Advanced Computing
10th International Conference, IACC 2020, Panaji, Goa, India, December 5–6, 2020, Revised Selected Papers, Part II
- 2021
- Book
- Editors
- Deepak Garg
- Kit Wong
- Jagannathan Sarangapani
- Suneet Kumar Gupta
- Publisher
- Springer Singapore
About this book
This two-volume set (CCIS 1367-1368) constitutes reviewed and selected papers from the 10th International Advanced Computing Conference, IACC 2020, held in December 2020.
The 65 full papers and 2 short papers presented in two volumes were thorougly reviewed and selected from 286 submissions. The papers are organized in the following topical sections: Application of Artificial Intelligence and Machine Learning in Healthcare; Using Natural Language Processing for Solving Text and Language related Applications; Using Different Neural Network Architectures for Interesting applications; Using AI for Plant and Animal related Applications.- Applications of Blockchain and IoT.- Use of Data Science for Building Intelligence Applications; Innovations in Advanced Network Systems; Advanced Algorithms for Miscellaneous Domains; New Approaches in Software Engineering.
Advertisement
Table of Contents
-
Frontmatter
-
Using AI for Plant and Animal Related Applications
-
Frontmatter
-
Tomato Leaf Disease Prediction Using Transfer Learning
R. Sangeetha, M. Mary Shanthi RaniThe chapter 'Tomato Leaf Disease Prediction Using Transfer Learning' delves into the application of deep learning, specifically transfer learning with CNNs, to predict tomato leaf diseases. It highlights the challenges of diagnosing diseases, particularly for inexperienced farmers, and showcases the potential of deep learning in surpassing human performance in visual recognition tasks. The research focuses on using pre-trained VGG16 and VGG19 models, fine-tuning them to classify diseases such as bacteria spot and septoria spot. The study compares the performance of these models, achieving high accuracy and demonstrating the effectiveness of transfer learning in agricultural applications. The chapter includes detailed methodologies, experimental results, and comparisons with existing work, making it a valuable resource for professionals interested in the intersection of machine learning and plant pathology.AI Generated
This summary of the content was generated with the help of AI.
AbstractTomato leaf is one of the essential edible food item which are affected by diseases. The advent of machine learning and deep learning methods has made crop diseases easier to recognize. The Deep Learning (DL) has emerged as a powerful technique in recent years for image processing and data analysis, with promising results. DL has been applied in various domains, including the field of agriculture. In deep learning, Convolution Neural Network (CNN) is one of the architectures are widely used. Transfer learning is a new approach in DL where the pre-trained models are used to train a new dataset for expediting the training process. This research work focuses on developing a Transfer Learning driven Prediction model for leaf disease detection. In this paper, a new composite and comprehensive prediction schemes specifically for tomato leaf disease analysis are developed. This Deep Neural Network (DNN) classifier is designed to build on the combination of two deep learning models, namely VGG16 and VGG19. The results were recorded in terms of Precision, Recall, F1-score and accuracy. The improved accuracy of Transfer Learning experimented in this reported research work vouches for its application in the prediction of Leaf Disease. -
Amur Tiger Detection for Wildlife Monitoring and Security
Shankho Boron Ghosh, Ketan Muddalkar, Balmukund Mishra, Deepak GargThe chapter delves into the critical issue of Amur tiger conservation, highlighting the decline in their population due to poaching and habitat loss. It introduces deep learning algorithms as a solution to traditional sensor-based methods, focusing on object detection techniques like Faster R-CNN and SSD. The study uses the ATRW dataset, emphasizing the need for high-variation datasets. The authors experiment with state-of-the-art models, such as SSDLite Mobilenet, achieving high accuracy and low latency. The chapter concludes by suggesting future improvements, including pose detection and deployment on mobile computers for real-time monitoring. The innovative approach and practical applications make this chapter a valuable resource for wildlife conservation and technological advancements in object detection.AI Generated
This summary of the content was generated with the help of AI.
AbstractIn our ecosystem, wildlife plays a key role in sustaining different natural processes in nature. So, protection and conservation of wildlife become vital, especially those which are on verge of being extinct. One such species is the Amur Tiger, which is categorized as endangered. The traditional method of Amur Tiger recognition is volunteer intensive and hence was difficult and time-consuming as well. Therefore in our project, we attempt to provide a more efficient and reliable method for detecting Amur tiger with the help of the recent advancements of neural networks and deep learning algorithms. However, deep learning algorithms require ample amounts of data for the training, in which ATRW (Amur Tiger Re-identification in the wild) datasets is the most suitable one with an adequate number of variations. This dataset contains 2485 images for training and 277 images for validation with their annotation in xml format for each instance of the tiger. To detect the Amur tiger, we have applied various state-of-the-art object detection algorithms on this dataset. Out of all the models applied on this dataset, SSDlite model achieves 0.955 mean Average Precision values, which is an outstanding performance of any deep learning models applied for detection tasks. In addition, out of all the models applied and available in the literature, SSDlite is one of the faster models which inturn have least inference time comparatively. -
Classification of Plant Species with Compound and Simple Leaves Using CNN Fusion Networks
P. G. Mary Sobha, Princy Ann ThomasThe chapter delves into the crucial role of plant species identification for ecosystem preservation and various applications. It highlights the challenges of conventional methods and the potential of deep learning, particularly CNNs, for this task. The proposed model employs a fusion of two CNN networks, VGG16, for feature extraction and an SVM classifier for classification. The use of real complex background leaf images and a new dataset, Realleaf, sets this research apart. The fusion model demonstrates superior accuracy, proving the effectiveness of combining CNN and SVM for plant species identification.AI Generated
This summary of the content was generated with the help of AI.
AbstractAn automatic plant species identification system could help to identify plant species very easily. Deep learning is an AI function which works like the human brain, with artificial neural networks. In neural networks, neuron nodes are connected like a web and used to extract higher-level features from input. Convolutional neural network (CNN), deep belief network (DBF) and recurrent neural network (RNN) etc. are deep learning networks and can extract more detailed information compared to conventional machine learning techniques [1, 10]. CNN is a very good choice for image processing, and it can work with large datasets efficiently. In our project VGG16 CNN is used to extract the features from leaf images of a simple and compound leaf. For the identification of plant species with simple and compound leaves with real complex background images, this paper proposes a fusion CNN model using original whole leaf images and patch images. A transfer learned VGG16 CNN was used for feature extraction and classification of real complex background images [1]. Feature extraction and classification are done with original (model1) and patch (model2) images separately with VGG16 CNN models. The feature maps from intermediate levels of model1 and model2 are taken, then concatenated and classified using SVM and KNN. The CNN-SVM model has the best performance over model1, model2 and CNN-KNN model. The proposed fusion model shows the efficiency of 98.6% accuracy in model evaluation and 90% accuracy in plant identification using complex background leaf images. -
A Deep Learning-Based Transfer Learning Approach for the Bird Species Classification
Tejalal Choudhary, Shubham Gujar, Kruti Panchal, Sarvjeet, Vipul Mishra, Anurag GoswamiThis chapter explores the application of deep learning and transfer learning techniques for the classification of bird species using high-resolution images. It introduces the problem of bird species identification, highlighting the importance of this task for wildlife conservation and environmental monitoring. The authors propose a deep learning-based transfer learning approach using Convolutional Neural Networks (CNNs) and evaluate the performance of different CNN architectures, including VGG16, ResNet50, and MobileNetV2. The chapter provides a detailed analysis of the experiments conducted and compares the accuracy and efficiency of these models, demonstrating the potential of this approach for practical applications such as bird detection and conservation efforts.AI Generated
This summary of the content was generated with the help of AI.
AbstractBirds are present in various scenarios that appear in different shapes, sizes, and colors. It is estimated that there exist around 10,000 different species all over the world. Watching birds is a common practice but identifying their species requires bird knowledge. Human ability to recognize the birds through images is much easier as compared to audio classification, so bird species classification through images is in more trend and accurate too. Traditional machine learning and audio-based approaches are not suitable for bird classification. In this paper, we proposed a transfer learning-based approach for the classification of the bird dataset of 200 bird species. We performed numerous experiments with different deep learning models and the experimental results suggest that the proposed transfer learning-based approach significantly performs better than other state-of-the-art methods.
-
-
Applications of Blockchain and IoT
-
Frontmatter
-
Integration of Explainable AI and Blockchain for Secure Storage of Human Readable Justifications for Credit Risk Assessment
Rahee Walambe, Ashwin Kolhatkar, Manas Ojha, Akash Kademani, Mihir Pandya, Sakshi Kathote, Ketan KotechaThe chapter delves into the integration of Explainable AI (XAI) and Blockchain technology for securely storing human-readable justifications in credit risk assessment. It discusses the importance of responsible AI systems and the need for interpretable models, particularly in sectors like finance. The authors propose a three-phase system that uses XAI to generate explanations for credit-scoring models and Blockchain to securely store these explanations. The system design, methodology, and results are detailed, showcasing the potential of this integration to enhance transparency, accountability, and trust in automated decision-making processes. The chapter also highlights the challenges and future directions in this field, making it a valuable resource for professionals seeking to understand the intersection of AI, blockchain, and financial risk assessment.AI Generated
This summary of the content was generated with the help of AI.
AbstractAutomated credit risk assessment is one of the most important applications of artificial intelligence. However, the traditional black-box AI models are no longer suitable due to the regulations imposed all over the world to ensure security, privacy and interpretability of the data. A notable example is the General Data Protection Regulation (GDPR). The problem is twofold; firstly, human-understandable explanations are required and secondly, these explanations must be stored securely to ensure the privacy of the user and should be accessible in the future if required. In this paper, we propose a system that leverages the secure and immutable nature of Blockchain to store machine learning model explanations for credit-scoring. Our proposed system aims to enable local interpretations of the global model to be publicly available to customers to access in a secure manner. A chronological account of the model decisions can be obtained per user, helping the user further understand the series of events leading to their loan approval decisions. Our results demonstrate the trustworthiness of an explained model prediction, with the security, reproducibility, traceability and transparency of Blockchain, providing the end-user with a way to securely request an explanation for the credit-scoring decision. -
Blockchain Based Approach for Managing Medical Practitioner Record: A Secured Design
Neetu Sharma, Rajesh RohillaThe chapter delves into the critical role of medical data in the healthcare industry and the challenges of integrating diverse medical records. It introduces blockchain technology as a solution to improve information security, trust, and data sharing. The proposed blockchain-based model for managing medical practitioner records (MPR) is designed to ensure accurate and tamper-proof storage of medical data, eliminating the need for third-party trust. The system allows medical institutions to securely transfer MPRs, and demanding entities to easily access and validate medical records. The architecture and implementation of the model are detailed, highlighting its potential to protect patients from fake treatments and ensure the integrity of medical records. A comparative analysis with prior works underscores the novelty and effectiveness of the proposed approach.AI Generated
This summary of the content was generated with the help of AI.
AbstractProper management of medical practitioner record (MPR) and its accuracy is essential for protecting public health. Medication errors can cause several physical injury and possible death to patients. Most of the medical institutes and organizations maintain such records locally in a centralised manner which can be steal by hackers for creating fake identities. Blockchain ledger has immutability, integrity and security benefits that allows patients to access accurate medical practitioner’s record and to take treatment from genuine practitioner only. Purpose of this study is to design a blockchain based immutable and secured model for MPR. This can be a very useful mechanism to guarantee right record availability at the time of registering medical practitioner for practice in any health organization in order to avoid fake medical Practitioner from playing with the public health. Proposed model enables medical institutions to transfer MPR content on the blockchain network and empowers demanding entities (hospitals, patients, medical colleges) to retrieve it to validate medical practitioner’s record. Unique IDs are given to each MPR by universally unique identifier (UUID) and the hash computation of blocks containing MPR data is done using secure hash algorithm (SHA)-256 algorithm. The developed prototype comprises four modules for different purposes such as MPR transferring, MPR block mining, MPR block verification and addition in blockchain and MPR retrieval. All modules of the created model are tested effectively and few test outcomes are presented. The test results shows that created model is suitable for MPR management. -
iSHAB: IoT-Enabled Smart Homes and Buildings
V. Lakshmi NarasimhanThe chapter 'iSHAB: IoT-Enabled Smart Homes and Buildings' delves into the transformative potential of the Internet of Things (IoT) in modernizing homes and buildings. It begins by discussing the historical evolution of structural standards and the need for new considerations such as eco-friendliness and energy efficiency. The paper defines 'smartness' and 'eco-friendliness' in the context of homes and buildings, providing practical examples of how IoT can enhance energy management, water usage, gardening, waste-water treatment, and earthquake protection. It also presents a private cloud-based information architecture for smart homes and buildings, evaluating its performance using parametric modeling. Additionally, the chapter addresses security and privacy issues in smart homes, emphasizing the importance of encryption and preventive measures. The comprehensive C-COM model and the evaluation of the iSHAB private cloud system are notable highlights, offering valuable insights into the future of smart home technology.AI Generated
This summary of the content was generated with the help of AI.
AbstractThis paper concerns the design of smart homes and buildings using coordinating Internet of Things (IoT). IoTs offer a variety of sensors and other smartness functions so that they can be deployed over a number of typical household scenarios and buildings to make them eco-friendly and intelligent. Such mechanisms include providing active management of energy, power generation, water, solid and liquid waste, bio-waste, garden and room space, besides offering earthquake monitoring and protection. This paper also presents a typical device model using a variant of UML, called C-COM used in energy industry. The devices are held over a private cloud based system, whose performance is also evaluated using parametric evaluation technique. The performance indices show that the system is viable even over typical households. It is also cautioned that while this system proposed a decent degree of security, privacy issues are matters yet to be addressed. -
Implementation of Blockchain Based Distributed Architecture for Enhancing Security and Privacy in Peer-To-Peer Networks
Kriti Patidar, Swapnil JainThis chapter delves into the implementation of blockchain technology in peer-to-peer networks to enhance security and privacy. It begins with an overview of blockchain technology and its evolution, highlighting its applications beyond the financial sector. The chapter then explores the properties of blockchain ledgers, distributed systems, and consensus mechanisms. A practical implementation of a blockchain-based distributed architecture in a P2P network is presented, showcasing the use of NodeJS, cryptographic techniques, and consensus algorithms. The implementation enhances data security and privacy through cryptographic methods and immutable transaction records. The chapter concludes by discussing the challenges and future directions of blockchain technology in decentralized systems, making it a valuable resource for professionals seeking to understand and implement blockchain solutions in various industries.AI Generated
This summary of the content was generated with the help of AI.
AbstractBlockchain technology has gained maximum popularity in recent years as compared to the other latest technologies. It works as immutable ledger in which the transactions are done in decentralized way. It has started spreading its wings now as it is not only used in financial applications but many other companies in the industry are trying to leverage the benefits offered by blockchain in their projects. As industries are committing towards blockchain based projects, there is a need of blockchain implementation and associated development. This paper introduces the basic requirements of blockchain development and proposes a generic blockchain that can be considered as a mould for implementation in different use cases. In this paper a distributed architecture is implemented using blockchain with improved security and privacy. The expected contribution of this paper is to provide an overview for researchers and developers to build applications utilizing the blockchain as a basis. -
Application of Neural Networks to Simulate a Monopole Microstrip Four-Tooth-Shaped Antenna
Zufar Kayumov, Dmitrii Tumakov, Angelina MarkinaThe chapter discusses the application of neural networks in simulating a monopole microstrip four-tooth-shaped antenna. It begins by highlighting the importance of microstrip antenna modeling and various approaches used in the field. The primary focus is on the use of multilayer perceptrons (MLPs) to solve both direct and inverse problems in antenna design. The direct problem involves predicting electrodynamic characteristics from geometric parameters, while the inverse problem involves reconstructing antenna geometry from these characteristics. The chapter delves into the training and evaluation of MLPs, comparing their performance with traditional regression models. It also explores the impact of neural network architecture on prediction accuracy, including the effects of hidden layers and neurons. The work concludes with a detailed analysis of the effectiveness of MLPs in antenna modeling and design, providing valuable insights for specialists in the field.AI Generated
This summary of the content was generated with the help of AI.
AbstractThe application of a multilayer perceptron (MLP) for calculating the electrodynamic characteristics of a monopole microstrip four-tooth-shaped antenna (direct problem) as well as for solving the problem of synthesizing a four-tooth-shaped antenna for given characteristics (inverse problem) is considered. A backpropagation algorithm is used to train MLP. Various MLP architectures are considered. The errors of MLP prediction of various electrodynamic characteristics of the antenna are estimated. The graphs of dependences of a network operation error on the number of neurons in the hidden layers are presented. The proposed MLP architecture with two hidden layers each having 16 neurons is found to give a sufficient accuracy. A conclusion is made regarding a more accurate determination of parameters with a given architecture using MLP in comparison with regression models.
-
-
Use of Data Science for Building Intelligence Applications
-
Frontmatter
-
RECA: Deriving the Rail Enterprise Confluence Architecture
V. Lakshmi NarasimhanThe chapter introduces the Rail Enterprise Confluence Architecture (RECA), a framework designed to harness the power of IoT sensors and big data analytics for enhancing railway systems. It begins by highlighting the manual monitoring practices prevalent in many countries and the potential of IoT and big data for real-time decision-making. The text delves into the issues at stake in rail data analytics, including data acquisition, management, and analysis techniques. It presents RECA as a systematic approach to capturing and integrating diverse data sources, emphasizing the use of a Federated Asset Registry and a heterogeneous Data Warehouse. The architecture is evaluated using a parametric model, demonstrating its viability and potential impact on the rail industry. The chapter concludes with a discussion on future research directions, emphasizing the importance of addressing security, privacy, and data provenance issues.AI Generated
This summary of the content was generated with the help of AI.
AbstractIn most countries, a number of data items pertaining to the running of trains and tracks are usually manually measured, monitored and analyzed by trained technicians, usually not in real-time. With the advent of a variety of sensors and Internet of Things (IoT), these sensors can be integrated over a network and data collected in real-time, analyzed and decisions made. This paper presents the mechanism to derive the Rail Enterprise Confluence Architecture (RECA) considering various railway data sources and their processes and procedures. RECA can be mounted over a private cloud and its core contains an ISO compliant Asset Registry along with data warehouse and various tools to perform data analytics in real-time. The performance of the system has been evaluated using parametric modeling and the indices indicate that the RECA system is practically viable and further, offers good value-add to rail data analytics over all type of train systems. -
Energy-Aware Edge Intelligence for Dynamic Intelligent Transportation Systems
Shajulin BenedictThe chapter delves into the crucial role of Intelligent Transportation Systems (ITS) in modern travel, highlighting the need for energy-efficient solutions to overcome challenges like data collection, privacy, and energy inefficiency. It introduces the EAEI framework, which predicts air quality values at tourism locations using edge nodes, significantly enhancing traveler experience and safety. The framework identifies an energy-optimal code version from a pool of prediction algorithms, ensuring efficient and accurate air quality predictions. Experiments validate the framework's effectiveness, demonstrating substantial energy savings and accurate predictions. The chapter concludes by emphasizing the importance of energy-conscious edge intelligence in revolutionizing intelligent transportation systems.AI Generated
This summary of the content was generated with the help of AI.
AbstractIntelligent Transportation System is propagating its roots among various researchers and smart city experts owing to the emergence of a wide range of applications or services, including connected cars. Next-generation tourism deliberately relies on its upcoming recommendation services that are fueled up with dynamic edge intelligence. This paper proposes an Energy-Aware Edge Intelligence (EAEI) framework for guiding tourists or automated vehicles in selecting air quality-aware tourism locations in an energy-efficient manner. EAEI collects the air quality parameter values through massive sensor networks; opts for an energy optimal prediction service to suggest air quality values; and, guides vehicles or tourists in cities. The proposed framework was evaluated at the IoT Cloud Research Lab and found to save over 90% of energy consumption in edge nodes. Besides, the article highlights the comprehensive need for the framework which inspires tens of thousands of solutions in the near future. -
A Single Criteria Ranking Technique for Schools Based on Results of Common Examination Using Clustering and Congenital Weights
Dillip RoutThis chapter addresses the challenge of ranking schools based on student performance in common examinations. Traditional methods, such as using average marks, fail to account for the volume of students and the distribution of marks. The proposed technique employs clustering to group students based on their performance and assigns congenital weights to each group. This approach reduces ties and provides a more nuanced ranking that reflects both the quality and quantity of student performance. The method is demonstrated through a real-world case study, showcasing its effectiveness and potential for broader application in educational data analysis.AI Generated
This summary of the content was generated with the help of AI.
AbstractIn this paper, an innovative clustering-based ranking technique is presented to rank the schools for a common examination. Particularly, the inputs to the problem are a set of schools, and a corresponding set of marks of the students appeared in the common examination. The expected output is the order of these schools on the basis of the marks of students. Note that the number of students participating from each school varies apart from various ranges of marks obtained by its students. Hence, ranking these schools is not straightforward as it is for the students who can be easily ranked by the marks only. An intuitive idea is to rank schools based on the average mark over all the students of each school. However, this will suppress the impact of the volume of students, and likely to consist of several ties. Also, it may discourage higher participation in the entrance and/or scholarship examinations where the school performance is also measured, especially barring the moderately meritorious students. Thus, a solution is adequate only if it additionally addresses the volume of students although the mark is the lone criterion to resolve the issues in an intuitive approach. The proposed ranking technique categorizes the students into disjoint groups generated through a cluster of marks. Then, the schools are ranked on scores that are computed based on the distribution of volumes and marks of the students in various groups. The ranking of schools for an undertaken case study shows that the ordering is adequate.
-
-
Innovations in Advanced Network Systems
-
Frontmatter
-
5G Software-Defined Heterogeneous Networks in Intra Tier with Sleeping Strategy
Rohit S. Waghmare, Hemlata Patil, Sujata KadamThis chapter delves into the advancements of 5G software-defined heterogeneous networks, focusing on intra-tier cooperation and sleeping strategies to address the growing demands of mobile users. It discusses the limitations of conventional cellular networks and the benefits of integrating small cell base stations (SCBs) to form heterogeneous networks. The use of a Software-Defined Network (SDN) controller and mobile edge computing servers is highlighted for efficient network management and real-time decision-making. The chapter also explores partial connectivity and load balancing techniques to enhance energy efficiency and reduce network congestion. Through extensive simulation results, the chapter demonstrates the improved performance of the proposed network model, including increased connectivity probability, load balancing, and reduced delay in packet transmission. The chapter concludes by emphasizing the transformative potential of SDN in managing and controlling networks dynamically, making it a must-read for professionals seeking to optimize network performance in the 5G era.AI Generated
This summary of the content was generated with the help of AI.
AbstractIn this work, we have studied connectivity probability, efficiency, load balancing technique, and delay of the heterogeneous networks. In the 5G Software-Defined Heterogeneous Networks with Intra tier cooperation, we have used a sleeping strategy that helps to save power and enhance the efficiency of the network. The load balancing technique is used to reduce the traffic of the network. The reduction of traffic automatically increases the efficiency of the network. -
Performance Investigation of MIMO-OFDM Based Next Generation Wireless Techniques
Balram Damodhar Timande, Manoj Kumar NigamThe chapter delves into the performance evaluation of MIMO-OFDM based next-generation wireless techniques, emphasizing the advantages and limitations of this technology. It begins by introducing OFDM as a prominent multicarrier modulation scheme, discussing its high data rates and efficiency. The text then explores the challenges posed by channel variations, fading, and interference in OFDM systems. To mitigate these issues, the chapter introduces MIMO antenna configurations and their capacity equations, showcasing the benefits of spatial diversity and improved channel capacity. The MIMO-OFDM system is presented as a promising solution for future wireless networks, offering high spectral efficiency and robustness against multipath fading. The chapter includes a detailed analysis of the MIMO antenna system model, with expressions for maximum SNR and average BER. Simulated results are provided, demonstrating the enhanced performance of MIMO-OFDM systems with variable symbol lengths and MRC techniques. The chapter concludes by highlighting the potential of MIMO-OFDM for reliable communication in next-generation wireless systems, making it a valuable resource for professionals seeking to understand and optimize these advanced technologies.AI Generated
This summary of the content was generated with the help of AI.
AbstractThe merger of the Multiple-input multiple-output (MIMO) antenna system with orthogonal-frequency-division-multiplexing (OFDM) is a prominent wireless technique for future wireless systems. This combination provides spatial diversity in the MIMO-OFDM system which minimizes the fading effect, improves channel capacity and data throughput, and also enhances the communication reliability. Due to multipath fading and channel impairment, the performance of the wireless link is degraded in terms of bit error rate (BER), and Symbol error rates (SER). Also due to high peak to average power ratio (PAPR) issue in the OFDM system significantly affects the orthogonality and hence results in inter-carrier interference (ICI). The Scarcity of bandwidth and the throughput in today’s high-speed digital communication is a primary issue, which can be improved with the combination of the MIMO-OFDM technique. OFDM scheme optimizes the system requirement and reduces complexity. It offers high spectral-efficiency with reliable wireless communication. The analytical and simulated result shows that the proposed scheme guarantees the reliability in wireless communication as well as significant enhancement in channel capacity which would meet the future wireless system requirements. -
DLC Re-Builder: Sketch Based Recognition and 2-D Conversion of Digital Logic Circuit
Maitreyi Sharma, Sonal Nipane, Rachita, Krupa N. Jariwala, Rasika KhadeThe chapter introduces the DLC Re-Builder, a system designed to recognize and convert hand-drawn digital logic circuits into 2D digital logic circuits. It leverages sketch-based recognition and 2D conversion to streamline the circuit design process. The system identifies basic objects such as logic gates and wires using object detection and replaces them with beautified versions. The chapter discusses the methodology, dataset collection, model training, and wire detection processes. It also covers the reconstruction of circuits and the generation of Boolean expressions and truth tables. The proposed methodology includes the use of deep learning models like YOLO and R-CNN for gate detection, with R-CNN showing superior performance. The chapter concludes by highlighting the application's accuracy and responsiveness, and outlines future work, including testing on larger datasets and enhancing the system's robustness.AI Generated
This summary of the content was generated with the help of AI.
AbstractSketch owing to its ability to convey ideas and ease of creation has found application in almost all spheres. From client presentations to documentations, sketching is ideal for conveying thoughts. Sketching of initial ideas or drafts for brainstorming sessions is a common practice. However, these rough sketches require remodelling to a formal format, which adds an overhead to the process. Thus, arises a need to efficiently create, store and update hand-drawn sketches. Computer interfaces for sketching, that can augment the process by maintaining fair drafts, recognizing objects in the sketch, analyzing and summarizing the sketch and more, have been developed. A system that does similar processing for logic circuits is aimed. The complete application is designed optimally to give an accuracy of 89.41% using R-CNN and 82.44% using YOLO. DLC Re-builder(Digital Logic Circuit Re-builder) is a system that would accept sketched logic circuits as input, recognize various components of the circuit and convert the sketch-based circuit into formal graphical format. It would also generate a boolean expression for the final reconstructed 2-D output along with the truth table. -
Design of I/O Interface for DDR2 SDRAM Transmitter Using gpdk 180 nm Technology
Jayashree C. NidagundiThe chapter delves into the design of an I/O interface for a DDR2 SDRAM transmitter, utilizing Cadence Virtuoso software and gpdk 180 nm technology. It discusses the specifications and block diagram of the interface, which includes components such as level shifters, power detectors, logic circuits, pre-drivers, and drivers. The methodology for designing these components is detailed, with a focus on ensuring compatibility and performance in high-speed data transfer. The chapter also presents DC and transient analysis results, demonstrating the effectiveness of the proposed design. This comprehensive approach makes the chapter a valuable resource for professionals seeking to optimize memory interface designs in high-speed applications.AI Generated
This summary of the content was generated with the help of AI.
AbstractIn the integrated environment or in integrated circuit, processor work with set of standard logic voltage level. The other components inside the chip may not use the same standard logic voltage level. Processor speed is usually faster than speed of other devices. So we cannot slow down and reduce the voltage levels of the processor for the sake of communication with other devices of the chip. Hence the user can design different devices with different voltage logic levels and speed. I/O blocks are used to assist the compatibility. This paper proposes the design of input output interface for maintaining voltage levels from 0.9 V to 1.8 V of dual data rate (DDR2) SDRAM transmitter block. Proposed work deals with design of level shifter, power detector circuit, pre-driver and driver circuit using Cadence Virtuoso with gpdk 180 nm technology. The characteristics of input output interface of DDR2 SDRAM transmitter are specified and designed. DC and AC analysis are performed with complete process corners. The results of proposed work like signalling, impedance, voltage, current and power specifications are observed and measured.
-
-
Advanced Algorithms for Miscellaneous Domains
-
Frontmatter
-
Improved SMO Based on Perturbation Rate in Local Leader Phase
Naveen Tanwar, Vishnu Prakash Sharma, Sandeep Kumar PuniaThe chapter focuses on improving the Spider-Monkey Optimization (SMO) algorithm by modifying the perturbation rate in the local leader phase. It begins by introducing nature-inspired algorithms and swarm intelligence, highlighting the SMO algorithm inspired by spider monkeys' food-searching behavior. The core of the chapter discusses the main steps of SMO, including initialization, local leader, global leader, and the proposed approach to enhance perturbation rate. The proposed approach is validated through benchmark problems and compared with other optimization algorithms like SMO, DE, and ABC, showing significant improvements in performance metrics such as mean function evaluation, success rate, error, and standard deviation. The chapter concludes by emphasizing the potential future applications of the improved SMO in machine learning for classification and prediction tasks.AI Generated
This summary of the content was generated with the help of AI.
AbstractThe paper, proposed a modified local leader phase of SMO based on perturbation rate and knowledge learn from experience of any spider monkey. “Improved SMO based on perturbation rate in local leader phase” and abbreviated as vSMO or variant of SMO is the name of proposed variant and approach gives more chance for travel or searching in the already searched area. To avoid poor divergence, paper allows incorporation of best known random direction for further improvement in perturbation rate that is, the amount of value by which perturbation rate changes on the basis of knowledge of any spider monkey which is obtained by experience during exploration of search space.Performance of this approach is tested over some bench mark functions out of which 26 functions including real world problems are showing that result is improved when compared with other methods. -
Generation of Pseudo Random Sequence Using Modified Newton Raphson Method
Aakash Paul, Shyamalendu KandarThe chapter introduces a groundbreaking approach to generating pseudo-random sequences using a modified Newton-Raphson method, deviating from traditional chaotic techniques. The method leverages the rapid convergence of the Newton-Raphson algorithm to generate sequences with high periodicity and unpredictability. The proposed technique has been rigorously tested using NIST randomness tests and various security analyses, demonstrating its potential for cryptographic applications. The chapter also highlights the method's sensitivity to initial conditions and its large key space, making it resistant to brute-force attacks. Furthermore, the chapter discusses the information entropy of the generated sequences, showcasing their high randomness and suitability for secure communications.AI Generated
This summary of the content was generated with the help of AI.
AbstractRandomness brings unpredictability and applicable to a number fields like science, statistic, gaming, gambling etc. True random number generator generates a random sequence based on unpredictable physical means. On contrary pseudo random sequence is generated by some mathematical equation (algorithm) using some initial parameters known as seed. Pseudo random sequence looks like random but can be retrieved by the seed feeding in the algorithm. Pseudo random sequence has wide applications in cryptography, simulation etc. Chaotic map is a preferable choice to the researcher community for pseudo random sequence generation due to its underlying properties like deterministic, long term unpredictability, high sensitivity to initial condition etc. Several non-chaotic techniques are also taking their ways for pseudo random sequence generation. In this current communication a non-chaotic method based on modified Newton Raphson technique is proposed to generate a pseudo random sequence. Several test results including randomness testing using NIST suite provide the proposed technique a strong base in the field of pseudo random sequence generation. -
Smoothening Junctions of Engineering Drawings Using C2 Continuity
Paramita DeThe chapter explores the necessity of converting pixel-based images into vector-based representations for engineering drawings, highlighting the advantages of vector images over raster images. It delves into the process of vectorization, including preprocessing steps such as binarization, noise removal, and text-graphics separation. The core of the chapter focuses on the smoothening of vectorized drawings using C2 continuous cubic Bézier curve approximation, which ensures the smooth representation of engineering drawings. The method is demonstrated through experimental results, showcasing the effectiveness of the C2 continuity in enhancing the visual quality of vectorized images. The chapter concludes by discussing the potential for future improvements, such as automating the curvature application process using deep learning models.AI Generated
This summary of the content was generated with the help of AI.
AbstractVector representation of images has numerous advantages such as storing, analyzing and editing of the same. The processed vector information can be used to reconstruct the images. This paper proposes a vectorization algorithm followed by smoothening of the drawings by applying C2 continuous Bézier curve approximation technique. For vectorization, text-graphics separation and skeletonization has been carried out on the input images followed by computing vector information based on the geometric features of different shapes. Junctions has been detected from the vector information and C2 continuity is applied on them for smoothening the shapes. The results of the proposed algorithm demonstrate that it can transform the input images into smoothen ones by individually correcting the shapes of different objects in the images. -
An Efficient Privacy Preserving Algorithm for Cloud Users
Manoj Kumar Shukla, Ashwani Kumar Dubey, Divya Upadhyay, Boris NovikovThe chapter discusses the critical issue of privacy and security in cloud-based messaging services, highlighting the vulnerabilities of existing encryption methods. It introduces a novel algorithm that combines Blowfish cryptography with Honey Encryption to provide enhanced security against brute force attacks. The proposed algorithm is analyzed for its efficiency, security, and scalability, demonstrating its superior performance compared to existing frameworks. The chapter also includes a detailed simulation study using AnyLogic, showcasing the practical implementation and performance analysis of the proposed system. The integration of these advanced techniques offers a robust solution for preserving privacy and improving security in cloud environments.AI Generated
This summary of the content was generated with the help of AI.
AbstractNow a day’s cloud-based instant message services are becoming popular. It can be an efficient way of communication even for events that require a certain level of security like banking, e-commerce applications, e-transactions etc. Cloud can provide a distributed notification of messages along with the queuing them to facilitate the transfer of messages between various applications and connected systems. Cloud message services can be as simple as a pay-as-you-go service, which does not require significant investment and protect its customers, providing a hassle-free and maintenance-free services platform. This research paper proposes a privacy-preserving approach to users utilizing cloud-based services. A cloud-based model is created and simulated. The proposed method is based on a multi-method approach where a virtual system is acting as an agent and is interconnected with other virtual systems. Inside a virtual system, a discrete event model of job processing and privacy preserving is defined as shown in the process flowchart. The proposed model is scalable. To preserve the privacy, Blowfish and Honey Encryption algorithm is used. This research aims to make the current messaging framework more secure, resilient and effective by the application of symmetric-key algorithm amalgamated with Honey key encryption. The performance of the proposed algorithm evaluated and found satisfactory results. -
An Upper Bound for Sorting with LRE
Sai Satwik Kuppili, Bhadrachalam Chitturi, Venkata Vyshnavi Ravella, C. K. Phani DattaThe chapter explores the problem of sorting permutations using a specific set of operations: LeftRotate, RightRotate, and Exchange (LRE). It defines the LRE operation and its generators, and sets out to find an upper bound on the number of moves required to sort a permutation. The study builds on previous research on similar operations and introduces two algorithms, LRE and LRE1, to achieve this goal. The chapter provides a detailed analysis of the algorithms, including lemmas and theorems proving their correctness and efficiency. Additionally, it presents an exhaustive search algorithm to compute the minimum number of moves for sorting permutations, with results shown for specific values of n. The chapter concludes by highlighting the significance of the findings and suggesting future research directions.AI Generated
This summary of the content was generated with the help of AI.
AbstractA permutation \(\pi \) over alphabet \(\varSigma = {1,2,3,\ldots ,n}\), is a sequence where every element x in \(\varSigma \) occurs exactly once. \(S_n\) is the symmetric group consisting of all permutations of length n defined over \(\varSigma \). \(I_n\) = \((1, 2, 3,\ldots , n)\) and \(R_n =(n, n-1, n-2,\ldots , 2, 1)\) are identity (i.e. sorted) and reverse permutations respectively. An operation, that we call as an LRE operation, has been defined in OEIS with identity A186752. This operation is constituted by three generators: left-rotation, right-rotation and transposition (1,2). We call transposition (1,2) that swaps the two leftmost elements as Exchange. The minimum number of moves required to transform \(R_n\) into \(I_n\) with LRE operation are known for \(n \le 11\) as listed in OEIS with sequence number A186752. For this problem no upper bound is known. OEIS sequence A186783 gives the conjectured diameter of the symmetric group \(S_n\) when generated by LRE operations [1]. The contributions of this article are: (a) The first non-trivial upper bound for the number of moves required to sort \(R_n\) with LRE; (b) a tighter upper bound for the number of moves required to sort \(R_n\) with LRE; and (c) the minimum number of moves required to sort \(R_{10}\) and \(R_{11}\) have been computed. Here we are computing an upper bound of the diameter of Cayley graph generated by LRE operation. Cayley graphs are employed in computer interconnection networks to model efficient parallel architectures. The diameter of the network corresponds to the maximum delay in the network. -
Programmable Joint Computing Filter for Low-Power and High-Performance Applications
Abhineet Bawa, Rama Kanta Choudhury, Chandra Kanta Samal, Navneet YadavThis chapter delves into the design and implementation of a Programmable Joint Computing Filter for low-power and high-performance applications. It addresses the growing demand for high-performance, low-power consuming filters in digital media and multimedia. The filter is designed using FPGA technology, with a focus on optimizing the computationally intensive operations in digital signal processing (DSP) through a convolution operation. The proposed architecture introduces a programmable joint accumulator (PJA) that minimizes the number of additions and multiplications, thereby reducing power consumption. A new carry-select adder is introduced, which is more efficient than traditional architectures. The implementation on an FPGA using VHDL is presented, showcasing improved power performance and reduced hardware requirements. The chapter also explores the potential of the proposed architecture in various applications, including DSPs, image processing, communication devices, and audio-video processing.AI Generated
This summary of the content was generated with the help of AI.
AbstractA high-performance programmable joint computing for low power and high-performance filter (PJA) is presented in this paper. It emphasizes on low power and high efficiency, which is reliable for filter operation. The architecture is designed based on CSHM which can be implemented effectively in vector-scalar products at the circuit level. The products of the premultiplier are shared with all A&As, which assist significantly in the performance of the system. A programmable digital10-tap PJA filter, which accepts the input signal and values of coefficients of up to 17 bits (signed), is designed using VHDL and implemented on the XilinxSpartan-7 XC7S100FGGA676FPGA. It contains a total of 64000 LUT (Look-Up Table) elements and is based on 28 nm HKMG (High K metal Gate) transistor. The implementation was done using Xilinx Vivado 2019.2, and the power is measured using Xilinx Power Analyzer. -
Novel Design Approach for Optimal Execution Plan and Strategy for Query Execution
Rajendra D. Gawali, Subhash K. ShindeThis chapter delves into the critical issue of query optimization in database systems, highlighting the importance of finding optimal execution plans to minimize costs. It begins with a literature survey of various query optimization techniques over the years, emphasizing the role of selectivity and cardinality estimation in reducing execution time. The proposed architecture aims to bypass the optimization phase by reusing previously optimized plans, employing feature extraction and similarity detection techniques to identify similar query instances. The experimental section demonstrates the application of these techniques on a sample dataset, showcasing the potential for significant performance improvements. The conclusion emphasizes the potential of machine learning and deep learning techniques to further enhance query optimizer performance.AI Generated
This summary of the content was generated with the help of AI.
AbstractQuery optimization is a challenging task for database management researchers. After parsing of queries during query processing, in query optimization step, various query execution plans are generated. The job of query optimizer is to propose an optimal plan that can evaluate the given relational expression at a reasonably lower cost. For every new input query instance, generating multiple execution plans and identifying an efficient optimal plan amongst is always challenging in terms of consumption of resources and costs associated with optimization. As the number of plans increases, it can take longer to find a good plan. Thus, to make query optimization practical and efficient, reusing the existing execution plans will provide the ideal solution for the new instances of equivalent old queries.In the paper, a novel design approach for execution of parametric query has been proposed, where query may have generic (reusable) and specific parameters. The heuristic transformations and query tree representations help to find the best plan among of all possible plans. This best plan is compared with plans stored in plan cache to find a equivalent generic plan. Feature extraction and similarity detection techniques are used to compare cached plan for reuse. If no generic plan is found, then by using dynamic programming heuristic search algorithm, cost of plan is computed before optimization and execution of query instance.
-
-
New Approaches in Software Engineering
-
Frontmatter
-
An Iterative and Incremental Approach to Address Regulatory Compliance Concerns in Requirements Engineering
Deepti Balaji Raykar, L. T. JayPrakash, K. V. DineshaThe chapter introduces an innovative approach to incorporate regulatory compliance into the requirements engineering phase of the Software Development Life Cycle (SDLC). It proposes a regulatory module that translates regulatory knowledge into logic and English text within the Software Requirements Specification (SRS). The module consists of five components: Identifying regulatory documents, Extracting regulatory requirements, Resolving conflicts, Requirement specification, and Mapping document. The authors advocate an iterative and incremental approach to generate the required outputs, including the SRS regulatory version and mapping document. The chapter also includes a proof of concept example using an SRS for an income tax calculation application in India, illustrating the practical application of the proposed solution. The work aims to limit the involvement of regulatory experts to the initial phase, ensuring that subsequent phases of the SDLC can proceed without continuous expert intervention.AI Generated
This summary of the content was generated with the help of AI.
AbstractWe propose to address regulatory compliance concerns for a software system in a module (regulatory module) in the requirements engineering phase. The inputs are the SRS document and an initial set of Regulatory Documents. The outputs of the regulatory module are the ‘SRS regulatory version’ and ‘mapping document’. SRS regulatory version will have regulatory requirements expressed as a set of logic and/or plain English text weaved into appropriate parts of the SRS. Mapping Document is used to maintain the SRS regulatory version whenever Regulatory requirements or SRS changes. Regulatory expert’s involvement is restricted only to the activities of the regulatory module in SDLC. The regulatory module contains components in it and generates outputs through a process workflow. The various components in the regulatory module are: Identifying Regulatory Documents, Extracting Regulatory Requirements, Resolving Conflicts, Requirement Specification, Mapping Document. We identify generic interface specifications for these components through an extensive literature survey. These generic interfaces are valid for any SRS and any Regulatory body. We start with the generic version of the components as the initial version of the components. The objective is to get a specific version applicable for a specific regulatory body and a specific SRS along with SRS regulatory version and mapping document. An incremental and iterative solution strategy is outlined to achieve this. We have worked on a proof of concept of the solution strategy for a use case of an SRS as input at the regulatory module level. In literature, the study is done at component levels and not at the regulatory module level. -
State Space Modeling of Earned Value Method for Iterative Enhancement Based Traditional Software Projects Tracking
Manoj Kumar Tyagi, Ajay Sikandar, Dheerendra Kumar Tyagi, Durgesh Kumar, Prashant Singh, Srinivasan Munisamy, L. S. S. ReddyThis chapter delves into the application of state space modeling to enhance the Earned Value Method (EVM) for tracking iterative software projects. Traditional tracking techniques often fall short in reliability, especially when dealing with partially completed tasks. The author introduces a state-space approach that effectively integrates these tasks, improving the overall accuracy and reliability of project status reporting. The methodology involves a detailed explanation of the Earned Value Method and the state-space approach, followed by a simulation study that demonstrates the effectiveness of the proposed model. The results show significant improvements in project tracking, particularly in terms of schedule and cost performance indices. The chapter concludes with implications for software development organizations and suggests future research directions to further enhance the model.AI Generated
This summary of the content was generated with the help of AI.
AbstractThe earned value method (EVM) is used as a standard tool for tracking of traditional software projects in order to report their status during execution. However, the status reported by EVM is not reliable as it has a fundamental limitation of not considering the partially completed tasks for the evaluation of the project status. This paper suggests considering the partially completed tasks for the earned value or budgeted cost of work performed (EV) calculation for project status evaluation. The state-space approach is used for modeling EVM, which provides foundation for integrating issues related to partially completed tasks effectively and explicit representation of the project status during planning and execution. The EVM is applied for tracking a project, and the tracking results obtained using Monte-Carlo simulation, with or without considering the partially completed tasks during execution, are presented to show the project’s execution behavior in comparison to planned behavior. -
Agile Planning and Tracking of Software Projects Using the State-Space Approach
Manoj Kumar Tyagi, Dheerendra Kumar Tyagi, Lalit Kumar Tyagi, Neha Tyagi, Durgesh Kumar, Ajay SikandarThe chapter discusses the importance of efficient project tracking in software development, highlighting the overhead associated with heavy-weight tracking techniques. It introduces a light-weight tracking technique based on the state-space approach, which considers both completed and partially completed tasks without maintaining state information for partially completed tasks. The technique is designed to report project status effectively during planning and execution, enhancing project visibility while minimizing resource consumption. The chapter also includes a simulation study to validate the technique's practicality and effectiveness in real-world scenarios.AI Generated
This summary of the content was generated with the help of AI.
AbstractAll projects are required to be tracked and reported their status information to the management for controlling their execution, and have limited resources in terms of budget and time, so agile projects are. Their tracking is an overhead for the software projects as the tracking is a management activity, not a technical one, for the development of the software product. Various tracking techniques have been developed for tracking of agile projects. We have classified them as heavy-weight and light-weight, in terms of the information maintained, or not, about which phase of the project lifecycle a partially completed task belongs to. Light-weight techniques consider the tasks which have been completed but do not consider the partially completed tasks; while heavy-weight ones consider the tasks which have been completed as well as the state information about which phase of the project lifecycle the partially completed tasks belong to. Light-weight techniques are not effective due to not considering the partially completed tasks and heavy-weight ones are not utilizing the resources efficiently. In this paper, a light-weight tracking technique is designed using the state-space approach, which considers both the completed tasks and the partially completed tasks but without maintaining the state information for the partially completed tasks in order to tracking the projects effectively and utilizing the resources efficiently. The application of the technique, using Monte-Carlo simulation, is demonstrated by tracking a project which was carried out with the agile software development philosophy.
-
-
Backmatter
- Title
- Advanced Computing
- Editors
-
Deepak Garg
Kit Wong
Jagannathan Sarangapani
Suneet Kumar Gupta
- Copyright Year
- 2021
- Publisher
- Springer Singapore
- Electronic ISBN
- 978-981-16-0404-1
- Print ISBN
- 978-981-16-0403-4
- DOI
- https://doi.org/10.1007/978-981-16-0404-1
Accessibility information for this book is coming soon. We're working to make it available as quickly as possible. Thank you for your patience.
