Traditional Detectors Research Articles

Adaptive Backdoor Attack against Deep Neural Networks

In recent years, the number of parameters of deep neural networks (DNNs) has been increasing rapidly. The training of DNNs is typically computation-intensive. As a result, many users leverage cloud computing and outsource their training procedures. Outsourcing computation results in a potential risk called backdoor attack, in which a welltrained DNN would perform abnormally on inputs with a certain trigger. Backdoor attacks can also be classified as attacks that exploit fake images. However, most backdoor attacks design a uniform trigger for all images, which can be easily detected and removed. In this paper, we propose a novel adaptive backdoor attack. We overcome this defect and design a generator to assign a unique trigger for each image depending on its texture. To achieve this goal, we use a texture complexity metric to create a special mask for each image, which forces the trigger to be embedded into the rich texture regions. The trigger is distributed in texture regions, which makes it invisible to humans. Besides the stealthiness of triggers, we limit the range of modification of backdoor models to evade detection. Experiments show that our method is efficient in multiple datasets, and traditional detectors cannot reveal the existence of a backdoor.

MetNet: A Novel Low-Complexity Neural Network-Aided Detection for Faster-Than-Nyquist (FTN) Signaling in ISI Channels

This paper studies the application of neural networks to Viterbi detection of FTN signals in an intersymbol interference (ISI) channel. The main contribution of this paper is to propose a receiver structure for detecting FTN signals in unknown static ISI channel. In particular, we propose a novel low-complexity neural network structure for calculating the branch metrics, and we explore its suitability for FTN signalling with channel uncertainty. We compare the proposed network, which we call the Metric Net (MetNet), to a benchmark neural network-based technique for metric calculation, the ViterbiNet, which was originally designed for ISI channels. The simulation results confirm that the MetNet outperforms the ViterbiNet, with two orders of magnitude lower complexity, and is much more resilient to channel uncertainty than the traditional Viterbi detector, which uses Euclidean distance for metric calculations. We further show that the MetNet exhibits robustness to being trained at mismatched SNR values and FTN pulse acceleration factors, meaning that the number of trained models required can be significantly reduced. Additionally, the results show that the proposed MetNet remains a favorable alternative at much higher levels of channel uncertainties. The results also reflect that we can generalize the MetNet to work with different channel models defined by different decaying factors. Finally, we show that we succeed in achieving a bandwidth efficiency gain of 33% due to FTN by using the MetNet in the presence of channel uncertainty.

Data-driven Approach for State Prediction and Detection of False Data Injection Attacks in Smart Grid

In a smart grid, state estimation (SE) is a very important component of energy management system. Its main functions include system SE and detection of cyber anomalies. Recently, it has been shown that conventional SE techniques are vulnerable to false data injection (FDI) attack, which is a sophisticated new class of attacks on data integrity in smart grid. The main contribution of this paper is to propose a new FDI attack detection technique using a new data-driven SE model, which is different from the traditional weighted least square based SE model. This SE model has a number of unique advantages compared with traditional SE models. First, the prediction technique can better maintain the inherent temporal correlations among consecutive measurement vectors. Second, the proposed SE model can learn the actual power system states. Finally, this paper shows that this SE model can be effectively used to detect FDI attacks that otherwise remain stealthy to traditional SE-based bad data detectors. The proposed FDI attack detection technique is evaluated on a number of standard bus systems. The performance of state prediction and the accuracy of FDI attack detection are benchmarked against the state-of-the-art techniques. Experimental results show that the proposed FDI attack detection technique has a higher detection rate compared with the existing techniques while reducing the false alarms significantly.

Early Detection of Heartbeat from Multimodal Data Using RPA Learning with KDNN-SAE

Heartbeat detection stays central to cardiovascular an electrocardiogram (ECG) is used to help with disease diagnosis and management. Existing Convolutional Neural Network (CNN)-based methods suffer from the less generalization problem thus; the effectiveness and robustness of the traditional heartbeat detector methods cannot be guaranteed. In contrast, this work proposes a heartbeat detector Krill based Deep Neural Network Stacked Auto Encoders (KDNN-SAE) that computes the disease before the exact heart rate by combining features from multiple ECG Signals. Heartbeats are classified independently and multiple signals are fused to estimate life threatening conditions earlier without any error in classification of heart beat. This work contained Training and testing stages, in the preparation part at first the Adaptive Filter Enthalpy-based Empirical Mode Decomposition (EMD) is utilized to eliminate the motion artifact in the signal. At that point, the robotic process automation (RPA) learning part extracts the effective features are extracted, and normalized the value of the feature then estimated utilizing the RPA loss function. At last KDNN-SAE prepared training for the data stored in the dataset. In the subsequent stage, input signal compute motion artifact and RPA Learning the evaluation part determines the detection of Heartbeat. So early diagnosis of heart failures is an essential factor. The results of the experiments show that our proposed method has a high score outcome of 0.9997. Comparable to the CIF, which reaches 0.9990. The CNN and Artificial Neural Network (ANN) had less score 0.95115 and 0.90147.

Information Retrieval With Chessboard-Shaped Topology for Hyperspectral Target Detection

Given a priori knowledge, hyperspectral target detection aims to locate objects of interest within specific scenes by utilizing differences in spectral characteristics among various land covers. However, for those traditional model-driven detectors with monotonic analytical mode, they perform mediocrely in the disassembly of hyperspectral image (HSI) data, failing to cope with real scenes with complexity. The discrepancy between fixed model assumptions and HSI data severely reduces detection effects, leading to the inability of such methods to mine deep-level features and adapt to the variability of imaging scenes. To overcome the limitations of traditional methods, we propose a chessboard-shaped topological framework for high-dimensional data structures to disassemble an HSI from both spatial and spectral dimensions adaptively. With hyperspectral target detection is refined into an information retrieval task in a topological space, a target detection method based on chessboard-shaped topology (CTTD) is proposed. In the topological space, latent and hidden data features of original images are presented in an intuitive way. Therefore, the differences in both spatial and spectral dimensions between the two classes of objects, namely target and background, are specifically amplified and exploited to perform the information retrieval task with superior performance. Extensive experimental results on benchmark HSI data sets demonstrate that CTTD can efficiently adapt to the variability of real scenes while extracting abundant and detailed information for accurate target localization. Moreover, both detection effects and computational efficiency exhibited by the proposed method provide a strong support for its popularization in practical applications.

Safety monitoring under stealthy sensor injection attacks using reachable sets

Stealthy sensor injection attacks are serious threats for industrial plants as they can compromise the plant's integrity without being detected by traditional fault detectors. In this manuscript, we study the possibility of revealing the presence of such attacks by monitoring only the control input. This approach consists in computing an ellipsoidal bound of the input reachable set. When the control input does not belong to this set, this means that a stealthy sensor injection attack is driving the plant to critical states. The problem of finding this ellipsoidal bound is posed as a convex optimization problem (convex cost with Linear Matrix Inequalities constraints). Our monitoring approach is tested in simulation.

Hyperspectral Anomaly Detection Based on Chessboard Topology

Without any prior information, hyperspectral anomaly detection is devoted to locating targets of interest within a specific scene by exploiting differences in spectral characteristics between various land covers. Traditional methods originated from the signal processing perspective, and most of them rely heavily on specific model assumptions. Due to the model-driven attributes, such methods cannot mine the deep-level features of data to adapt to the variability of scenes, and cannot fully extract the information of land covers contained in images to accurately separate anomalies from background. By independently designing a chessboard-shaped topological framework that avoids making any distribution assumptions but directly mines high-dimensional data features to break through the limitations of traditional detectors, this paper proposes a novel chessboard topology-based anomaly detection (CTAD) method, which constructs a chessboard-shaped topology to dissect images and extract detailed information of land covers adaptively, thereby enabling highly accurate detection. Extensive experimental results on HSIs in real scenes demonstrate that the proposed CTAD can be adapted to the variability of scenes by autonomously learning data features, and exhibits strong generalization and detection capabilities, facilitating practical applications.

Sea Surface Target Detection Using Global False Alarm Controllable Adaptive Boosting Based on Correlation Features

In the complex marine environment, traditional detectors based on the classifiers cannot guarantee the global false alarm control. In this paper, we propose a detector based on dual channel convolutional neural network (DC-CNN) and global false alarm controllable adaptive boosting tree (GFAC-A), which is shortened as DC-CNN-GFAC-A. DC-CNN focuses on the correlation features of radar echoes in time domain and frequency domain, better use multi-dimensional features and show a better feature extraction ability. Through the application of GFAC-A, the false alarm rate is introduced into the algorithm combining decision tree and AdaBoost to achieve the high-performance detection of global false alarm controllable for high-dimensional features. It is heuristics. The combination of DC-CNN and GFAC-A solves the disadvantage that current classifiers cannot meet the conditions of good performance and low false alarm. First, the connectivity information is obtained by using the frequency domain amplitude characteristics of radar echo data, and the recursive information is obtained by using the nonlinear recursive time series characteristics. And the dual-channel datasets of targets and clutter are built. Then, DC-CNN is built to extract and fuse high-dimensional features to obtain feature vectors of targets and clutter. Besides, the performance comparison of different neural network model combinations is carried out. Finally, compared with the traditional threshold-controllable classifiers, the proposed GFAC-A classifier achieves the high detection performance under the global controlled false alarm. The results show that DC-CNN-GFAC-A can achieve 96.491% detection accuracy when the false alarm rate is 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> , which is superior to other detections.

Novel FDIs-based data manipulation and its detection in smart meters’ electricity theft scenarios

Non-technical loss is a serious issue around the globe. Consumers manipulate their smart meter (SM) data to under-report their readings for financial benefit. Various manipulation techniques are used. This paper highlights novel false data injection (FDIs) techniques, which are used to manipulate the smart meter data. These techniques are introduced in comparison to six theft cases. Furthermore, various features are engineered to analyze the variance, complexity, and distribution of the manipulated data. The variance and complexity are created in data distribution when FDIs and theft cases are used to poison SM data, which is investigated through skewness and kurtosis analysis. Furthermore, to tackle the data imbalance issue, the proximity weighted synthetic oversampling (ProWsyn) technique is used. Moreover, a hybrid attentionLSTMInception is introduced, which is an integration of attention layers, LSTM, and inception blocks to tackle data dimensionality, misclassification, and high false positive rate issues. The proposed hybrid model outperforms the traditional theft detectors and achieves an accuracy of 0.95%, precision 0.97%, recall 0.94%, F1 score 0.96%, and area under-the-curve (AUC) score 0.98%.

Neighborhood representative for improving outlier detectors

Over the decades, traditional outlier detectors have ignored the group-level factor when calculating outlier scores for objects in data by evaluating only the object-level factor, failing to capture the collective outliers. To mitigate this issue, we present a framework called neighborhood representative (NR), which empowers all the existing outlier detectors to efficiently detect outliers, including collective outliers, while maintaining their computational integrity. It achieves this by selecting representative objects, scoring these objects, then applies the score of the representative objects to its collective objects. Without altering existing detectors, NR is compatible with existing detectors, while improving performance on eleven real world datasets with +8 % (0.72 to 0.78 AUC) on average relative to twelve state-of-the-art outlier detectors. The implementation of NR can be found via www.OutlierNet.com for reproducibility.

Robust Keypoint Detection and Matching on Fisheye Images by Self-Supervised Learning.

Accurate image feature point detection and matching are essential to computer vision tasks such as panoramic image stitching and 3D reconstruction. However, ordinary feature point approaches cannot be directly applied to fisheye images due to their large distortion, which makes the ordinary camera model unable to adapt. To address such a problem, this paper proposes a self-supervised learning method for feature point detection and matching on fisheye images. This method utilizes a Siamese network to automatically learn the correspondence of feature points across transformed image pairs to avoid high annotation costs. Due to the scarcity of the fisheye image dataset, a two-stage viewpoint transform pipeline is also adopted for image augmentation to increase the data variety. Furthermore, this method adopts both deformable convolution and contrastive learning loss to improve the feature extraction and description of distorted image regions. Compared with traditional feature point detectors and matchers, this method has been demonstrated with superior performance on fisheye images.

Stealthy replay attack detection of 3-DOF helicopter benchmark system using dynamic watermarking approach

Cyber security for unmanned systems has drawn tremendous research attention in recent years owing to its priority in ensuring system performance and safety. This paper presents a dynamic watermarking-based active detection approach for unmanned helicopter system under stealthy replay attacks. First, the dynamic model of the 3-DOF helicopter and the replay attack model are constructed, respectively. Statistical analysis is conducted using Kalman filter residuals to interpret the concealment of stealthy replay attacks for a system under intrusion. Then, as traditional chi-square detector is unable to detect stealthy replay attacks, an active detector by incorporating dynamic watermarking with the chi-square detector is explored. Finally, by conducting simulations of the detection procedures under different test scenarios comparatively using traditional chi-square detector and the proposed approach, the effectiveness and merits of the watermarking approach is verified sufficiently. The watermarking-based approach is promising in comprehensive detections against stealthy replay attacks for enhancing system security.

Geospatial Cognitive Networking Protocols and Sensing Algorithms for 5G NR Beamforming

Although 5G New Radio (NR) has created new opportunities for cognitive radio networks, its increased physical layer security and flexibility limit the usefulness of traditional cognitive detectors such as energy and blind control channel algorithms. This paper presents CASINO-NR, a novel framework for establishing a cognitive self-reliant secondary network with no additional physical infrastructure, collaboration from the primary network nodes, and software or hardware changes to the existing 5G network. CASINO-NR includes a novel beam detection algorithm that finds and ranks 5G NR synchronization signals to determine geospatially non-interfering beams for secondary communications. We compare the developed beam detector with multiple existing approaches for sensitivity to interference and phase distortions. We also apply power control to prevent interference on neighboring beams. CASINO-NR is analyzed against the estimated throughput capacity and capabilities of other cognitive detectors found in literature. Finally, we examine an experimental beamforming example to demonstrate our beam detection algorithm and present a case for geospatial resources for cognitive radio communications.

Flexible perovskite scintillators and detectors for X-ray detection.

X-ray detection and imaging technology has been rapidly developed for various fields since 1895, offering great opportunities to scientific and industrial communities. Particularly, flexible X-ray detectors have drawn numerous attention in medical-related applications, solving the uniform issues of traditional rigid X-ray detectors. Out of all the potential materials, metal halide perovskites (MHPs) have been emerged as excellent candidates as flexible X-ray scintillators and detectors owing to the advantages including low temperature solution processable, strong X-ray absorption coefficient, large mobility lifetime product and tunable bandgap. In this review, the recent advances of MHP-based flexible X-ray detectors are comprehensively summarized, focusing on the scalable synthesis technologies of materials and diverse device architectures, and covering both direct and indirect X-ray detection. A brief outlook that highlights the current challenges impeding the commercialization of flexible MHP-based X-ray detectors is also included with possible solutions to those problem being provided.

Experiment analysis of non-uniformity measurement by array detector scanning system

Solar simulator is used to analysis characteristic of the solar cells. The non-uniformity is its major performance. The traditional non-uniformity measurement calls single detector method. The paper’s objective is to design and construct an array detector scanning system and to determine the optimal scanning time to achieve the lowest uncertainty. To investigate the non-uniformity by our proposed method and the traditional method, our detector consisted of eight photodiodes mounted on an arm of a linear motion lead screw to guide the detector scaning onto the lighting area. A microcontroller applied for controling and measuring light irradiance in 64 points corresponding to IEC 60904-9 standard. The results showed that the array detector scanned at a speed of 33.33 mm/s to obtain the non-uniformity with the lowest uncertainty, less than 0.6%. Analysis results of the non-uniformity obtained from our system on the test areas of (mm×mm) 156×156, 166×166 and 200×200 compared with the single detector. It showed that the mean absolute error was 1.27. Our system provided a lower uncertainty than the traditional method. The measurement accuracy was acceptable. The advantage is for testing on different test areas within a single device. The measurement time is around 1/32 of the traditional method.

High-performances ultraviolet photodetector based on vertical van der Waals heterostructures

High-performance ultraviolet (UV) photodetectors play a very important role in many fields, especially in the military, biomedical and other fields. [1]In recent years, many studies have realized ultraviolet photodetectors of 2D layered materials, overcome the problems of traditional ultraviolet detectors that are large and use high voltages. [1]Up to now, most of these works use atomically thin layers and simple p-n van der Waals (vdW) heterostructures, which have difficulty meeting the conditions of high sensitivity and ultrafast response at the same time. we report the double p-n van der Waals (vdW) heterostructure built on a large electrode. The two p-n junctions connected in parallel were proven to be able to effectively separate photo-generated carriers and suitable for ultraviolet light. This new type of photodetector exhibits competitive performance, including high R up to 254.8 A/W under UV light, and fast photoresponse τr = 7.9 μs and τd = 3.9 μs. These results provide an ideal platform for realizing highly sensitive UV photodetectors.

Particle physics readout electronics and novel detector technologies for neutron science

Traditional thermal neutron detectors are based on Helium-3 as conversion and detection material due to its large neutron cross-section. The difficult availability of this converter and the simultaneous higher demand for neutron detectors lead to the investigation of alternatives. Solid converters with a large thermal cross-section as boron or gadolinium, which emit charged particles when hit by a neutron are a prominent type. Similar to the concepts in particle physics, the techniques then rely on charged particle detection. A prime example is the Gadolinium Gas Electron Multiplier (GdGEM) detector for the NMX instrument at ESS.We present three further novel concepts for neutron detectors. By modifying existing detector principles with a solid converter and employing recent high-granular readout electronics from particle physics we aim at developing thermal neutron detectors beyond the state-of-the-art. The first concept combines the Timepix3 data-driven, self-triggered readout chip with a boronated Micro-Channel Plate. The second concept is a Time Projection Chamber with boronated scintillator walls. The third concept combines boronated segmented GEMs with a VMM3a-based strip readout.This proceeding introduces the novel concepts, outlines how they could go beyond the state-of-the-art, sketches the current progress and obstacles overcome and, where available, presents first measurement results. The project, even though at an intermediate stage, shows the high value of technology transfer between disciplines and the originality required to start such a process.

Low-energy X-ray spectroscopy with RGB-HD SiPMs coupled to CsI(Tl) scintillator

The continuous improvement of Silicon Photomultiplier (SiPM) characteristics opens new perspectives for the application of these detectors in middle to low energy X-ray scintillation light readout. Compared to traditional PET detectors this application poses an additional challenge because of the much lower number of photons generated in the scintillator. Many SiPM characteristics influence the energy resolution of the detected radiation, including Photon Detection Efficiency (PDE), primary noise and Excess Noise Factor (ENF). The most recent development of the RGB-HD SiPM technology at Fondazione Bruno Kessler (FBK, Italy) provides a PDE up to 40% at 550 nm while keeping the Dark Count Rate (DCR) below 500 kc.p.s./mm2. A low DCR value is fundamental in improving the energy resolution for soft X-rays. In this work we coupled a 1×1 mm2 RGB-HD SiPM to a 0.9×0.9×2 mm3 CsI(Tl) crystal allowing the detection of 55Fe 5.9 keV photons with an energy resolution of 39% FWHM. A 4×4 mm2 RGB-HD SiPM with a 3×3×5 mm3 CsI(Tl) was used for the readout of a 57Co source (6.4 keV, 14.4 keV and 122 keV) in order to prove the dynamic range of the system. The energy resolution of these three peaks shows different trends with respect to the SiPM bias and to the signal integration time. This is related to the fact that SiPM parameters such as PDE, DCR and ENF, and readout parameters such as integration time, become more or less relevant depending on the energy of the primary photon.

Performance of semi-insulating metal-semiconductor-metal GaN prototype devices as ionizing radiation detector

We describe the fabrication and characterization of semi-insulating GaN devices for the detection of ionizing radiation with applications in high radiation environment. We present the DC characterization and the signal response of such devices from Am-241 α-source. The detector prototypes show up to 80% charge collection efficiency with bias voltages as low as -40V. Wide band gap semiconductors like synthetic diamond, GaN, SiC have gained immense importance in the field of charged particle detection due to their low intrinsic noise and high radiation tolerance. The ability of suppressing the thermal noise generated by intrinsic carriers, makes the fabrication of these devices simpler when compared to the multi-step lithography required for doping of traditional silicon-based solid-state detectors. We have made interdigitated metal-semiconductor-metal(MSM) device on a 3μm in thick GaN epitaxial layer grown on sapphire substrate by MOCVD technique. We employed Ni/Pt/Au metal stack for Schottky contact with finger width of 4μm, spacing of 8μm and finger length of 120μm. Details of the experimental set-up for fabrication and characterization will be presented.

Quantitative determination of major alkaloids in areca nut products by high-performance liquid chromatography coupled with ion mobility spectrometry.

Areca nut is a popular and addictive food as well as a traditional herbal medicine in many countries. Areca nut contains alkaloids including arecoline, guvacine, and arecaidine, which are the major bioactive compounds in areca products. Areca alkaloids can be carcinogenic, and thus sensitive and specific analytical methods are urgently desired for the identification and quantification of these compounds. High-performance liquid chromatography-based methods are often preferred, but areca alkaloids do not have chromophores, and detection using a traditional UV detector can be difficult. The complexity of areca sample extracts can also lead to the co-elution of peaks leading to poor quantitative performance. We report here high-performance liquid chromatography coupled with an ion mobility spectrometer for sensitive determination of areca alkaloids in various products including areca nut, areca nut products, and herbal oral liquid. An X-Bridge reversed-phase C18 column was used in the experiment and was combined with high-performance liquid chromatography coupled to an ion mobility spectrometer system. A custom-made adjustable post-column splitter acted as an interface between the high-performance liquid chromatography and the ion mobility spectrometer; it also acted as the electrospray ionization source. The mobile phase was methanol and 0.5% ammonium hydroxide. The results demonstrate that the splitter can afford a wide range of split ratios that match the ion mobility spectrometer ionization source while keeping the separation efficiency of high-performance liquid chromatography. Three major alkaloid compounds were then accurately determined using the resulting method without dativization steps. Many coeluted high-performance liquid chromatography peaks are effectively separated in the ion mobility spectrometer dimension, which in turn improved the quantification accuracy.