Article on side-channel attacks in IEEE TCAD

C. Luo, Y. Fei, A. A. Ding and P. Closas, “Comprehensive Side-Channel Power Analysis of XTS-AES,” in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. doi: 10.1109/TCAD.2018.2878171

Abstract: XTS-AES is an advanced mode of AES for data protection of sector-based devices. It features two secret keys instead of one, and an additional tweak for each data block. These characteristics make the mode not only resistant against cryptoanalysis attacks, but also more challenging for side-channel attack. In this paper, we comprehensively analyze the side-channel power leakage of various XTS-AES implementations and invent effective attacks. We first run a simple power analysis of a software implementation. For a hardware implementation on FPGA, we analyze side-channel leakage of the particular modular multiplication in XTS-AES mode. In addition, we utilize the relationship between two consecutive block tweaks and propose a method to work around the masking of ciphertext by the tweak. These attacks are verified on an FPGA implementation of XTS-AES. The results show that XTS-AES is susceptible to side-channel power analysis attacks, and therefore dedicated protections are required for security of XTS-AES in storage devices.

Article on Robust Multimodal Data Fusion in IEEE Sensors Letters

Our work got recently accepted:

B. Kadioglu, I. Yildiz, P. Closas, M. B. Fried-Oken, D. Erdogmus, “Robust Fusion of c-VEP and Gaze”, IEEE Sensors Letters, accepted!

Abstract: Brain computer interfaces (BCIs) are one of the developing technologies, serving as a communication interface for people with neuromuscular disorders. Electroencephalography (EEG) and gaze signals are among the commonly used inputs for the user intent classification problem arising in BCIs. Fusing different types of input modalities, i.e. EEG and gaze, is an obvious but effective solution for achieving high performance on this problem. Even though there are some simplistic approaches for fusing these two evidences, a more effective method is required for classification performances and speeds suitable for real-life scenarios. One of the main problems that is left unrecognized is highly noisy real-life data. In the context of the BCI framework utilized in this work, noisy data stem from user error in the form of tracking a nontarget stimuli, which in turn results in misleading EEG and gaze signals. We propose a method for fusing aforementioned evidences in a probabilistic manner that is highly robust against noisy data. We show the performance of the proposed method on real EEG and gaze data for different configurations of noise control variables. Compared to the regular fusion method, robust method achieves up to 15 % higher classification accuracy.

Article on Robust PCA in IEEE JSTSP

The paper “M-Estimation Based Subspace Learning for Brain Computer Interfaces” was recently accepted for publication:

B. Kadioglu, I. Yildiz, P. Closas, D. Erdogmus, “M-Estimation Based Subspace Learning for Brain Computer Interfaces,” in IEEE Journal of Selected Topics in Signal Processing, accepted!

Abstract: Many problems in signal processing, statistical learning, and data science can be posed as the problem of learning lower dimensional representation of the data. Particularly, we consider the Brain Computer Interface (BCI) application where electroencephalography (EEG) data is used to determine user’s intent to type letters through stimulation with rapid serial visual presentations (RSVP). Such a typing system requires dimensionality reduction and classification abilities. The former is achieved through Principal Component Analysis (PCA), while the latter involves Regularized Discriminant Analysis (RDA). Remarkably, EEG recordings are prone to contain user-produced artifacts such as eye blinks, jaw contractions, or scalp movements. In this article, we present a methodology to fully robustify the aforementioned BCI application, enhancing its suitability in challenging recording scenarios. We consider a solution for Robust PCA (RPCA) of fully observed data with outlying samples based on M-estimation theory, as well as a similar methodology for Robust RDA (RRDA). Our proposed algorithm iteratively yields robust mean vector and covariance matrix estimates, and then applies eigen analysis on the estimated robust covariance matrix. The result is a principled way of dealing with outliers in the data that is simple, yet effective in delivering remarkable classification performance. The methodology is validated with real EEG data and compared to the state-of-the-art RSVP Keyboard™, with a thorough experimental setup being recorded and described.

NSF SaTC project funded

We were awarded a $160K NSF grant for “Securing GNSS-based infrastructures” within the Secure and Trustworthy Cyberspace (SaTC) program.

Abstract Source: NSF

This project develops novel anti-jamming techniques for Global Navigation Satellite Systems (GNSS) that are effective, yet computationally affordable. GNSS is ubiquitous in civilian, security and defense applications, causing a growing dependence on such technology for position and timing purposes, particularly in critical infrastructures. The threat of a potential disruption of GNSS is real and can lead to catastrophic consequences. This project studies methods to secure GNSS receivers from jamming interference, and doing so within size, weight, and power (SWAP) requirements. Existing solutions are either bulky and not cost-effective, such as those based on antenna array technology, or specifically adapted to an interference type. In addition, most of these solutions require the detection and classification of the interference before mitigating its effects, which constitutes a single point of error in the process. This project will investigate GNSS receivers that are resilient to interference without requiring detection and classification, by leveraging robust statistics to design methods that require few modifications with respect to state-of-the-art receiver architectures, keeping SWAP requirements comparable to those from standard GNSS receivers. The findings will be implemented and validated on an end-to-end GNSS software-defined radio receiver, successfully transitioning research into practice. Educational activities are closely integrated with this research agenda, including a course developed by the principal investigator and outreach activities.

This research advances knowledge of how robust statistics can be leveraged to design cost-effective and efficient mitigation techniques for anti-jamming GNSS. The main premise of the project is that most interference sources have a sparse representation, on which they can be seen as outliers to the nominal signal model. Tools from robust statistics are then used to discard those outliers in a sound manner, identifying and substituting specific critical operations in GNSS processing. This approach avoids the need for detecting and estimating interference, processes which can cause errors. The project envisions a lightweight, yet robust, GNSS receiver that can be easily adopted in substitution of current GNSS receivers that are supporting operation of critical infrastructures. It will enable reliable and precise anti-jamming technology with drastic SWAP and cost improvements. Particularly, the project will provide a GNSS receiver solution that can cope with common jamming interference. The development of such receiver enhancements, along with their validation in a software receiver, will allow for large-scale deployments of GNSS receivers that are more resilient and reliable.

Geoffrey Cideron completed his stay in the lab

Over the past months (May-August 2018) we hosted Geoffrey Cideron (MS student at École Central de Lille, France) in the lab. We jointly worked in topics related to statistical signal processing and machine learning. His research visit was very stimulating and we hope it was a great experience for Geoffrey. We wish him all the best in his future endeavours and hope to see Geoffrey soon.

A bientôt ami!

Haoqing presented his MS thesis

Haoqing successfully defended his MS thesis on July 26, 2018. Congrats!

H. Li, “Robust Interference Mitigation method for GNSS anti-jamming,” Advisor: Prof. P. Closas, Master Thesis, Electrical and Computer Engineering, Northeastern University. Boston MA, USA, July 2018.

The lab is glad to welcome him as a PhD student, starting Fall 2018. We look forward to do some exciting research.

Article accepted in Sensors

As a result of a collaboration with Daniele Borio (JRC, EU) and partial results from Haoqing’s MS thesis, we got accepted the paper “Huber’s Non-Linearity for GNSS Interference Mitigation” in the Sensors journal. Check it out at http://www.mdpi.com/1424-8220/18/7/2217/htm 

D. Borio, H. Li, P. Closas, “Huber’s Non-Linearity for GNSS Interference Mitigation”, Sensors 2018, 18(7), 2217; doi:10.3390/s18072217

Abstract: Satellite-based navigation is prevalent in both commercial applications and critical infrastructures, providing precise position and time referencing. As a consequence, interference to such systems can have repercussions on a plethora of fields. Additionally, Privacy Preserving Devices (PPD)—jamming devices—are relatively inexpensive and easy to obtain, potentially denying the service in a wide geographical area. Current jamming mitigation technology is based on interference cancellation approaches, requiring the detection and estimation of the interference waveform. Recently, the Robust Interference Mitigation (RIM) framework was proposed, which leverages results in robust statistics by treating the jamming signal as an outlier. It has the advantage of rejecting jamming signals without detecting or estimating its waveform. In this paper, we extend the framework to situations where the jammer is sparse in some transformed domain other than the time domain. Additionally, we analyse the use of Huber’s non-linearity within RIM and derive its loss of efficiency. We compare its performance to state-of-the-art techniques and to other RIM solutions, with both synthetic and real signals, showing remarkable results.

IEEE/ION PLANS 2018 Best Paper in Track Award

We are glad to announce that our paper was awarded with the Best Paper in Track Award at the IEEE/ION Position Location and Navigation Symposium (PLANS).

Jordi Vilà-Valls, James T. Curran, Pau Closas, Carles Fernández-Prades and Javier Arribas, “On-line Model Learning for Adaptive GNSS Ionospheric Scintillation Estimation and Mitigation”, in the Proceedings of the IEEE/ION Position Location and Navigation Symposium (PLANS ’18), 23-26 April 2018, Monterey, CA, USA.

This conference is jointly sponsored by the IEEE Aerospace and Electronics Systems Society (AESS), and the Institute of Navigation (ION). IEEE is the world’s largest professional engineering organization. The ION is the world’s premier professional organization for the advancement of positioning, navigation and timing.

Article accepted in Navigation

Our article “Ziv-Zakai Bound for Direct Position Estimation” was recently accepted:

A. Gusi-Amigó, P. Closas, A. Mallat, and L. Vandendorpe, “Ziv-Zakai Bound for Direct Position Estimation”, Navigation, accepted!

Abstract: Direct Position Estimation (DPE) has arisen as an appealing alternative to the conventional two steps positioning (2SP) approach. The usual metric to assess the performance of estimators is the Cramer-Rao Bound (CRB). However, the ´ CRB is only accurate in the high signal-to-noise ratio (SNR) region. In the lower SNR regions, high estimation errors take place and the variance of the estimates approach the a priori domain of the parameter. The DPE approach is expected to overcome the 2SP approach in such conditions. With the objective to throw light on this aspect, this paper shows the derivation of an approximate Ziv-Zakai Bound (ZZB) for the DPE approach and a mean square error (MSE) performance approximation based on the ZZB for the conventional approach. The derived bounds are compared with their associated CRBs and also with the performance of the corresponding Maximum Likelihood Estimator (MLE).

Article accepted in IET Radar, Sonar & Navigation

The article “Complex Signum Non-Linearity for Robust GNSS Interference Mitigation” was accepted for publication in the IET Radar, Sonar and Navigation journal.

D. Borio, P. Closas, “Complex Signum Non-Linearity for Robust GNSS Interference Mitigation”, IET Radar, Sonar and Navigation, accepted!

Abstract: The performance of a Global Navigation Satellite System (GNSS) receiver can be significantly degraded in the presence of pulsed interference and jamming. In this article we leverage on tools from robust statistics to enhance the receiver performance, with jamming signals treated as outliers to the nominal, interference-free, model. Particularly, the signal samples are pre-processed with a Zero-Memory Non-Linearity (ZMNL), which limits the impact of pulsed inference in a principled way. A possible approach for the design of such ZMNL is provided the M-estimator framework when the noise at the receiver input is modelled with a heavy-tailed distribution. This approach is adopted in this paper and the complex signum non-linearity is analysed. This ZMNL is obtained by considering a complex Laplacian noise. This choice is discussed and analysed in the context of GNSS receivers under jamming. The impact of the complex signum non-linearity is theoretically analysed under nominal conditions, that is, in the absence of interference. Theoretical results are supported by Monte Carlo simulations. Real GNSS signals, collected in the presence of jamming, are used to demonstrate the advantages brought by the complex signum non-linearity. Theoretical and experimental results demonstrate the benefits of robust GNSS signal processing and of the complex signum non-linearity