PyTorchKR Papers

최신 논문

총 3,337편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2006.03772

A precise machine learning aided algorithm for land subsidence or upheave prediction from GNSS time series

This paper is aimed at the problem of predicting the land subsidence or upheave in an area, using GNSS position time series. Since machine learning algorithms have presented themselves as strong prediction tools in different fields of science, we employ them to predict the next values of the GNSS position time series. For this reason, we present an algorithm that takes advantage of the machine learning algorithms for the prediction of positions in a GNSS time series. The proposed algorithm has two steps-preprocessing and prediction. In the preprocessing phase, the periodic tidal and atmospheric signals in the time series are removed and coordinates are transferred to the local coordinate system. In the prediction phase, eight different machine learning algorithms are used, namely, multilayer perceptron, Bayesian neural network, radial basis functions, Gaussian processes, k-nearest neighbor, generalized regression neural network, classification and regression trees, and support vector regression. We show the superiority of the Gaussian processes algorithm, compared to other methods, in 14 different real GNSS time series studies. The proposed algorithm can achieve up to 4 millimeters in accuracy, with the average accuracy as 2 centimeters across all time series.

arXiv:2008.02333

Fast Position-Aided MIMO Beam Training via Noisy Tensor Completion

In this paper, a data-driven position-aided approach is proposed to reduce the training overhead in MIMO systems, by leveraging side information and on-the-field measurements. A data tensor is constructed by collecting beam-training measurements on a subset of positions and beams, and a hybrid noisy tensor completion (HNTC) algorithm is proposed to predict the received power across the coverage area, which exploits both the spatial smoothness and the low-rank property of MIMO channels. A recommendation algorithm based on the completed tensor, beam subset selection (BSS), is proposed to achieve fast and accurate beam-training. Besides, a grouping-based BSS algorithm is proposed to combat the detrimental effect of noisy positional information. Numerical results evaluated with the Quadriga channel simulator at 60 GHz millimeter-wave channels show that the proposed BSS recommendation algorithm in combination with HNTC achieve accurate received power predictions, enabling beam-alignment with small overhead: given power measurements on 40% of possible discretized positions, HNTC-based BSS attains a probability of correct alignment of 91%, with only 2% of trained beams, as opposed to a state-of-the-art position-aided beam-alignment scheme which achieves 54% correct alignment in the same configuration. Finally, an online HNTC method via warm-start is proposed, that alleviates the computational complexity by 50%, with no degradation in prediction accuracy.

arXiv:2009.03557

Securing Mobile Multiuser Transmissions with UAVs in the Presence of Multiple Eavesdroppers

This paper discusses the problem of securing the transmissions of multiple ground users against eavesdropping attacks. We propose and optimize the deployment of a single unmanned aerial vehicle (UAV), which serves as an aerial relay between the user cluster and the base station. The focus is on maximizing the secrecy energy efficiency by jointly optimizing the uplink transmission powers of the ground users and the position of the UAV. The joint optimization problem is nonconvex; therefore we split it into two subproblems and solve them using an iterative algorithm.

arXiv:2011.12571

Group Delay Measurements of Multicore Fibers with Correlation Optical Time Domain Reflectometry

Several multi-core fibers (MCF) were characterized using Correlation Optical Time Domain Reflectometry (C-OTDR) in terms of propagation delay and polarization mode dispersion (PMD). The results show that the propagation delay in the cores depends on the position of the core in the fiber and that the differential delay between the cores varies with temperature.

arXiv:2011.10177

Flight Sensor Data and Beamforming based Integrated UAV Tracking with Channel Estimation using Gaussian Process Regression

With explosively increasing demands for unmanned aerial vehicle (UAV) applications, reliable link acquisition for serving UAVs is required. Considering the dynamic characteristics of UAV, it is hugely challenging to persist a reliable link without beam misalignment. In this paper, we propose a flight sensor data and beamforming signal based integrated UAV tracking scheme to deal with this problem. The proposed scheme provides a compatible integrated system considering the practical specification of the flight sensor data and the beamforming pilot signal. The UAV position tracking is comprised of two steps: 1) UAV position prediction by the flight sensor data and 2) position update with the beamforming signal using Gaussian process regression (GPR) method, which is a nonparametric machine learning. The flight sensor data can assist ground station (GS) or UAV nodes in designing the precoding and the receive beamforming matrix with drastically reduced overheads. The beamforming signal can accomplish high beamforming gain to be maintained even when the flight sensor data is absent. Therefore, the proposed scheme can support the moving target continuously by utilizing these two signals. The simulation results are provided to confirm that the proposed scheme outperforms other conventional beam tracking schemes. We also derive 3-dimensional (3D) beamforming gain and spectral efficiency (SE) from the mean absolute error (MAE) of the angular value estimation, which can be used as beamforming performance metrics of the data transmission link in advance.

arXiv:2011.03846

RFEye in the Sky

We introduce RFEye, a generalized technique to locate signals independent of the waveform, using a single Unmanned Aerial Vehicle (UAV) equipped with only one omnidirectional antenna. This is achieved by acquiring signals from uncoordinated positions within a sphere of 1-meter radius at two nearby locations and formulating an asynchronous, distributed receiver beamforming at the UAV to compute the Direction of Arrival (DoA) from the unknown transmitter. The proposed method includes four steps: 1) Blind detection and extraction of unique signature in the signal to be localized, 2) Asynchronous signal acquisition and conditioning, 3) DoA calculation by creating a virtual distributed antenna array at UAV and 4) Obtaining position fix of emitter using DoA from two locations. These steps are analyzed for various sources of error, computational complexity and compared with widely used signal subspace-based DoA estimation algorithms. RFEye is implemented using an Intel-Aero UAV, equipped with a USRP B205 software-defined radio to acquire signals from a ground emitter. Practical outdoor experiments show that RFEye achieves a median accuracy of 1.03m in 2D and 2.5m in 3D for Wi-Fi, and 1.15m in 2D and 2.7m in 3D for LoRa (Long Range) waveforms, and is robust to external factors like wind and UAV position errors.

arXiv:2011.02730

Feasibility Study of UAV-Assisted Anti-Jamming Positioning

As the cost and technical difficulty of jamming devices continue to decrease, jamming has become one of the major threats to positioning service. Unfortunately, most conventional technologies are vulnerable to jamming attacks due to their inherent shortcomings like weak signal strength and unfavorable anchor geometry. Thanks to the high operational flexibility, unmanned aerial vehicle (UAV) could fly close to users to enhance signal strength while maintaining a satisfactory geometry, making it a potential solution to the above challenges. In this article, we propose a UAV-assisted anti-jamming positioning system, in which multiple UAVs first utilize time-difference-of-arrival (TDoA) measurements from ground reference stations and double-response two-way ranging (DR-TWR) measurements from UAV-to-UAV links to perform self-localization as well as clock synchronization, and then act as anchor nodes to provide TDoA positioning service for ground users in the presence of jamming. To evaluate the feasibility and performance of the proposed system, we first derive the Cramer-Rao lower bound (CRLB) of UAV self-localization. Then, the impacts of UAV position uncertainty and synchronization errors caused by jamming on positioning service are modeled, and the theoretical root-mean-square error (RMSE) of user position estimate is further derived. Numerical results demonstrate that the proposed system is a promising alternative to existing positioning systems when their services are disrupted by jamming. The most notable advantage of the proposed system is that it is fully compatible with existing user equipment (UE) and positioning methods.

arXiv:2011.02338

Soft Attention Convolutional Neural Networks for Rare Event Detection in Sequences

Automated event detection in the sequences is an important aspect of temporal data analytics. The events can be in the form of peaks, changes in data distribution, changes of spectral characteristics etc. In this work, we propose a Soft-Attention Convolutional Neural Network (CNN) based approach for rare event detection in sequences. For the purpose of demonstration, we experiment with well logs where we aim to detect events depicting the changes in the geological layers (a.k.a. well tops/markers). Well logs (single or multivariate) are inputted to a soft attention CNN and a model is trained to locate the marker position. Attention mechanism enables the machine to relatively scale the relevant log features for the task. Experimental results show that our approach is able to locate the rare events with high precision.

arXiv:2011.01558

Enhanced RSS-based UAV Localization via Trajectory and Multi-base Stations

To improve the localization precision of unmanned aerial vehicle (UAV), a novel framework is established by jointly utilizing multiple measurements of received signal strength (RSS) from multiple base stations (BSs) and multiple points on trajectory. First, a joint maximum likelihood (ML) of exploiting both trajectory information and multi-BSs is proposed. To reduce its high complexity, two low-complexity localization methods are designed. The first method is from BS to trajectory (BST), called LCSL-BST. First, fixing the nth BS, by exploiting multiple measurements along trajectory, the position of UAV is computed by ML rule. Finally, all computed positions of UAV for different BSs are combined to form the resulting position. The second method reverses the order, called LCSL-TBS. We also derive the Cramer-Rao lower boundary (CRLB) of the joint ML method. From simulation results, we can see that the proposed joint ML and separate LCSL-BST methods have made a significant improvement over conventional ML method without use of trajectory knowledge in terms of location performance. The former achieves the joint CRLB and the latter is of low-complexity.

arXiv:2012.07545

Pointing Error Analysis of Optically Pre-Amplified Pulse Position Modulation Receivers

We present analytical results on the effect of pointing errors on the average bit-error probability (ABEP) of optically pre-amplified pulse-position modulation (PPM) receivers. The results show that the beam width plays a key role in the ABEP and that a significant power penalty is introduced by utilizing sub-optimal widths, especially when pointing errors incorporate a jitter component. We also present the optimisation the beam width for a number of pointing error scenarios and show that increasing beam widths are required as additional signal energy becomes available. The optimal beam width is also affected by the PPM modulation order and the optical noise modes, with more energy efficient (higher modulation order and lower noise) systems allowing for broader beams.

arXiv:2102.10561

Detection of Winding Axial Deformation in Power Transformers by UWB Radar Imaging

In this paper, a novel method for detecting transformer winding axial displacement has been presented. In this method, which is based on UWB radar imaging, a UWB pulse is transmitted to the transformer winding and the reflection from it is received and recorded. By changing the antenna position this process is repeated on several points along the axis of the winding. The measured signals are mapped to a 2-D image of the winding. By analyzing this image, the axial displacement of the winding can detected and the magnitude of it is determined with an acceptable precision. Simulation results are provided to verify the proposed method.

arXiv:2104.06314

Energy-Efficient 3D Deployment of Aerial Access Points in a UAV Communication System

In this letter, we propose an energy-efficient 3-dimensional placement of multiple aerial access points (AAPs), in the desired area, acting as flying base stations for uplink communication from a set of ground user equipment (UE). The globally optimal energy-efficient vertical position of AAPs is derived analytically by considering the inter-cell interference and AAP energy consumption. The horizontal position of AAPs which maximize the packing density of the AAP coverage area are determined using a novel regular polygon-based AAP placement algorithm. We also determine the maximum number of non-interfering AAPs that can be placed in the desired area. The effect of the AAP energy consumption on the optimal placement and the analytic findings are verified via numerical simulations.

arXiv:2104.06728공식 코드 · Papers with Code

Adversarial Sticker: A Stealthy Attack Method in the Physical World

To assess the vulnerability of deep learning in the physical world, recent works introduce adversarial patches and apply them on different tasks. In this paper, we propose another kind of adversarial patch: the Meaningful Adversarial Sticker, a physically feasible and stealthy attack method by using real stickers existing in our life. Unlike the previous adversarial patches by designing perturbations, our method manipulates the sticker's pasting position and rotation angle on the objects to perform physical attacks. Because the position and rotation angle are less affected by the printing loss and color distortion, adversarial stickers can keep good attacking performance in the physical world. Besides, to make adversarial stickers more practical in real scenes, we conduct attacks in the black-box setting with the limited information rather than the white-box setting with all the details of threat models. To effectively solve for the sticker's parameters, we design the Region based Heuristic Differential Evolution Algorithm, which utilizes the new-found regional aggregation of effective solutions and the adaptive adjustment strategy of the evaluation criteria. Our method is comprehensively verified in the face recognition and then extended to the image retrieval and traffic sign recognition. Extensive experiments show the proposed method is effective and efficient in complex physical conditions and has a good generalization for different tasks.

arXiv:2104.04567

Light-weight sleep monitoring: electrode distance matters more than placement for automatic scoring

Modern sleep monitoring development is shifting towards the use of unobtrusive sensors combined with algorithms for automatic sleep scoring. Many different combinations of wet and dry electrodes, ear-centered, forehead-mounted or headband-inspired designs have been proposed, alongside an ever growing variety of machine learning algorithms for automatic sleep scoring. In this paper, we compare 13 different, realistic sensor setups derived from the same data set and analysed with the same pipeline. We find that all setups which include both a lateral and an EOG derivation show similar, state-of-the-art performance, with average Cohen's kappa values of at least 0.80. This indicates that electrode distance, rather than position, is important for accurate sleep scoring. Finally, based on the results presented, we argue that with the current competitive performance of automated staging approaches, there is an urgent need for establishing an improved benchmark beyond current single human rater scoring.

arXiv:2104.04972

Data-driven predictive control with estimated prediction matrices and integral action

This paper presents a data-driven approach to the design of predictive controllers. The prediction matrices utilized in standard model predictive control (MPC) algorithms are typically constructed using knowledge of a system model such as, state-space or input-output models. Instead, we directly estimate the prediction matrices relating future outputs with current and future inputs from measured data, off-line. On-line, the developed data--driven predictive controller reduces to solving a quadratic program with a similar structure and complexity as linear MPC. Additionally, we develop a new procedure for estimating prediction matrices from data for predictive controllers with integral action, corresponding to the rate-based formulation of linear MPC. The effectiveness of the developed data-driven predictive controller is illustrated on position control of a linear motor model.

arXiv:2104.05107

Towards a Collective Agenda on AI for Earth Science Data Analysis

In the last years we have witnessed the fields of geosciences and remote sensing and artificial intelligence to become closer. Thanks to both the massive availability of observational data, improved simulations, and algorithmic advances, these disciplines have found common objectives and challenges to advance the modeling and understanding of the Earth system. Despite such great opportunities, we also observed a worrying tendency to remain in disciplinary comfort zones applying recent advances from artificial intelligence on well resolved remote sensing problems. Here we take a position on research directions where we think the interface between these fields will have the most impact and become potential game changers. In our declared agenda for AI on Earth sciences, we aim to inspire researchers, especially the younger generations, to tackle these challenges for a real advance of remote sensing and the geosciences.

arXiv:2104.05015

Temporal Consistency Two-Stream CNN for Human Motion Prediction

Fusion is critical for a two-stream network. In this paper, we propose a novel temporal fusion (TF) module to fuse the two-stream joints' information to predict human motion, including a temporal concatenation and a reinforcement trajectory spatial-temporal (TST) block, specifically designed to keep prediction temporal consistency. In particular, the temporal concatenation keeps the temporal consistency of preliminary predictions from two streams. Meanwhile, the TST block improves the spatial-temporal feature coupling. However, the TF module can increase the temporal continuities between the first predicted pose and the given poses and between each predicted pose. The fusion is based on a two-stream network that consists of a dynamic velocity stream (V-Stream) and a static position stream (P-Stream) because we found that the joints' velocity information improves the short-term prediction, while the joints' position information is better at long-term prediction, and they are complementary in motion prediction. Finally, our approach achieves impressive results on three benchmark datasets, including H3.6M, CMU-Mocap, and 3DPW in both short-term and long-term predictions, confirming its effectiveness and efficiency.

arXiv:2104.04843

Error Propagation in Satellite Multi-image Geometry

This paper describes an investigation of the source of geospatial error in digital surface models (DSMs) constructed from multiple satellite images. In this study the uncertainty in surface geometry is separated into two spatial components; global error that affects the absolute position of the surface, and local error that varies from surface point to surface point. The global error component is caused by inaccuracy in the satellite imaging process, mainly due to uncertainty in the satellite position and orientation (pose) during image collection. A key result of the investigation is a new algorithm for determining the absolute geoposition of the DSM that takes into account the pose covariance of each satellite during image collection. This covariance information is used to weigh the evidence from each image in the computation of the global position of the DSM. The use of covariance information significantly decreases the overall uncertainty in global position. The paper also describes an approach to the prediction of local error in the DSM surface. The observed variance in surface position within a single stereo surface reconstruction defines the local horizontal error. The variance in the fused set of elevations from multiple stereo pairs at a single DSM location defines the local vertical error. These accuracy predictions are compared to ground truth provided by LiDAR scans of the same geographic region of interest.

arXiv:2104.04517

AdCOFE: Advanced Contextual Feature Extraction in Conversations for emotion classification

Emotion recognition in conversations is an important step in various virtual chat bots which require opinion-based feedback, like in social media threads, online support and many more applications. Current Emotion recognition in conversations models face issues like (a) loss of contextual information in between two dialogues of a conversation, (b) failure to give appropriate importance to significant tokens in each utterance and (c) inability to pass on the emotional information from previous utterances.The proposed model of Advanced Contextual Feature Extraction (AdCOFE) addresses these issues by performing unique feature extraction using knowledge graphs, sentiment lexicons and phrases of natural language at all levels (word and position embedding) of the utterances. Experiments on the Emotion recognition in conversations dataset show that AdCOFE is beneficial in capturing emotions in conversations.