Research on Automatic Target Recognition in High-Performance Reconfigurable Computing

高性能可重构计算中的自动目标识别研究

• 批准号: 0819919
• 负责人: Christopher Doss
• 金额: $ 4.99万
• 依托单位: North Carolina Agricultural & Technical State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-06-01 至 2010-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0819919&HistoricalAwards=false
• 关键词: Research; Automatic; Target; Recognition; Performance

The proposed work will support the research of the Center for High-Performance Reconfigurable Computing (CHREC) by using its tools for mapping algorithms onto reconfigurable hardware. The algorithms to be tested are the preprocessing and feature extraction portion of automatic target recognition (ATR) algorithms. The objective of the proposed work is study the computational complexities associated with hidden Markov model (HMM) algorithms when applied to ATR. The proposed research will allow investigation of application areas that are not supported by the current CHREC industry members. The proposed work will provide access to cutting edge tools for mapping signal processing algorithms onto reconfigurable hardware. This set of skills is becoming more prevalent in the telecommunication and defense industries. The proposed research will be performed at a Historically Black College/University (HBCU), and will utilize undergraduates who are underrepresented in the STEM fields and give them with exposure to research activities.

拟议的工作将通过使用其工具将算法映射到可重构硬件来支持高性能可重构计算中心（CHREC）的研究。 待测试的算法是自动目标识别（ATR）算法的预处理和特征提取部分。 本文的目的是研究应用于 ATR 时与隐马尔可夫模型 (HMM) 算法相关的计算复杂性。 拟议的研究将允许调查当前 CHREC 行业成员不支持的应用领域。 拟议的工作将提供最先进的工具，用于将信号处理算法映射到可重新配置的硬件上。 这套技能在电信和国防工业中变得越来越普遍。 拟议的研究将在一所传统黑人学院/大学 (HBCU) 中进行，并将利用在 STEM 领域代表性不足的本科生，让他们接触研究活动。