Mitarbeiter

M. Sc. Anand Dubey

Kontakt

  • E-Mail:
  • Telefon: 09131/85-27190
  • Fax: 09131/85-28730
  • Raum: 04.236
  • Cauerstraße 9
    91058 Erlangen

Über Anand Dubey

Biography

In February, 2018 Anand Dubey finished his Master's Degree in Automotive Software Engineering at the Technical  University of Chemnitz. Since September 2018 he is working as a research assistant at the Institute for Electronics Engineering as part of the team's Circuits, Systems & Hardware Test (CST). He has worked for 3.5 years in Automotive Industry in the area of embedded programming, computer vision and machine learning. Currently, he is working for Prystine project. 

Areas of Interest

  • Radar Signal Processing
  • Machine Learning Algorithms

Open thesis projetcs

Just call me!

Publikationen

2020

  • A. Dubey, J. Fuchs, M. Lübke, R. Weigel, and F. Lurz, "Generative Adversial Network based Extended Target Detection for Automotive MIMO Radar" in 2018 International Conference on Radar (RADAR), Washington DC, USA, 2020 (to be published). [Bibtex]
    @inproceedings{dubey2020,
    abstract = {In recent years, the automotive radar systems has gained substantial interest for different applications of autonomous driving. The performance of most applications likes classification and tracking directly relies on accurate target detection. The state-of-the-art detection pipeline is vulnerable to multi-path reflections, clutter noise, interference from another radar and leads to false or ghost detections. To address this issue, an end-to-end target detection pipeline using a residual based U-Net architecture is proposed. In contrast to the conventional approach, the network directly generates the detection map from range-Doppler map. The network uses a generative adversarial training over multiple real world measurements. We demonstrate that the proposed network can learn effectively to detect extended targets and shows significant improvement under increased noise floor in comparison to the state-of-the-art detection techniques.},
    author = {Dubey, Anand and Fuchs, Jonas and Lübke, Maximilian and Weigel, Robert and Lurz, Fabian},
    language = {English},
    booktitle = {2018 International Conference on Radar (RADAR)},
    cris = {https://cris.fau.de/converis/publicweb/publication/231325915},
    year = {2020},
    month = {04},
    day = {27},
    eventdate = {2020-04-27/2020-05-01},
    faupublication = {yes},
    note = {unpublished},
    peerreviewed = {automatic},
    title = {Generative Adversial Network based Extended Target Detection for Automotive MIMO Radar},
    venue = {Washington DC, USA},
    }
  • J. Fuchs, A. Dubey, M. Lübke, R. Weigel, and F. Lurz, "Automotive Radar Interference Mitigation using a Convolutional Autoencoder" in 2018 International Conference on Radar (RADAR), Washington, DC, USA, 2020 (to be published). [Bibtex]
    @inproceedings{fuchs2020,
    abstract = {Automotive radar interference imposes big challenges on signal processing algorithms as it raises the noise floor and consequently lowers the detection probability. With limited frequency bands and increasing number of sensors per car, avoidance techniques such as frequency hopping or beamforming quickly become insufficient. Detect-and-repair strategies have been studied intensively for the automotive field, to reconstruct the affected signal samples. However depending on the type of interference, reconstruction of the time domain signals is a highly non-trivial task, which can affect following signal processing modules. In this work an autoencoder based convolutional neural network is proposed to perform image based denoising. Interference mitigation is phrased as a denoising task directly on the range-Doppler spectrum. The neural networks shows significant improvement with respect to signal-to-noiseplus- interference ratio in comparison to other state-of-the-art mitigation techniques, while better preserving phase information of the spectrum compared to other techniques.
    }, author = {Fuchs, Jonas and Dubey, Anand and Lübke, Maximilian and Weigel, Robert and Lurz, Fabian}, language = {English}, booktitle = {2018 International Conference on Radar (RADAR)}, cris = {https://cris.fau.de/converis/publicweb/publication/231966704}, year = {2020}, month = {04}, day = {27}, eventdate = {2020-04-27/2020-05-01}, faupublication = {yes}, note = {unpublished}, peerreviewed = {unknown}, title = {Automotive Radar Interference Mitigation using a Convolutional Autoencoder}, type = {Konferenzschrift}, venue = {Washington, DC, USA}, }
  • M. Lübke, J. Fuchs, V. Shatov, A. Dubey, R. Weigel, and F. Lurz, "Combining Radar and Communication at 77 GHz Using a CDMA Technique" in 2020 IEEE MTT-S International Conference on Microwaves for Intelligent Mobility (ICMIM), Linz, Austria, 2020 (to be published). [Bibtex]
    @inproceedings{luebke2020,
    author = {Lübke, Maximilian and Fuchs, Jonas and Shatov, Victor and Dubey, Anand and Weigel, Robert and Lurz, Fabian},
    booktitle = {2020 IEEE MTT-S International Conference on Microwaves for Intelligent Mobility (ICMIM)},
    cris = {https://cris.fau.de/converis/publicweb/publication/234501926},
    year = {2020},
    month = {04},
    day = {20},
    eventdate = {2020-04-20/2020-04-22},
    faupublication = {yes},
    note = {unpublished},
    peerreviewed = {automatic},
    title = {Combining Radar and Communication at 77 GHz Using a CDMA Technique},
    venue = {Linz, Austria},
    }
  • A. Dubey, J. Fuchs, T. Reißland, R. Weigel, and F. Lurz, "Uncertainty Analysis of Deep Neural Network for Classification of Vulnerable Road Users using micro-Doppler" in IEEE Topical Conference on Wireless Sensors and Sensor Networks (WiSNet), San Antonio, Texas, USA, 2020 (to be published). [Bibtex]
    @inproceedings{dubey2020a,
    abstract = {Unlike optical imaging, it’s difficult to extract descriptive features from radar data for problems like classification of different targets. This paper takes the advantage of different neural network based architectures such as convolutional neural networks and long-short term memory to propose an end-to-end framework for classification of vulnerable road users. To make the network’s prediction more reliable for automotive applications, a new concept of network uncertainty is introduced to the defined architectures. The signal processing tool chain described in this paper achieves higher accuracy than state-of-the-art algorithms while maintaining latency requirement for automotive applications.},
    author = {Dubey, Anand and Fuchs, Jonas and Reißland, Torsten and Weigel, Robert and Lurz, Fabian},
    language = {English},
    booktitle = {IEEE Topical Conference on Wireless Sensors and Sensor Networks (WiSNet)},
    cris = {https://cris.fau.de/converis/publicweb/publication/229613648},
    year = {2020},
    month = {01},
    day = {26},
    eventdate = {2020-01-26/2020-01-29},
    faupublication = {yes},
    keywords = {Autonomous Driving,Automotive Radar,Deep Neural Networks,Mico-Doppler,VRU Classification.},
    note = {unpublished},
    peerreviewed = {automatic},
    title = {Uncertainty Analysis of Deep Neural Network for Classification of Vulnerable Road Users using micro-Doppler},
    venue = {San Antonio, Texas, USA},
    }

COPYRIGHT NOTICE: Copyright and all rights of the material above are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by the appropriate copyright. The material may not be reposted without the explicit permission of the copyright holder.

COPYRIGHT NOTICE FOR IEEE PUBLICATIONS: © IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

COPYRIGHT NOTICE FOR EUMA PUBLICATIONS: © EUMA. Personal use of this material is permitted. Permission from European Microwave Association(EUMA) must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.