2020 Sensor Signal Processing for Defence Conference (SSPD)最新文献

筛选
英文 中文
Learning Entropy of Adaptive Filters via Clustering Techniques 基于聚类技术的自适应滤波器熵学习
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272138
I. Bukovský, G. Dohnal, P. Steinbauer, Ondrej Budik, K. Ichiji, Homma Noriyasu
{"title":"Learning Entropy of Adaptive Filters via Clustering Techniques","authors":"I. Bukovský, G. Dohnal, P. Steinbauer, Ondrej Budik, K. Ichiji, Homma Noriyasu","doi":"10.1109/SSPD47486.2020.9272138","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272138","url":null,"abstract":"Learning Entropy (LE) was initially introduced as a measure for sample point novelty by unusually large learning effort of an online learning system. The key concept is that LE is based on pre-training and further online learning, and the novelty measure is not necessarily correlated to the prediction error. Most recently, the idea of LE was revised as a novel non-probabilistic, i.e., machine-learning-based information measure. This measure is high when a learning system is not familiar with a given data point, so the learning activity to learn novel data points is unusual (regardless of the prediction error), i.e., the learning increments display unusual patterns during adaptation. In this paper, we propose concepts of the learning state and the learning state space so that LE can be approximated via neighbourhood analysis in the learning space. Further, two novel clustering-based techniques for approximation of sample point LE are proposed. The first one is based on the sum of K nearest neighbour distances. The second one is based on multiscale neighbourhood cumulative sum. Also, we preprocess the learning space with dimensionality reduction that is promising for research of LE even with neural networks and potentially with deep neural networks. The performance of novelty detection with the clustering-based sample point LE with dimensionality reduction is compared to the original algorithms of LE, and its potentials are discussed.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"284 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122965803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Detection of Linear Frequency Modulation, Phase-Coded and Multicarrier Radar Waveforms in Electronic Warfare Context 电子战环境下线性调频、相位编码和多载波雷达波形的检测
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272129
Antoine Foucault, Cédric Cornu, A. Khenchaf, F. Comblet
{"title":"Detection of Linear Frequency Modulation, Phase-Coded and Multicarrier Radar Waveforms in Electronic Warfare Context","authors":"Antoine Foucault, Cédric Cornu, A. Khenchaf, F. Comblet","doi":"10.1109/SSPD47486.2020.9272129","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272129","url":null,"abstract":"Interception of radar signals is analyzed. The waveforms of interest are the Linear Frequency Modulation (LFM), the Phase-Coded (PC) and multicarrier (MC) waveforms. The problematic is to perform detection with real-time requirement and the possibility to integrate on the longest pulse width (LFM). The well-known ambiguity function is proposed as quadratic time-frequency detector which is able to help waveform recognition among the analyzed set of waveforms. Its detection performances are evaluated and some estimators are proposed to facilitate the intrapulse estimation step.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"327 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124299084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Fast Surface Detection Using Single-Photon Detection Events 使用单光子探测事件的快速表面探测
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272104
Abderrahim Halimi, A. Wallace, G. Buller, S. Mclaughlin
{"title":"Fast Surface Detection Using Single-Photon Detection Events","authors":"Abderrahim Halimi, A. Wallace, G. Buller, S. Mclaughlin","doi":"10.1109/SSPD47486.2020.9272104","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272104","url":null,"abstract":"This paper presents a fast object detection algorithm for 3D single-photon Lidar data. Lidar imaging acquires time-of-flight (ToFs) events in different spatial locations to build a 3D image of the observed objects. However, high ambient light or obscurants, might affect the reconstruction quality of the 3D scene. This paper proposes a solution by first detecting the pixels containing photons reflected from a object/surface, allowing a higher level processing of the data while only accounting for informative pixels. In contrast to histogram based approaches, the proposed algorithm operates on the detected photon events allowing a reduction in memory requirements and computational times. A Bayesian approach is considered leading to analytical estimates that can be computed efficiently. Results on simulated and real data highlight the benefit of the proposed approach when compared to a state-of-the-art algorithm based on histogram of counts.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"154 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127227065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Tail of Distribution GAN (TailGAN): GenerativeAdversarial-Network-Based Boundary Formation 分布的尾部GAN (TailGAN):基于生成对抗网络的边界形成
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272100
Nikolaos Dionelis
{"title":"Tail of Distribution GAN (TailGAN): GenerativeAdversarial-Network-Based Boundary Formation","authors":"Nikolaos Dionelis","doi":"10.1109/SSPD47486.2020.9272100","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272100","url":null,"abstract":"Generative Adversarial Networks (GANs) are a powerful methodology and can be used for unsupervised anomaly detection, where current techniques have limitations such as the accurate and robust detection of anomalies near the tail of a distribution. GANs generally do not guarantee the existence of a probability density and are susceptible to mode collapse, while few GANs use likelihood to reduce mode collapse. In this paper, we create a GAN-based tail formation model for anomaly detection, the Tail of distribution GAN (TailGAN), to generate samples on the tail of the data distribution and detect anomalies near the support boundary. Using TailGAN, we use maximum entropy regularization and leverage GANs for anomaly detection. Using GANs that learn the probability of the underlying distribution has advantages in improving the anomaly detection methodology by allowing us to devise a generator for boundary samples, and use this model to characterize anomalies. TailGAN addresses supports with disjoint components and achieves competitive performance on images. We evaluate TailGAN for identifying Out-of-Distribution (OoD) data and its performance evaluated on MNIST, CIFAR-10, Baggage X-Ray, and OoD data shows competitiveness compared to methods from the literature.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130197414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 12
Classifying LPI signals with transfer learning on CNN architectures 基于CNN架构的迁移学习LPI信号分类
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272032
Bunlong Lay, A. Charlish
{"title":"Classifying LPI signals with transfer learning on CNN architectures","authors":"Bunlong Lay, A. Charlish","doi":"10.1109/SSPD47486.2020.9272032","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272032","url":null,"abstract":"Due to the increased deployment of low probability of intercept radar systems, recognition and classification of low probability of intercept signals has developed an increased importance for electronic warfare systems. Recent results showed that combining time-frequency transformations such as Choi-Williams distribution with convolutional neural networks yield high accuracy. Since training convolutional neural networks is a time consuming task, we propose to use transfer learning on pre-trained convolutional neural network architectures. Furthermore, we compare these retrained neural network to the neural network trained with randomly initialized weights. We will demonstrate that the required training time for the transfer learning method is significantly shorter. Moreover, classifying time-frequency images based on Choi-Williams distribution achieves for both weight initialization methods an accuracy of over 99%.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132983227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 6
Sensing and Automation in the Future Maritime Environment 未来海洋环境中的传感与自动化
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272152
D. Sternlicht
{"title":"Sensing and Automation in the Future Maritime Environment","authors":"D. Sternlicht","doi":"10.1109/SSPD47486.2020.9272152","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272152","url":null,"abstract":"In this emerging era of great power competition, the goal of outpacing potential adversaries in the development of military technology takes on a new urgency. Evolving capabilities in sensing and automation are driven by a trade space that includes range and lethality versus close engagement and survivability; finders versus hiders; centralized command/control versus asset independence/dispersion; and planning and judgement versus reaction and autonomy. This paper explores this trade space first by describing sensing and automation innovations demonstrated during the 1990/1991 Gulf War and shortly thereafter, followed by discussion on current and emerging game-changing technologies. Capabilities projected for near and far term advantage include: weapons systems ensuring long-range lethality; unmanned cooperative networks of offboard systems; artificial intelligence and machine learning; and exploitation of advanced materials and quantum technologies. These will play a vital role in realizing a networked force of manned and unmanned systems with the ability to sense, comprehend, communicate, predict, plan, and take appropriate action in the future maritime environment.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117105363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Optimising Network Architectures for Provable Adversarial Robustness 优化可证明的对抗鲁棒性网络架构
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272169
H. Gouk, Timothy M. Hospedales
{"title":"Optimising Network Architectures for Provable Adversarial Robustness","authors":"H. Gouk, Timothy M. Hospedales","doi":"10.1109/SSPD47486.2020.9272169","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272169","url":null,"abstract":"Existing Lipschitz-based provable defences to adversarial examples only cover the $ell_{2}$ threat model. We introduce the first bound that makes use of Lipschitz continuity to provide a more general guarantee for threat models based on any $ell_{p}$ norm. Additionally, a new strategy is proposed for designing network architectures that exhibit superior provable adversarial robustness over conventional convolutional neural networks. Experiments are conducted to validate our theoretical contributions, show that the assumptions made during the design of our novel architecture hold in practice, and quantity the empirical robustness of several Lipschitz-based adversarial defence methods.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132772860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Information-Theoretic Compressive Measurement Design for Micro-Doppler Signatures 微多普勒信号的信息论压缩测量设计
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272136
Fraser K. Coutts, J. Thompson, B. Mulgrew
{"title":"Information-Theoretic Compressive Measurement Design for Micro-Doppler Signatures","authors":"Fraser K. Coutts, J. Thompson, B. Mulgrew","doi":"10.1109/SSPD47486.2020.9272136","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272136","url":null,"abstract":"In this work, we utilise gradient-ascent multi-objective optimisation within an information-theoretic compressive sensing framework to classify micro-Doppler (m-D) signatures in the presence of structured input noise. The proposed framework has the potential to simultaneously detect the class of a primary source exhibiting m-D features in its radar return and the class of a secondary, coincident source with its own m-D signature. We demonstrate through simulations with real radar return data that there is a configurable trade-off between the classification accuracies for the two sources and that, given a sufficient number of compressive measurements, the performance for the secondary source can be improved without substantially impacting the classification accuracy for the primary source.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126473262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Robust Source Number Estimation Based on Denoising Preprocessing 基于去噪预处理的鲁棒信源数估计
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272068
K. Ichige, Shohei Hamada, Katsuhisa Kashiwagi, Nobuya Arakawa, Ryo Saito
{"title":"Robust Source Number Estimation Based on Denoising Preprocessing","authors":"K. Ichige, Shohei Hamada, Katsuhisa Kashiwagi, Nobuya Arakawa, Ryo Saito","doi":"10.1109/SSPD47486.2020.9272068","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272068","url":null,"abstract":"This paper presents an accurate source number detection method for array antennas or MIMO (Multi-Input Multi-Output) radars. DOA (Direction of arrival) estimation is one of the significant techniques in high-speed wireless communication and radar imaging. Most of the representative DOA estimation methods requires the source number information in advance, however the classical methods AIC or MDL are often used. Those methods often fails to estimate source number in severe environments like low SNR or large transmission power difference. Recall that received signals are often bandlimited or narrowband signals, the proposed method first applies denoising preprocessing by removing undesired components, and compare the original and denoised signal information. Performance of the proposed method is evaluated through computer simulation.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126314621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Multimodal Learning for Early Detection of Explosive Sounds using Relative Spectral Distribution 基于相对谱分布的多模态学习早期检测爆炸声音
2020 Sensor Signal Processing for Defence Conference (SSPD) Pub Date : 2020-09-01 DOI: 10.1109/SSPD47486.2020.9272145
Vishwajeet Shukla, Mayank Singour
{"title":"Multimodal Learning for Early Detection of Explosive Sounds using Relative Spectral Distribution","authors":"Vishwajeet Shukla, Mayank Singour","doi":"10.1109/SSPD47486.2020.9272145","DOIUrl":"https://doi.org/10.1109/SSPD47486.2020.9272145","url":null,"abstract":"Human safety is a major challenge in the modern world. Bomb blasts and public shooting incidents are on the rise. Early recognition of such occurrences by detecting corresponding blaring audio events could help in providing swifter aid and reducing overall devastation. Hence rapid and accurate identification of related audio events is a crucial task for public safety. In this paper we propose a novel approach of Multimodal learning for the automated detection of explosive sounds such as Gunfire, Machine gun, Fusillade, Artillery Fire, Cap Gun and Explosion sounds for audio surveillance task. We present a deep feature stacking based method to classify explosive and non-explosive sounds substantially outperforming existing methods. We also discuss the challenges involved in explosive sound detection based on the inferences observed.","PeriodicalId":234896,"journal":{"name":"2020 Sensor Signal Processing for Defence Conference (SSPD)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125179271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书