Search

Classical Siddha medicine recommends the usage of functionally parallel substitute raw drugs in the scarcity of the original material. The concept of standard raw drug substitution is designated by the Tamil term Matru sarakku, which emphasizes selecting alternatives either in herbal or non-herbal sources based on certain attributes. Functionally similar materials are suggested for balancing the deficit based on such parameters. Literature sources from Siddha medicine and other allied subjects were explored and documented for the inclusive understanding of the concept of ideal substitution. Few of the standard substitutes mentioned in the classical Siddha literature were evaluated in a scientific account for its justification. An outlook on different specimens as described in the ancient texts of Tamil medicine indicates the usage of substitutes from numerous plants, animals, and other metallo-mineral ingredients. The rationale of selecting alternatives primarily depends on the equivalent following traditional attributes like organoleptic entities, potency, division, general properties, specific actions, and medicinal uses of the material shared between the raw drugs. This must undergo a systematic evaluation by pharmacognostic and phytochemical studies to justify current practices of substitution. A systematic document in these lines will give a proper guideline for the effective employment of substitute drugs in the current scenario of dwindling official botanical sources for many Traditional Siddha formulations.

The paper is regarding the automating of driver’s license testing system and updating the results to the person through website and conjointly through registered email. Usually, while driving test the person who requested for license have to be compelled to show his driving skills ahead of the authorities. The person need to operate the vehicle according to several rules. If he fails, he/she are knocked out and have to appear for the driving test next time. The Officials observe mistakes of the applicant physically. The proposed solution for the automation of existing manual test method permits the elimination of intervention of humans and improves the accuracy of driving test thereby going paperless, with Driving Skill Evaluation System. In the proposed system, we have a tendency of taking data from sensor as inputs from hardware simulator and stores into the database. In this system, the person participating in the test are obsereved by sensors. Therefore weather the person is qualified or not is informed to the applicant as well as the authorities. Gradual increase in number of road hazards are due to less practice in driving and illegal driving license given to the unskilled drivers by taking bribe. To beat this drawback, automated driving license test will be advantageous. This solution is introduced for ensuring the quality in approving in license to enhance safety.

Agriculture sector is recognized as the backbone of the Indian economy that plays a crucial role in the growth of the nation’s economy. It imparts on weather and other environmental aspects. Some of the factors on which agriculture is reliant are Soil, climate, flooding, fertilizers, temperature, precipitation, crops, insecticides, and herb. The soil fertility is dependent on these factors and hence difficult to predict. However, the Agriculture sector in India is facing the severe problem of increasing crop productivity. Farmers lack the essential knowledge of nutrient content of the soil, selection of crop best suited for the soil and they also lack efficient methods for predicting crop well in advance so that appropriate methods have been used to improve crop productivity. This paper presents different Supervised Machine Learning Algorithms such as Decision tree, K-Nearest Neighbor (KNN), Support Vector Machine (SVM) to predict the fertility of soil based on macro-nutrients and micro-nutrients status found in the dataset. Supervised Machine Learning algorithms are applied on the training dataset and are tested with the test dataset, and the implementation of these algorithms is done using R Tool. The performance analysis of these algorithms is done using different evaluation metrics like mean absolute error, cross-validation, and accuracy. Result analysis shows that the Decision tree is produced the best accuracy of 99% with a very less mean square error (MSE) rate.

An interactive Student Information system in the cloud is an one stop portal on the internet where the students can get answers to their queries and look up relevant information. The users can check their personal details, communicate with faculties and avail a ”livechat” to get their queries solved. They are also able to access relevant information regarding different activities and events including and not limited to ”attendance”, ”timetable”, ”subject list”, ”college events”, etc. Cloud technologies reduce the challenges faced by traditional systems such as costs associated with development and maintenance. The student support system is ubiquitous in nature since it is hosted in the cloud and uses the power of cloud computing to make the application secure, scalable, and fast. Along with the power of cloud computing, good User Interface/User Experience(UI/UX) is focused upon to ensure that the end-user has a great hassle-free experience. Load balancers are used to handle huge traffic the website might face. User experience principles are used in order to make the portal easy to use.

The core aspiration of this proposed work is to classify Tamil characters inscribed in the palm leaf manuscript using an Artificial Neural Network. Tamil palm leaf manuscript characters in the form of images were processed and segmented using contour-based convex hull bounding box segmentation. The segmented characters were transformed into two forms: Binary Coded Value and the Gray-Level Co-occurrence Matrix (GLCM) feature. The features extracted from the segmented characters were trained by the proposed method of the Modified Adaptive Backpropagation Network (MABPN) algorithm with Shannon activation function. Weight initialization plays an important role in the Backpropagation Neural Network, and hence Nguyen-Widrow weight initialization was introduced to initialize the weights instead of random weight initialization in the proposed method. The models evaluated are MABPN with Shannon activation function using Nguyen-Widrow weight initialization in two forms of input: Binary Coded Value and GLCM feature extracted values. The proposed method with GLCM features as input gave a promising result over binary coded transform.

The fact that pancreatic cancer has a low life expectancy, that is only 9% of people survive five years, makes a diagnosis catastrophic. The majority of patients are diagnosed late in life, where care choices are minimal. Early diagnosis of pancreatic cancer will greatly increase a person’s chances of survival. Accurate PC staging will help doctors have the right treatment plan for PC patients at different stages, as well as the diagnostic measures needed for a quicker cancer recovery. In this proposed project, ultrasound images will be analyzed. The noise in the image is minimised using the Median Filter. In the next step, Gray Level Co-occurrence Matrix (GLCM) and Discrete Wavelet Transform (DWT)are used to extract related features. Following this extraction step, the refined characteristics are fed into a Probabilistic Neural Network (PNN) neural network classifier, which determines whether or not cancer is present. Metrics such as sensitivity, precision, and specificity are used in experimental computation.

Medical Image segmentation is complex and challenging task due to the distinct features of biomedical images. MRI scan is commonly used in treating brain, prostate, uterus cancer, ankle and foot cancers. The proposed research work utilizes the real medical images of uterus cancer, brain tumour and throat cancer images obtained from Sagar hospital. MRI images are likely to suffer from the noises such as Gaussian noise, salt pepper noise and speckle noise. Henceforth, noise removal is a very important task to make the image clear with high accuracy and suitable for further diagnosis process. In our paper, the proposed tumour image processing involves three stages 1) converting RGB images to grayscale, 2) pre-processing the images using Gaussian, median and anisotropic filters for noise removal and image enhancement 3) Tumour segmentation using watershed and active contour method. The proposed research work has compared the results of watershed and active contour method. Also, it leverages the ability to find the area and perimeter values of tumour, which indicates the size of brain tumours that provide more accuracy compared to other existing algorithms. Watershed algorithm gives better results, when compared to active contour method, since tumour area using watershed segmentation is more accurate than active contour method. Keywords

Hardware Trojans are malicious modules causing vulnerabilities in designs. Secured hardware designs are desirable in almost all applications. So, it is important to make a trustworthy design that actually exposes malfunctions when a Trojan is present in it. Recently, ring oscillator based detection methods are gaining prominence as they help in detecting Trojans accurately. In this work, a non-destructive method of Trojan detection by modifying the circuit paths into oscillators is proposed. The change in frequencies of ring oscillators upon taking the process corners into account, indicate the presence of Trojans. Since Transient Effect Ring Oscillators (TERO) are also emerging as a good alternative to classical ring oscillators in Trojan detection, an effort is made to analyze the detection capability. Evaluation is done using ISCAS'85 benchmark circuits. Comparison is done in terms of frequency and findings indicate that TERO based Trojan detection is precise. Evaluation is carried out using Xilinx Vivado and ModelSim platforms.

The skin cell which grows abnormally is known as skin cancer and it is caused due to sun rays in the uncovered skin. It spreads within a week, so the early diagnosis of skin cancer is required. Skin Cancer Classification (SCC) based on ranklet transform, co-occurrence features and random forest classifier is presented in this paper. Dermoscopic image in the PH2 database is used in this study for performance evaluation. Initially, the dermoscopic images of three categories normal benign and malignant images are preprocessed to smooth the images. Then the images are given to ranklet transform for decomposition. It produces subband coefficients. The Ranklet Features based Co-occurrence Matrix (RFCM) is used to extract the features and stored in database. The classification is made by Random Forest Classifier (RFC). The result shows better classification accuracy of 93.5% sensitivity is 92% and specificity is 95% is obtained by RFCM and RFC.

Particle Swarm Optimization (PSO) algorithm has been applied in electromagnetics to design microwave absorbers. Generally, microwave absorbers are used for absorbing the electromagnetic radiation caused due to numerous electronic equipments and is being extensively used in stealth technology. The main aim of this paper is to find and analyze the minimized maximum reflection coefficient over a range of frequency and angle of incidence for a fixed number of layers and polarization. An improvised PSO algorithm has been suggested by utilizing a velocity restriction factor that intelligently searches for the optimum solution. The pareto principle with an improvisation in social and cognitive parameters has also been applied. The algorithm succeeded in finding better values of reflection coefficient for the microwave absorber structures comparatively. Based on the pareto principle a form of mutation technique is also used for better convergence. The results have been compared and tabulated for various combinations of the microwave absorber structure and the thickness of each layer is also optimized for a predefined database.

Security and testability are the most important factors affecting designing for testability. Scan chain based testing is a standard DfT (Design for Testability) due to its simple design and low cost. But this method can act as back door, through which the hacker can retrieve the sensitive information through side channel attack. Therefore we developed an efficient and inexpensive LFSR (linear feedback shift register) based secured architecture through which it provides predominant security without effecting testability. The experimental result leads to a low area and power overhead with a secure methodology.

Due to the globalization it is easy to pirate ICs, overbuild IC, and insert Trojan, etc. To protect industries from the illegal production, several methods have been introduced. Logic locking is a method, which locks the functionality of the circuits by using keys. In conventional logic locking method, a single key locks the circuit whereas in weighted logic method the circuit is locked by multiple keys instead of a single key. Normally at highest fault impact location, key gates are inserted, each key gate has corresponding key input. In this work, the comparison between the normal fault impact type insertion and weighted logic locking type insertion are presented. It is observed that there is an increase in the hamming distance (HD), which is the corruptibility of output. The weighted logic locking method which is immune to key sensitization attack and achieves 50% HD and reduces the execution time. Weighted logic locking is a generic method and can be merged with any of the existing methods and thereby gain advantages.

Muti-user resource allocation for cooperative OFDM systems is proposed for use in multi-user situations. The objective is to minimize the total two-stage transmission power for each user for a given transmission rate, formulate the optimization problem and proposes several algorithms for the allocation of various resources. The main aim is to allocate sub-channels for minimizing the total power of transmission.

In a cloud storage, while deduplication enables optimal storage space utilization, it also incurs a substantial overhead in maintaining the metadata, namely, the Fingerprint Index and the File Recipe. As this metadata is huge, it has to be stored in the disk which causes considerable read-latency in Deduplicated Cloud Storage (DCS). In order to improve this read-latency, it would be highly beneficial to prefetch the relevant fingerprints in a cache. Many existing research solutions have utilized either the spatial locality or the similarity among the files to prefetch the relevant fingerprints. However, the DCS that has been designed and implemented in this paper is intended to cater to non-backup workloads that do not exhibit significant spatial locality or similarity among the files. Hence, this paper proposes a suitable alternative prefetching approach that mines the pattern of client read accesses to find the most frequently accessed files. The proposed prefetching approach has been implemented and incorporated in the DCS. The experimental investigations indicate that the proposed prefetching approach improves the cache hit rates by 140% and increases the read throughput by 88% when compared with the Extreme Binning approach (Bhagwat, D., Eshghi, K. Long, D.D.E. and Lillibridge, M. (2009) Extreme Binning: Scalable, Parallel Deduplication for Chunk-based File Backup. Proc. MASCOTS'09, London, UK, September 21–23, pp. 1–9. IEEE) while incurring only a marginal computational overhead of 1.7 s.

With the increase in outsourcing design and fabrication, malicious third-party vendors often insert hardware Trojan (HT) in the integrated Circuits(IC). It is difficult to identify these Trojans since the nature and characteristics of each Trojan differ significantly. Any method developed for HT detection is limited by its capacity on dealing with varied types of Trojans. The main purpose of this study is to show using deep learning (DL), this problem can be dealt with some extent and the effect of deep neural network (DNN) when it is realized on field programmable gate array (FPGA). In this paper, we propose a comparison of accuracy in finding faults on ISCAS’85 benchmark circuits between random forest classifier and DNN. Further for the faster processing time and less power consumption, the network is implemented on FPGA. The results show the performance of deep neural network gets better when a large number of nets are used and faster in the execution of the algorithm. Also, the speedup of the neuron is 100x times better when implemented on FPGA with 15.32% of resource utilization and provides less power consumption than GPU.

Vehicular Ad-Hoc Networks (VANET) are wireless networks that does not require any fixed infrastructure, concerning the traffic safety applications warning messages have to be quickly and smartly to be disseminated in order to reduce the required dissemination time and to increase the number of vehicles receiving the traffic information. Adaptive techniques are not suitable to warn more number of vehicles and they consider only on the vehicles speed, density and position. To improve the warning message propagation process a protocol called Profile-driven Adaptive Warning Dissemination Scheme (PAWDS) is used, and it evaluates the vehicles which are in the dangerous position then send warning messages immediately. PAWDS modifies some of the key parameters of the propagation process which does not detect the vehicles in the dangerous position. The simulation is accomplished by the integration of VANET MOBISIM (MOBIlity SIMulation) Tool and NS-2(Network Simulation) Tool. The X-graph is used to analyze the performance of the network.

This paper focuses on improving the classification accuracy for supervised learning in areas of application with very few training data and with extremely available high dimensionality. This paper proposes a framework which acts as a decision support system incorporating both feature selection and feature extraction to improvise the classification accuracy. The feature selection technique comprises redundancy elimination and relevance analysis. Feature subset selection problems eliminate features which are redundant by using correlation-based maximum spanning tree. But, the eliminated features may contain useful information which may contribute in determining the target or class labels. The principal components are extracted from the eliminated features and they are complemented with the selected features to perform classification. The superiority of the proposed method over other feature selection methods, in terms of computational complexity and classification accuracy, is established extensively on various datasets.

As the functionality in real-time embedded systems becoming complex, there has been a demand for higher computation capability, exploitation of parallelism and effective usage of the resources. Further, technological limitations in uniprocessor in terms of power consumption, instruction level parallelism reaching saturation, delay in access of memory blocks; directed towards emergence of multicore. Multicore design has its challenges as well. Increase in number cores has raised the demand for proper load distribution, parallelizing existing sequential codes, enabling effective communication and synchronization between cores, memory and I/O devices. This paper brings out the demand for effective load distribution with analyzes and discussion about the various task allocation techniques and algorithms associated with decentralized task scheduling technique for multicore systems. This paper also addresses on the multithreaded architecture, where parallel tasks are formulated from sequential code blocks and finally on the techniques to parallelize the sequential code block.

Reducing mutual coupling between elements of antennas are more attractive in the electromagnetic and antenna community. The complementary split-ringresonators (SRRs) are introduced to reduce the mutual coupling between two microstrip patch antennas that radiate in the same frequency band. The new unit cell consists of two complementary SRR. The adjustment in SRR improves the rejection response in terms of bandwidth and suppression. The filtering characteristics of the band-gap structure are investigated using dispersion analysis. This SRR implementation used for Wimax applications at a resonant frequency of 5.8GHz. The design simulation and optimization is done by ADS(Advanced Design System) software. The simulated return loss of this proposed antenna is 7dB. The proposed resonator structures are broadband and also they can be used to minimize coupling and co-channel interference in antennas.

The information processing capability of artificial neural networks (ANNs) is closely related to its architecture and connection weights. Various algorithms have been used to get an optimized structure suited for various applications. The paper does a comparative survey on effectiveness of evolving artificial neural networks based on the particle swarm optimization (PSO) algorithm over other algorithms.A description of this type of system tested on real problems in various domains has also been quoted. The results show that ANNs evolved by PSOANN have good accuracy and generalization ability.