordingly to entice voluntary blood donors to meet demand. In this research, the pattern of blood donors’ behaviors is based on factors influencing blood donation decision that is conducted using online questionnaire. To find out the potential individuals to become the blood donor the factors like altruistic values, knowledge in blood donation, perceived risks, attitudes toward blood donation, and intention to donate blood, are analyzed. To predict whether individual person is a donor or not from the data given by the person, Naive Bayes technique and K-nearest neighbors (KNN) algorithm are used. The results indicate that the accuracy value for KNN is higher than the Naive Bayes algorithm. The database can be used to track potential blood donors.

Worldwide, cancer occupies the second position in the list of diseases associated with the high mortality rate. Cancer is the condition, where in the body immune system will not be having any control over the population of the cells and their division process. Oral cancer is the cancer associated with the oral cavity or the mouth area. The detection of cancer from the photomicrographs is done manually by the pathologists and oncologists. This paper tries to automate the detection process by locating the keratin pearls which are rose like patterns found in the photomicrographs of malignant tissues. This is achieved through the Tensorflow object detection API freely available through the github repository. We have used four different models used for the coco data set. The analysis of the same is presented here.

Recommendation systems (RSs) are used by different e-commerce sites like Amazon, eBay, etc., for suggesting relevant recommendations based upon users’ preferences or items purchased by people of similar interests. Recommendation systems help the users to identify the items which may be worth to buy. This is referred to as top-N recommendations. So, the ultimate goal of RS is to find out which item or product is more relevant as per user preference. In this paper, our main focus is to elaborate the recommendation process with follow-up of book recommender systems in the existing works and challenges associated with them. In order to deal with one of the recommendation challenges, we proposed book recommendation approach based on identification of opinion leaders in detected communities of social networks by using information related to their interests, preferences, age and online available attributes on social networks. This approach can be helpful in solving user cold-start problem, i.e. generating recommendations of books to new user in the absence of availability of purchase history. The same approach can be used for solving an item cold-start problem also, if none of the users had rated it till now.

In the biomedical field, huge amounts of data have been produced day by day. These data drives the development of the biomedical area researches in so many ways. This paper mainly focusing on biomedical named entity recognition (NER) with the aim to enhance the performance through deep learning. Impressive results in natural language processing are made possible by deep learning techniques. Deep learning enables us to use them for NLP tasks and producing huge differences in accuracy compared to traditional methods. NER is a crucial initial step in information extraction in the biomedical domain. Here we use RNN, LSTM, and GRU on GENIA version 3.02 corpus and achieves an F score of 90%, which is better than the most state-of-the-art systems.

Advances in GPU, parallel computing, and deep neural network made rapid growth in the field of machine learning and computer vision. In this paper, we try to explore the convolution neural network to classify animals. The convolution neural network is a powerful machine learning tool which is trained using a large collection of diverse images. In this paper, we combine convolutional neural network and SVM for classification of animals. The animal images are trained using AlexNet pretrained convolution neural network. Further, the extracted features are fed into multiclass SVM classifier for the purpose of classification. To evaluate the performance of our system, we have conducted extensive experimentation on our own dataset of 5000 images with 50 classes, each class containing 100 images. From the results, we can easily observe that the proposed method has achieved a good classification rate compared to the works in the literature.

Wire electrical discharge machining (WEDM) is a thermoelectrical machining process capable of accurately machining parts with varying hardness or intricate shapes. Improvement of the process productivity by avoiding wire breakage is one of the main research fields in wire-cut EDM. In a wire-cut EDM, the electrode is the only element that requires frequent changes due to failure. In order to replace the wire electrode in time, there is an essential need to keep a watch on the condition of the wire electrode during the machining process. Texture of the machined surface is dramatically affected by the worn electrode. Electrode status monitoring can be done by analyzing the surface of machined component. The measurements of surface roughness by traditional devices are time consuming and also obtained by scratching the surface of components. Consequently, these problems can be overcome by the vision system. In this work, to determine the surface roughness of the WEDM components, a machine vision system has been utilized. In order to check the effectiveness of the vision-based results, various surface roughnesses were produced on WEDM using Design of Experiments technique. Stylus-based parameter Ra was acquired and compared with vision-based parameter (Ga). The experimental result indicates that with a reasonable accuracy and by using vision system, surface roughness could be predicted. Results clearly indicate that wire electrode status monitoring in WEDM can be successfully carried out by analyzing the image of surfaces.

The theme of this paper is to analyze how a student can perform in an examination by answering the questions by selecting from various units. This paper focuses mainly on the common units of the questions to which students can make their maximum attempts to write the answers and the units to which the students will make rare attempts. It is considered to be one of the attempts made in this regard which will enlighten teaching community as well as college management to focus where exactly risks will be there to both students and teachers. From last 4 years, AICTE has made mandatory rules to all the technical education institutes in this regard as a part of accreditation certificate. It helps in framing the syllabus as well as setting the question papers by considering number of chapters for a given unit and number of units for a given subject. The main theme behind this performance analysis lies with the units with which maximum students will make an attempt to answer which in turn improve the confidence level of the teacher in setting the type of questions and framing trickiness in the questions. Here, there were two approaches which have been suggested both in traditional manual method and soft computing method. In the manual method, a program application was written which makes the group of the students based on the total marks scored by the students. In soft computing method, it was used with the basic feed-forward neural network to group the students by taking the split marks of individual units of the student. These splits marks will be considered as features to train the network. At the end, a fruitful positive result was received in both manual methods and soft computing methods.

From the initial stages of Internet till now, the number of Internet users has increased drastically over time. In today’s era, it is not just about the Internet of people, more smart and independent devices connected together have come along giving rise to a new technology “Internet of things.” The Internet of things has become a stride in Internet evolution. It saddles the insights of billions of sensors and connected devices that gather enormous data to make relevant decisions. To keep up the pace with this new wave on the Internet, changes in the current TCP-/IP-based Internet model were made and also new future Internet architecture NDN is introduced. The principle reason for this work is to show which Internet architecture is more qualified to meet the needs for future Web and Internet of things. This work will help in determining the gaps that need to fill for these Internet architectures (IPv6 and NDN) in order to meet the growing needs of Internet of things.

A machine learning approach to identify the risk factors of preterm birth (PTB) is presented. Dataset used for analysis is collected from the local hospitals of Mysuru. For the analysis, the main risk factors considered are age, number of times pregnant, diabetes, obesity, and hypertension. The dataset contains the details of women having either diabetes mellitus (DM) or developed gestational diabetes mellitus (GDM) in pregnancy. Two different prediction models, namely support vector machine (SVM) with linear and nonlinear kernels and logistic regression, are used. Data-imbalanced problem is handled by using Synthetic Minority Over-sampling Technique (SMOTE). The performances of the models are measured using accuracy, specificity, and sensitivity.

Biometrics plays a vital role in security through authentication and verification. Securing the biometric template is the current challenge field of research. As intruders attack the biometric system which affects the applications which are being used for human authentication. To name a few border security, access control surveillance and so on. Vulnerabilities too affect the protected templates at the time of biometric verification process. Biometric templates can be protected by using the techniques like transformation or cancelable biometrics, biometrics cryptosystem, and hybridization. Most of the vulnerabilities or attacks are resolved by developing hybridization process like combining biometrics template with user knowledge keys or passwords for secure authentication. The challenges in template protection are like alignment of template, selection of pseudonymous identifiers (PI), and auxiliary data (AD). An attempt is made to review the literature in this direction and present few issues and challenges in the field of biometric template protection.

In recent years, data stream has been considered as one of the primary sources of big data. Data stream has grown very rapidly in the last decades. Data stream environment has many features distinguishing the batch learning data which arrives on the fly with high speed. Data stream mining has attracted research focus due to its presence in many real-time applications such as telecommunication, networking, and banking. One of the most important challenges in data stream is the distribution of data is changing continuously which is leading to the phenomenon called “concept drift.” Another issue for streaming data is dealing with imbalanced class in the dataset. Many classification algorithms have been made to cope with the concept drift; however, many of them are dealing with the drift from the balanced data. In this paper, we propose a model called “CD2A: Concept Drift Detection Approach Toward Imbalanced Data Stream” which aims to handle the imbalanced data and detect the concept drift and behave equally with different types of drift. The algorithm was evaluated on real and synthetic dataset and compared with leading edge methods AWE, SMOTE, SERA, and OOB. Our method performs significantly better average prediction accuracy than the other compared methods.

Traditionally employed Human-to-Human (H2H) and Human-to-Machine (H2M) communications have recently been reintegrated by a new trend known as Internet of Things (IoT). These will captivate new clients and increase in the data flowing in their networks. Internet of Things enables Machine-to-Machine (M2M) communication without any human intervention, hence offers many challenges. In this model, the device self-sustainability due to limited energy capabilities presents a great threat for all Internet of Things devices. In this paper, we have conducted a brief literature survey on Internet of Things that will explain about IoT concept, challenges, and energy harvesting.

The span of modern medical imaging provides new and efficient techniques for segmentation of liver that are used by the clinicians to view in order to diagnose, monitor and treat liver diseases. Liver cancer is one of the most prominent diseases which cause death. Extraction of liver in different modalities is a difficult task because of its varying shape, similarity between organ intensities and variability in liver region intensities. In this review paper, a study has been carried out on liver segmentation in CT and MRI images with different methodologies and datasets. The observation has been made to highlight the merits, demerits and performance metrics of different works published.

Network forensics is one of the investigation areas, focusing on capturing the packets, monitoring the packets, recording the packet’s data, and analyzing the network traffic. Network forensics helps to trace back the malicious/suspect packets during the communication. Network intrusion prevention system/network intrusion detection system (NIPS/NIDS) are specialized methods to find out the malicious attackers. Network forensics is an extension of network cryptography. The objective of network forensics is collecting the evidence during trace attackers. This paper shows the best methodology for investigation strategy and data acquisition tools.

Sentiment analysis which is otherwise also called sentiment mining or opinion mining is the process of ascertaining and categorizing the positive, negative, or neutral opinion of the speaker or writer about a specific product, service, etc., in essence. The development of the user-generated content in social media has opened new prospects to extract knowledge from the opinions. Sentiment analysis categorizes the opinion of a writer into positive, negative, and neutral. This paper presents a detailed study on sentiment analysis process, various tools used for data collection, importance and the sources of user data. The importance of machine learning in sentiment analysis with various preprocessing steps and the ontology for sentiment is also discussed.

Handwritten recognition is an emerging field in the pattern recognition. Extraction of Sanskrit palm manuscript is challenging task and complicated in character recognition compared to other languages. In order to extract relevant information from the manuscript images efficiently, each step of the process requires reduction of irrelevant data such as noise on the images. This paper explains the importance of preprocessing step to determine the noise reduction, binarization, optimal binarization text, and character segmentation technique for palm manuscript. The objective of this paper is to present a survey of the existing methods for character recognition on palm manuscript which have been developed recently, to handle the historical document images.

Evolution of the Internet in the past decade resulted in generation of voluminous data in all sectors. Due to these advents, the people have new ways of expressing their opinions about anything in the form of tweets, blog posts, online discussion forums, status updates, etc. Sentiment analysis deals with the process of computationally identifying and categorizing opinions expressed in a piece of text, especially in order to determine whether the writer’s attitude toward a particular topic is positive, negative, or neutral. Knowing the opinion of customers is very important for any business. Hence, in this paper, we analyze the reviews given by the customers of the restaurant with the help of machine learning classification algorithms. This paper mainly focuses on the implementation of various classification algorithms and their performance analysis. The simulation results showed that SVM classifier resulted in the highest accuracy of 94.56% for the given dataset.

Non-communicable diseases (NCDs), in our study, include diabetes, hypertension and obesity, represent the global burden of diseases and cause deaths each year mainly in rural areas. Diabetic retinopathy (DR) is the prime root of irreversible blindness in the modern world. It refers to the retinal changes seen in patients with diabetes mellitus. In this paper, we present a novel approach to identify diabetic patients and predict the risk of diabetic retinopathy in diabetics using Naïve Bayes, z-ordering and kNN techniques. The results show that there is reduction in misclassification error and improvement in the efficiency of classification which help in improved screening of the dreaded complication of diabetes that aids in its early treatment.

Mobile analytics plays a significant role in investigating the information collected from a different business organization. It gives a specialized perspective of information gathered from a mobile device. The different mobile analytics techniques gather and represent information in various ways. The information gathered through mobile analytics is converted into data such that an organization can utilize for its benefits. This data identifies with the conduct of a user or client concerning any application in the mobile device. By investigating this conduct, association can know the purchasing behavior and different preferences of clients and offer services or items to them appropriately. Along with this, mobile analytics causes associations to pick up user loyalty and increase authoritative benefit. Hence, the aim of this paper is to provide the working overview, types, issues, tools, challenges related to mobile analytics.

The tumor is an unwanted tissue that develops in breast. Breast cancer may include a swelling, change in shape of the breast, rashes on the skin, and the watery-like substance coming from the nipple—American Cancer Society (Breast cancer, 2011) [1]. As the cancer spreads, there may be different symptoms like pain in breast and swelling. In conventional two-dimensional mammography, overlapping of the tissues is very considerable problem. The new methodology digital breast tomosynthesis is used for the identification of the breast masses. The input images are preprocessed using adaptive median filtering technique which reduces the impulse noise. In the next stage, image is segmented using Gaussian mixture model (GMM) where GMM is a category of the clustering algorithm. Next, the images are subjected to feature extraction. And finally, it is considered for feature classification using probabilistic neural network (PNN) classifier.

Among many biometrics such as face, iris, fingerprint and others, periocular region has the advantages over other biometrics because it is non-intrusive and serves as a balance between iris or eye region (very stringent, small area) and the whole face region (very relaxed large area). Research has shown that this is the region which does not get affected much because of various poses, aging, expression, facial changes and other artifacts, which otherwise would change to a large variation. This region can be captured using the similar setups used for obtaining face and iris images. Active research has been carried out on this topic since past few years due to its obvious advantages over face and iris biometrics in unconstrained and uncooperative scenarios. Many researchers have explored periocular biometrics involving both visible (VIS) and infrared (IR) spectrum images. For a system to work for 24/7 (such as in surveillance scenarios), the registration process may depend on the daytime VIS periocular images (or any mug shot image) and the testing or recognition process may occur in the nighttime involving only IR periocular images. This gives rise to a challenging research problem called the cross-spectral matching of images where VIS images are used for registration or as gallery images and IR images are used for testing or recognition process and vice versa. After intensive research of more than two decades on face and iris biometrics in cross-spectral domain, a number of researchers have now focused their work on matching heterogeneous (cross-spectral) periocular images. Though a number of surveys have been made on existing periocular biometric research, no study has been done on its cross-spectral aspect. This paper analyzes and reviews current state-of-the-art techniques in cross-spectral periocular recognition including various methodologies, databases, their protocols and current state-of-the-art recognition performances.

The aim of this paper is to determine the presence or absence of ossification of adductor sesamoid bone, which indicates if a child has attained puberty or not along with other physiological parameters of indicators of growth, using a computerized, automatic method of assessment. The Greulich and Pyle method is used in medical and dental sciences as a gold standard for assessment of skeletal age. This study describes the implementation of a computerized and automatic method for the identification of adductor sesamoid, using the numerical computing environment MATLAB. The input is the hand–wrist radiographic image, and the output is a Boolean true or false value indicating the presence of the adductor sesamoid bone in the input radiographic image. The onset of the appearance of adductor sesamoid indicates the onset of puberty which is important to diagnose for growth modification treatment in medicine and orthodontics.

In this paper, we focus on key points detection of a person. We judge the abnormal behavior of the person by detecting the motion of key points of that person. In the starting frame, we select a bounding box. If some numbers of key points are moved out of the bounding box, then the system gives an alert message. So here we will take you through a new approach for detecting an anomaly in the surveillance video. The surveillance in these days has only capable of saving the videos on a hard disk. These are not yet the best solution. The crime record shows that the crimes are recorded. How ever, the criminals escape from the punishment due to mishandling of the recorded videos. The project aims to store only the anomalous behavior, by storing the picture with date and time. This kind of alternative not only decreases the disk space but also provides intelligence to the system. The techniques used to achieve this project are comparable to state of art techniques in the computer vision domain.

Power dissipation is a significant issue in many digital and VLSI systems. Adiabatic logic is a promising technique in minimizing the power dissipation, and positive feedback adiabatic logic (PFAL) proves to be efficient. The arithmetic operations in the digital systems are incomplete without the use of adders and multipliers. In this paper, a 16-bit square root carry select adder (SQRT CSLA) is implemented using ripple carry adder (RCA). The limitation of power and area in SQRT CSLA using RCA is overcome by incorporating Binary to Excess-1 Converter (BEC) in place of RCA. An 8 × 8 Wallace tree multiplier (WTM) is implemented using the concept of carry-save addition. The limitation of area in WTM is overcome by implementing reduced complexity WTM (RCWTM). The adders and multipliers are realized in both static CMOS and PFAL in Cadence Virtuoso 180 nm technology and simulated in Spectre. The static CMOS-based SQRT CSLA using BEC dissipates 50.25% less power as compared to SQRT CSLA using RCA which makes SQRT CSLA using BEC a better choice w.r.t. power dissipation and area. The PFAL-based SQRT CSLA using RCA and SQRT CSLA using BEC dissipates 54.5 and 83.5% less power as compared to static CMOS designs. PFAL-based RCWTM dissipates 81.8% less power than the static CMOS design. Circuits designed using PFAL dissipates less power as compared to those designed using static CMOS logic with a tradeoff in area.

For the airborne radar systems, resolution has always been a challenge. Fine resolution can be achieved by increasing the aperture of the antenna. The carrying capacity and the streamline structure of the aircraft are the limiting factors. Considering the limitations, the aperture of the antenna cannot be increased after a limit. Therefore, a special category of radars called as Pulse-Doppler radars (PD) are usually used for airborne and spaceborne radar applications. PD radars possess the capability of detecting the targets in the mist of background clutter. Pulse compression is employed at the radar receiver to obtain the fine range resolution. However, it becomes difficult for the radar to resolve two targets present at the same distance. In this scenario, the angular resolution has to be considered for the detection. Doppler Beam Sharpening (DBS) is a technique used to observe and improve the azimuth resolution of the radar. In this paper, the DBS technique is presented to resolve the targets at the same range.

A semantic representation of text helps us understand the lexical association between words. Capturing these associations becomes an integral part of perceiving any language. One such fundamental property that expresses these associations is ‘similarity.’ This property shows consistency in comprehending similar words in the same manner. However, seldom it falls short on specific tasks where the lexical similarity in itself is not sufficient enough to validate the semantic representations. In this paper, the objective is to capture such semantic distinctions. It is based on the shared task ‘capturing discriminative attributes’ conducted in SemEval-2018. Our team participated in the task and held an F1 score of 0.658 with GloVe representation. An extension to this work is taken up in this paper where a new embedding known as ConceptNet Numberbatch is explored. The ConceptNet word embedding in overall showed improvement in the scores with the previous rule-based feature representation. The model is further tuned over certain hyperparameters, which improved the score to as much as 6%. A comparison is also put forth here with another prominent embedding like FastText. The ConceptNet model achieved a near par score with the state of the art, based only on a simple ensemble of features as representation.

Presently, stock market is the most important part of economy of the country and acts as key driver for its growth. Since gold is oldest form of currency, it drives more attention of people to invest in gold stock market. As we know stock market is volatile in nature and risk inevitable. To maximize profit, we need a model which predicts stock in upcoming future. A lot of data have to be dealt for précised prediction, and machine learning will be faultless. This paper uses artificial neural network (ANN) for predicting the fluctuation in gold price. It takes historical data and predicts the price for next day. The aim was to build a model which forecasts price with maximum precision and also helps user to maximize their profit. Model has achieved success but forecasting gold stock market prices is highly complicated and depends on series of events such as festival, political event, and marriage seasons. They also depend on other commodities like Sensex, Nifty, crude oil, and so on. Model will not be capable of taking these events into account as they fall randomly, but people who regularly invest in stock market can give better advice.

This paper discusses recognition of digits in Kannada language using an open-source speech recognition tool, Kaldi. The system considers small digit corpora with numbers ranging from 0 to 9 and 4480 samples with a set of fourteen speakers. The monophone and triphone models of the corpora for Kannada language are investigated, and a significant decrease in the word error rate is observed while using the triphone modelling using Kaldi toolkit. The feature extraction is done with Mel-frequency Cepstral coefficients (MFCC). The word error rate (WER) is achieved for this corpus and compared with that achieved the HTK toolkit. A WER of 9% and 6%, respectively, is achieved with monophone modelling and triphone modelling using Kaldi toolkit, and with the same corpus, a WER of 10% is applied in HTK toolkit. A preliminary research of the speech recognition using Kaldi toolkit is reported in this paper.

Clogging of traffic in streets is one of the biggest challenges faced in all huge and growing urban areas. To solve this quagmire of vehicular transit control at pathway crossings, an intelligent fully automatic real-time traffic control system is proposed in this paper. It uses RF and GPS technology to detect congestion by determining the average velocity of the vehicles and the vehicle count and GSM module to communicate the level of congestion with the central traffic kiosk which changes the traffic signal based on the level of congestion.

Recently, the need for programmable networks has drawn the interest of industrialists and academicians to develop a programmable networking paradigm called software-defined network (SDN). It is an effort made to separate network intelligence (control plane) from forwarding hardware (data plane). This paper provides a clear perspective of working of SDN and an open interface protocol called OpenFlow(OF). Researchers provide a broad insight into the working of SDN and various challenges faced while implementing it such as scalability, controller bottleneck, load balancing in distributed controller environment, routing and security. This paper focuses on security issues of SDN. We discuss different scenarios at which SDN is vulnerable to attacks and solutions to such attacks. Possible security attacks in the data plane, control plane and the interface between them are elaborated.

In this paper, comparative performance analysis of fuzzy logic controller and PID controller for space vector pulse width modulation (SVPWM)-based inverter-fed 150hp (horse power) three-phase induction motor which is used in the cable industry (Ravicab Cables Private Limited) at Bidadi is presented. In this cable industry, 611 Nm load torque is considered as full load and 305.5 Nm load torque is considered as half load for the three-phase induction motor. Proportional–integral–derivative (PID) controller-based voltage frequency drive (VFD) is used in this industry to control the speed of 150hp induction motor. Since VFD used in this industry is affected by several disturbances, robust speed controller is needed to be interfaced with the three-phase induction motor. In order to interface and identify the robust controller, this paper deals with simulation and comparison chart. As a part of the simulation, initially PID controller with real-time data which is used in this industry is interfaced with the 150hp induction motor under disturbance environment. Besides, the performance of this PID controller is analysed. Then, the proposed fuzzy logic controller is interfaced with the 150hp induction motor under disturbance environment. Moreover, the performance of this fuzzy logic controller is analysed. At the end, to identify the robust controller, the comparison chart is made.

Image compression and progressive transmission (PIT) is a technique used as an alternative solution to the communication problem, where there is a need to transmit huge data such as in medical image transmission. The technique divides image that is to be transmitted into several phases and effectively provides a fairly accurate reconstruction of the original image in every phase. This paper proposes a novel architecture that combines bit-plane slicing, DD-dual-tree DWT, and SPIHT. The bit-plane slicing separates the image into eight planes; each bit plane is then decomposed with wavelet. The wavelet coefficients in different sub-bands are then encoded using SPIHT and transmitted progressively and reconstructed by combining the data received in each phase with the previous phase data. Experimental results have shown that the proposed technique has better efficiency. Even at the lowest SPIHT rate of 0.1 bits/pixel for a 512 × 512 gray scale medical image, the reconstructed image shows higher quality at third phase itself, which makes it an efficient choice for transmission of medical images even at low speed communication channel.

Memory is one of the fundamental components of the modern computers. The need for low-power, faster non-volatile memories has been increasing due to compact and increased chip density. In this paper, the faster volatile Schmitt trigger SRAM cell with improved read and write operation is made non-volatile by inclusion of memristors. Multi-threshold CMOS (MTCMOS) technique is applied to reduce the overall power consumption of the circuit. The Schmitt trigger (ST) implementation improved the switching characteristics, reduced the leakage power, and gave improved static noise margin (SNM). The addition of memristor to the Schmitt trigger SRAM cell made the cell non-volatile and further increased the SNM. The MTCMOS non-volatile ST SRAM reduced average power consumption by 56.67% and increased SNM value by 58.65%.

To meet the time-critical and fail-safe requirements of automotive applications, there are several hardware and software components which are implemented on multiple electronic control units (ECU). With the recent trends in automotive technology, the vehicular functions are increasing, especially the advanced powertrain control demands more sophisticated intelligent hardware and software algorithms to be integrated into the electronic control units. The complex timing systems of advanced powertrain/e-powertrain requires maximum accuracy and precision with synchronization, especially while processing the complex sensors and actuators. In this paper, we will discuss about the implementation of innovative—intelligent phase-locked loops (IPLL) VHDL model for the advanced engine control applications. The IPLL module is designed to provide the high-resolution sensor output in normal operating mode and in addition, it shall retain the synchronization among the systems during the disturbance or loss of high-frequency sensor signal. This is achieved by generating the IPLL output through the self-learning hardware, even though when there is no valid sensor signal available. Hence the virtual synchronization with accurate positions of the engine is maintained.

In all processing systems, multiplication is one of the computation-intensive operations demanding more resources. Hence, multiplication operations demand more time, power and resources. One of the better solutions is Mitchell’s algorithm. Mitchell-based logarithmic multiplier is used as alternative approach which improves the speed, at the cost of accuracy. This paper presents the logarithmic multiplier implementation using the FinFETs. The hardware level simulation is done in Cadence Virtuoso using 18 nm technology. Comparison of power consumption of logarithmic multiplier using MOSFETs and FinFETs is presented. A 93.69% power reduction is seen in the proposed design as compared with the previous work.

In terms of memory, multivalued logic can be the fitting logic for the existing binary logic. Ternary logic contains three symbols in place of two symbols used in the binary logic, i.e., 0, 1, 2. More information can be stored with the help of these three symbols. SRAM cell is widely used in the digital circuit. The SRAM cell designed using the ternary logic can be used in the design of large memory arrays designed using ternary logic. The traditional ternary inverter which is used in the design of the traditional ternary SRAM cell is unable to store the proper values for the second state, there is a voltage level drop, which in turn affects the data read/write value of the SRAM cell designed using this traditional ternary inverter. Hence there is a need to design the ternary inverter cell which can give the proper output voltage level of all three states of the ternary logic. The level shift ternary inverter is designed to fulfill this disadvantage. The ternary inverter is designed in order to achieve the ideal DC characteristics, and the same level shift ternary inverter is used in the design of level shift ternary SRAM. This ternary SRAM stores the data properly at read/write signal. The traditional ternary inverter and traditional ternary SRAM, level shift ternary inverter and level shift ternary SRAM are implemented in Cadence 45 nm technology. The traditional ternary inverter consumes 2.37 μW power, and the level shift ternary SRAM consumes 2.43 μW power. The traditional ternary SRAM consumes 3.012 μW and level shift ternary SRAM consumes 3.14 μW. At the cost of a little bit increased power and the number of transistors, the traditional ternary SRAM can be replaced with level shift ternary SRAM. This level shift ternary SRAM stores all the voltage levels at all three levels. The same level shift ternary SRAM cell can be used for the design of large memory arrays.

Machining performance monitoring is the utilization of different sensors to determine the condition of processes. Machine vision system has been used to monitor the state of both cutting tool and workpiece during turning process. The turning experiments on Inconel 718 material have been performed in a precision lathe using coated carbide cutting tool in dry conditions. Cutting tool and machined surface images were acquired using machine vision. Image features of machined surface and cutting tool were extracted by processing the images. Image features such as wear area and perimeter have been considered to characterize tool wear state; consequently, machined surface state was characterized by means of image histogram frequency. Further, trends have been plotted with image features extracted from both tool and machined surfaces. Results indicate that monitoring of turning performance could be effectively accomplished by plotted trends.

The wireless sensor networks are the decentralized and self-configuring network in which sensor nodes can sense the surrounding environment and pass the information to the base station. Due to decentralized nature and distant deployment, the consumption of energy is one of the major issues of wireless sensor networks. To reduce consumption of energy in wireless sensors, hierarchal clustering is the efficient type of clustering technique. In this research work, WEMER protocol is implemented and improved to increase lifetime of sensor networks. In the improved WEMER protocol, whole network is divided into many clusters, and for each cluster, one cluster head is selected. The proposed WEMER protocol is implemented in MATLAB. The simulation results show that proposed WEMER protocol has less number of dead nodes send more number of packets and less energy consumption.

Ultrasound is used imaging modality for the diagnosis and diseases. In past decades, the advances in technology have become crucial imaging modality, due to its suppleness and non-invasive. Ultrasound image uses high-frequency sound waves to contrasting reflection signals created when a light emission is anticipated into the body the convenience of ultrasound image is corrupted by the noise identified as dot (speckle). The speckle model depends on the image tissue and different image parameters. The noise shows in ultrasound picture influences edges and subtle elements contain the differentiation and determination. The reasons for speckle noise reduction in ultrasound image (i) The ultrasound images are improved for human interpretation (ii) The speckle noise reduction is the preprocessing step for segmentation and registration in ultrasound image processing tasks. The objective of paper is to give different techniques have been used to decrease speckle noise in ultrasound image.

This paper provides a detailed analysis of the gm/ID design methodology for low-power applications. A systematic procedure is proposed to fix the current and transistor dimensions of analog circuits so as to meet specifications such as gain–bandwidth while optimizing power and area. The paper also provides an explanation as how short channel effects are taken into consideration when designing analog circuits. A simple differential amplifier is used to illustrate the methodology.

One serious problem that all the developing nations are facing today is injuries and death of animals due to road accidents. Report says that, there are around 300,000 collisions per year. However, many of the databases exclude accidents that have vehicle damage less than $1,000 ( https://apiar.org.au/wpcontent/uploads/2017/07/23_APJCECT_ICT-268-281.Pdf , [1]). Accidents lead to the reduction in wildlife. Eventually, this may lead to reduction and endangered of rare species. A system has to be designed to overcome this problem. In this paper, the problem is addressed by focusing on designing an IoT-based system which will perform two functions one is alerting the driver whenever an animal is nearer to the vehicle and the second one is alerting the driver whenever he exceeds the speed limit, especially in the forest region.

RAW is a digital file contains the camera-captured image data regarding the sensor pixels values and text information. The raw is being highlighted as digital negative and varies with the formats, which depend on hardware manufacturer. The raw processing is significant to ignore the duplication of data, to economize the space needed, to ease image file operations, and to have an uninterrupted capturing. The image quality is the substantial parametric quantity which determines the visual of the captured raw. The most extreme resolution with no inbuilt compression (raw) results in high image from any digital camera. The proposed workflow is to extract the contents of raw sensor information from the raw files and processing and displaying the information in image format. The raw test files were gathered from cameras by different manufactures. The MATLAB R2016a has been used for executing the workflow and analysis purpose. The display quality is ensured by the performance parametric—Quality of Image Improvement (QOII), and also the file size reduction ratio was analyzed.

This paper proposes an image processing approach for compression of ECG signals based on 2D compression standards. This will explore both inter-beat and intra-beat redundancies that exist in the ECG signal leading to higher compression ratio (CR) as compared to 1D signal compression standards which explore only the inter-beat redundancies. The proposed method is twofold: In the first step, ECG signal is preprocessed and QRS detection is used to detect the peaks. In the second step, baseline wander is removed and a 2D array of data is obtained through the cut-and-align beat approach. Further beat reordering is done to arrange the ECG array depending upon the similarities available in the adjacent beats. Then ECG signal is compressed by first applying the lossless compression scheme called the 2D Run Length Encoding (RLE), and then a variant of discrete wavelet transform (DWT) called set partitioning in hierarchical trees (SPIHT) is applied to further compress the ECG signal. The proposed method is evaluated on the selected data from MITs Beth Israel Hospital, and it was conceded that this method surpasses some of the prevailing methods in the literature by attaining a higher compression ratio (CR) and moderate percentage-root-mean-square difference (PRD).

The usage of autonomous robots is getting increased day by day. Most of the applications are moving toward automation with the help of robots. This paper mainly focuses on service robots and understanding their working environments. A few robotic scenarios are created using Webots tool, and the data from there are collected as a sequence of images and lidar sensor values. The lidar values are collected with both single layer and multilayer. The environments are analyzed with the help of the collected data. The collected multimodal data are preprocessed in order to reduce the number of features. After that, the collected data are sorted out to suitably characterize each environment, and the machine learning techniques are applied to classify the environments. Different machine learning algorithms like Naive Bayes classifier, support vector machine, decision-tree-like random forest tree, and simple logistic regression are used for the classification, and results are compared with each other.

The embedded system is the combination of electronics hardware and software. Its applications intermingle uninterruptedly with the surrounding environment through different types of sensors and actuators. In electronics market, the plethora of embedded controller boards is available for engineers/developers to automate the world around them. The vision of this paper is to compare three controller boards, e.g., Arduino, Intel Galileo, and mbed LPC1768 through the case studies of implemented projects of automation: Automatic Auditorium System Using Arduino Board IOT-Based LCD Gadget using mbed LPC1768 Home Automation using Intel Galileo Board Analysis performed for the important parameters reliability, efficiency, and the user-friendly nature of latest development boards. It will help to engineers and professional developers to easily choose their microcontroller to develop any project or automation system/device.

The Internet is definitely the greatest contributor to globalization which has brought together the whole world and has made every nook and corner attainable. The conventional IP networks have not been able to serve the purpose of a simplified infrastructure even after their extensive acquisition. Service provider networks are not fully fledged in providing (1) fast switching in the core network without any routing lookup, load balancing and retort to faults, (2) reorganize the network according to prevailing network policies and terms simultaneously. Software-defined network (SDN) which operates on OpenFlow protocol is a revolution in networking which aims to remodel the traditional network by disintegrating the data and control planes. It bifurcates the control logic from the various network devices like routers and switches, by providing a core supervisory control logic for the entire network and delivering aptness to code the network. In traditional routing, the information is flooded to the entire network which causes overutilization of resources, high bandwidth requirements, and many other drawbacks. Compared to the legacy routing, SDN is more effective in route computations and provides complete control for packet transmission. The paper proposes a method to find the shortest path routing between the source and destination using the Bellman–Ford routing algorithm. Secondly, the routing emulations for various network topologies are presented and compared using Mininet which implements routing for SDN. The comparative analysis of both the scenarios shows that the routing algorithm proposed in this paper contributes utmost QoS.

Demand of wireless sensor networks has increased dramatically due to their significant use in various real-time industrial, military, and medical field applications. Due to their vast applications, WSNs have become an attractive field of research. These sensor networks are easy to deploy and can efficiently monitor the area and environment, but due to limited resources, such as memory, battery capacity, and computation capacity, it becomes a challenging task. In this work, we focus on the network lifetime enhancement by developing particle swarm optimization (PSO)-based technique. In order to achieve the desired performance, the complete proposed model is divided into various stages where first of all, sensor nodes are deployed randomly. Later, cluster head selection is performed followed by the shortest path identification. In order to minimize the energy consumption, we apply multi-objective PSO scheme. However, fitness function computation suffers from the slow convergence which leads to the premature solution resulting in degraded communication performance. In order to address this issue, we present a new fitness function computation which considers residual energy parameter and formulates a new energy consumption model for each node which helps to optimize the power consumption during data transmission and reception by considering the sleeping phases of sensor nodes. An extensive simulation study is carried out using MATLAB simulation tool, and the performance of the proposed approach is compared with the state-of-the-art techniques. Results and discussion of this approach show that the proposed approach can achieve better performance in terms of QoS, packet delivery rate, and network lifetime enhancement.

This work explores partial efficiency with effective structure based on fundus image to characterize among disease and normal images. The execution of LBP in the process of a surface description as retinal image had it investigated also contrasted and different description. The objective is to separate DR and typical fundus images investigating the surface of the retina foundation and keeping away from a past sore division. For each experiment, several classifiers were tested on an average sensitivity and specificity higher than 0.86 in all the cases, and almost of 1 and 0.99, respectively, for DR detection were achieved. These outcomes recommend that the strategy exhibited in this paper is a powerful calculation for portraying retina surface and can be valuable for the analysis which helps to framework for retinal disease screening.

Orthogonal Frequency Division Multiplexing (OFDM) is a promising multicarrier modulation technique used for high data rate communication. However, one of the major drawbacks is it suffers from high peak-to-average power ratio (PAPR). This paper is a study on different types of PAPR reduction techniques and proposes a new hybrid DCT with companding PAPR reduction technique. It is shown that proposed method is simple and provides better PAPR reduction compared to DCT, companding and conventional OFDM system.

The paper describes the implementation of systolic array-based hardware accelerator for multilayer perceptrons (MLP) on FPGA. Full precision hardware implementation of neural network increases resource utilization. Therefore, it is difficult to fit large neural networks on FPGA. Moreover, these implementations have high power consumption. Neural networks are implemented with numerous multiply and accumulate (MAC) units. The multipliers in these MAC units are expensive in terms of power. Algorithms have been proposed which quantize the weights and eliminate the need of multipliers in a neural network without compromising much on classification accuracy. The algorithms replace MAC units with simple accumulators. Quantized weights minimize the weight storage requirements. Quantizing inputs and constraining activations along with weights simplify the adder as well as further reduce the resource utilization. A systolic array-based architecture of neural network has been implemented on FPGA. The architecture has been modified according to Binary Connect and Ternary Connect algorithms which quantize the weights into two and three levels, respectively. The final variant of the architecture has been designed and implemented with quantized inputs, Ternary connect algorithm and activations constrained to +1 and −1. All the implementations have been verified with MNIST data set. Classification accuracy of hardware implementations has been found comparable with its software counterparts. The designed hardware accelerator has achieved reduction in flip-flop utilization by 7.5 times compared to the basic hardware implementation of neural network with high precision weights, inputs and normal MAC units. The power consumption also has got reduced by half and the delay of critical path decreased by three times. Thus, larger neural networks can be implemented on FPGA that can run at high frequencies with less power.

The industry has undergone three revolutions mechanization, electrification, and informatization and now it is the era of fourth industrial revolution the Internet of things also called Industrial revolution 4.0 which is predicted its way into factory. The main vision of the fourth revolution is smart factory with globally networked and real-time capable industrial production. As a part of these, companies are aiming to create intelligent devices, storage systems, and supplies in industrial production. These intelligent devices in turn exchange data, perform action, control each other and also the products are aware of their current state. Furthermore, the challenge is to control of device in network with increase in their number and RT (real time) capability. The industrial communication network plays a vital role in automation industry. The industrial communication provides an effective means of data exchange, data controllability, and flexibility to connect other devices. The paper gives an introduction to industrial communication network and various industrial Ethernet protocols for real-time data exchange in automation industry. The paper mainly explains the effective role of EtherCAT protocol for the development of smart factory.

Allocation of resource and binding it to functional unit at high-level synthesis an optimal problem to minimize the area and performance in terms of resource sharing and binding is presented in this paper. The paper presents the comparative analysis of nature-inspired computation techniques for resource allocation and binding: 1. Evolutionary-based computation: genetic algorithm. 2. Swarm intelligence-based computation: particle swarm optimization. The comparative analysis of the results shows genetic algorithm surpasses particle swarm optimization in providing the precise mapping between the operation and functional unit sharing with zero errors in resource allocation.

Tortuosity and retina blood vessel dilation are important symptoms of plus disease in retinopathy of prematurity. This paper presents differential geometrical method for tortuosity measurement. The problem of tortuosity evaluation is formulated as one-dimensional differential geometrical curvature characterization. The vessel network extracted from retinal image is subjected to boundary extraction and individual vessel boundaries are extracted as planar curves, further these curves are segmented and differential curvature is computed at segment level and at vessel level for the individual vessels. The method is tested and validated on the available public data and local data set. Vessels with considerable tortuosity are found to be having significant curvature variation compared to the normal vessels.

Objective: A set of different nanocomposite hybrid materials of Sr0.5 Y0.1 Cu0.3 has been synthesized through co-precipitation synthetic method using Cetyl Trimethyl Ammonium Bromide (CTAB) and Tri ethyl amine (TEA) as surfactants. The different derivatives of indole Schiff base compounds (5a–5e) were doped to the nanocomposites in molar ratio of 1–2%. The entire hybrid material isolated by forming gelatinous network and calcinated. The hybrid nanomaterials were investigated for their semiconductor and dielectric properties. Method: Nanocompositional hybrid materials were prepared by co-precipitation method with the composition of Sr0.5 Y0.1 Cu0.3 with doping of indole Schiff base derivative (ISB) of molar ratio 1–2%, and the reduction is done using surface reduction methods gelated precipitate of hybrid nanomaterials were calcinated, characterized by XRD, FT-IR, UV–Vis and SEM spectroscopic techniques. The semiconductor and dielectric properties of pelletalized samples were measured with respect to various temperatures and frequency. The precursor material used for the synthesis is strontium carbonate, yttrium oxide and cuprous chloride which are reduced in presence of CTAB and urea. After isolating the hybrid materials ISB derivatives (5a–5e) were doped and the aqueous mixture of bluish-reddish precipitate was filtered, washed with ethyl alcohol and octanol mixture to remove any impurities present with the precipitate, dried at 50–120 °C, heat treated at 650–750 °C and obtained pure nanocomposite. Findings: Initial spectroscopic studies showed that grain size of copper was 20–30 nm in diameter. XRD pattern demonstrated the formation of trigonal copper and dopant addition considerably affects the crystal structure. The dielectric constant were measured and compared with undoped hybrid material. The band alignment and band gap obtained from the ISB material was further felicitating the fissile ejection of the electrons and ensuring the semiconductor properties of the materials.

Ever since its introduction, wireless sensor networks have enabled us to do things which were previously unimaginable by simplifying more tasks and enhancing quality of life to millions of technology-dependent groups of people. Just like every other cutting edge technology, WSN also has its own needs and challenges with respect to that of the usage or transmission of data packets. This study paper is an attempt to introduce the research community to some of the clustering approaches followed in WSN along with the challenges that one should be familiar with before working on clustering approaches in WSN.

Information security and its research are getting significant, and therefore, the steganographic technique is used in high level to send data secretly using a cover image such that its presence cannot be detected. In his paper, an improved RGB image steganographic technique which is a novel approach for hiding the secret image based on compressive sensing algorithm has been proposed. In the proposed approach, the RGB planes of payload image are extracted and compressed; then, the coefficients are reshaped and discrete cosine transform (DCT) is applied over 2 * 2 matrices. The compressed payload image is embedded with chaotically chosen random keys in the segmented RGB planes of cover image. The proposed method results in significant improvements with high PSNR value and payload capacity.

This paper presents the design and implementation of a GSM-based dual-tone multi-frequency (DTMF) controller for remotely monitoring of electrical appliances. The present system provides an added feature like the secure connection with a passcode and reliable control of various devices without perturbation. The device can be controlled from any location using a mobile phone. For this, one has to make a call from a cell phone to control the electrical appliances. Since mobile phones are available with everyone, it will be much easier to employ this technology in practical life. In this paper, the main focus is on the controlling of more devices without ambiguity. The developed system is cost effective and can be used to control up to 128 devices with a secured connection unlike the 8–12 devices in the conventional systems.

In this paper, a photovoltaic model is designed and developed for generating 110 V (RMS) is presented. The voltage obtained out of the photovoltaic (PV) module is comparatively less and variable in nature. The model presented in this paper consists of three voltage fed-back boost converters that are designed and modeled using state-space averaging technique (SSA) in order to obtain a higher and stabilized constant DC output voltage. This boosted output is fed to an asymmetrical 15-level multilevel inverter to convert DC to AC. The switching angles for the 15-level H-bridge multilevel inverter are designed using equal area criteria. The designed circuit is developed using IRFP250N and IRF840 MOSFETs. The simulation and hardware results compared in the result section convey that the proposed design is able to produce 110 V (RMS) for variable loads.

Multilevel inverters (MLI) are suitable for high power and medium voltage applications. Number of topologies has been introduced among which the cascaded H-bridge (CHB) inverter is widely used because of its simplicity. It consists of number of power semiconductor switches and DC voltage sources to generate step voltage waveform. The use of separate DC voltage source for each H-bridge allows the CHB inverter to be easily integrated with photovoltaic systems. Depending upon DC voltage source, the CHB inverter has two types: symmetrical CHB inverter which consists of equal DC voltage sources and asymmetrical CHB inverter with unequal DC voltage sources. Symmetrical inverter topology suffers from increased number of power devices, complexity and losses with the increase in number of output levels, whereas asymmetrical topology uses less number of power devices for producing higher level output voltage waveform. In this paper, analysis and simulation results for 31-level symmetrical and asymmetrical CHB MLI are presented.

At the present time, the use of nonconventional energy resources is having a huge demand as the conventional energy sources are depleting in nature and causing more pollution which ensures the need of renewable energy resources. This paper deals with the design of a cuk-sepic converter with a three-phase inverter feeding a nonlinear load. Harmonics will be introduced to the system due to the use of nonlinear load which will affect the equipment in both source and load end, so there is a need of filter to mitigate the harmonics. Here, a shunt active filter is used which will reduce the harmonics introduced in the circuit. The desired low harmonic three-phase voltage is obtained as per the IEEE standard of below 5% as the filter results in 1.33% total harmonic distortion (THD). The simulations are carried out and analyzed by using PSIM software.

With the increased emphasis on both reliability and functionality, shape memory alloys (SMAs) are fast becoming an enabling technology capturing the attention of engineers and scientists worldwide. The thermal-electrical-mechanical dynamics of SMA are nonlinear and hysteretic in nature, possessing a problem for the researchers to model the actuator. The increased range of applications and better realisation of SMA actuators have led to the research on modelling of SMA’s thermo-mechanical response. The paper discusses various SMA actuator modelling approaches such as Preisach model, Fermi–Dirac statistics, Duhem hysteresis model and Brinson model and attempts to elucidate their advantages and limitations through Simulink-based models and simulation results.

Isolated energy systems are gaining popularity for supplying power to remote and rural areas. Wind energy conversion systems (WECS) are a good choice among various options. For efficient operation, hill climbing search (HCS) maximum power point tracking (MPPT) algorithm is applied to extract maximum power from wind turbine. This paper presents the simulation study and harmonic analysis of self-excited induction generator-based WECS for a nonlinear load by adapting the maximum power point tracking mechanism.

Among the different multilevel topologies, the modular multilevel converter (MMC) has now become a subject of intense research, where it offers some interesting and useful features. In this project, DC/DC converters are proposed for the HVDC-based systems to reduce the transmission losses. The full-bridge converters and three-level flying capacitor circuits are combined and integrated by employing the phase shift control scheme which can be easily applied to achieve 0-voltage and proposed for the high step-down and HVDC-based systems, and also which has a capability to generate the high-quality power under various conditions. Importantly, in this project the voltage auto-balance ability among the cascaded modules is achieved by the flying capacitor which removes the additional components or control loops, and it also allows the operation at higher frequencies and at higher input voltages without scarifying the efficiency. The neutral point clamped (NPC) converters and flying capacitor-based converters are the major multilevel topologies for the high-power and high-voltage applications. Zero-voltage switching (ZVS) performance for both the leading and lagging switches can be provided to reduce the switching losses by adopting the phase-shift control scheme. The time sequence of the leading leg in the phase-shift-controlled full-bridge converters is kept constant, and only the phase of the lagging leg is shifted to regulate the output voltage. A high DC voltage is required for the DC-based distribution and micro-grid systems to improve the delivery power capability which reduces the transmission losses. The switch voltage stress is reduced, and thus, the circuit reliability is enhanced in this project. The MMC concept can be easily extended to N-stage converter to satisfy the high-voltage applications with low-rated voltage switches. The circuit operation and converter performance are analyzed by simulations. The result of input voltage sharing across the capacitors is of 300 V. Thus, the input voltage auto-balance is achieved excellently. The input voltage is stepped down to 53 V. The modes of operations and converter performance are analyzed and simulated by using P-Sim software. The three-stage converter is designed to increase the step-down ratio when compared to two-stage converter. This operation is similar to the two-stage converter.

Stochastic performance of an electric arc furnace (EAF) and its associated power quality constraints has drawn the attention of the researcher’s. In this paper, the power quality disturbances are investigated by characteristic modeling of an electric arc furnace and shunt active filter is incorporated with the system to suppress the effect of these disturbances. The nonlinearity of an EAF load results in distortion of the fundamental component leading to the occurrence of voltage flicker, harmonics, and inter-harmonics. The simulation studies are carried out in MATLAB (Simulink).

In the present era, electricity is one of the essential needs of human society. Transmission lines are used in conventional power system which is expensive and less efficient. Transmission line losses account for the major reduction in transmission efficiency due to long transmission lines spanning several kilometres. One best solution for this old-age problem is to have wireless power transmission (WPT). Not only in transmission, can any application module can be made compact and energy efficient using WPT concept. In WPT, AC voltage is rectified, converted to high-frequency AC, and transmitted using inductive power transfer or capacitive power transfer approach. Inductive power transfer system contains transmitter and receiver coils designed for high-frequency AC voltage. Therefore, the coils’ size is reduced and they can be made compact. In this paper, a WPT scheme is proposed for LED lighting which can be used in public information display system. The system is designed to drive 100 W LED display unit. Class-DE inverter is considered and series–parallel resonance is used in transmitter and receiver coil sides, respectively. The simulation results are presented.

Solar is one of the most widely used energy sources due to its ease of availability and preferred for the generation of electricity. This paper presents a bidirectional power conversion from DC to AC by using non-complementary active clamp circuits. It consists of the bidirectional flyback converter. In order to interface the grid with low-voltage energy storage through single-stage power conversion, the bidirectional converter is used to convert low voltage directly into the grid voltage and regulates the grid current. By using the non-complementary operation approach, the flyback converter avoids the sudden increase in voltage and also decreases the energy loss. And hence by using single-stage power conversion approach and non-complementary active clamp circuits, the flyback converter attains high power efficiency and low total harmonic distortion. A control algorithm is developed to facilitate the bidirectional single power conversion. The flyback converter assures high voltage quality by this control algorithm. The simulation is done by using the PSIM software and analyzed in detail.

This paper proposes performance study of series, parallel and series–parallel resonant converter topologies for solar PV application. The performance studies are carried out considering the switching losses and power factor improvement, ZVS and ZCS switching techniques are implemented in each simulation model of DC–DC resonant converter topologies. The solar PV panel is introduced, and its simulation model is connected to DC–DC resonant converter input terminals, and complete system is simulated for the performance study under different topologies. The simulation tool PSIM 9.1.1 is used.

Multilevel inverters are gaining much greater attention lately as a result of increasing demand for high power inverter units in the industrial applications. Reduction of harmonics and minimizing the THD from the MLI output is considered as a very important task in order to achieve the better performance and to prevent the device from getting damaged. This paper presents the performance study of different soft computing method for CMLI using SHEPWM technique. Genetic algorithmic (GA), Particle Swarm optimization (PSO) techniques are applied MLI to calculating the switching angles. The switching angles are determined such that the constraints of SHEPWM are met, which controls the switching of the CMLI thereby eliminating the lower order harmonics and minimizing the Total Harmonic Distortion (THD) while maintaining the required fundamental voltage. The switching angles are calculated offline and are then used in MATLAB to generate the pulses required for CMLI. A detailed simulation study work is carried out using MATLAB/SIMULINK.

Induction motors are extensively used motor type for various industrial applications for the reason that they are robust, simple in structure, and efficient. On the other hand, induction motors are prone to different faults during their lifetime due to hostile environments. If the fault is not detected in its rudimentary phase, it may cause unexpected shut down of the entire system and colossal loss in industry. It is conspicuous that scope of this field is huge. This work presents detection of internal and external faults of induction motor. S-Transformation, which is superior as compared to CWT and STFT as it does not contain any cross terms, is used for bearing fault detection, and random forest, an algorithm which is easy to implement and requires minimum memory, is used for detection of external faults. The fault can be detected with more accuracy in premature state leads to improve the reliability of the system.

For effective utilization of irradiation’s falling on the solar photovoltaic panel, several maximum power point (MPPT) techniques are used. The comparative analysis of perturb and observe (P&O) and fuzzy logic methods for MPPT is presented in this paper. The modeling technique employing fuzzy logic is simplified to enable it to track power efficiently.

A DC–DC converter with high voltage gain is advantageous in most of the industrial applications such as uninterruptable power supply, vehicular system and conversion of low output renewable energy resources. Many DC–DC converters with high voltage gain are already available, but there are some issues need to be addressed like complexity of circuit, high ripple at input current, large stress across power semiconductor devices. In this paper, comparisons of different high-gain converters are carried out. Among all the compared converters, modified converter with dual output port has more gain, very useful and more advantageous compare to other converters. The results are carried out in the PSIM environment.

Power electronic transformer (P.E.T.) and solid-state transformer (SST) are one of the promising technologies in medium and high power conversion systems. In case of controlling power quality for the various load connected, P.E.T. and SST perform greatly in comparison with conventional line frequency transformer (CLFT) With advancements in high power switches and magnetic materials, P.E.T. can reach the efficiency almost equal to CLFT in distribution as well as traction system. P.E.T. eliminates all the limitation that CLFT faces in region of power quality maintenance and power transfer. Over the past two decades, researches and field trial studies are conducted to explore the challenges faced by conventional P.E.T. models and improved them to face all sort of applications in electrical systems. This paper aims to review the essential requirements of P.E.T. and SST modules both in traction and distribution systems. Basic design topologies of both SST and P.E.T. modules are also reviewed. There is also tabulation of all the models recently manufactured by companies for railways and distributions. Finally, this paper discusses the latest go-through in advanced topologies of P.E.T. and their architectural designs.

Voltage is a fundamental element in the quality of power supply, may rise/drop depending on the reactive power balance; hence, it has become extremely important to manage the reactive power balance for voltage control in the off-grid hybrid power system (OGHPS). This paper investigates the application of bacteria forging algorithm and genetic algorithm optimization algorithm to design an optimal control for voltage stability of off-grid hybrid power system. The off-grid hybrid power system considered in this work consists of an induction generator for wind power system, photovoltaic (PV) system with inverter, synchronous generator for diesel power generation, and composite load. The over-rated PV inverter has ample amount of VAR capacity while sourcing PV real power. Two control structures are incorporated in this work, to regulate load bus voltage. The first control structure is for controlling the total reactive power requirement of the system that by controlling inverter voltage magnitude for sourcing required reactive power to the system, and the second control structure is for controlling the SG excitation and hence the load bus terminal voltage. Both control structures have proportional-integral (PI) controller with a single input. In order to coordinate VAR/voltage control, the controller parameters tuned optimally and simultaneously using bacterial forging and genetic algorithm-based optimization method. Small signal model of all components of OGHPS is simulated in Simulink, tested for reactive load disturbance, and/or wind power input disturbance of different magnitudes for voltage stability. All system state variables are examined to evaluate the effectiveness of proposed optimal coordinated controls.

There are a plethora of challenges faced due to the usage of conventional energy sources. Our objective is to optimize the performance and working of an existing on-grid solar power plant by focusing on common problems found in such large-scale installations. The suggested proposals for improvements were based on three identified areas: power generation and performance ratio (PR), running costs, and maintenance costs. We have used PVsyst, a software package, for simulating and analysing the working of the power plant. SCADA data was obtained from the plant along with generated sets of data from PVsyst and was further used to develop our proposals for improvement. Four improvements were suggested, which are “method to improve PV panel cleaning efficiency for dry tropical regions”, “Simulink-based estimation of partial shading loss reduction using herbicides for vegetation management”, “procedure to suggest new optimum tilt for a seasonal tilt arrangement”, and “automated switching and lighting circuit using two-way relay switch”.

Electrical power quality portrays an imperative part of providing power effectively to the consumers. As power turns out to be more fundamental and significant asset for the whole world, the quality used at all its level will be critical for the steady and also for the efficient working of the equipment. As there is an increase in the consumption of power, the production of quality power is a challenge in power engineering. Therefore, it is important to address the issues that affect the quality of the power. Hence, to address these issues, distinctly voltage swell, sag, transients, and distortions due to harmonics, Hilbert–Huang transform is utilized for identification of distortions and support vector machine is employed for classification. The data is also collected from the substation and the analysis is accomplished by estimating the performance of empirical mode decomposition, ensemble empirical mode decomposition, and complete ensemble empirical mode decomposition.

Voltage stability has a major concern in power system operation. It is the ability of the power system to maintain acceptable voltages at all buses in the system under normal conditions and after being subjected to a disturbance. Voltage instability may result in voltage collapse of the system. Hence, assessment of voltage stability is important. Implementation of new equipment including high-power electronics-based technologies such as flexible AC transmission systems (FACTS) has become essential for improvement of operation and control of power systems. The project work aims at the enhancement of voltage stability in the radial distribution system by using shunt capacitor and FACTS controller. A stability index named line stability indicator (LSI) is formulated for voltage stability analysis. This indicator is tested on a standard IEEE 33 bus radial distribution system. The indicator is used to find the weak lines in the system. Placement of shunt capacitor and FACTS controller at the receiving end side of the weak bus results in improvement of voltage profile of the system. Cuckoo search (CS) algorithm is applied for optimal sizing of shunt capacitor and FACTS controller. Program is coded in MATLAB for the enhancement of voltage stability in the radial distribution system.

Day by day employing Distributed Generation (DG) is increasing and it is becoming an indispensable small capacity generation in the distribution system. It is cost effective, eco-friendly and it can enhance the reliability of the distribution network. This paper proposes a technique for optimal sizing and siting of DG, using modified Particle Swarm Optimization technique. It is proposed for optimal placement and sizing of DG. Since the objective is both reduction of losses and voltage profile improvement the multi objective function is chosen and by choosing appropriately desired level of emphasis is given to both the objectives. Also, by using an index Multi Objective Ranking Index (MORI) the best combination of loss reduction and voltage profile improvement is obtained. The effectiveness of the methodology is tested on standard IEEE-33 bus system and results are presented.