A5052850765- Neural Networks and Applications
- Evolutionary Algorithms and Applications
- Protein Structure and Dynamics
- Advancements in Semiconductor Devices and Circuit Design
- Machine Learning in Bioinformatics
- Wireless Communication Networks Research
- Coding theory and cryptography
- Advanced Wireless Communication Techniques
- Heat Transfer and Optimization
- Machine Learning and Data Classification
- Control Systems and Identification
- Sensor Technology and Measurement Systems
- Advanced Computational Techniques and Applications
- Semiconductor materials and devices
- Fuzzy Logic and Control Systems
- Advanced Sensor Technologies Research
- Computational Drug Discovery Methods
- Silicon Carbide Semiconductor Technologies
- Metaheuristic Optimization Algorithms Research
- Imbalanced Data Classification Techniques
- Machine Learning and Algorithms
Cochin University of Science and Technology
2021-2023
University of Science, Art and Technology
2015
Iowa State University
2011
Yahoo (United Kingdom)
2009
Advanced Micro Devices (United States)
2002
Texas Instruments (United States)
2002
The key performance advantages and challenges of SOI CMOS for ULSI applications are discussed in detail. Included is an insightful analysis comparing the benefits technologies over its bulk-Si counterpart. hysteretic trends a floating-body PD/SOI inverter circuit uniquely characterized using Teradyne J971 system; charge-dump self-heating effects shown to be under control advanced 0.13 /spl mu/m device technology. Future technology opportunities described that could provide viable roadmap...
Positive Example based learners reduce human annotation effort significantly by removing the burden of labeling negative examples. Various methods have been proposed in literature for building classifiers using positive and unlabeled However, we empirically observe that classification accuracy state art degrades as number labeled examples decreases. In this paper, propose an active learning method to address issue. The learns starting from a handful positively large Experimental results on...
The linear characteristics of temperature sensors are affected by material properties, aging, self-heating, lead resistance, and external interference. This study aims to improve sensor overcoming power consumption response time limitations. An evolutionary optimised log-ratiometric function is implemented for linearisation using a translinear circuit. differential evolution optimisation strategy used find the optimum values linearising parameters in non-linear function. Further, these...
A comparative analysis of Linearization technique using two prominent neural network approaches is presented. Generally linearizing the non-linear input/output characteristics most sensors available hardware/software indeed a formidable task. The ANN employs different feed forward radial basis function and Levenberg-Marquardt algorithm, for automatic adjustments weights biases to arrive within minimum mean square error number iterations. performance comparison algorithms linearization NTC...
Linear response characteristics of sensors are affected by ageing and changes in material properties. However, existing methods for improving these disadvantaged recurrent full-scale errors insufficient speed. Accordingly, this study aims to reduce temperature sensor non-linearity through implementation evolutionary optimised non-linear functions via a translinear-based application-specific integrated circuit. Moreover, the analogue computation method is compared proposed field programmable...
In a supervised learning system the combination of fuzzy and neural network (NN) balancing complexity by modification membership functions hidden nodes. This paper presents for classification based on Vapnik Chervonenkis Dimensions (VC) structuring layers. A strong mathematical structure with quartic equation is developed in this work to optimize lower bound nodes network. Experimentally it analyzed Thyroid dataset from UCI Machine Learning Repository using MATLAB R2018 (a) software.
The linearity of thermo-resistive sensors is affected by changes in the properties materials, lead resistances, and self-healing ability. However, existing methods employed for linearization are ineffective, given their high error rate prolonged computational time. As a result, this paper discusses using field-programmable gate array(FPGA) to implement an evolutionary optimized ratiometric log-ratiometric function sensor linearization. optimal values linearizing parameters non-linear were...
The acquisition problem in the wideband-CDMA (WCDMA) system proposed by 3rd Generation Partnership Project (3GPP) to ITU consists of determining both timing and identity particular scrambling code being received. A fast search mechanism tackle this problem, based on cyclically permutable (CP) codes is Sriram Hosur (see IEEE Inter. Conf. Communications (ICC), June 1999) for frequency-division-duplex (FDD) component WCDMA. CP FDD mode WCDMA are not optimum time-division duplex (TDD) have be...