{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:09:20Z","timestamp":1791392960383,"version":"4.3.3"},"reference-count":35,"publisher":"Association for Computing Machinery (ACM)","issue":"5","license":[{"start":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T00:00:00Z","timestamp":1787702400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["U23A20317, 62572182, 62576051"],"award-info":[{"award-number":["U23A20317, 62572182, 62576051"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100004735","name":"Natural Science Foundation of Hunan Province","doi-asserted-by":"crossref","award":["2023JJ10016"],"award-info":[{"award-number":["2023JJ10016"]}],"id":[{"id":"10.13039\/501100004735","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2026,9,30]]},"abstract":"<jats:p>Convolutional Neural Network (CNN) has achieved great success in various fields of machine vision, such as image classification and recognition, image segmentation, and video analysis. In specific applications, it is often necessary to customize the network structure by altering image size, convolution kernel size, pooling size, network architecture, and the number of network layers. These customizations pose significant challenges to the architecture of CNN accelerator especially in real-time systems, and as the number of these network variables increases, the internal convolution calculations become extremely demanding. The inference speed and energy consumption of CNN are becoming more and more important, which requires the design of the accelerator to adapt to different network architecture and run efficiently. Field-programmable gate arrays (FPGAs) is an ideal choice for a CNN accelerator due to its high programmability and low power consumption. To address the aforementioned challenges, we present NRUCA, a novel network reconfigurable universal CNN accelerator based on FPGA. On one hand, we have designed a flexible architecture that can dynamically configure CNN\u2019s network structure parameters. These parameters can be sent to the FPGA via a configuration file, enabling the same design to run different CNN networks and allowing to modify on the fly. This forms the basis of the architecture\u2019s adaptability presented in this article. On the other hand, we employ a multi-channel parallel and efficient pipeline matrix multiplication architecture to implement the convolution and fully-connected layers, which constitute the majority of the computational load in CNNs. By utilizing an innovative inverted matrix multiplication algorithm and an interleaving cache data method, we reduce the internal cache required for multi-channel data. Furthermore, most of the intermediate calculation data does not need to be output to DDR, significantly improving the operation efficiency of the accelerator. We also fully leverage the FPGA chip architecture to compile multiple calculation kernels, which can be flexibly scheduled and combined to cater to various application scenarios. Based on the flexible and efficient overall design architecture, the layout of various resources of FPGA is balanced, which enables the designed FPGA project to compile four acceleration kernels with the running clock up to 300MHz in Xilinx Vitis and test them on Xilinx Alveo U250. We verified the adaptability and acceleration capabilities of our FPGA project using LeNet, AlexNet, VGG11, VGG13, VGG16, and VGG19 networks. Experiments demonstrated that our FPGA architecture achieves 33X and 35X energy savings compared with the Intel Xeon 5220R CPU, and 1.08X and 1.05X energy savings compared with the Nvidia Tesla P100 GPU, when accelerating AlexNet and VGG16, respectively.<\/jats:p>","DOI":"10.1145\/3779218","type":"journal-article","created":{"date-parts":[[2026,1,23]],"date-time":"2026-01-23T21:08:37Z","timestamp":1769202517000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["A Design of Network Reconfigurable Universal CNN Accelerator Based on FPGA"],"prefix":"10.1145","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4764-6103","authenticated-orcid":false,"given":"Wenhua","family":"Ye","sequence":"first","affiliation":[{"name":"College of Information Science and Engineering, Hunan University","place":["Changsha, China"]},{"name":"China Electronics Technology Group Corporation 36th Research Institute","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5505-9989","authenticated-orcid":false,"given":"Huan","family":"Li","sequence":"additional","affiliation":[{"name":"College of Information Science and Engineering, Hunan University","place":["Changsha, China"]},{"name":"China Electronics Technology Group Corporation 36th Research Institute","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5871-9136","authenticated-orcid":false,"given":"Xu","family":"Zhou","sequence":"additional","affiliation":[{"name":"College of Information Science and Engineering, Hunan University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0884-4051","authenticated-orcid":false,"given":"Huilong","family":"Pi","sequence":"additional","affiliation":[{"name":"College of Information Science and Engineering, Hunan University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2635-7716","authenticated-orcid":false,"given":"Kenli","family":"Li","sequence":"additional","affiliation":[{"name":"College of Information Science and Engineering, Hunan University","place":["Changsha, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,8,26]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/3020078.3021738"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSII.2018.2865896"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1145\/3570928"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/3385414"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1145\/3519598"},{"key":"e_1_3_2_7_2","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1109\/FCCM51124.2021.00029","volume-title":"2021 IEEE 29th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","author":"Deng Huipeng","year":"2021","unstructured":"Huipeng Deng, Jian Wang, Huafeng Ye, Shanlin Xiao, Xiangyu Meng, and Zhiyi Yu. 2021. 3D-VNPU: A flexible accelerator for 2D\/3D CNNs on FPGA. In 2021 IEEE 29th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). IEEE, 181\u2013185."},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1201\/9781003162810-13"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2017.25"},{"key":"e_1_3_2_10_2","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1109\/FCCM51124.2021.00026","volume-title":"2021 IEEE 29th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","author":"Jiang Chao","year":"2021","unstructured":"Chao Jiang, David Ojika, Bhavesh Patel, and Herman Lam. 2021. Optimized FPGA-based deep learning accelerator for sparse CNN using high bandwidth memory. In 2021 IEEE 29th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). IEEE, 157\u2013164."},{"key":"e_1_3_2_11_2","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky Alex","year":"2012","unstructured":"Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. 2012. Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems 25 (2012), 1\u20139.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2824280"},{"issue":"5","key":"e_1_3_2_14_2","first-page":"1436","article-title":"Block convolution: Toward memory-efficient inference of large-scale CNNs on FPGA","volume":"41","author":"Li Gang","year":"2021","unstructured":"Gang Li, Zejian Liu, Fanrong Li, and Jian Cheng. 2021. Block convolution: Toward memory-efficient inference of large-scale CNNs on FPGA. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 41, 5 (2021), 1436\u20131447.","journal-title":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems"},{"key":"e_1_3_2_15_2","first-page":"1","volume-title":"2016 26th International Conference on Field Programmable Logic and Applications (FPL)","author":"Li Huimin","year":"2016","unstructured":"Huimin Li, Xitian Fan, Li Jiao, Wei Cao, Xuegong Zhou, and Lingli Wang. 2016. A high performance FPGA-based accelerator for large-scale convolutional neural networks. In 2016 26th International Conference on Field Programmable Logic and Applications (FPL). IEEE, 1\u20139."},{"key":"e_1_3_2_16_2","first-page":"1","volume-title":"FSP Workshop 2018; Fifth International Workshop on FPGAs for Software Programmers","author":"Noronha Daniel H.","year":"2018","unstructured":"Daniel H. Noronha, Bahar Salehpour, and Steven J. E. Wilton. 2018. LeFlow: Enabling flexible FPGA high-level synthesis of tensorflow deep neural networks. In FSP Workshop 2018; Fifth International Workshop on FPGAs for Software Programmers. VDE, 1\u20138."},{"issue":"11","key":"e_1_3_2_17_2","first-page":"1","article-title":"Accelerating deep convolutional neural networks using specialized hardware","volume":"2","author":"Ovtcharov Kalin","year":"2015","unstructured":"Kalin Ovtcharov, Olatunji Ruwase, Joo-Young Kim, Jeremy Fowers, Karin Strauss, and Eric S. Chung. 2015. Accelerating deep convolutional neural networks using specialized hardware. Microsoft Research Whitepaper 2, 11 (2015), 1\u20134.","journal-title":"Microsoft Research Whitepaper"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD.2013.6657019"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847265"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASAP.2009.25"},{"key":"e_1_3_2_21_2","doi-asserted-by":"publisher","DOI":"10.1145\/3140659.3080221"},{"key":"e_1_3_2_22_2","unstructured":"Karen Simonyan and Andrew Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556. Retrieved from https:\/\/arxiv.org\/abs\/1409.1556"},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847276"},{"issue":"1","key":"e_1_3_2_24_2","doi-asserted-by":"crossref","first-page":"208","DOI":"10.1109\/TNNLS.2022.3172941","article-title":"EDCompress: Energy-aware model compression for dataflows","volume":"35","author":"Wang Zhehui","year":"2022","unstructured":"Zhehui Wang, Tao Luo, Rick Siow Mong Goh, and Joey Tianyi Zhou. 2022. EDCompress: Energy-aware model compression for dataflows. IEEE Transactions on Neural Networks and Learning Systems 35, 1 (2022), 208\u2013220.","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1145\/3061639.3062207"},{"issue":"1","key":"e_1_3_2_26_2","first-page":"1","article-title":"Low-precision floating-point arithmetic for high-performance FPGA-based CNN acceleration","volume":"15","author":"Wu Chen","year":"2021","unstructured":"Chen Wu, Mingyu Wang, Xinyuan Chu, Kun Wang, and Lei He. 2021. Low-precision floating-point arithmetic for high-performance FPGA-based CNN acceleration. ACM Transactions on Reconfigurable Technology and Systems (TRETS) 15, 1 (2021), 1\u201321.","journal-title":"ACM Transactions on Reconfigurable Technology and Systems (TRETS)"},{"key":"e_1_3_2_27_2","volume-title":"Vitis High-Level Synthesis User Guide","year":"2022","unstructured":"Xilinx, Inc. 2022. Vitis High-Level Synthesis User Guide. Xilinx, Inc., San Jose, CA, USA. [Online]. Retrieved from https:\/\/docs.amd.com\/r\/2022.2-English\/ug1399-vitis-hls"},{"key":"e_1_3_2_28_2","volume-title":"Vitis Unified Software Platform Documentation","year":"2022","unstructured":"Xilinx, Inc. 2022. Vitis Unified Software Platform Documentation. Xilinx, Inc., San Jose, CA, USA. [Online]. Retrieved from https:\/\/docs.amd.com\/r\/2022.2-English\/ug1393-vitis-application-acceleration"},{"key":"e_1_3_2_29_2","volume-title":"Alveo U250 Data Center Accelerator Card User Guide","year":"2023","unstructured":"Xilinx, Inc. 2023. Alveo U250 Data Center Accelerator Card User Guide. Xilinx, Inc., San Jose, CA, USA. [Online]. Retrieved from https:\/\/docs.amd.com\/r\/en-US\/ug1301-getting-started-guide-alveo-accelerator-cards"},{"key":"e_1_3_2_30_2","doi-asserted-by":"crossref","first-page":"90","DOI":"10.1109\/FCCM60383.2024.00019","volume-title":"2024 IEEE 32nd Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","author":"Yang Shuxin","year":"2024","unstructured":"Shuxin Yang, Chenchen Ding, Mingqiang Huang, Kai Li, Chenghao Li, Zikun Wei, Sixiao Huang, Jingyao Dong, Liuyang Zhang, and Hao Yu. 2024. LAMPS: A layer-wised mixed-precision-and-sparsity accelerator for NAS-optimized CNNs on FPGA. In 2024 IEEE 32nd Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM). IEEE, 90\u201396."},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/2684746.2689060"},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2017.2785257"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2019.8702071"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3240765.3240801"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1145\/3400302.3415609"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2020.3002779"}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3779218","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T15:41:42Z","timestamp":1787758902000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3779218"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,26]]},"references-count":35,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,9,30]]}},"alternative-id":["10.1145\/3779218"],"URL":"https:\/\/doi.org\/10.1145\/3779218","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"value":"1539-9087","type":"print"},{"value":"1558-3465","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,26]]},"assertion":[{"value":"2025-01-17","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-13","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-26","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}