In the rapidly evolving digital world, network security is a critical concern. Traditional security measures often fail to detect unknown attacks, making anomaly-based Network Intrusion Detection Systems (NIDS) using Machine Learning (ML) vital. However, these systems face challenges such as computational complexity and misclassification errors. This paper presents ZW-IDS, an innovative approach to enhance anomaly-based NIDS performance. We propose a two-layer classification NIDS integrating zero-watermarking with data provenance and ML. The first layer uses Support Vector Machines (SVM) with ensemble learning model for feature selection. The second layer generates unique zero-watermarks for each data packet using data provenance information. This approach aims to reduce false alarms, improve computational efficiency, and boost NIDS classification performance. We evaluate ZW-IDS using the CICIDS2017 dataset and compare its performance with other multi-method ML and Deep Learning (DL) solutions.

This paper presents an improved method for estimating the accuracy of a model based on images intended for prediction, enhancing the standard Detection of Classifier Inconsistencies (DCI) method. The conventional DCI method typically requires a large enough set of images from the same source to provide accurate estimations, which limits its practicality. Our enhanced approach overcomes this limitation by generating a set of images from a single original image, thereby enabling the application of the standard DCI method without requiring more than one target image. This method ensures that the generated images maintain the statistical properties of the original, preserving any embedded steganographic messages, through the use of non-destructive image manipulations such as flips, rotations, and shifts. Experimental results demonstrate that our method produces results comparable to those of the traditional DCI method, effectively estimating model accuracy with as few as 32 generated images. The robustness of our approach is also confirmed in challenging scenarios involving cover source mismatch (CSM), making it a viable solution for real-world applications.

The widespread use of digital platforms, prioritising content based on engagement metrics and rewarding content creators accordingly, has contributed to the proliferation of disinformation and its far-reaching social and political impact. In addition, digital platforms often operate as black boxes, concealing their decision-making processes from users and prioritizing investor interests over ethical and social considerations. Consequently, this has contributed to the erosion of general trust in verification systems. To mitigate this issue, our project proposes a two-stage verification system. The first stage allows media industries to watermark their image and video content. The second stage involves implementing a machine-learning-based manipulation detection system for suspicious content. We present findings from an international user experience study, where potential online news consumers verified the authenticity of images on a prototype version of our system. In this paper, we reflect on critical issues of explainability addressed by participants in our user study and how we addressed this issue in the platform’s design.

Abstract In this paper, an analysis on hybrid Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) for image watermarking is carried out to investigate the effect of a deeper level of the SVD on imperceptibility and robustness to resist common signal processing and geometric attacks. For this purpose, we have designed two hybrid watermarking schemes, the first one with DWT and first level of SVD, whereas, in the second scheme, the same design is employed with a second level of SVD. In this experiment, a comprehensive analysis is performed on the two designed schemes and the effect of robustness and imperceptibility is compared in the first and second levels of SVD in each DWT sub-band. Having analyzed more than 100 medical and non-medical images in standard datasets and real medical samples of patients, the experimental outcomes show a remarkable increase in both imperceptibility and robustness in the second level of SVD, in comparison to the first level. In addition, the achieved result shows that the SVD2 scheme offers the highest imperceptibility in the LL sub-band (more than 60 dB on average PSNR), with satisfactory robustness against noise attacks, but less persistence in some geometric attacks such as cropping. For the HH sub-band, strong robustness against all types of tested of attacks is obtained, though its imperceptibility is slightly lower than the achieved PSNR in the LL sub-band. In HH sub-band, an average growth of 5 dB in PSNR and 2% in NC can be observed from the second level of SVD in comparison to the first level. These results make SVD2 a good candidate for content protection, especially for medical images.

Privacy protection for personal data and fairness in automated decisions are fundamental requirements for responsible Machine Learning. Both may be enforced through data preprocessing and share a common target: data should remain useful for a task, while becoming uninformative of the sensitive information. The intrinsic connection between privacy and fairness implies that modifications performed to guarantee one of these goals, may have an effect on the other, e.g., hiding a sensitive attribute from a classification algorithm might prevent a biased decision rule having such attribute as a criterion. This work resides at the intersection of algorithmic fairness and privacy. We show how the two goals are compatible, and may be simultaneously achieved, with a small loss in predictive performance. Our results are competitive with both state-of-the-art fairness correcting algorithms and hybrid privacy-fairness methods. Experiments were performed on three widely used benchmark datasets: Adult Income , COMPAS, and German Credit .

In this paper, we propose a robust method for detecting guilty actors in image steganography while effectively addressing the Cover Source Mismatch (CSM) problem, which arises when classifying images from one source using a classifier trained on images from another source. Designed for an actor-based scenario, our method combines the use of Detection of Classifier Inconsistencies (DCI) prediction with EfficientNet neural networks for feature extraction, and a Gradient Boosting Machine for the final classification. The proposed approach successfully determines whether an actor is innocent or guilty, or if they should be discarded due to excessive CSM. We show that the method remains reliable even in scenarios with high CSM, consistently achieving accuracy above 80% and outperforming the baseline method. This novel approach contributes to the field of steganalysis by offering a practical and efficient solution for handling CSM and detecting guilty actors in real-world applications.

Steganalysis is a collection of techniques used to detect whether secret information is embedded in a carrier using steganography. Most of the existing steganalytic methods are based on machine learning, which requires training a classifier with “laboratory” data. However, applying machine-learning classification to a new data source is challenging, since there is typically a mismatch between the training and the testing sets. In addition, other sources of uncertainty affect the steganlytic process, including the mismatch between the targeted and the actual steganographic algorithms, unknown parameters –such as the message length– and having a mixture of several algorithms and parameters, which would constitute a realistic scenario. This article presents subsequent embedding as a valuable strategy that can be incorporated into modern steganalysis. Although this solution has been applied in previous works, a theoretical basis for this strategy was missing. Here, we cover this research gap by introducing the “directionality” property of features concerning data embedding. Once a consistent theoretical framework sustains this strategy, new practical applications are also described and tested against standard steganography, moving steganalysis closer to real-world conditions.

Digital media have changed the classical model of mass media that considers the transmitter of a message and a passive receiver, to a model where users of the digital media can appropriate the contents, recreate, and circulate them. In this context, online social media are a suitable circuit for the distribution of fake news and the spread of disinformation. Particularly, photo and video editing tools and recent advances in artificial intelligence allow non-professionals to easily counterfeit multimedia documents and create deep fakes. To avoid the spread of disinformation, some online social media deploy methods to filter fake content. Although this can be an effective method, its centralized approach gives an enormous power to the manager of these services. Considering the above, this paper outlines the main principles and research approach of the ongoing DISSIMILAR project, which is focused on the detection of fake news on social media platforms using information hiding techniques, in particular, digital watermarking, combined with machine learning approaches.

The insecure growth of Internet-of-Things (IoT) can threaten its promising benefits to our daily life activities. Weak designs, low computational capabilities, and faulty protocol implementations are just a few examples that explain why IoT devices are nowadays highly prone to cyber-attacks. In this survey paper, we review approaches addressing this problem. We focus on machine learning-based solutions as a representative trend in the related literature. We survey and classify Machine Learning (ML)-based techniques that are suitable for the construction of Intrusion Detection Systems (IDS) for IoT. We contribute with a detailed classification of each approach based on our own taxonomy. Open issues and research challenges are also discussed and provided.

Real-time mobility data is useful for several applications such as planning transports in metropolitan areas or localizing services in towns. However, if such data is collected without any privacy protection it may reveal sensible locations and pose safety risks to an individual associated to it. Thus, mobility data must be anonymized preferably at the time of collection. In this paper, we consider the SwapMob algorithm that mitigates privacy risks by swapping partial trajectories. We formalize the concept of sufficient sanitizer and show that the SwapMob algorithm is a sufficient sanitizer for various statistical decision problems. That is, it preserves the aggregate information of the spatial database in the form of sufficient statistics and also provides privacy to the individuals. This may be used for personalized assistants taking advantage of users’ locations, so they can ensure user privacy while providing accurate response to the user requirements. We measure the privacy provided by SwapMob as the Adversary Information Gain, which measures the capability of an adversary to leverage his knowledge of exact data points to infer a larger segment of the sanitized trajectory. We test the utility of the data obtained after applying SwapMob sanitization in terms of Origin-Destination matrices, a fundamental tool in transportation modelling.

In this paper, a methodology to detect inconsistencies in classification-based image steganalysis is presented. The proposed approach uses two classifiers: the usual one, trained with a set formed by cover and stego images, and a second classifier trained with the set obtained after embedding additional random messages into theoriginal training set. When the decisions of these two classifiers are not consistent, we know that the prediction is not reliable. The number of inconsistencies in the predictions of a testing set may indicate that the classifier is not performing correctly in the testing scenario. This occurs, for example, in case of cover source mismatch,or when we are trying to detect a steganographic method that theclassifier is no capable of modelling accurately. We also show how the number of inconsistencies can be used to predict the reliability of the classifier (classification errors).

In this study, the authors present a new steganographic technique called random least significant bits of pitch and Fourier magnitude steganography (RLPFS). It is based on hiding a secret speech coded by mixed excitation linear prediction (MELP) speech coder in speech bitstream (cover signal), which is also encoded by MELP coder. First, the RLPFS leaks the hidden speech in the following modes: pitch‐based steganography, Fourier magnitude‐based steganography or both. These modes are selected randomly. Second, during transmission, the stego speech, the mode number, and the number of embedded bits would be transmitted either through a covert channel created in the transmission protocol or through the cover speech. In this work, the authors have dealt with the challenge of embedding a secret speech into a cover speech coded by a very low bit rate speech coder while maintaining a reasonable level of speech quality. They have shown that RLPFS was able to create hidden channels with maximum steganographic bandwidths up to 266.64 bit/s at the cost of a steganographic noise between 0.031 and 0.62 mean opinion score. Also, this study takes into account the security of the parameters, the synchronisation of the receiver to deal with a packet loss during transmission and the resistance of the proposed method against steganalysis.

Abstract Mobility data mining can improve decision making, from planning transports in metropolitan areas to localizing services in towns. However, unrestricted access to such data may reveal sensible locations and pose safety risks if the data is associated to a specific moving individual. This is one of the many reasons to consider trajectory anonymization. Some anonymization methods rely on grouping individual registers on a database and publishing summaries in such a way that individual information is protected inside the group. Other approaches consist of adding noise, such as differential privacy, in a way that the presence of an individual cannot be inferred from the data. In this paper, we present a perturbative anonymization method based on swapping segments for trajectory data (SwapMob). It preserves the aggregate information of the spatial database and at the same time, provides anonymity to the individuals. We have performed tests on a set of GPS trajectories of 10,357 taxis during the period of Feb. 2 to Feb. 8, 2008, within Beijing. We show that home addresses and POIs of specific individuals cannot be inferred after anonymizing them with SwapMob, and remark that the aggregate mobility data is preserved without changes, such as the average length of trajectories or the number of cars and their directions on any given zone at a specific time.

The increased use of location-aware devices, such as smartphones, generates a large amount of trajectory data. These data can be useful in several domains, like marketing, path modeling, localization of an epidemic focus, and so on. Nevertheless, since trajectory information contains personal mobility data, improper use or publication of trajectory data can threaten users’ privacy. It may reveal sensitive details like habits of behavior, religious beliefs, and sexual preferences. Therefore, many users might be unwilling to share their trajectory data without a previous anonymization process. Currently, several proposals to address this problem can be found in the literature. These solutions focus on anonymizing data before its publication, i.e., when they are already stored in the server database. Nevertheless, we argue that this approach gives the user no control about the information she shares. For this reason, we propose anonymizing data in the users’ mobile devices, before they are sent to a third party. This article extends our previous work which was, to the best of our knowledge, the first one to anonymize data at the client side, allowing users to select the amount and accuracy of shared data. In this article, we describe an improved version of the protocol, and we include the implementation together with an analysis of the results obtained after the simulation with real trajectory data.

Differential privacy is a popular privacy model within the research community because of the strong privacy guarantee it offers, namely that the presence or absence of any individual in a data set does not significantly influence the results of analyses on the data set. However, enforcing this strict guarantee in practice significantly distorts data and/or limits data uses, thus diminishing the analytical utility of the differentially private results. In an attempt to address this shortcoming, several relaxations of differential privacy have been proposed that trade off privacy guarantees for improved data utility. In this paper, we argue that the standard formalization of differential privacy is stricter than required by the intuitive privacy guarantee it seeks. In particular, the standard formalization requires indistinguishability of results between any pair of neighbor data sets, while indistinguishability between the actual data set and its neighbor data sets should be enough. This limits the data controller's ability to adjust the level of protection to the actual data, hence resulting in significant accuracy loss. In this respect, we propose individual differential privacy, an alternative differential privacy notion that offers the same privacy guarantees as standard differential privacy to individuals (even though not to groups of individuals). This new notion allows the data controller to adjust the distortion to the actual data set, which results in less distortion and more analytical accuracy. We propose several mechanisms to attain individual differential privacy and we compare the new notion against standard differential privacy in terms of the accuracy of the analytical results.

Recombined fingerprinting has been recently proposed as a scalable alternative to the traditional client-server model for anonymous fingerprinting. In recombined fingerprinting, the contents are distributed in P2P fashion from a few seed nodes to the rest of buyers. However, the solution of the previous works requires that a hash of the whole fingerprint is also constructed during distribution in such a way that two different codes must be used (one at segment level and one at hash level). This solution is impractical for two reasons: (1) the fingerprint's length is increased, and (2) the construction of the fingerprints through recombination must be supervised in order to obtain a valid hash-level codeword. This work contributes with a solution to this problem, consisting of embedding collusion-resistant codewords only at segment level and removing the need for hash-level encoding. The proposed solution not only results in shorter fingerprints, but also simplifies the distribution protocol since supervision is no longer required. The resulting system is shown to work for four different state-of-the-art collusion-resistant codes by means of thousands of simulations.

This paper presents a novel high-capacity audio watermarking system to embed data and extract them in a bit-exact manner by changing some of the magnitudes of the FFT spectrum. The key idea is to divide the FFT spectrum into short frames and change the magnitude of the selected FFT samples using Fibonacci numbers. Taking advantage of Fibonacci numbers, it is possible to change the frequency samples adaptively. In fact, the suggested technique guarantees and proves, mathematically, that the maximum change is less than 61% of the related FFT sample and the average error for each sample is 25%. Using the closest Fibonacci number to FFT magnitudes results in a robust and transparent technique. On top of very remarkable capacity, transparency and robustness, this scheme provides two parameters which facilitate the regulation of these properties. The experimental results show that the method has a high capacity (700 bps to 3 kbps), without significant perceptual distortion (ODG is about -1) and provides robustness against common audio signal processing such as echo, added noise, filtering, and MPEG compression (MP3). In addition to the experimental results, the fidelity of suggested system is proved mathematically.

Anonymous fingerprint has been suggested as a convenient solution for the legal distribution of multimedia contents with copyright protection whilst preserving the privacy of buyers, whose identities are only revealed in case of illegal re-distribution. However, most of the existing anonymous fingerprinting protocols are impractical for two main reasons: 1) the use of complex time-consuming protocols and/or homomorphic encryption of the content, and 2) a unicast approach for distribution that does not scale for a large number of buyers. This paper stems from a previous proposal of recombined fingerprints which overcomes some of these drawbacks. However, the recombined fingerprint approach requires a complex graph search for traitor tracing, which needs the participation of other buyers, and honest proxies in its P2P distribution scenario. This paper focuses on removing these disadvantages resulting in an efficient, scalable, privacy-preserving and P2P-based fingerprinting system.

This article presents a fast and low complexity audio watermarking system which is very suitable for mobile device applications. The key idea is to divide the FFT spectrum into short frames and change the magnitude of the selected FFT samples using Fibonacci numbers. The suggested technique guarantees and proves, mathematically, that the maximum change is less than 61% of the related FFT sample and the average error for each sample is 25%. Using the closest Fibonacci number to FFT magnitudes results in a robust and transparent technique. On top of very remarkable capacity, transparency and robustness, this scheme provides two parameters which facilitate the regulation of these properties. The experimental results show that the method has a high capacity (700 to 3 kbps), without significant perceptual distortion (ODG is about - 1) and provides robustness against common audio signal processing such as echo, added noise, filtering and MPEG compression (MP3). In addition to the experimental results the fidelity of suggested system is proved mathematically.

This article presents a novel high capacity audio watermarking system to embed data and extract them in a bit-exact manner by changing some of the magnitudes of the FFT spectrum. The key idea is to divide the FFT spectrum into short frames and change the magnitude of the selected FFT samples using Fibonacci numbers. The suggested technique guarantees and proves, mathematically, that the maximum change is less than 61% of the related FFT sample and the average error for each sample is 25%. Using the closest Fibonacci number to FFT magnitudes results in a robust and transparent technique. On top of very remarkable capacity, transparency and robustness, this scheme provides two parameters which facilitate the regulation of these properties. The experimental results show that the method has a high capacity (700 to 3 kbps), without significant perceptual distortion (ODG is about -- 1) and provides robustness against common audio signal processing such as echo, added noise, filtering and MPEG compression (MP3). In addition to the experimental results the fidelity of suggested system is proved mathematically.

When selling electronic content, the merchant would like each buyer to receive a different copy of the content fingerprinted with a serial number, in order to be able to trace redistributors should illegal redistribution happen. On the other hand, the merchant would like content distribution to be as scalable as possible, in order for mass transactions to be possible. Multicast content distribution fails to satisfy the first requirement: all receivers get exactly the same copy of the content, which makes it difficult to trace illegal redistributors. Unicast distribution of fingerprinted content, on the other hand, fails to satisfy the second requirement: for each buyer, the merchant needs to compute a fingerprint and establish a connection. P2P content distribution is a third option combining the strengths of multicast and unicast: the merchant needs to establish unicast connections only with a few seed buyers; on the other hand, with a suitable fingerprinting mechanism, illegal redistributors can still be identified and honest buyers can stay anonymous. We present a P2P content distribution scheme with such an anonymous fingerprinting mechanism, which is inspired in the way DNA sequences combine and spread from ancestors to descendants.

This paper presents two edge detectors and a preprocessing algorithm for histogram based data hiding schemes. The proposed technique takes advantage of the edge detectors to segment the image into plain and textured areas. The plain areas are selected for embedding where a histogram based scheme is used to embed information in these areas. Since in the plain areas, pixel intensities are close to each other, with the proposed preprocessing algorithm the number of shifted pixels for the same amount of hidden data capacity are decreased which results in a better transparency. To validate the efficiency of the technique two of the best known histogram based schemes which use prediction and interpolation are implemented and results are compared with and without the preprocessing scheme. The experimental results show that the histogram based methods with preprocessing algorithm, under the same capacity, have better transparency than the schemes without preprocessing. This scheme improves capacity, even by 200%, at equal distortion, or about 4 dB improvement in PSNR, at the same hiding capacity.

This paper proposes a high capacity audio watermarking algorithm in the logarithm domain based on the absolute threshold of hearing (ATH) of the human auditory system (HAS) which makes this scheme a novel technique. The key idea is to divide the selected frequency band into short frames and quantize the samples based on the HAS. Apart from remarkable capacity, transparency and robustness, this scheme provides three parameters (frequency band, scale factor, and frame size) which facilitate the regulation of the watermarking properties. The experimental results show that the method has a high capacity (800 to 7000 bits per second), without significant perceptual distortion (ODG is greater than - 1) and provides robustness against common audio signal processing such as added noise, filtering and MPEG compression (MP3).

In this paper we introduce a highly efficient reversible data hiding technique. It is based on dividing the image into tiles and shifting the histograms of each image tile between its minimum and maximum frequency. Data are then inserted at the pixel level with the largest frequency to maximize data hiding capacity. It exploits the special properties of medical images, where the histogram of their non-overlapping image tiles mostly peak around some gray values and the rest of the spectrum is mainly empty. The zeros (or minima) and peaks (maxima) of the histograms of the image tiles are then relocated to embed the data. The grey values of some pixels are therefore modified. High capacity, high fidelity, reversibility and multiple data insertions are the key requirements of data hiding in medical images. We show how histograms of image tiles of medical images can be exploited to achieve these requirements. Compared with data hiding method in the whole image, our scheme can result in 30%-200% capacity improvement with still better image quality, depending on the medical image content.

More and more people and organisations embrace Free Software (FS) and Open Standards (OS). However a lack of knowledge holds back their massive adoption. The Free Technology Academy will address this by setting up a virtual campus offering course modules on these topics to become a showcase of a virtual campus based on FS, OS and the use of Open Educational Resources. This distance learning programme will enable IT professionals, students, teachers and decision makers to upgrade knowledge and acquire relevant skills on free technologies. The FTA is realised by an international consortium and welcomes other interested parties to join the network.

This paper describes a study regarding the influence of a watermarking scheme based on near-lossless JPEG2000 compression in the performance of lossless or near-lossless image compression methods. Due to the nature of ultraspectral imaging applications, the imperceptibility property becomes more important than capacity or robustness, which determines the design of the watermarking scheme. Four strategies for band watermarking are discussed, depending on which bands are selected for embedding the mark and how. Preliminary results show that both intra-granule and inter-granule band capacity and imperceptibility are very variable, thus an adaptive watermarking band strategy must be used. On the other hand, the proposed watermarking scheme does not alter the obtained compression ratio for watermarked images when compared to the original ones, for a reasonable capacity and an imperceptible quality loss.

This paper provides a study concerning the suitability of well-known image coding techniques originally devised for lossy compression of still natural images when applied to lossless compression of ultraspectral sounder data. An ultraspectral sounder generates an unprecedented amount of 3D data, consisting of two spatial and one spectral dimensions; with ultraspectral sounder data, better inference of atmospheric, cloud and surface parameters is feasible. Here we present the experimental results of five widespread wavelet-based coding techniques, namely EZW, IC, SPIHT, JPEG2000 and CCSDS-IDC. Since the considered still image coding techniques are 2D in nature, but the ultraspectral data is 3D, a preprocessing step is applied to convert the two spatial dimensions into a single one. We are also interested in analyzing the benefits of applying some pre-processing step (e.g., linear prediction or bias adjusted reordering) prior to the coding process in order to further exploit the spectral correlation, which is much stronger than the spatial correlation.