Lab 10 Signal noise compression

Paper Review

      Paper topic: Adaptive noise cancellation using LMS algorithm.
     Paper was based on the minimization of the squared Euclidean norm of the difference weight vector under a stability constraint to remove the noise from sample. To this purpose, the Lagrangian methodology has been used in order to propose a nonlinear adaptation rule defined in terms of the product of differential inputs and errors which means a generalization of the normalized (N)LMS algorithm. The proposed method yields better tracking ability in this context as shown in the experiments which are carried out on the AURORA 2 and 3 speech database.

Patent rewiev:
Patent no. 5,682,341
Topic:
           Adaptive signal processing using NETWORK/LMS algorithm
          
    In this patent Newton/least mean square algorithm was adapted to remove signal distortion and noise. The adaptive signal processor is used to get the autocorrelation matrix by executing DFT and IDFT,and to apply the derived inverse matrix to an adaptive signal processing thereby reducing the computation to derive the inverse matrix reducing the cost.
Paper link-https://drive.google.com/open?id=0B9yXRTSFouTyaXRYUEVVNVJtMHM

Plagarism report-https://drive.google.com/open?id=0B9yXRTSFouTyaXRYUEVVNVJtMHM

Lab9-Basic application on DSP processors

It was a hardware based demo experiment. We used the hardware kit for the first time. The kit used was TMS320F28375. Instructions for arithmetic, Logicsl and Shift operations were performed. The changes in the register values before and after execution were noted. The experiment was demonstratd by our seniors. Also a program was voice modulation was demonstrated.

Lab 8-FIR filter design using frequency sampling method

  In this experiment, we were required to design a digital FIR filter using Frequency Sampling Method (FSM). The code was written in SCILAB. In this method, desired frequency response is sampled and the samples obtained are taken as DFT coefficients, and then h(n) is calculated using Inverse DFT. The input parameters, just like the previous two experiments, were taken and the magnitude response was plotted. Both LPF and HPF were computed using Scilab. We observed that order of FIR filter was greater than IIR filter for the same input parameters

Lab 7- FIR filter design using windowing method

This experiment aim was to design an FIR filter, using the Windowing method. In this method, the desired impulse response is multiplied with window function w(n) to obtain h(n) which after Z-transfrom yields the transfer function H(z). There are primarily 5 window functions - Rectangular, Bartlett, Hamming, Hanning and Blackman. The values of Attenuation in Stop band (As) and Pass band (Ap) as well as Pass band frequency, Stop band frequency and sampling frequency was given as input. The program was executed for an LPF.

Lab 5-Butterworth filter design

In this experiment digital filter was design using the analog filter. The transfer function was found using Laplace domain and using Bilinear Transform​ Method we got the transfer function in Z-domain. For both high pass filter and low pass filter poles lies in the unit circle. Response close to ideal filter were observed when order was in the range of 10 and above.

Lab 6- Digital Chebyshev filter design

Digital Chebyshev filter was design using analog filter and given input parameters.In this experiment we given proper comments to codes.We design Chebyshev-1 filter having ripples in passband and no ripple in stopband where ripple indicates the order of filter.The minimum order for Chebyshev filter is 2 because we didn't get the ripple if order is less than two.

Two types of Chebyshev filter were design:

Chebyshev low pass filter 

Chebyshev high pass filter 

Lab 4- Overlap add method and Overlap save method

Real-time signals converted into the digital sample are very large in the length which can make the processing speed much slower and complex.Therefore to make the processing speed faster the samples are taken in blocks and then output for blocks are taken one by one hence the process is called block processing technique. So in this experiment we perform two of the block processing technique
1. Overlap add method.
2. Overlap save method.