数字信号处理解读.pptVIP

Lecture 4 The Fast Fourier Transform For example: The following figure compares the number of complex multiplications required by DFTs and radix-2 FFTs as a function of the number of input data points N. When N = 512, the DFT requires 200 times more complex multiplications than those needed by the FFT. When N = 8192, the DFT must calculate 1000 complex multiplications for each complex multiplication in the FFT! Number of complex multiplications in the DFT and the radix-2 FFT as a function of N. Lecture 4 The Fast Fourier Transform (1) (2) Periodicity (3)Reducibility Quick Example For N=4 3 Redundancy Analysis in DFT Lecture 4 The Fast Fourier Transform A straight forward derivation of the FFT proceeds with the separation of the input data sequence x(n) into two parts, In other word, the x(n) is segmented into its even and odd indexed elements. n= 0 1 2 3 4 5 … … r= 0 1 2 3 4 5 … … X2(r)=x(2r+1) X(1) X(3) X(5) X(7) … … … … X1(r)= x(2r) X(0) X(2) X(4) X(6) …. … … … 4 The principle of Decimation- in-Time radix-2 FFT Lecture 4 The Fast Fourier Transform One can see from the above: G(k) and H(k) are N/2-point DFT. We have two N/2 summations whose results can be combined to give us the N-point DFT. Lecture 4 The Fast Fourier Transform The G(k) and H(k) represent the upper and down half of the DFT outputs respectively. For case N=8, we have Lecture 4 The Fast Fourier Transform As mentioned above, We have two N/2 summations whose results can be combined to give us the N-point DFT. Similarly, N/2-point DFT can be subdivided into two N/4-point DFTs. It gives the FFT its capacity to greatly reduce the number of necessary multiplications to implement DFTs. Lecture 4 The Fast Fourier Transform Full decimation-in-time FFT implementation of an 8-point DFT Lecture 4 The Fast Fourier Transform Table 5-1 Input Index Bit Reversal for an N-point FFT 5 Input Index Bit Reversal The decimation-in-time phrase refers to how we broke the DFT input samples