CAJ | 학술논문

　　该文针对现有大点数快速傅里叶变换(FFT)在TS201处理器上的实现没有充分考虑Cache丢失对执行效率影响的问题，提出了改进型 Winograd 算法的实现方法。该改进型方法通过优化行列读取方法，最大程度利用Cache的读写特点，避免了三次显性转置；并通过重构蝶形运算，隐藏了乘铰链因子。实例测试与现有处理方法对比结果表明，Cache优化的大点数FFT执行速度有了明显提高，可用于雷达处理系统中的脉冲压缩的快速实现。
　　해문침대현유대점수쾌속부리협변환(FFT)재TS201처리기상적실현몰유충분고필Cache주실대집행효솔영향적문제，제출료개진형 Winograd 산법적실현방법。해개진형방법통과우화행렬독취방법，최대정도이용Cache적독사특점，피면료삼차현성전치；병통과중구접형운산，은장료승교련인자。실례측시여현유처리방법대비결과표명，Cache우화적대점수FFT집행속도유료명현제고，가용우뢰체처리계통중적맥충압축적쾌속실현。
This paper proposes an improved method for Winograd algorithm to solve the problem that the existing methods of long sequences Fast Fourier Transform (FFT) on the TS201 processor does not take full account of the Cache’s miss influence on efficiency. The new method makes maximum use of the Cache’s advantages in reading and writing by optimizing the access method of rows and columns to avoid three explicitly matrix transposition, and hiding the twiddle factor multiplication by reconfiguration butterfly computation. Test results show that the performance of Cache-optimized implementation of FFT is significantly improved, and it can be used for fast acquisition of pulse-compression in radar system.