--- _id: '45899' author: - first_name: Alexander full_name: Boschmann, Alexander last_name: Boschmann - first_name: Lennart full_name: Clausing, Lennart id: '74287' last_name: Clausing orcid: 0000-0003-3789-6034 - first_name: Felix full_name: Jentzsch, Felix id: '55631' last_name: Jentzsch orcid: 0000-0003-4987-5708 - first_name: Hassan full_name: Ghasemzadeh Mohammadi, Hassan id: '61186' last_name: Ghasemzadeh Mohammadi - first_name: Marco full_name: Platzner, Marco id: '398' last_name: Platzner citation: ama: 'Boschmann A, Clausing L, Jentzsch F, Ghasemzadeh Mohammadi H, Platzner M. Flexible Industrial Analytics on Reconfigurable Systems-On-Chip. In: Haake C-J, Meyer auf der Heide F, Platzner M, Wachsmuth H, Wehrheim H, eds. On-The-Fly Computing -- Individualized IT-Services in Dynamic Markets. Vol 412. Verlagsschriftenreihe des Heinz Nixdorf Instituts. Heinz Nixdorf Institut, Universität Paderborn; 2023:225-236. doi:10.5281/zenodo.8068713' apa: Boschmann, A., Clausing, L., Jentzsch, F., Ghasemzadeh Mohammadi, H., & Platzner, M. (2023). Flexible Industrial Analytics on Reconfigurable Systems-On-Chip. In C.-J. Haake, F. Meyer auf der Heide, M. Platzner, H. Wachsmuth, & H. Wehrheim (Eds.), On-The-Fly Computing -- Individualized IT-services in dynamic markets (Vol. 412, pp. 225–236). Heinz Nixdorf Institut, Universität Paderborn. https://doi.org/10.5281/zenodo.8068713 bibtex: '@inbook{Boschmann_Clausing_Jentzsch_Ghasemzadeh Mohammadi_Platzner_2023, place={Paderborn}, series={Verlagsschriftenreihe des Heinz Nixdorf Instituts}, title={Flexible Industrial Analytics on Reconfigurable Systems-On-Chip}, volume={412}, DOI={10.5281/zenodo.8068713}, booktitle={On-The-Fly Computing -- Individualized IT-services in dynamic markets}, publisher={Heinz Nixdorf Institut, Universität Paderborn}, author={Boschmann, Alexander and Clausing, Lennart and Jentzsch, Felix and Ghasemzadeh Mohammadi, Hassan and Platzner, Marco}, editor={Haake, Claus-Jochen and Meyer auf der Heide, Friedhelm and Platzner, Marco and Wachsmuth, Henning and Wehrheim, Heike}, year={2023}, pages={225–236}, collection={Verlagsschriftenreihe des Heinz Nixdorf Instituts} }' chicago: 'Boschmann, Alexander, Lennart Clausing, Felix Jentzsch, Hassan Ghasemzadeh Mohammadi, and Marco Platzner. “Flexible Industrial Analytics on Reconfigurable Systems-On-Chip.” In On-The-Fly Computing -- Individualized IT-Services in Dynamic Markets, edited by Claus-Jochen Haake, Friedhelm Meyer auf der Heide, Marco Platzner, Henning Wachsmuth, and Heike Wehrheim, 412:225–36. Verlagsschriftenreihe Des Heinz Nixdorf Instituts. Paderborn: Heinz Nixdorf Institut, Universität Paderborn, 2023. https://doi.org/10.5281/zenodo.8068713.' ieee: 'A. Boschmann, L. Clausing, F. Jentzsch, H. Ghasemzadeh Mohammadi, and M. Platzner, “Flexible Industrial Analytics on Reconfigurable Systems-On-Chip,” in On-The-Fly Computing -- Individualized IT-services in dynamic markets, vol. 412, C.-J. Haake, F. Meyer auf der Heide, M. Platzner, H. Wachsmuth, and H. Wehrheim, Eds. Paderborn: Heinz Nixdorf Institut, Universität Paderborn, 2023, pp. 225–236.' mla: Boschmann, Alexander, et al. “Flexible Industrial Analytics on Reconfigurable Systems-On-Chip.” On-The-Fly Computing -- Individualized IT-Services in Dynamic Markets, edited by Claus-Jochen Haake et al., vol. 412, Heinz Nixdorf Institut, Universität Paderborn, 2023, pp. 225–36, doi:10.5281/zenodo.8068713. short: 'A. Boschmann, L. Clausing, F. Jentzsch, H. Ghasemzadeh Mohammadi, M. Platzner, in: C.-J. Haake, F. Meyer auf der Heide, M. Platzner, H. Wachsmuth, H. Wehrheim (Eds.), On-The-Fly Computing -- Individualized IT-Services in Dynamic Markets, Heinz Nixdorf Institut, Universität Paderborn, Paderborn, 2023, pp. 225–236.' date_created: 2023-07-07T08:36:58Z date_updated: 2023-07-07T11:15:06Z ddc: - '004' department: - _id: '7' doi: 10.5281/zenodo.8068713 editor: - first_name: Claus-Jochen full_name: Haake, Claus-Jochen last_name: Haake - first_name: Friedhelm full_name: Meyer auf der Heide, Friedhelm last_name: Meyer auf der Heide - first_name: Marco full_name: Platzner, Marco last_name: Platzner - first_name: Henning full_name: Wachsmuth, Henning last_name: Wachsmuth - first_name: Heike full_name: Wehrheim, Heike last_name: Wehrheim file: - access_level: open_access content_type: application/pdf creator: florida date_created: 2023-07-07T08:36:40Z date_updated: 2023-07-07T11:14:23Z file_id: '45900' file_name: T1-Chapter-SFB-Buch-Final.pdf file_size: 468973 relation: main_file file_date_updated: 2023-07-07T11:14:23Z has_accepted_license: '1' intvolume: ' 412' language: - iso: eng oa: '1' page: 225-236 place: Paderborn project: - _id: '1' grant_number: '160364472' name: 'SFB 901: SFB 901: On-The-Fly Computing - Individualisierte IT-Dienstleistungen in dynamischen Märkten ' - _id: '82' name: 'SFB 901 - T: SFB 901 - Project Area T' - _id: '83' name: 'SFB 901 - T1: SFB 901 -Subproject T1' publication: On-The-Fly Computing -- Individualized IT-services in dynamic markets publisher: Heinz Nixdorf Institut, Universität Paderborn series_title: Verlagsschriftenreihe des Heinz Nixdorf Instituts status: public title: Flexible Industrial Analytics on Reconfigurable Systems-On-Chip type: book_chapter user_id: '477' volume: 412 year: '2023' ... --- _id: '33990' abstract: - lang: eng text: Deep neural networks (DNNs) are penetrating into a broad spectrum of applications and replacing manual algorithmic implementations, including the radio frequency communications domain with classical signal processing algorithms. However, the high throughput (gigasamples per second) and low latency requirements of this application domain pose a significant hurdle for adopting computationally demanding DNNs. In this article, we explore highly specialized DNN inference accelerator approaches on field-programmable gate arrays (FPGAs) for RadioML modulation classification. Using an automated end-to-end flow for the generation of the FPGA solution, we can easily explore a spectrum of solutions that optimize for different design targets, including accuracy, power efficiency, resources, throughput, and latency. By leveraging reduced precision arithmetic and customized streaming dataflow, we demonstrate a solution that meets the application requirements and outperforms alternative FPGA efforts by 3.5x in terms of throughput. Against modern embedded graphics processing units (GPUs), we measure >10x higher throughput and >100x lower latency under comparable accuracy and power envelopes. author: - first_name: Felix full_name: Jentzsch, Felix id: '55631' last_name: Jentzsch orcid: 0000-0003-4987-5708 - first_name: Yaman full_name: Umuroglu, Yaman last_name: Umuroglu - first_name: Alessandro full_name: Pappalardo, Alessandro last_name: Pappalardo - first_name: Michaela full_name: Blott, Michaela last_name: Blott - first_name: Marco full_name: Platzner, Marco id: '398' last_name: Platzner citation: ama: 'Jentzsch F, Umuroglu Y, Pappalardo A, Blott M, Platzner M. RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures. IEEE Micro. 2022;42(6):125-133. doi:10.1109/MM.2022.3202091' apa: 'Jentzsch, F., Umuroglu, Y., Pappalardo, A., Blott, M., & Platzner, M. (2022). RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures. IEEE Micro, 42(6), 125–133. https://doi.org/10.1109/MM.2022.3202091' bibtex: '@article{Jentzsch_Umuroglu_Pappalardo_Blott_Platzner_2022, title={RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures}, volume={42}, DOI={10.1109/MM.2022.3202091}, number={6}, journal={IEEE Micro}, publisher={IEEE}, author={Jentzsch, Felix and Umuroglu, Yaman and Pappalardo, Alessandro and Blott, Michaela and Platzner, Marco}, year={2022}, pages={125–133} }' chicago: 'Jentzsch, Felix, Yaman Umuroglu, Alessandro Pappalardo, Michaela Blott, and Marco Platzner. “RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures.” IEEE Micro 42, no. 6 (2022): 125–33. https://doi.org/10.1109/MM.2022.3202091.' ieee: 'F. Jentzsch, Y. Umuroglu, A. Pappalardo, M. Blott, and M. Platzner, “RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures,” IEEE Micro, vol. 42, no. 6, pp. 125–133, 2022, doi: 10.1109/MM.2022.3202091.' mla: 'Jentzsch, Felix, et al. “RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures.” IEEE Micro, vol. 42, no. 6, IEEE, 2022, pp. 125–33, doi:10.1109/MM.2022.3202091.' short: F. Jentzsch, Y. Umuroglu, A. Pappalardo, M. Blott, M. Platzner, IEEE Micro 42 (2022) 125–133. date_created: 2022-11-03T14:42:16Z date_updated: 2023-04-04T15:09:17Z department: - _id: '78' doi: 10.1109/MM.2022.3202091 intvolume: ' 42' issue: '6' language: - iso: eng main_file_link: - url: https://ieeexplore.ieee.org/document/9933377 page: 125-133 project: - _id: '52' name: 'PC2: Computing Resources Provided by the Paderborn Center for Parallel Computing' publication: IEEE Micro publication_status: published publisher: IEEE status: public title: 'RadioML Meets FINN: Enabling Future RF Applications With FPGA Streaming Architectures' type: journal_article user_id: '55631' volume: 42 year: '2022' ... --- _id: '30908' author: - first_name: Hassan full_name: Ghasemzadeh Mohammadi, Hassan id: '61186' last_name: Ghasemzadeh Mohammadi - first_name: Felix full_name: Jentzsch, Felix id: '55631' last_name: Jentzsch orcid: 0000-0003-4987-5708 - first_name: Maurice full_name: Kuschel, Maurice last_name: Kuschel - first_name: 'Rahil ' full_name: 'Arshad, Rahil ' last_name: Arshad - first_name: Sneha full_name: Rautmare, Sneha last_name: Rautmare - first_name: Suraj full_name: Manjunatha, Suraj last_name: Manjunatha - first_name: Marco full_name: Platzner, Marco id: '398' last_name: Platzner - first_name: Alexander full_name: Boschmann, Alexander last_name: Boschmann - first_name: 'Dirk ' full_name: 'Schollbach, Dirk ' last_name: Schollbach citation: ama: 'Ghasemzadeh Mohammadi H, Jentzsch F, Kuschel M, et al. FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics. In: Machine Learning and Principles and Practice of Knowledge Discovery in Databases. Springer; 2021. doi:https://doi.org/10.1007/978-3-030-93736-2_27' apa: 'Ghasemzadeh Mohammadi, H., Jentzsch, F., Kuschel, M., Arshad, R., Rautmare, S., Manjunatha, S., Platzner, M., Boschmann, A., & Schollbach, D. (2021). FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics. Machine Learning and Principles and Practice of Knowledge Discovery in Databases. https://doi.org/10.1007/978-3-030-93736-2_27' bibtex: '@inproceedings{Ghasemzadeh Mohammadi_Jentzsch_Kuschel_Arshad_Rautmare_Manjunatha_Platzner_Boschmann_Schollbach_2021, title={FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics}, DOI={https://doi.org/10.1007/978-3-030-93736-2_27}, booktitle={ Machine Learning and Principles and Practice of Knowledge Discovery in Databases}, publisher={Springer}, author={Ghasemzadeh Mohammadi, Hassan and Jentzsch, Felix and Kuschel, Maurice and Arshad, Rahil and Rautmare, Sneha and Manjunatha, Suraj and Platzner, Marco and Boschmann, Alexander and Schollbach, Dirk }, year={2021} }' chicago: 'Ghasemzadeh Mohammadi, Hassan, Felix Jentzsch, Maurice Kuschel, Rahil Arshad, Sneha Rautmare, Suraj Manjunatha, Marco Platzner, Alexander Boschmann, and Dirk Schollbach. “FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics.” In Machine Learning and Principles and Practice of Knowledge Discovery in Databases. Springer, 2021. https://doi.org/10.1007/978-3-030-93736-2_27.' ieee: 'H. Ghasemzadeh Mohammadi et al., “FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics,” 2021, doi: https://doi.org/10.1007/978-3-030-93736-2_27.' mla: 'Ghasemzadeh Mohammadi, Hassan, et al. “FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics.” Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Springer, 2021, doi:https://doi.org/10.1007/978-3-030-93736-2_27.' short: 'H. Ghasemzadeh Mohammadi, F. Jentzsch, M. Kuschel, R. Arshad, S. Rautmare, S. Manjunatha, M. Platzner, A. Boschmann, D. Schollbach, in: Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Springer, 2021.' date_created: 2022-04-18T10:16:55Z date_updated: 2023-09-15T15:09:07Z department: - _id: '78' doi: https://doi.org/10.1007/978-3-030-93736-2_27 language: - iso: eng project: - _id: '83' name: 'SFB 901 - T1: SFB 901 -Subproject T1' - _id: '1' grant_number: '160364472' name: 'SFB 901: SFB 901: On-The-Fly Computing - Individualisierte IT-Dienstleistungen in dynamischen Märkten ' - _id: '82' name: 'SFB 901 - T: SFB 901 - Project Area T' publication: ' Machine Learning and Principles and Practice of Knowledge Discovery in Databases' publisher: Springer status: public title: 'FLight: FPGA Acceleration of Lightweight DNN Model Inference in Industrial Analytics' type: conference user_id: '477' year: '2021' ...