Search

We propose a novel malware detection system for critical embedded and cyber-physical systems (CPS). The system exploits electromagnetic (EM) side-channel signals from the device to detect malicious activity. During training, the system models EM emanations from an uncompromised device using a neural network. These EM patterns act as fingerprints for the normal program activity. Next, we continuously monitor the target device’s EM emanations. Any deviation in the device’s activity causes a variation in the EM fingerprint, which in turn violates the trained model, and is reported as an anomalous activity. The system can monitor the target device remotely (without any physical contact), and does not require any modification to the monitored system. We evaluate the system with different malware behavior (DDoS, ransomware, and code modification) on different applications using an Altera Nios-II soft-processor. Experimental evaluation reveals that our framework can detect DDoS and ransomware with 100% accuracy (AUC = 1.0), and stealthier code modification (which is roughly a 5 μ s long attack) with an AUC ≈ 0.99, from distances up to 3 m. In addition, we execute control-flow hijack, DDoS, and ransomware on different applications using an A13-OLinuXino—a Cortex A8 ARM processor single board computer with Debian Linux OS. Furthermore, we evaluate the practicality and the robustness of our system on a medical CPS, implemented using two different devices (TS-7250 and A13-OLinuXino), while executing control-flow hijack attack. Our evaluations show that our framework can detect these attacks with perfect accuracy.

This paper evaluates and compares the performance of K-Nearest Neighbors (KNN), Support Vector Machine (SVM) and Sparse Representation Classifier (SRC) for recognition of isolated Arabic handwritten characters. The proposed framework converts the gray-scale character image to a binary image through Otsu thresholding, and size-normalizes the binary image for feature extraction. Next, we exploit image down-sampling and the histogram of image gradients as features for image classification and apply fusion (combination) of these features to improve the recognition accuracy. The performance of the proposed system is evaluated on Isolated Farsi/Arabic Handwritten Character Database (IFHCDB) – a large dataset containing gray scale character images. Experimental results reveal that the histogram of gradient consistently outperforms down-sampling based features, and the fusion of these two feature sets achieves the best performance. Likewise, SRC and SVM both outperform KNN, with the latter performing the best among the three. Finally, we achieved a commanding accuracy of 93.71% in character recognition with fusion of features classified by SVM, where 92.06% and 91.10% is achieved by SRC and KNN respectively.

Establishing trust for an execution environment is an important problem, and practical solutions for it rely on attestation, where an untrusted system (prover) computes a response to a challenge sent by the trusted system (verifier). The response typically is a checksum of the prover's program, which the verifier checks against expected values for a "clean" (trustworthy) system. The main challenge in attestation is that, in addition to checking the response, the verifier also needs to verify the integrity of the response computation. On higher-end processors, this integrity is verified cryptographically, using dedicated trusted hardware. On embedded systems, however, constraints prevent the use of such hardware support. Instead, a popular approach is to use the request-to-response time as a way to establish confidence. However, the overall request-to-response time provides only one coarse-grained measurement from which the integrity of the attestation is to be inferred, and even that is noisy because it includes the network latency and/or variations due to micro-architectural events. Thus, the attestation is vulnerable to attacks where the adversary has tampered with response computation, but the resulting additional computation time is small relative to the overall request-to-response time. In this paper, we make a key observation that execution-time measurement is only one example of using externally measurable side-channel information, and that other side-channels, some of which can provide much finer-grain information about the computation, can be used. As a proof of concept, we propose EMMA, a novel method for attestation that leverages electromagnetic side-channel signals that are emanated by the system during response computation, to confirm that the device has, upon receiving the challenge, actually computed the response using the valid program code for that computation. This new approach requires physical proximity, but imposes no overhead to the system, and provides accurate monitoring during the attestation. We implement EMMA on a popular embedded system, Arduino UNO, and evaluate our system with a wide range of attacks on attestation integrity. Our results show that EMMA can successfully detect these attacks with high accuracy. We compare our method with the existing methods and show how EMMA outperforms them in terms of security guarantees, scalability, and robustness.

In previous work, we introduced zero-overhead profiling (ZOP), a technique that leverages the electromagnetic emissions generated by the computer hardware to profile a program without instrumenting it. Although effective, ZOP has several shortcomings: it requires test inputs that achieve extensive code coverage for its training phase; it predicts path profiles instead of complete execution traces; and its predictions can suffer unrecoverable accuracy losses. In this paper, we present zero-overhead path prediction (ZOP-2), an approach that extends ZOP and addresses its limitations. First, ZOP-2 achieves high coverage during training through progressive symbolic execution (PSE)---symbolic execution of increasingly small program fragments. Second, ZOP-2 predicts complete execution traces, rather than path profiles. Finally, ZOP-2 mitigates the problem of path mispredictions by using a stateless approach that can recover from prediction errors. We evaluated our approach on a set of benchmarks with promising results; for the cases considered, (1) ZOP-2 achieved over 90% path prediction accuracy, and (2) PSE covered feasible paths missed by traditional symbolic execution, thus boosting ZOP-2's accuracy.

Side-channel signals have long been used in cryptanalysis, and recently they have also been utilized as a way to monitor program execution without involving the monitored system in its own monitoring. Both of these use-cases for side-channel analysis have seen steady improvement, allowing ever-smaller deviations in program behavior to be monitored (to track program behavior and/or identify anomalies) or exploited (to steal sensitive information). However, there is still very little intuition about where the limits for this are, e.g. whether a single-instruction or a single-bit difference can realistically be recovered from the signal. In this paper, we use a popular open-source cryptographic software package as a test subject to demonstrate that, with enough training data, enough signal bandwidth, and enough signal-to-noise ratio, the decision of branch instructions that cause even single-instruction-differences in program execution can be recovered from the electromagnetic (EM) emanations of an IoT/embedded system. We additionally show that, in cryptographic implementations where branch decisions contain information about the secret key, nearly all such information can be extracted from the signal that corresponds to only a single cryptographic operation (e.g. encryption). Finally, we analyze how the received signal bandwidth, the amount of training, and the signal-to-noise ratio (SNR) affect the accuracy of side-channel-based reconstruction of individual branch decisions that occur during program execution.

Wireless Capsule Endoscopy (WCE) has emerged as a popular non-invasive imaging tool for inspection of human Gastrointestinal (GI) tract. In order to identify the location of an anomaly or intestinal disease, the physicians need to know the exact location of the endoscopic capsule which influences the treatment plan. In this paper, we present a displacement estimation technique based on feature point tracking which utilizes the images captured by a commercial capsule, named PillCam. The proposed displacement calculation approach is tested using a virtual testbed. Results show that, with assistance of ASIFT-RANSAC algorithms, the proposed algorithm is able to estimate the linear displacement of the endoscopic capsule with an accuracy of 93.7% on average.

We present a novel approach for abnormal breast mass classification from digitized mammography images. The proposed framework exploits rotation invariant uniform Local Binary Pattern (LBP) as texture feature. These features are classified using Support Vector Machine (SVM). In addition, we take advantage of the breast mammograms taken from multiple views or angles. We classify breast scans from ‘cranial-caudal’ view and ‘mediolateral-oblique’ view separately, and combine these classification scores to make an improved diagnosis. This reduces the classification error, and achieves higher recognition rate than that of either views individually. The proposed computer aided diagnosis system was evaluated on DDSM (Digital Database for Screening Mammography) data set, and was able to achieve a classification accuracy of 74%.

Local Binary Pattern (LBP) is a simple yet robust texture descriptor that has been widely used in many computer vision applications including face recognition. In this paper, we exploit LBP for handwritten Bangla numeral recognition. We classify Bangla digits from their LBP histograms using K Nearest Neighbors (KNN) classifier. The performance of three different variations of LBP - the basic LBP, the uniform LBP and the simplified LBP was investigated. The proposed OCR system was evaluated on the off-line handwritten Bangla numeral database CMATERdb 3.1.1, and achieved an excellent accuracy of 96:7% character recognition rate.

We present a framework for handwritten Bangla digit recognition using Sparse Representation Classifier. The classifier assumes that a test sample can be represented as a linear combination of the train samples from its native class. Hence, a test sample can be represented using a dictionary constructed from the train samples. The most sparse linear representation of the test sample in terms of this dictionary can be efficiently computed through l 1 -minimization, and can be exploited to classify the test sample. We applied Sparse Representation Classifier on the image zone density, an image domain statistical feature extracted from the character image, to classify the Bangla numerals. This is a novel approach for Bangla Optical Character Recognition, and demonstrates an excellent accuracy of 94% on the off-line handwritten Bangla numeral database CMATERdb 3.1.1. This result is promising, and should be investigated further.

Elastography is a promising technique for detection and classification of abnormal growth and tumors in soft tissues. In this paper we propose a simple yet effective optimization to speed up the time-domain indirect strain estimation. The proposed method exploits the displacement values of the neighbours to predict the displacement and utilizes this predicted value to define an adaptive search region for finding the best match between pre and post compression signals. This Proposed method is faster than other block matching algorithms such as SAD, SSD or cross-correlation. We have evaluated the method using performance parameters such as elastrographic Signal to Noise Ratio (SNRe), elastrographic Contrast to Noise Ratio (CNRe), elastrographic Peak-Signal to Noise Ratio (PSNRe) and Mean Structural Similarity (MSSIM). For lower percentage of strain, the proposed method faster demonstrated similar performance to the other methods.

Cell segmentation in microscopic images is inherently challenging due to the embedded optical artifacts and the overlapping of cells. Proper segmentation can help for shape analysis, motion tracking and cell counting. We present a framework for cell segmentation and counting by detection of cell centroids in microscopic images. The method is specifically designed for counting circular cells with a high probability of occlusion. The proposed algorithm has been implemented and evaluated on images of fluorescent cell population, collected from the Broad Bioimage Benchmark Collection (www.broad.mit.edu/bbbc), with different degrees of overlap probability. The experimental results show an excellent accuracy of 92% for cell counting even at a very high 60% overlap probability.

We present the Adaptive Vector Pattern Matching (AVPM) method, a novel method for the detection of vortical structures specifically designed for velocity encoded 4D PCMRI datasets. AVPM is based on vector pattern matching combined with robust orientation estimation. This combination provides for a simple yet robust algorithm, which is a priori axial flow invariant. We demonstrate these properties by comparing the performance of AVPM with Heiberg's Vector Pattern Matching algorithm.