{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,8]],"date-time":"2026-04-08T16:39:18Z","timestamp":1775666358865,"version":"3.50.1"},"reference-count":18,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2024,7,31]],"date-time":"2024-07-31T00:00:00Z","timestamp":1722384000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,7,31]],"date-time":"2024-07-31T00:00:00Z","timestamp":1722384000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Real-Time Image Proc"],"published-print":{"date-parts":[[2024,8]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>In-memory computing (IMC) represents a promising approach to reducing latency and enhancing the energy efficiency of operations required for calculating convolution products of images. This study proposes a fully differential current-mode architecture for computing image convolutions across all four quadrants, intended for deep learning applications within CMOS imagers utilizing IMC near the CMOS sensor. This architecture processes analog signals provided by a CMOS sensor without the need for analog-to-digital conversion. Furthermore, it eliminates the necessity for data transfer between memory and analog operators as convolutions are computed within modified SRAM memory. The paper suggests modifying the structure of a CMOS SRAM cell by incorporating transistors capable of performing multiplications between binary (\u22121 or +1) weights and analog signals. Modified SRAM cells can be interconnected to sum the multiplication results obtained from individual cells. This approach facilitates connecting current inputs to different SRAM cells, offering highly scalable and parallelized calculations. For this study, a configurable module comprising nine modified SRAM cells with peripheral circuitry has been designed to calculate the convolution product on each pixel of an image using a <jats:inline-formula><jats:alternatives><jats:tex-math>$$3 \\times 3$$<\/jats:tex-math><mml:math xmlns:mml=\"http:\/\/www.w3.org\/1998\/Math\/MathML\">\n                  <mml:mrow>\n                    <mml:mn>3<\/mml:mn>\n                    <mml:mo>\u00d7<\/mml:mo>\n                    <mml:mn>3<\/mml:mn>\n                  <\/mml:mrow>\n                <\/mml:math><\/jats:alternatives><\/jats:inline-formula> mask with binary values (\u22121 or 1). Subsequently, an IMC module has been designed to perform 16 convolution operations in parallel, with input currents shared among the 16 modules. This configuration enables the computation of 16 convolutions simultaneously, processing a column per cycle. A digital control circuit manages both the readout or memorization of digital weights, as well as the multiply and add operations in real-time. The architecture underwent testing by performing convolutions between binary masks of 3\u2009\u00d7\u20093 values and images of 32\u2009\u00d7\u200932 pixels to assess accuracy and scalability when two IMC modules are vertically integrated. Convolution weights are stored locally as 1-bit digital values. The circuit was synthesized in 180\u00a0nm CMOS technology, and simulation results indicate its capability to perform a complete convolution in 3.2\u00a0ms, achieving an efficiency of 11,522 1-b TOPS\/W (1-b tera-operations per second per watt) with a similarity to ideal processing of 96%.<\/jats:p>","DOI":"10.1007\/s11554-024-01520-x","type":"journal-article","created":{"date-parts":[[2024,7,31]],"date-time":"2024-07-31T19:21:47Z","timestamp":1722453707000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["A real-time and energy-efficient SRAM with mixed-signal in-memory computing near CMOS sensors"],"prefix":"10.1007","volume":"21","author":[{"given":"Jose-Angel","family":"Diaz-Madrid","sequence":"first","affiliation":[]},{"given":"Gines","family":"Domenech-Asensi","sequence":"additional","affiliation":[]},{"given":"Ramon","family":"Ruiz-Merino","sequence":"additional","affiliation":[]},{"given":"Juan-Francisco","family":"Zapata-Perez","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2024,7,31]]},"reference":[{"issue":"3","key":"1520_CR1","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1109\/MSSC.2019.2922889","volume":"11","author":"N Verma","year":"2019","unstructured":"Verma, N., et al.: In-memory computing: advances and prospects. IEEE Solid-State Circuits Mag. 11(3), 43\u201355 (2019). https:\/\/doi.org\/10.1109\/MSSC.2019.2922889","journal-title":"IEEE Solid-State Circuits Mag."},{"key":"1520_CR2","doi-asserted-by":"publisher","unstructured":"Shanbhag, N.R., Roy, S.K.: Comprehending in-memory computing trends via proper benchmarking. In: 2022 IEEE Custom Integrated Circuits Conference (CICC), Newport Beach, CA, USA, 2022, pp. 1\u20137. https:\/\/doi.org\/10.1109\/CICC53496.2022.9772817","DOI":"10.1109\/CICC53496.2022.9772817"},{"issue":"2","key":"1520_CR3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3365837","volume":"16","author":"HAD Nguyen","year":"2020","unstructured":"Nguyen, H.A.D., Yu, J., Lebdeh, M.A., Taouil, M., Hamdioui, S., Catthoor, F.: A classification of memory-centric computing. ACM J. Emerg. Technol. Comput. Syst. 16(2), 1\u201326 (2020). https:\/\/doi.org\/10.1145\/3365837","journal-title":"ACM J. Emerg. Technol. Comput. Syst."},{"issue":"3","key":"1520_CR4","doi-asserted-by":"publisher","first-page":"031401","DOI":"10.1088\/1674-4926\/43\/3\/031401","volume":"43","author":"Z Lin","year":"2022","unstructured":"Lin, Z., Tong, Z., Zhang, J., Wang, F., Xu, T., Zhao, Y., Wu, X., Peng, C., Lu, W., Zhao, Q.: A review on SRAM-based computing in-memory: circuits, functions, and applications. J. Semicond. 43(3), 031401 (2022). https:\/\/doi.org\/10.1088\/1674-4926\/43\/3\/031401","journal-title":"J. Semicond."},{"issue":"6","key":"1520_CR5","doi-asserted-by":"publisher","first-page":"1789","DOI":"10.1109\/JSSC.2019.2899730","volume":"54","author":"H Valavi","year":"2019","unstructured":"Valavi, H., Ramadge, P.J., Nestler, E., Verma, N.: A 64-tile 2.4-Mb in-memory-computing CNN accelerator employing charge-domain compute. IEEE J. Solid-State Circuits 54(6), 1789\u20131799 (2019). https:\/\/doi.org\/10.1109\/JSSC.2019.2899730","journal-title":"IEEE J. Solid-State Circuits"},{"issue":"1","key":"1520_CR6","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1109\/JSSC.2018.2880918","volume":"54","author":"A Biswas","year":"2019","unstructured":"Biswas, A., Chandrakasan, A.P.: CONV-SRAM: an energy-efficient SRAM with in-memory dot-product computation for low-power convolutional neural networks. IEEE J. Solid-State Circuits 54(1), 217\u2013230 (2019). https:\/\/doi.org\/10.1109\/JSSC.2018.2880918","journal-title":"IEEE J. Solid-State Circuits"},{"issue":"6","key":"1520_CR7","doi-asserted-by":"publisher","first-page":"1924","DOI":"10.1109\/JSSC.2021.3056447","volume":"56","author":"Z Chen","year":"2021","unstructured":"Chen, Z., et al.: CAP-RAM: a charge-domain in-memory computing 6T-SRAM for accurate and precision-programmable CNN inference. IEEE J. Solid-State Circuits 56(6), 1924\u20131935 (2021). https:\/\/doi.org\/10.1109\/JSSC.2021.3056447","journal-title":"IEEE J. Solid-State Circuits"},{"issue":"8","key":"1520_CR8","doi-asserted-by":"publisher","first-page":"2521","DOI":"10.1109\/TCSI.2020.2981901","volume":"67","author":"M Ali","year":"2020","unstructured":"Ali, M., Jaiswal, A., Kodge, S., Agrawal, A., Chakraborty, I., Roy, K.: IMAC: in-memory multi-bit multiplication and accumulation in 6T SRAM array. IEEE Trans. Circuits Syst. I Regul. Pap. 67(8), 2521\u20132531 (2020). https:\/\/doi.org\/10.1109\/TCSI.2020.2981901","journal-title":"IEEE Trans. Circuits Syst. I Regul. Pap."},{"issue":"3","key":"1520_CR9","doi-asserted-by":"publisher","first-page":"710","DOI":"10.1109\/JSEN.2013.2286823","volume":"14","author":"R Njuguna","year":"2014","unstructured":"Njuguna, R., Gruev, V.: Current-mode CMOS imaging sensor with velocity saturation mode of operation and feedback mechanism. IEEE Sens. J. 14(3), 710\u2013721 (2014). https:\/\/doi.org\/10.1109\/JSEN.2013.2286823","journal-title":"IEEE Sens. J."},{"key":"1520_CR10","doi-asserted-by":"publisher","unstructured":"D\u00edaz-Madrid, J.A., Asensi, G.D., Ruiz-Merino, R., Zapata-P\u00e9rez, J., Mart\u00ednez-\u00c1lvarez, J.J.: Mixed signal multiply and adder parallel circuit for deep learning convolution operations. In: 2020 IEEE International Symposium on Circuits and Systems (ISCAS), Seville, Spain, 2020, pp. 1\u20135. https:\/\/doi.org\/10.1109\/ISCAS45731.2020.9180757","DOI":"10.1109\/ISCAS45731.2020.9180757"},{"key":"1520_CR11","doi-asserted-by":"publisher","unstructured":"Chen, T., Botimer, J., Chou, T., Zhang, Z.: An Sram-based accelerator for solving partial differential equations. In: 2019 IEEE Custom Integrated Circuits Conference (CICC), Austin, TX, USA, 2019, pp. 1\u20134. https:\/\/doi.org\/10.1109\/CICC.2019.8780234","DOI":"10.1109\/CICC.2019.8780234"},{"key":"1520_CR12","doi-asserted-by":"publisher","unstructured":"Papistas, I.A., et al.: A 22 nm, 1540 TOP\/s\/W, 12.1 TOP\/s\/mm2 in-memory analog matrix-vector-multiplier for DNN acceleration. In: 2021 IEEE Custom Integrated Circuits Conference (CICC), Austin, TX, USA, 2021, pp. 1\u20132. https:\/\/doi.org\/10.1109\/CICC51472.2021.9431575","DOI":"10.1109\/CICC51472.2021.9431575"},{"key":"1520_CR13","doi-asserted-by":"publisher","unstructured":"Zhao, Y., Zhang, M., He, P., Yan, Z., Chan, C.-H., Martins, R.P.: A double-mode sparse compute-in-memory macro with reconfigurable single and dual layer computation. In: 2023 IEEE Custom Integrated Circuits Conference (CICC), San Antonio, TX, USA, 2023, pp. 1\u20132. https:\/\/doi.org\/10.1109\/CICC57935.2023.10121308","DOI":"10.1109\/CICC57935.2023.10121308"},{"key":"1520_CR14","doi-asserted-by":"publisher","unstructured":"Ueyoshi, K., et al.: DIANA: an end-to-end energy-efficient digital and ANAlog hybrid neural network SoC. In: 2022 IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, USA, 2022, pp. 1\u20133. https:\/\/doi.org\/10.1109\/ISSCC42614.2022.9731716","DOI":"10.1109\/ISSCC42614.2022.9731716"},{"key":"1520_CR15","doi-asserted-by":"publisher","unstructured":"Lee, J., Valavi, H., Tang, Y., Verma, N.: Fully row\/column-parallel in-memory computing SRAM macro employing capacitor-based mixed-signal computation with 5-b inputs. In: 2021 Symposium on VLSI Circuits, Kyoto, Japan, 2021, pp. 1\u20132. https:\/\/doi.org\/10.23919\/VLSICircuits52068.2021.9492444","DOI":"10.23919\/VLSICircuits52068.2021.9492444"},{"key":"1520_CR16","doi-asserted-by":"publisher","unstructured":"Dong, Q., et al.: 15.3 A 351 TOPS\/W and 372.4 GOPS compute-in-memory SRAM macro in 7 nm FinFET CMOS for machine-learning applications. In: 2020 IEEE International Solid-State Circuits Conference\u2014(ISSCC), San Francisco, CA, USA, 2020, pp. 242\u2013244. https:\/\/doi.org\/10.1109\/ISSCC19947.2020.9062985","DOI":"10.1109\/ISSCC19947.2020.9062985"},{"key":"1520_CR17","doi-asserted-by":"publisher","unstructured":"Hsieh, S.-E., et al.: 7.6 A 70.85\u201386.27 TOPS\/W PVT-insensitive 8b word-wise ACIM with post-processing relaxation. In: 2023 IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, USA, 2023, pp. 136\u2013138. https:\/\/doi.org\/10.1109\/ISSCC42615.2023.10067335","DOI":"10.1109\/ISSCC42615.2023.10067335"},{"key":"1520_CR18","doi-asserted-by":"publisher","unstructured":"Yue, J., et al.: A 5.6\u201389.9 TOPS\/W heterogeneous computing-in-memory SoC with high-utilization producer-consumer architecture and high-frequency read-free CIM macro. In: 2023 IEEE Symposium on VLSI Technology and Circuits (VLSI Technology and Circuits), Kyoto, Japan, 2023, pp. 1\u20132. https:\/\/doi.org\/10.23919\/VLSITechnologyandCir57934.2023.10185315","DOI":"10.23919\/VLSITechnologyandCir57934.2023.10185315"}],"container-title":["Journal of Real-Time Image Processing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11554-024-01520-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11554-024-01520-x\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11554-024-01520-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,27]],"date-time":"2024-08-27T16:02:26Z","timestamp":1724774546000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11554-024-01520-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,7,31]]},"references-count":18,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2024,8]]}},"alternative-id":["1520"],"URL":"https:\/\/doi.org\/10.1007\/s11554-024-01520-x","relation":{},"ISSN":["1861-8200","1861-8219"],"issn-type":[{"value":"1861-8200","type":"print"},{"value":"1861-8219","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,7,31]]},"assertion":[{"value":"21 February 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"17 July 2024","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"31 July 2024","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"143"}}