近传感器处理中基于注意力的安全特征提取:正在研究中

2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS) Pub Date : 2020-09-20 DOI:10.1109/CODESISSS51650.2020.9244036

Pankaj Bhowmik, Md Jubaer Hossain Pantho, S. Saha, C. Bobda

{"title":"近传感器处理中基于注意力的安全特征提取:正在研究中","authors":"Pankaj Bhowmik, Md Jubaer Hossain Pantho, S. Saha, C. Bobda","doi":"10.1109/CODESISSS51650.2020.9244036","DOIUrl":null,"url":null,"abstract":"This paper presents a secure hardware architecture of an image sensor to accelerate feature extraction using region-level parallelism. For each logical region, the design includes a region processing unit (RPU) with an attention module (AM). The AM activates the processing in the RPU if there are no spatiotemporal redundancies. It reduces power consumption and data volume by utilizing the concepts of predictive coding. Also, every RPU has a crypto-core driven by the AM to withstand against adversaries. Simulation results show we can save 89.70% power with a significant speedup.","PeriodicalId":437802,"journal":{"name":"2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Attention-Based Secure Feature Extraction in Near Sensor Processing: Work-in-Progress\",\"authors\":\"Pankaj Bhowmik, Md Jubaer Hossain Pantho, S. Saha, C. Bobda\",\"doi\":\"10.1109/CODESISSS51650.2020.9244036\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents a secure hardware architecture of an image sensor to accelerate feature extraction using region-level parallelism. For each logical region, the design includes a region processing unit (RPU) with an attention module (AM). The AM activates the processing in the RPU if there are no spatiotemporal redundancies. It reduces power consumption and data volume by utilizing the concepts of predictive coding. Also, every RPU has a crypto-core driven by the AM to withstand against adversaries. Simulation results show we can save 89.70% power with a significant speedup.\",\"PeriodicalId\":437802,\"journal\":{\"name\":\"2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)\",\"volume\":\"11 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-09-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/CODESISSS51650.2020.9244036\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CODESISSS51650.2020.9244036","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

本文提出了一种安全的图像传感器硬件结构，利用区域级并行性来加速特征提取。对于每个逻辑区域，该设计包括一个带有注意模块(AM)的区域处理单元(RPU)。如果没有时空冗余，AM将激活RPU中的处理。它通过利用预测编码的概念来降低功耗和数据量。此外，每个RPU都有一个由AM驱动的加密核心，以抵御对手。仿真结果表明，该方法可以节省89.70%的功耗，并有显著的加速。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Attention-Based Secure Feature Extraction in Near Sensor Processing: Work-in-Progress

This paper presents a secure hardware architecture of an image sensor to accelerate feature extraction using region-level parallelism. For each logical region, the design includes a region processing unit (RPU) with an attention module (AM). The AM activates the processing in the RPU if there are no spatiotemporal redundancies. It reduces power consumption and data volume by utilizing the concepts of predictive coding. Also, every RPU has a crypto-core driven by the AM to withstand against adversaries. Simulation results show we can save 89.70% power with a significant speedup.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2020 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)

自引率

0.00%

发文量