OSM OAM

We have performed filtering of long data sequence using overlap-add(OAM) and overlap save (OSM) method.These techniques are block processing techniques. Because input data sequence is divided into small blocks on which operation is performed. This operation in OAM is linear convolution and circular convolution in OSM.OSM and OAM are used to find the output of  FIR filter. 

FFT

We performed Fast fourier transform in 3rd experiment . In FFT,The procedure of  performing the butterfly computations and returning the result to memory is repeated many times .this repetition takes place ( (N log2 (N))/2 times ) in the computation of an N-point DFT. For DSP processor radix 2 or 4 decimation in frequency, FFT algorithm are preferable in terms of speed and accuracy. 

DFT

Second lab of DSPP was about discrete fourier transform i.e DFT and IDFT which are the computational method for frequency analysis of signal . DFT gives periodic result with period equal to N(length of i/p signal). DFT is practical implementation of DTFT .Spectrum of DFT is in the frequency range of 0 to 2π .N-point DFT values csn be computed in a total of N^2 Complex multiplication and N(N-1) complex addition.  

convolution and correlation

The first experiment performed in DSPP lab was convolution and correlation with the help of c language. we used input signal x(n) of length L and h(n) of  M .Linear convolution gives output signal of length L+M-1.Where as circular convolution gives max(L,M). Correlation is used to find degree of similarity between two signals.