Implementation Challenges in FFNN Architecture Design for PQ Analysis on FPGA | Chapter 2 | Current Approaches in Engineering Research and Technology Vol. 7

The present study addresses the implementation challenges in Feedforward neural network (FFNN) architecture design using the most efficient resources. Power quality (PQ) issues are a major concern in the present day. FPGAs (Field programmable gate arrays) are essential in PQ analysis, particularly in smart meters for data processing, storage, and transmission. Artificial neural networks are selected for PQ event classification as they are found to be more robust once they are trained with a large number of data sets. FPGA's reconfigurability, which makes use of abundant hardware resources to create intricate and time-sensitive data processing units, is one of its greatest features. Due to data loss in the data channel unit caused by the FPGA architecture's support for fixed point arithmetic, the PQ event detection module and classification model must be realized with greater accuracy than software implementation approaches. Most of the work reported, with FFNN (Feedforward neural network) structure occupying a large number of multipliers and adders for classification, most of the work reported has not addressed to minimize the data path resources for FFNN instead have addressed in improving classification accuracy. Based on these issues, this paper addresses the implementation challenges in FFNN architecture design by proposing improved and fast architectures. The proposed FFNN architecture design uses optimum resources. The FFNN-based classifier is designed to perform PQ event detection and classification with 99.5% accuracy. The FFNN processor operates at the maximum frequency of 238 MHz. The present paper addresses the hardware implementation of FFNN cores on the FPGA platform. Interfacing FFNN with all other glue logic modules will require FIFO architecture and a data synchronization network.  The proposed designs for FFNN can be used as IP cores for any signal and control applications.

 

 

Author(s) Details

Prathibha Ekanthaiah
Department of Electrical and Electronics Engineering, Channabasaveshwara Institute of Technology, Gubbi, Tumkur, Karnataka, India.

Likhitha R.

Department of Electrical and Electronics Engineering, Nitte Meenakshi Institute of Technology, Bangalore, Karnataka, India.

 

Chandrakala B. M.
Department of Information Science & Engineering, Dayananda Sagar College of Engineering, Bangalore, Karnataka, India.

 

Radha B. N.
Department of Electrical and Electronics Engineering, Channabasaveshwara Institute of Technology, Gubbi, Tumkur, Karnataka, India.

 

Yashaswini C. S.
Technology, Gubbi, Tumkur, Karnataka, India.

 

Vinaya B. Koradoo
Department of Electrical and Electronics Engineering, Channabasaveshwara Institute of Technology, Gubbi, Tumkur, Karnataka, India.

 

Please see the book here:- https://doi.org/10.9734/bpi/caert/v7/1410                                                 

A Standardized Threshold Approach for Outlier Detection Using Python Algorithms| Chapter 2 | Mathematics and Computer Science: Research Updates Vol. 6

This study aims to standardise threshold techniques and dataset preprocessing steps to improve outlier detection using Python-based algorithms. The objective of the study is to get results that are more precise in the algorithm for most types of datasets. The methodology involved using samples datasets and testing the results when using the normal thresholds of the python outliers detection algorithm, and comparing that results with the results have been done by using the generalize threshold which is mean-median. The results obtained from the supervised results showed that when standardising the threshold using the formula (Mean-Med) produced more precise and more generalised outcomes. The study also applied algorithms that use quartiles (Q1, Median, Q3) and found that adjusting the first quartile to 15% instead of the standard 25% helped to better isolate outliers in the lower range. Similarly, modifying the third quartile threshold to 80% instead of 75% provided more effective detection of upper-range outliers. More precise and more generalize results were obtained when using the formula in the python algorithms use threshold or normal thresholds, which are 0.1 to 2.5 as datasets threshold, compare to use the formula (Mean-Med). The study used some sample datasets for analysis and indicated the potential for applying the method to many other unsupervised datasets in future research.

 

Author(s) Details

 Zahir Saif Alhashami
Ministry of Education, Oman.

 

Please see the book here:-  https://doi.org/10.9734/bpi/mcsru/v6/5746

Unlocking Fingerprint Intelligence: Extracting Ridge, Minutiae, and DWT Features | Chapter 7 | New Horizons of Science, Technology and Culture Vol. 2

A fingerprint image captures the unique spatial pattern of ridges and valleys on the human fingertip, serving as a powerful and widely adopted A fingerprint image is a digital representation of the intricate and unique spatial configuration of ridges and valleys found on the human fingertip. These patterns are distinct for every individual and remain virtually unchanged throughout a person’s life, making fingerprints one of the most reliable forms of biometric identification. Their inherent individuality and permanence have made fingerprint recognition systems indispensable across a wide range of applications, including forensic investigations, national identity verification programs, secure access control systems, and personal device authentication.

 

This chapter, inspired by advancements in biometric sciences, anthropometry, and computational pattern recognition, investigates refined and more efficient methods of extracting rich and diverse features from fingerprint images. The emphasis lies on enhancing the accuracy and reliability of classification and identification tasks by focusing on extracting ridge information, minutiae patterns, and Discrete Wavelet Transform (DWT) features. These advanced features not only facilitate accurate fingerprint matching but also enable the derivation of soft biometric indicators such as gender, age, and potentially even blood type. This opens up promising avenues for the development of lightweight, non-invasive, and cost-effective biometric classification systems, which are particularly valuable in resource-constrained settings.

 

Fingerprint-based systems offer several key advantages over other biometric modalities, such as iris scans, facial recognition, or voice analysis. They typically require less storage space, involve simpler data acquisition procedures, and demand relatively low computational resources. These attributes make fingerprint recognition an ideal candidate for large-scale biometric applications, especially in densely populated or economically limited regions. The methodology explored in this chapter revolves around an automated framework that systematically analyses the spatial and structural features of fingerprints. It leverages both spatial domain and frequency domain analysis techniques to achieve a high-fidelity representation of fingerprint traits. By focusing on the precise extraction of ridge flows and minutiae points such as bifurcations and endings, and further enriching this representation with wavelet-based descriptors, the approach ensures robustness across various acquisition conditions, including variations in scanner types, resolutions, lighting conditions, and finger orientation.

 

Unlike traditional systems that rely solely on features such as ridge counts, thickness, and basic minutiae, the proposed approach employs enhanced feature extraction strategies that significantly improve identification accuracy. Through the application of Discrete Wavelet Transform, fingerprint images are analysed at multiple resolutions, enabling the capture of both global and fine-grained local details. This multiresolution capability allows the system to identify subtle fingerprint variations that might otherwise go undetected, making the overall classification more precise and resilient. Further, ridge structure is assessed in terms of quantifiable metrics like minimum, maximum, and average ridge lengths across the fingerprint. These measurements add another layer of distinguishing information, especially useful in scenarios where individuals have similar minutiae layouts but different ridge formations. In addition to these features, a rich set of minutiae descriptors-such as the number of ridge bifurcations, ridge endings, and total minutiae points-is extracted to enhance the discriminatory capability of the system.

 

By integrating these diverse features-spatial, geometric, and frequency-based-into a cohesive fingerprint recognition pipeline, this chapter presents a powerful and holistic approach to automated fingerprint classification. The comprehensive feature representation facilitates accurate and efficient matching of test fingerprint samples with stored templates, making the system suitable for use in high-security environments, law enforcement databases, and scalable authentication solutions.

 

In essence, this chapter contributes to the ongoing evolution of biometric technologies by introducing refined feature extraction techniques that enhance the reliability and versatility of fingerprint-based systems. The emphasis on improved accuracy, computational efficiency, and adaptability underscores the relevance of these techniques in shaping the future of secure and intelligent biometric identification.

 

Author(s) Details

 

Sayed Abulhasan Quadri
SECAB Institute of Engineering and Technology, Vijayapura, India.

 

Chandrakant P. Divate
SECAB Institute of Engineering and Technology, Vijayapura, India.

 

Tabasum Guledgudd
SECAB Institute of Engineering and Technology, Vijayapura, India.

 

Sayed Abdulhayan
PACE College, Mangalore, India.

 

 

Please see the book here:- https://doi.org/10.9734/bpi/nhstc/v2/5744

Justification of the TOA Method for Lightning Positioning and Algorithm of Application | Chapter 5 | Recent Progress in Science and Technology Vol. 7

 The paper considers a numerical problem fundamental the problem of positioning a electrical discharge discharge by the TOA order for three stations. The mathematical question is formulated in this manner. Three points are fixed on the unit circle. Three spherical slices are introduced, arising out of the given points in the direction of few unknown point, whose position on the circle must be determined. The necessity must be met: the distance each point to an unknown point must add up to the sum of the distance of the segment arising out of this point, and some increment, the alike for all three segments. In the item, the conditions for the solvability of the question are established utilizing the methods of spherical arithmetic and vector arithmetic. It is proved that this lines problem is always explainable when they are fulfilled. The number of answers is two, except in rare cases place there is singular solution. An algorithm for the mathematical solution of the question of positioning a fulmination discharge has been developed.

Author(s) Details:

Yu. R. Shpadi,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan and Institute of Mathematics and Mathematical Modeling, Almaty, Kazakhstan.

A. S. Inchin,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

A. Yu. Lozbin,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

B. Aliyeva,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

M. Yu. Shpadi,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

G. Ayazbayev,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

R. Bykayev,
Institute of Space Techniques and Technologies, Almaty, Kazakhstan.

Please see the link here: https://stm.bookpi.org/RPST-V7/article/view/9951

Breast Cancer Likelihood Prediction Using Bayesian Networks | Chapter 5 | Current Innovations in Medicine and Medical Science Vol. 4

 Worldwide, breast cancer is the most prevalent type of cancer among women. With the help of medical professionals and machine learning algorithms, early detection of breast cancer can result in earlier diagnosis, treatment, patient prolongation, lower treatment costs, and fewer fatalities. To accomplish this, we investigated various methods to improve the accuracy for predicting the likelihood and early diagnosis of breast cancer in this study. Bayesian Network machine learning models were examined to assess the vast medical data. We employed the Breast Cancer Surveillance Consortium (BCSC) dataset comprising 53,370 screening records, the machine learning package, and WEKA (Waikato Environment for Knowledge Analysis). The variable "breast cancer history" was the primary class variable to be predicted in this dataset of thirteen variables, with "class 0" denoting a negative breast cancer diagnosis and "class 1" denoting a positive breast cancer diagnostic. Three separate Bayesian networks—K2 (Hill Climbing Algorithm), TAN (Tree Augmented Naive Bayes), and Simulated Annealing—were produced by various structure-determination algorithms. Using the performance metrics Recall and Precision, we compared the Bayes Networks' performances and related network structures. Our findings show that TAN, K2, and the Bayesian Network produced by Simulated Annealing (SA) had the best overall prediction performances. For Class 1, K2 had the highest Recall, followed by SA and TAN, and TAN had the highest Precision. K2 had the lowest Recall and highest Precision. When the network architectures of the three Bayesian Networks were compared, it became clear that the SA had the highest number of connections between variable nodes and the most complicated network structure, followed by TAN and K2.

Author(s) Details:

G. Nanda,
School of Engineering Technology, Purdue University, West Lafayette, IN-47907, USA.

R. Sundararajan,
School of Engineering Technology, Purdue University, West Lafayette, IN-47907, USA.

Please see the link here: https://stm.bookpi.org/CIMMS-V4/article/view/8394

The QoS in Big Data| Chapter 8 | Current Overview on Science and Technology Research Vol. 2

 The information on the Internet is developing enormously and this dramatic development is it self-turning into a test to access, interaction and show the legitimate data. As the data is developing so does, the innovation and the calculations to get to the information is likewise developing. The rate at which data is developing on the Internet is far more prominent than the advancement of new calculations that are getting to this information. As right now, developing information is of heterogeneous kind which implies it contains blend of text, sound, video, designs which is making a confined admittance concerning the innovation stylish. This enormous amount of heterogeneous information is for the most part known as Big Data.

Author(s) Details:

Masroof Amin,
Department of Computer Technology and Business Management, Maharishi Markandeshwar Deemed to be University, Mullana, Ambala, 133207, Haryana, India.

Munishwar Rai,
Department of Computer Technology and Business Management, Maharishi Markandeshwar Deemed to be University, Mullana, Ambala, 133207, Haryana, India.

Please see the link here: https://stm.bookpi.org/COSTR-V2/article/view/7991

Development of a Stochastic Multicriteria Algorithm for Generating Waste Management Facility Expansion Alternatives | Chapter 11 | Technological Innovation in Engineering Research Vol. 5

When addressing waste management (WM) planning issues, it is frequently useful to present a large number of quantifiably excellent solutions that offer distinctive, contrasting views. This is because, when supporting decision models are required, WM planning frequently entails complex problems with a wide range of performance targets and design criteria that are challenging to identify and explain. The produced alternatives must satisfy all of the aforementioned system requirements while being maximally distinct from one another in the needed decision space. Using modelling to generate alternatives (MGA), you may produce as many different sets of solutions as you can. Simulation-optimization techniques have been routinely used to tackle computationally difficult stochastic WM issues. The stochastic multicriteria MGA strategy for WM planning is described in this study, and it may provide sets of maximally varied alternatives for any simulation-optimization technique that employs a population-based solution algorithm. This algorithmic approach is computationally effective since it offers the required number of maximally diversified solution alternatives in a single computing run of the procedure. The effectiveness of this stochastic MGA technique is demonstrated on a "real-world" waste management facility expansion instance.

Author (s) Details:

Julian Scott Yeomans,
OMIS Area, Schulich School of Business, York University, Toronto, M3J 1P3, Canada.

Please see the link here: https://stm.bookpi.org/TIER-V5/article/view/7470

Study about Rule Mining for Multiple Tables with Fuzzy Data | Chapter 05 | Novel Research Aspects in Mathematical and Computer Science Vol. 1

 The study looks on mining association rules in databases with a large number of tables and fuzzy data, as well as their taxonomy. Many data mining algorithms have been created to cope with databases comprised of a single table with fuzzy taxonomic structures built on top of it. This study makes use of fuzzy data from many tables that were constructed using ER models. Each entity table keeps track of all attributes associated with a specific item, while the relationship table keeps track of relationships between different entities. The study's main purpose is to handle a large number of tables at different levels. The goal of the research is to combine the previously published methods Extended Apriori and Apriori star to build a new algorithm. The study will aid in the identification of relevant outcomes from ambiguous data in database tables.

Author(S) Details

Praveen Arora
Jagan Institute of Management Studies, Rohini, Delhi, India.

Priyanka Gandhi
Jagan Institute of Management Studies, Rohini, Delhi, India.

Geeta Sharma
Jagan Institute of Management Studies, Rohini, Delhi, India.

Sanjive Saxena
Jagan Institute of Management Studies, Rohini, Delhi, India.

View Book:- https://stm.bookpi.org/NRAMCS-V1/article/view/6551

Anomaly Detection of Outlier Features from Spatio-temporal Databases of Landsat-8 Sensor, using Cloud Computing Platform | Chapter 12 | Current Topics on Mathematics and Computer Science Vol. 2

 There has recently been a surge in the number of research articles published in peer-reviewed journals about machine learning and specialized algorithms for feature identification, feature selection, and feature extraction studies. This article distinguishes proof-of-concept applications from domains such as computer vision, remote sensing, image processing, and geospatial database technology. Using satellite imagery from the Landsat-8 sensor for rendering in multimedia and scalable vector processing modes, The article validates the fundamentals and principles of digital image analysis. The article describes the RSVM and DAFE scientific methods in the cloud computing platform in detail using a user-defined algorithm and a scientific approach. It is proposed to introduce data analytics nuances in distributed computing and parallel databases.

Author (S) Details

C. Rajabhushanam
Department of Computer Science Engineering, Bharath Institute of Higher Education and Research, Selaiyur, Chennai, Tamilnadu, India.

View Book :- https://stm.bookpi.org/CTMCS-V2/article/view/1784