Theses and Dissertations
Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/1
Browse
939 results
Search Results
Item Open Access Performance Assessment of Edge Traffic Distribution Routing Algorithm for Graphene Based Network-on-Chip(2021) Gupta, Yatin Kumar; Gohel, Bakul; Agrawal, YashNetwork-on-chip (NoC) has evolved as new paradigm for high-dense interconnect configurations in advanced integrated circuit designs. The increasing numbers of transistor cores with decrease in chip area is the leading motivation behind employment of NoC over SoC architectures. NoC can be addressed a move ahead from computation-centric to communication-centric design and the implementation of scalable communication structures. NoC provides re-configurable interconnections between the different cores in SoC design. It maximizes data transfer speed and reduction in wiring congestion. For further effective enhancing performance of NoCs, it is investigated that incorporation of graphene material can be good for realizing interconnects. As the graphene has the remarkable physical properties it is one of the most important emerging research material for not only the front-end but also for the back-end devices. In this work, edge traffic distribution (ETD) algorithm is explored along with magnificent graphene based interconnects for NoC design. Performance parameters considered are delay, power, energy, and throughput. It is investigated that the ETD routing algorithm leads to reduced delay, higher throughput, and smaller packet loss. Further, it is also analyzed that if the copper based router-to-router link of a mesh based NoC is replaced by a grahene based link then it leads to smaller energy consumption whenever there is a flit transfer from one router to the other. The assessment of NoC structures has been performed using Noxim and SPICE electronic design automation tools.Item Open Access Adversarial Defense Using Partial Pseudorandom Encryption(2021) Kalgutkar, Amruta; Joshi, M. V.Machine Learning models like Deep neural networks are vulnerable to adversarial attacks. Carefully crafted adversarial examples force a learned classifier to misclassify the input which can be correctly classified by a human observer. In this thesis, we present a novel approach for defense against such Adversarial attacks. We train and test the model on transformed images in black-box and gray-box scenarios. Here, we propose a transformation technique that partially encrypts every image before training and testing using the Rivest–Shamir–Adleman (RSA) , an asymmetric-key encryption algorithm for visual encryption. The internal structure of the system and the keys generated by RSA are secret. We encrypt only those pixels which are generated by a pseudorandom number generator with a pre-decided secret seed. The images encrypted with such transformation are extremely difficult to decrypt and to launch adaptive adversarial attacks or transferability attacks which makes this visual defense technique against adversarial attack robust. As the field of Adversarial machine learning (AML) is still under study, researchers have not attempted such an approach of training the model on encrypted images for robust learning. State-of-the-art defense techniques are effective but they are computationally expensive and still will not guarantee total security. This idea of partial encryption maintains features and asymmetric key encryption makes it difficult for adversary to guess encryption parameters. This makes the technique novel and hence out-performs state-of-the-art defense techniques.Item Open Access Phase Based Methods for Various Speech Applications(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Pusuluri, Aditya; Patil, Hemant A.Vocal communication plays a fundamental role in human interaction and expression.Right from the first cry to adult speech, the signal conveys information aboutthe well-being of the individual. Lack of coordination between the speech musclesand the brain leads to voice pathologies. Some pathologies related to infants areAsphyxia, Sudden Death Syndrome (SIDS), etc. The other voice pathologies thataffect the speech production systems are dysarthria, cerebral palsy, and parkinson�sdisease.Dysarthria, a neurological motor speech disorder, is characterized by impairedspeech intelligibility that can vary across severity-levels. This works focuses onexploring the importance of Modified Group Delay Cepstral Coefficients (MDGCC)-based features in capturing the distinctive acoustic characteristics associated withdysarthric severity-level classification, particularly for irregularities in speech.Convolutional Neural Network (CNN) and traditional Gaussian Mixture Model(GMM) are used as the classification models in this study. MGDCC is comparedwith state-of-the-art magnitude-based features, namely, Mel Frequency CepstralCoefficients (MFCC) and Linear Frequency Cepstral Coefficients (LFCC). In addition,this work also analyzed the noise robustness of MGDCC. To that effect,experiments were performed on various noise types and SNR levels, where thephenomenal performance of MGDCC over other feature sets was reported. Further,this study also analyses the cross-database scenarios for dysarthric severitylevelclassification. Analysis of Voice onset Time (VOT) and experiments wereperformed using MGDCC to detect dysarthric speech against normal speech. Further,the performance of MGDCC was then compared with baseline features usingprecision, recall, and F-1 score and finally, the latency period was analysed forpractical deployment of the system.This work also explores the application of phase-based features on the emotionrecognition task and pop noise detection. As technological advancementsprogress, dependence on machines is inevitable. Therefore, to facilitate effectiveinteraction between humans and machines, it has become crucial to develop proficienttechniques for Speech Emotion Recognition (SER). The MGDCC featureset is compared against MFCC and LFCC features using a CNN classifier and theLeave One Speaker Out technique. Furthermore, due to the ability of MGDCCto capture the information in low-frequency regions and due to the fact that popnoise occurs at lower frequencies, the application of phase-based features on voiceliveness detection is performed. The results are obtained from a CNN classifierusing the 5-Fold cross-validation metric and are compared against MFCC andLFCC feature sets.This work proposed the time averaging-based features in order to understandthe amount of information being captured across the temporal axis as there wouldnot be many temporal variations in a cry signal. The research conducted in thisstudy utilizes a 10-fold stratified cross-validation approach with machine learningclassifiers, specifically Support Vector Machine (SVM), K-Nearest Neighbor(KNN), and Random Forest (RF). This work also showcased CQT-based Constant-Q Harmonic coefficient (CQHC) and Constant-Q Pitch coefficients (CQPC) for theclassification of infant cry into normal and pathology as an effective representationof the spectral and pitch components of a spectrum together is not achievedleaving scope for improvement. The results are compared by considering theMFCC, LFCC, and CQCC feature sets as the baseline features using machinelearning and deep learning classifiers, such as Convolutional Neural Networks(CNN), Gaussian Mixture Models (GMM), and Support Vector Machines (SVM)with 5-Fold cross-validation accuracy as the metric.Item Open Access A Spectrally Efficient MIMO System with Sparse Matrix Precoding(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Yadav, Prabhanshu; Vasavada, YashThis thesis proposes a novel technique of sparse matrix-based precoding at thetransmitter of a Multiple Input Multiple Output (MIMO) system. We proposedtwo sparse matrix precoded MIMO systems. Our first proposal improves thespectral efficiency beyond the existing spectral efficiency of Precoding-aided SpatialModulation (PSM-MIMO) system. Our second proposal increases spectralefficiency compared to an existing MIMO system.Both proposals use a two-stage precoding approach in which the conventionalzero-forcing (ZF) MIMO precoder, which inverts the matrix MIMO channel, iscombined with a sparse matrix precoding. With the conventional ZF precoder, thedegrees of freedom (DoF) available at the transmitter equals the number of antennasat the receiver. By adding another layer of precoding using a sparse matrix,we increase the DoF at the transmitter, thereby facilitating an increase in spectralefficiency. We demonstrate proof of the concept (PoC) by simulation-driven experiments.Our PoC is based on the ML (Maximum Likelihood) detection at thereceiver. ML detection has quite high complexity. We propose a belief propagationalgorithm at the receiver which is more practical to implement in a real-worldsystem. The belief propagation algorithm leverages the sparseness of the precodingmatrix and has low computational complexity.Item Open Access Shadow Detection and Removal from video using Deep Learning(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Dodiya, Krutika; Khare, Manish; Gohel, BakulThe removal of shadow from images is crucial in computer vision as it can enhancethe interpretability and visual quality of images. This research work proposesa cascade U-Net architecture for the shadow removal, consisting of twostages of U-Net Architecture. In the first stage, a U-Net is trained using theshadow images and their corresponding ground truth to predict the shadow freeimages. The second stage uses the predicted shadow free images and groundtruth as input to another U-Net, which further refines the shadow removal results.This cascade U-Net architecture enables the model to learn and refine theshadow removal progressively, leveraging both the initial predictions and groundtruth.Experimental evaluations on benchmark datasets demonstrate that our approachachieves notably good performance in both qualitative and quantitative evaluations.By using both objective metrics such as Structural Similarity Index(SSIM),and Root mean Square Error (RMSE), and subjective evaluations where humanobservers rate the quality of the shadow removal results, our approach was foundto outperform other state-of-the-art methods. Overall, our proposed cascade UNetarchitecture offers a promising solution for the shadow removal that canimprove image quality and interpretabilityItem Open Access Analysing User Reviews for Evaluating Game Playability of Mobile Gaming Apps(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Thakar, Swapnil; Tiwari, SaurabhThe playability of a game depends on the players� experience in terms of functionality,usability, and satisfaction. Mobile gaming has recently evolved because ofthe availability of suitable hardware, configurable mobile devices, and the abilityto download games from the Android and iOS platforms. Most online gamingstores allow customers to submit their reviews about gameplay, issues, and functionalitiespublicly. Game developers can better grasp such consumer issues byexamining player feedback and increasing how well-liked a game is among players.We have mapped the playability of S�nchez�s model with Schwartz�s theoryof human values and analyzed 20,346 user/player reviews from the top 15 gameapps in the Google Play Store. We have also created a labelled dataset of eachplayability category of S�nchez�s model. Finally, we applied a machine learningmodel to support the automatic classification of a review to a specific playabilitycategory violation. Our analysis shows that 30% of the reviews show human valuesviolations, consequently affecting game playability. We found that Socialism isthe most violated and Emotion is the least violated value category. We also foundthat only 18% of the user reviews received responses from the game app developersfor the value violations. Using fine-grained feature extraction, we found thetop 42 functionalities, issues, and concerns for the violations. The analysis resultsof our study give developers a foundation for creating apps that consider users�values for ensuring better playability of mobile game apps.Item Open Access Privacy-Preserving Iris Based Authentication System(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Agrawal, Radha; Singh, Priyanka; Joshi, Manjunath V.Biometric authentication systems have gained immense popularity due to theirability to provide secure and convenient authentication. However, the leakageof sensitive biometric data can compromise an individual�s privacy and security.To address this issue, a privacy-preserving biometric authentication system basedon iris data is proposed in this paper. The framework exploits the homomorphicproperties to process encrypted data, thereby ensuring the privacy of sensitivedata, even while using the services of third-party cloud service providers (CSPs).In the initial stage of the experiment, we encrypt the data, and comparison wasdone by using hamming distance, but after completion of the first experiment,we realized that data can be morphed through an insecure channel by using multipleattacks to overcome this we have proposed framework were morphing isperformed on the iris data by using a man-in-the-middle attack. Two iris identificationAlgorithms are proposed, with a success rate of over 60% and a false matchrate of 5%, and are vulnerable to morph attacks. We also examine how comparablethe original and morphed iris images must be. Using original images, we presentour findings for morphing iris detection. The proposed privacy-preserving biometricauthentication system offers a robust framework that minimizes time complexitycompared to other state-of-the-art approaches. This framework ensuresthe privacy of sensitive data and provides a secure biometric authentication system.Item Open Access On the Robustness of Federated Learning towards Various Attacks(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Yagnik, Shrey Devenkumar; Singh, Priyanka; Joshi, Manjunath V.A study based on Federated Learning (FL), i.e., a kind of decentralized learningthat consists of local training among the clients, and the central server returnsthe federated average. Deep learning models have been used in numeroussecurity-critical settings since they have performed well on various tasks. Here,we study different kinds of attacks on FL. FL has become a popular distributedtraining method because it enables users to work with large datasets without sharingthem. Once the model has been trained using data on local devices, only theupdated model parameters are sent to the central server. The FL approach is distributed.Thus, someone could launch an attack to influence the model�s behavior.In this work, we conducted the study for a Backdoor attack, a black-box attackwhere we added a few poisonous instances to check the model�s behavior duringtest time. Also, we conducted three types of White-Box attacks, i.e., Fast GradientSign Method (FGSM), Carlini-Wagner (CW), and DeepFool. We conductedvarious experiments using the standard CIFAR10 dataset to alter the model�s behavior.We used ResNet20 and DenseNet as the Deep Neural Networks. Wefound some adversarial samples upon which the required perturbation is addedto fool the model upon giving the misclassifications. This decentralized approachto training can make it more difficult for attackers to access the training data, butit can also introduce new vulnerabilities that attackers can exploit. We found outthat the expected behavior of the model could be compromised without havingmuch difference in the training accuracy.Item Open Access Semantic Segmentation Based Object Detection for Autonomous Driving(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Prajapati, Harsh; Maiti, Tapas KumarThis research focuses on solving the autonomous driving problem which is necessaryto fulfill the increasing demand of autonomous systems in today�s world.The key aspect in addressing this challenge is the real-time identification andrecognition of objects within the driving environment. To accomplish this, weemploy the semantic segmentation technique, integrating computer vision, machinelearning, deep learning, the PyTorch framework, image processing, and therobot operating system (ROS). Our approach involves creating an experimentalsetup using an edge device, specifically a Raspberry Pi, in conjunction with theROS framework. By deploying a deep learning model on the edge device, we aimto build a robust and efficient autonomous system that can accurately identifyand recognize objects in real time.Item Open Access Image Processing Using Digital Programming on FPGA(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Kachchhi, Hardi; Agrawal, Yash; Khare, ManishImage processing is a way to transform an image into digital form and after thatperform some operations on it that helps to improve images for human interpretationand extract useful information from it. It is essential for a wide range ofapplications. It allows for enhancing and restoring images, extracting featuresfor object recognition, compressing images for efficient storage and transmission,analyzing images for computer vision tasks, enabling medical diagnostics andtreatment, and interpreting data from remote sensing.Field Programmable Gate Array (FPGA) is preferred for image processing dueto their parallel processing capabilities, reconfigurability, low latency, energy efficiency,pipelining support, customization options, real-time processing capabilities,and ease of integration. These advantages make FPGAs a powerful tool forimplementing high-performance and efficient image processing solutions acrossvarious applications.To implement various filters in Image processing, we have developed a methodthat performs various edge detection techniques using FPGAs and displaying theimage on the monitor through Video Graphics Array (VGA)Controller. Edge detectionfilters and blurring filters are an indispensable part of Image processing invarious fields due to their ability to extract information, enhance visual quality,and enable decision-making based on visual data .