Search

The wireless sensor network is gaining significant attention because of its ubiquitous deployability nature. In general, the role of the wireless sensor network is remarkable in smart city applications for data sensing, collecting, and transmitting. Advanced metering infrastructure is an automatic system for the reading of electricity consumption by individual users. The reading of data from meters normally communicated over a wireless medium. Apart from collecting and sensing, routing is one of the major attributes in the AMI network. Successful communication is possible when the link and node are in the proper state. In this paper, we propose a fault-tolerant routing mechanism named Grid Topological Routing scheme to provide an efficient route in case of encountering a faulty path. The simulation shows better performance against the Routing Protocol for Low-Power and Lossy Networks and Ad-Hoc On-Demand Vector algorithm in terms of power consumption and packet delivery ratio. This model is explicitly designed for the smart grid in India. [ABSTRACT FROM AUTHOR]

The classification of recurrence and non recurrence of Hepato Cellular carcinoma (HCC) outcome after Radio Frequency Ablation therapy is a critical task. Multiple time series clinical liver cancer dataset is collected from different dataset and time interval. A merging algorithm is used to merge all attributes collected from different sources in multiple time periods. In order to preserve the originality of information, statistical measures of each attribute is calculated and considered them as additional attributes for accurate prediction. However the merged dataset is unbalanced, in which, the number of samples from HCC recurrence class is much smaller than from HCC non recurrence. The feature weighting scheme select optimal features and parameter of classifiers are sequentially obtained from multiple iterations which causes higher computation time. In this paper, an efficient sampling approach is proposed using Inverse Random under Sampling (IRUS) to overcome class imbalance issue. IRUS under sample the majority class which creates a number of distinct partitions with a boundary separated minority and majority class samples. Additionally an optimization approach is proposed using Artificial Plant Optimization (APO) algorithm to select optimal features and parameters of classifiers to improve the effectiveness and efficiency of classification. The optimization approach reduces the number of iteration and computation time for feature selection and parameter selection for classifiers which classify the recurrence and non recurrence of HCC. Classify patients with HCC and without HCC based on optimal features and parameters by Support Vector Machine (SVM) and Random Forest(RF) classifiers. Finally, the experimental results are conducted to prove the effectiveness of the proposed method over existing method in terms of accuracy, specificity, sensitivity and balanced accuracy. [ABSTRACT FROM AUTHOR]

Due to lack of a central bureaucrat in mobile ad hoc networks, the security of the network becomes serious issue. During malicious attacks, according to the motivation of intruder the severity of the threat varies. It may lead to loss of data, energy or throughput. This paper proposes a lightweight Intruder Node Detection and Isolation Action mechanism (INDIA) using feature extraction, feature optimization and classification techniques. The indirect and direct trust features are extracted from each node and the total trust feature is computed by combining them. The trust features are extracted from each node of MANET and these features are optimized using Particle Swarm Optimization (PSO) algorithm as feature optimization technique. These optimized feature sets are then classified using Neural Networks (NN) classifier which identifies the intruder node. The performance of the proposed methodology is studied in terms of various parameters such as success rate in packet delivery, delay in communication and the amount of energy consumption for identifying and isolating the intruder. [ABSTRACT FROM AUTHOR]

This paper presents a hybrid model to better predict Indian summer monsoon rainfall. The algorithm considers suitable techniques for processing dense datasets. The proposed three-step algorithm comprises closed itemset generation-based association rule mining for feature selection, cluster membership for dimensionality reduction, and simple logistic function for prediction. The application of predicting rainfall into flood, excess, normal, deficit, and drought based on 36 predictors consisting of land and ocean variables is presented. Results show good accuracy in the considered study period of 37years (1969-2005). [ABSTRACT FROM AUTHOR]

The present paper reports a study, where growing hierarchical self-organising map (GHSOM) has been applied to achieve a visual cluster analysis to the Indian rainfall dataset consisting of 142 years of Indian rainfall data so that the yearly rainfall can be segregated into small groups to visualise the pattern of clustering behaviour of yearly rainfall due to changes in monthly rainfall for each year. Also, through support vector machine (SVM), it has been observed that generation of clusters impacts positively on the prediction of the Indian summer monsoon rainfall. Results have been presented through statistical and graphical analyses. [ABSTRACT FROM AUTHOR]

Two algorithms are outlined, each of which has interesting features for modeling of spatial variability of rock depth. In this paper, reduced level of rock at Bangalore, India, is arrived from the 652 boreholes data in the area covering 220 sq⋅km. Support vector machine (SVM) and relevance vector machine (RVM) have been utilized to predict the reduced level of rock in the subsurface of Bangalore and to study the spatial variability of the rock depth. The support vector machine (SVM) that is firmly based on the theory of statistical learning theory uses regression technique by introducing ε-insensitive loss function has been adopted. RVM is a probabilistic model similar to the widespread SVM, but where the training takes place in a Bayesian framework. Prediction results show the ability of learning machine to build accurate models for spatial variability of rock depth with strong predictive capabilities. The paper also highlights the capability of RVM over the SVM model. [ABSTRACT FROM AUTHOR]

This paper present the results of a study made to assess the grades of iron ore deposits in parts of Noamundi area using the hyperspectral image data (EO-1 Hyperion). The study involves hyperspectral image data, preprocessing (removal of atmospheric effects), spectral curve generation from the image, extraction of certain spectral parameters and spectral data analysis. Atmospheric correction was carried out by using algorithms involving atmospheric analysis of spectral hyper-cubes and Quick Atmospheric Correction modules. Spectral curves generated for the pixels that contain iron ores resemble the spectral signature of iron ores, i.e., very strong absorption at 850-900nm region is observed. The strength of the absorption varies spatially at different pixels. Hence, spectral parameters such as radius of curvature of NIR absorption trough (750-1000nm), position of NIR absorption trough and distance from a reference(100% reflectance) line have been derived for all the image-derived spectra and were compared with the concentration of iron in the ore samples of mines. Empirical models were generated by relating the image-derived spectral parameters of iron ore locations and geochemistry of the samples collected from the same location which shows a reasonable match. Strong negative correlation is seen between the radius of curvature of the NIR absorption trough and concentration of iron with a R value of 0.809. The position of the NIR absorption trough is seen to shift towards longer wavelengths with decreasing iron content and the distance from the reference line increases with increasing iron content. Thus, this study indicates that it is feasible to discriminate the grades of iron ores from the spectral parameters derived from Hyperspectral satellite images. [ABSTRACT FROM AUTHOR]

general neural network based methodology that has the potential to replace the computationally-intensive site-specific seismic analysis of structures is proposed in this paper. The basic framework of the methodology consists of a feed forward back propagation neural network algorithm with one hidden layer to represent the seismic potential of a region and soil amplification effects. The methodology is implemented and verified with parameters corresponding to Delhi city in India. For this purpose, strong ground motions are generated at bedrock level for a chosen site in Delhi due to earthquakes considered to originate from the central seismic gap of the Himalayan belt using necessary geological as well as geotechnical data. Surface level ground motions and corresponding site-specific response spectra are obtained by using a one-dimensional equivalent linear wave propagation model. Spectral acceleration values are considered as a target parameter to verify the performance of the methodology. Numerical studies carried out to validate the proposed methodology show that the errors in predicted spectral acceleration values are within acceptable limits for design purposes. The methodology is general in the sense that it can be applied to other seismically vulnerable regions and also can be updated by including more parameters depending on the state-of-the-art in the subject. [ABSTRACT FROM AUTHOR]

The positional accuracy of the Global Positioning System (GPS) is limited due to several error sources. The major error is ionosphere. By augmenting the GPS, the Category I (CAT I) Precision Approach (PA) requirements can be achieved. The Space-Based Augmentation System (SBAS) in India is known as GPS Aided Geo Augmented Navigation (GAGAN). One of the prominent errors in GAGAN that limits the positional accuracy is instrumental biases. Calibration of these biases is particularly important in achieving the CAT I PA landings. In this paper, a new algorithm is proposed to estimate the instrumental biases by modelling the TEC using 4th order polynomial. The algorithm uses values corresponding to a single station for one month period and the results confirm the validity of the algorithm. The experimental results indicate that the estimation precision of the satellite-plus-receiver instrumental bias is of the order of ±0.17 nsec. The observed mean bias error is of the order −3.638 nsec and −4.71 nsec for satellite 1 and 31 respectively. It is found that results are consistent over the period. [ABSTRACT FROM AUTHOR]

Road traffic accidents are a major social concern as well as a crucial issue for the public in recent days due to the risk factors involved. Analysing and identifying the major risk factors of road accident is still a challenging task. In this paper, a fuzzy Context-free Grammar (FCFG)-based association rule mining (ARM) technique is proposed to categorize a heterogeneous road accident dataset into two categories based on the critical factors such as total number of accidents (TA), persons killed (PK) and persons injured (PI). The role of the fuzzy grammar in this paper is to govern the entire algorithm using the prescribed grammar rules to proceed further. The considered road accident dataset does not have class labels; hence there is a need to assign class labels for the available data instance. The accident data with assigned class labels are given as input to K-nearest neighbour (KNN) machine learning algorithm in order to train the classifier for testing purpose. Further, the collected test data from the user are utilized by the KNN classifier for carrying out the performance analysis of the proposed algorithm. The case study is conducted on the National Highway roads, India, to examine the proposed approach. The experimentations are executed for road accident records using MATLAB software and the analysis is made using the following performance measures: accuracy, recall or sensitivity, precision or specificity and F1 score. A comparative study is accomplished with existing algorithms in order to show that the proposed algorithm works with improved accuracy of more than 83%. The results suggested that the road users are responsible for the acceptance or rejection of safe or un-safe roads, respectively. [ABSTRACT FROM AUTHOR]

Typically, the acquired renal ultrasound image includes a course of speckle noises. This paper primarily investigates an approach for the detection of renal calculi by processing those raw US images with the help of a meta-heuristic SVM classifier. One of the major downsides of involving Ultrasound images in medical analysis is the prevalence of Speckle Noises. An Adaptive Mean Median Filter approach has been introduced in the work to get rid of the speckle noises to the maximum extent ever in the literature. Segmentation is performed by employing conventional K-Means and GLCM features were extracted for classification using a meta-heuristic SVM classifier. The proposed methodology investigates with a Real-time Acquired Dataset of Mithra Scans, Tamilnadu, India comprises of 250 clinical Ultra-Sound Kidney Images of which 150 are having Calculi and the rest are Healthy. With the experimental results, the proposed meta-heuristic SVM classifier have performed better in noisy images while comparing with other conventional methods considered in the literature. It exhibits an Accuracy of 98.8% with a FAR rate of 1.8 for FRR as high as 3.3. The results clearly proposed that the novel AMM-PSO-SVM could be a promising technique for object or foreign body detection in a medical imaging application that uses ultrasound imaging. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy (DR) is an eye disease, which affects the people who are all having the diabetes for more than 10 years. The ophthalmologist identifies when the dilated eye exam causes severe in any one of the following deviations in the retina: changes in blood vessels, leaking blood vessels, newly grown blood vessels, swelling of the macula, changes in the lens, and damages to the nerve tissue. It can eventually lead to vision loss. The early detection of DR prevents the cause of blindness. In this paper, we propose the retinal image segmentation and extraction of blood vessels by morphological processing, thresholding, edge detection, and adaptive histogram equalization. For the automatic diagnosis of DR from the fundus image, we also developed a network with the convolutional neural network architecture for accurately classifying its severity. By using high-end graphical processor unit (GPU), we trained this network on the publicly available dataset such as DRIVE, DIARETDB0, and DIARETDB1_v1, and the images collected from the Aravind Eye Hospital, Coimbatore, India. Our proposed CNN achieves a sensitivity of 98%, a specificity of 93%, and an accuracy of 96.9% containing a database of 854 images. [ABSTRACT FROM AUTHOR]