共查询到20条相似文献,搜索用时 46 毫秒
1.
本文提出一种FFT新算法,其计算量不大于现有的各种基2DFT算法.然后,与Winograd小DFT(4,8,16点)结合使用,得出一种计算DFT的最快速算法. 相似文献
2.
3.
FFT是数字信号处理最重要的算法之一,论文分析了常规的2N点按时间抽选的实序列FFT运算的基本原理,介绍了一种改进的算法,算法将奇数序列和偶数序列部分开计算,并提取旋转因子的公因子,大大减少了计算过程中的加法和乘法的个数和旋转因子的引用次数,并在实际的DSP平台上进行了实现,实验数据表明,该算法在运算效率和复杂度上都较传统FFT算法有较大的改进。 相似文献
4.
快速傅里叶变换(FFT)是减少离散傅里叶变换(DFT)计算时间的算法。而在无线/移动通信系统中无线通信算法和多媒体应用处理算法中存在大量的矩阵或向量运算,均可以由DLP计算实现。本文研究的FFT算法就存在大量的矩阵运算,通过对FFT矩阵算法的分析,本文提出了在DLP计算模式下通过阵列计算机来实现FFT的快速算法,在MATLAB仿真平台上进行了传统算法与改进之后算法的比较,提出了进一步减少运算时间的FFT并行算法。 相似文献
5.
针对高速实时信号处理的需要,提出了一种对任意长度序列进行FFT的快速改进算法。通过对FFT处理前数据添零个数和DFT分解参数的优化选择,显著降低了FFT处理的运算量。结合频域脉冲压缩等信号处理实例,探讨了该算法在高速DSP上实现时的资源分配、程序编程以及传输I/O瓶颈问题,分别提出了具体的解决方法,并在实际DSP系统中测试了这种改进算法的性能指标,将其和普通算法的性能作了比较。 相似文献
6.
7.
8.
本文结合一个离散傅里叶变换DFT计算实例,分析了DFT与连续时间傅里叶变换FT之间的联系,以较为直观的方式阐述了由FT到DFT的演变过程,以及该过程关键环节对最终结果的影响,主要包括谱泄漏,折叠效应,栅栏效应以及谱折移等.教学实践表明,通过该实例讲述DFT对FT的逼近问题,能有效加深学生对DFT的认识和理解. 相似文献
9.
10.
在分析FFT算法的基本原理的基础上,介绍TFFT算法在TMS320LF2407上的实现及其应用。FFT算法程序采用汇编语言编写,可通过C语言函数调用。实验表明,该方法具有实现简单、高效、且扩展性和兼容性强等特点。 相似文献
11.
王中德 《电子科学学刊(英文版)》1991,8(1):60-67
Starting from an index mapping for one to multi-dimensions, a general in-placeand in-order prime factor FFT algorithm is proposed in this paper. In comparing with existingprime factor FFT algorithms, this algorithm saves about half of the required storage capacityand possesses a higher efficiency. In addition, this algorithm can easily implement the DFT andIDFT in a single subroutine, 相似文献
12.
13.
A simple algorithm for the evaluation of discrete Fourier transforms (DFT) and discrete cosine transforms (DCT) is presented. This approach, based on the divide and conquer technique, achieves a substantial decrease in the number of additions when compared to currently used FFT algorithms (30% for a DFT on real data, 15% for a DFT on complex data and 25% for a DCT) and keeps the same number of multiplications as the best known FFT algorithms. The simple structure of the algorithm and the fact that it is best suited for real data (one does not have to take a transform of two real sequences simultaneously anymore) should lead to efficient implementations and to a wide range of applications. 相似文献
14.
本文从一维到多维的下标变换出发,得到了一种通用顺序,即位素因子FFT算法。与现在的素因子FFT算法相比较,这种算法不仅节省了约一半内存,而且有更高的计算效率。此外,这种算法能很方便地将逆变换也包括在同一程序内。 相似文献
15.
16.
J.A Johnston 《Signal processing》1984,6(1):61-66
One method of computing a radix-2 N-point DFT uses N/2 butterflies in parallel, interconnected by a perfect shuffle mapping. For the radix-2 case the multipliers required by each butterfly at each stage can be computed from those in the previous stage. This note extends the method to the radix-4 DFT. 相似文献
17.
BruunFFT在TMS320C30上的实现 总被引:1,自引:0,他引:1
本文研究了BruunFFT在TMS32C30上的实现。文章从蝶形运算、循环控制和数字据存取等方面提出了有效的实现方法和技术,产生的程序运算速度快于基2、基4等其它FFT。 相似文献
18.
19.
A. P. Clark A. Aftelak M. Savjani C. Ang 《International Journal of Satellite Communications and Networking》1990,8(5):349-361
This paper describes a novel technique for the measurement and correction of the Doppler shift (frequency offset) in a received QPSK signal that has been transmitted over a satellite link. The satellite is here assumed to be in a non-geostationary orbit, and correction of the Doppler shift is required to be achieved as soon as the satellite rises above the horizon, when the Doppler shift may be very high and the signal/noise ratio very low. The receiver now has no prior knowledge of the transmitted data, so that a non-data-aided system must be used, capable of operating under seriously adverse conditions. The selected technique measures the DFTs of successive blocks of samples of the received and demodulated signal, and from the DFTs it derives a control signal for the VCO, which generates the in-phase and quadrature reference carriers for the coherent demodulator. The frequency of the latter is adjusted to reduce the residual Doppler shift in the demodulated signal to a sufficiently small value, so that correct phase synchronization of the VCO on to the received signal carrier can then be achieved by means of an appropriate conventional system (not considered here). The paper describes the new technique of Doppler shift correction, and presents the results of computer-simulation tests to measure the time taken to correct an initial Doppler shift of 5 kHz in the received QPSK signal, at very low signal/noise ratios. 相似文献
20.
本文将直线法与快速付里叶变换和网络分解技术相结合,首次提出了一种新的快速算法,这一方法首先对任意入射角平面波激励下无限周期结构的Helmholtz方程进行变换,用直线法离散新方程,其阻抗元素利用FFT进行计算,再结合网络分解技术求出电流分布。 相似文献