http://repository.iiitd.edu.in/xmlui/handle/123456789/713 2026-04-11T02:46:27Z http://repository.iiitd.edu.in/xmlui/handle/123456789/796 Estimation and concealment of forensic multimedia signatures Mehrish, Ambuj; Subramanyam, A V (Advisor) The remarkable evolution of digital imaging techniques, processing and sharing in the past decades has spurred the penetration of multimedia into our lives. Unaccountable and ubiquitous use of multimedia brings severe issues and challenges about its origin and veracity. For instance, due to the growth of social media and instant messaging applications, the circulation of tampered content has become an unavoidable reality. The waning credibility of digital content has also lead to unfavorable consequences in terms of political, economic and social issues. Therefore, to address the source of origin and processing history-related issues of multimedia content, the scientific community has focused its attention on digital multimedia forensics techniques. Two of our major contributions belong to this class of forensics algorithms. On the other hand, to counter forensics algorithms, a parallel area of adversarial signal processing has also gained a lot of momentum. These counter-forensics algorithms are designed to hide the fingerprints left behind by image processing operations, thereby degrading the performance of forensics detectors. Our third major contribution belongs to this class of counter forensics techniques. In our first contribution, we address the problem of establishing the link between a given image to its source camera device. This problem is called Source Camera Identification. In order to achieve this, the photo response non-uniformity (PRNU) characteristic of the sensor is exploited. However, the existing techniques suffer from the problem of noise induced during in-camera image processing. This noise can suppress the PRNU leading to poor camera identification performance. To this end, we propose a novel algorithm for robust estimation of PRNU from probabilistically obtained raw data. Since not all cameras provide raw data as their output, we compute raw data from the JPEG output using a probabilistic color de-rendering procedure. The estimated raw data is modeled as a Poisson process, and Maximum Likelihood Estimation is used for PRNU estimation. We also extend the estimated PRNU for tampering detection. The extensive experimental analysis performed on thousands of patches from various cameras reveal state-of-the-art performance. We also demonstrate the robustness of estimated PRNU by accurate tampering localization. In our second contribution, we analyse the counter forensic algorithm for Contrast Enhancement, which is a common post-processing step in image tampering. The existing algorithms only consider the spatial domain. In our work, we consider changes in both spatial and DCT domain into account and obtain enhanced images such that the statistical properties are similar to natural images. Unlike the conventional techniques which lead to artifacts that can be captured by forensics detectors, the proposed algorithm suppresses the detectable artefacts. In our experiments, we demonstrate a significant performance degradation for deep learning as well as steganalysis-DCT feature based detectors. We also compute the popular image quality assessment metrics and show that the proposed model generates better visual quality images compared to the existing counter forensics techniques. Nowadays cameras are also used in personal and commercial vehicles which pose a different problem of linking a given video to the vehicle in which the camera was mounted. This is useful for various applications, for example, insurance companies can authenticate the origin of video before processing the claim. In a different scenario of illegitimate video upload on the web, the video can be traced back to the car it originated from. To this end, we state our third contribution, in which we introduce the new area of multimedia vehicle forensics. We propose an algorithm for linking a dash-cam video to a specific car. Inspired by human gait bio-metrics, we observe that the subtle motion pattern of every vehicle can serve as its unique signature. We extract motion blur from dash-cam videos, which encode the motion pattern of the car. Experimental results on hours of dash-cam videos of several cars show the effectiveness of our approach. We further investigate the adversarial process of forging the signature of the vehicle and propose a forensics method to detect such forgery. 2019-12-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/719 Deep dictionary learning Singhal, Vanika; Majumdar, Angshul (Advisor) Currently there are three basic frameworks in deep learning - stacked autoencoders (SAE), deep belief network (DBN) and convolutional neural network (CNN); SAE and DBN can be applied to arbitrary inputs but CNN can only be applied to natural signals having local correlations (speech, image, ECG, EEG etc.). I am working on developing a new framework for deep learning – deep dictionary learning (DDL). Just as SAE uses autoencoders as basic units and DBN uses restricted Boltzmann machines, DDL uses dictionaries as the basic unit. In lay man’s terms, DDL is formed by stacking one dictionary after another such that the output (features) from the shallower layer feeds into the next (deeper) layer as input. The initial work on DDL was a greedy sub-optimal solution, i.e. each of the layers were solved separately. My first work has been on proposing an optimal solution to jointly learn all the layers. This was a solution for unsupervised feature extraction using DDL. Later I worked on a supervised version of deep dictionary learning with a plug-and-play approach. The supervised version is general enough to perform classification, multi-label classification and regression. Proposed supervised deep dictionary learning for multi-label classification has been used for solving a practical problem of Non-Intrusive Load Monitoring (NILM). We also proposed a technique called deep blind compressed sensing which combines the analytic power of deep learning with reconstruction ability of compressed sensing. The objective of this work was to classify biomedical signals from their compressive measures. Next, we work on Siamese DDL networks. These are usually required for verification problems in biometrics. We applied them on face verification problem in disguise detection and kinship verification. It can also be applied in a variety of other situations. For example in estimating dense depth from an image and sparse depth with a complete learning based approach. Or in problems arising in BP estimation from multiple sources (multi-channel ECG and PPG). Finally, we worked on deeply coupled dictionary learning. These networks are used to generate a linear mapping between samples of different domains. Existing approaches learn a linear map between the source and target domains. We propose to use deep dictionary learning to solve for complex mapping. Such networks can be used in applications like image denoising, image super-resolution, image reconstruction etc. 2019-08-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/714 End-to-end performance analysis of MIMO-OFDM based wireless transceivers in the presence of hardware impairments Aggarwal, Parag; Bohara, Vivek Ashok (Advisor) Wireless communication systems based on multi-antenna multi-carrier modulation techniques such as multi-input-multi-output (MIMO) orthogonal frequency division multiplexing (OFDM) have become the state-of-art method for a broadband data transfer. However, it is still not enough as a standalone to satisfy the ever increasing demand for high data rate transmission. To fulfill this demand, new features such as carrier aggregation (CA) and advanced multi-antenna techniques have been proposed by 3rd Generation Partnership Project (3GPP) in Long Term Evolution-Advanced (LTE-A). On exploiting the advantages of OFDM with CA and advanced multi-antenna techniques, the efficiency of the cellular networks can be significantly improved. However, OFDM signal suffers from high peak-to-average-power ratio (PAPR) problem, and high power amplifiers (HPAs) induce nonlinear distortions to such a high PAPR signal. Furthermore, another hardware impairment, i.e., phase noise caused by non-ideal oscillators significantly degrades the efficiency and performance of the system. These impairments severely limit the gains provided by the OFDM-based systems as compared to the linear systems. Hence, it is imperative to take into account the impact of nonlinear HPA and non-ideal oscillator while evaluating the performance of the OFDM-based systems. As a consequence, this dissertation is mainly focused on analyzing the performance of OFDM based CA and multi-user (MU) MIMO systems in the presence of hardware impairments. The first part of this dissertation analyzes the impact of nonlinear HPA on multi-band CA-MIMO-OFDM system. The nonlinear behavior of the HPA is modeled by a multi-band generalized memory polynomial (MB-GMP) model. It is shown that the received symbol after down-conversion can be canonically decomposed into complex attenuation factor and additive nonlinear noise. The generalized mathematical expressions of complex attenuation factor and variance of nonlinear noise for any number of aggregated bands with any nonlinearity order and any memory depth of HPA are derived. From the derived expressions, an analytical methodology is proposed to obtain the received signal-to-distortion-plus-noise ratio (SDNR), symbol error rate (SER) and error vector magnitude (EVM) of the nonlinear multi-band CAvii MIMO-OFDM system. The proposed work also provides valuable insights on the impact of number of aggregated carriers on the error performance of a nonlinear CA-MIMO-OFDM system. In the second part, the performance of CA dual-band (DB) MU-MIMO-OFDM system in the presence of nonlinear HPA is investigated. The HPA nonlinearity is modeled by a two-dimensional GMP model. A transmit preprocessing technique is employed to mitigate the effects of inter-user interference. The generalized expression for a signal-to-interference-plusnoise ratio (SINR) for nonlinear DB MU-MIMO-OFDM system is derived and further utilized to obtain an analytical framework to evaluate the performance of the aforementioned system in terms of SER and average capacity. Furthermore, it is shown that the nonlinear interference is a function of inter-modulation and cross-modulation products from every user and increases as the number of users increases, thus significantly deteriorating the performance of the DB MU-MIMO-OFDM system. A theoretical framework to study the joint impact of non-ideal oscillators and nonlinear HPAs on the downlink MU-MIMO-OFDM system is proposed in the next part of this dissertation. It is shown that the received data symbol affected by phase noise and HPA nonlinearity can be canonically decomposed into following: 1) the desired signal multiplied by an attenuation factor and common phase error (CPE), 2) inter-channel interference (ICI), and 3) additive nonlinear distortion noise which is uncorrelated with the desired signal. The mathematical expression of an instantaneous signal-to-distortion-plus-interference-plus-noise ratio (SDINR) for MU-MIMO-OFDM system impaired with phase noise and nonlinear HPA is derived and then used to obtain the closed-form expression for both SER and overall capacity. In the end, this dissertation presents an end-to-end analytical framework to evaluate the performance of the cascaded digital predistorter (DPD) and HPA structure on the MIMO-OFDM system. A joint nonlinear polynomial (JNP) model is proposed to represent the characteristics of the cascaded DPD+HPA architecture. It is shown that the received symbol after the MIMOOFDM demodulation consists of residual noise and a multiplicative factor. By modeling the residual noise as a complex Gaussian process, an analytical methodology is adopted to obtain the received SDNR of a MIMO-OFDM system in the presence of both DPD and HPA. Further by utilizing the expression of SDNR, a closed-form expression of SER is also derived for a multi-path Rayleigh fading channel. It is shown that the SER of the MIMO-OFDM system in the presence of cascaded DPD+HPA architecture significantly improves and approaches to a linear MIMO-OFDM system. The proposed frameworks in this dissertation can easily be utilized in various wireless standards and will be useful for a communication engineer to design a link budget for MIMO-OFDM based wireless systems without performing extensive simulations or tedious experiments. 2019-05-01T00:00:00Z