高级检索
    张克环 任小西 李仁发 凌纯清. 频率直接数字合成器的一种存储优化方法研究[J]. 计算机研究与发展, 2010, 47(5): 956-961.
    引用本文: 张克环 任小西 李仁发 凌纯清. 频率直接数字合成器的一种存储优化方法研究[J]. 计算机研究与发展, 2010, 47(5): 956-961.
    Zhang Kehuan, Ren Xiaoxi, Li Renfa, and Ling Chunqing. A Storage Optimization Method for Frequency Direct Digital Synthesizer[J]. Journal of Computer Research and Development, 2010, 47(5): 956-961.
    Citation: Zhang Kehuan, Ren Xiaoxi, Li Renfa, and Ling Chunqing. A Storage Optimization Method for Frequency Direct Digital Synthesizer[J]. Journal of Computer Research and Development, 2010, 47(5): 956-961.

    频率直接数字合成器的一种存储优化方法研究

    A Storage Optimization Method for Frequency Direct Digital Synthesizer

    • 摘要: 直接数字频率合成(DDS)是一种先进的频率合成技术.正弦相位-振幅映射是DDS的核心部件之一.正弦查找表方案需要较多片上存储器资源,成本较高,因而,提出了一种基于相邻采样点振幅增量的改进方案.分析改进方案的资源开销,介绍其实现技术,并以FPGA为开发平台实现了改进方案.从功能、误差、最高工作频率、资源开销等方面对改进方案进行了评估.结果表明:改进方案以较小硬件资源额外开销为代价,节约了大量片上存储器资源,同时保证最高工作频率不变.

       

      Abstract: Direct digital synthesis (DDS) is one type of advanced frequency synthesizing technology which uses the sine phase-amplitude mapping table stored in the read-only memory (ROM) as its core component to convert the phase value into amplitude value. However, this look-up table is large and will occupy lots of precious on-chip transistor resources. In order to reduce the consumption of on-chip memory resource, proposed in this paper is a new and optimized mapping scheme which stores only the amplitude difference between two adjacent phase sampling points instead of the full values, and uses an accumulator to add this amplitude difference with the amplitude at previous phase point. Analysis results show that the new scheme uses only twenty percent of the resources used by current technology. The implementation technology is discussed and finally a prototype is built on FPGA to demonstrate the feasibility and the performance. The max frequency is also evaluated on the prototype system. The evaluation results show that: when compared with the traditional mapping and storage scheme, the proposed new scheme can save a large amount of on-chip RAM resource, and is able to achieve the same maximum working frequency, but with a minimum cost by requiring only a little more logical resource.

       

    /

    返回文章
    返回