Basic operations on DSP processor

Basic operation on DSP Processor

     All previous labs performed by us were either mathematical calculation and its execution using c language on compiler or simulation using SCILAB. But to know what is real time, perform of our system and what is its response for given input, we need to test a real time system. For that we go for Emulation of the system. 
   All previous labs performed by us were either mathematical calculation and its execution using c language on compiler or simulation using SCILAB. But to know what is real time, perform of our system and what is its response for given input, we need to test a real time system. For that we go for Emulation of the system. 

DSP application on one dimentional signal

Patent review: DIGITAL SIGNAL PROCESSOR ENHANCED PULSE WIDTH MODULATION                             AMPLIFIER

Application no.10/106,544

Patent no.US,6,683,494,B2

Date of patent:Jan27,2014

Summary:
A pulse width modulator(PWM) audio power amplifier with a switch mode power convertor is provided that may adjust the duty cycle of the switch mode power converter based on output current.The PWM audio power amplifier may include digital signal processor to process input signal and control switch mode power converter to minimize distortion in amplified input signal.The control within PWM audio power amplifier may include digital feed for ward open loop control or may combine digital feed forward digital open loop with analog negative feedback closed loop control
to minimize distortion in output signal.
Link:http:https://patents.google.com/patent/US6683494B2/en


IEEE Paper review: DSP BASED IMPLEMENTATION OF A DIGITALLY-CONTROLLED                                              SINGLE PHASE PWM INVERTER FOR UPS
Authors: H.J. JIANG, Y. QIN , S. S DU,Z.Y. YU, S. CHOUDHURY
Publisher: IEEE


Summary:
This paper presents a new method to realize a digitally-controlled Pulse Width Modulation (PWM) inverter used in an on-line Uninterruptible Power Supply (UPS). This digitally-controlled PWM inverter has been implemented using a Digital Signal Processor (DSP). This implementation includes PWM pulse generation, voltage feedback loop (outer loop) and average current feedback loop (inner loop). Simulation and experimental results are given to verify the DSP based PWM inverter design and implementation.
Link:http://ieeexplore.ieee.org/document/793502/

     

FIR filter using FSM

FIR filter design using frequency sampling method

The aim of the experiment was to design digital filter using frequency sampling method. The magnitude and phase spectrum were plotted for LPF and HPF. 

It was observed that the phase plot is similar. Also, if the order of LPF and HPF are same, then the phase plot for both the filters is also same. The observed and calculated values of As and Ap were verified.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

FIR filter using windowing method

FIR filter design using windowing method

The aim of the experiment was to design digital filter using windowing method and study the spectrum. The input parameters were passband attenuation(Ap), stopband attenuation(As), passband frequency(Fp), stopband frequency(Fs) and sampling frequency(F). The magnitude and phase spectrum for LPF and BPF using Hanning window were plotted. 

The observed and calculated values of Ap and As were compared. Thus, the values were verified. The phase spectrum was observed to be linear.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

Butterworth filter

Butterworth filter

This experiment was also performed using Scilab. The input parameters were passband attenuation(Ap), stopband attenuation(As), passband frequency(Fp), stopband frequency(Fs) and sampling frequency(F).The theoretical and observed values of Ap and As were compared.

It was observed that all the poles lie in the unit circle for high pass and low pass digital filter. Thus, the filters are stable. Greater the order of filter, greater is the accuracy in the theoretical and observed values. 

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

Chebyshev filter

Chebyshev filter

This was the first experiment performed using Scilab. Digital chebyshev filter was designed using BLT method.

The input parameters were pass band attenuation(Ap), stop band attenuation(As), pass band frequency(Fp), stop band frequency(Fs) and sampling frequency(F). The observed and theoretical values of Ap and As were compared.

The slight difference in the values were due to variations in constant values and inaccuracies in measurement. More the order of the filter, less will be the difference between observed and theoretical Ap and As values.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

OAM and OSM

 
     The Fast Fourier Transform Algorithm had an important limitation in which it required all input                 values to be available for obtaining an output.

For a real world signal which can be arbitrarily long, waiting for the entire signal to arrive and get stored would result in massive delays and also increase the cost of storage equipment. Here OAM and OSM come into picture. The inputs are very large in these methods. They are used for decreasing the output delay. 

 A long input data sequence was taken and the impulse response was taken of a smaller length i.e. 3. The length of decomposed signal came to be 5 whereas the length of output decomposed signal came to be 8.

 We concluded that for inputs having large lengths OAM and OSM are useful for getting outputs with reduced delay.This experiment took some time as the concept was new to us.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

DFT

The aim of this experiment was to perform Discrete Fourier Transform.

We observed that as the value of N (length of signal) increases, frequency spacing increases and error decreases.

We plotted the magnitude spectrum when N=4 and also after zero padding i.e when N=8. The magnitude spectrum was found to be symmetric.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

FFT

The aim of the experiment was to perform fast fourier transform of a 4 point sequence. The program was written in C language.

In this experiment, we observed that the number of calculations in FFT are less than that in DFT. Thus, FFT is faster.The number of calculations to implement the DFT equation directly is proportional to N*N, where N is the number of data points. The FFT algorithm reduces this to a number proportional to NlogN where the log is to base 2.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM

Convolution and Correlation

The aim of the experiment was to study linear convolution, circular convolution and correlation. In circular convolution, aliasing effect i.e folding of sequence was observed. 

We concluded that in linear convolution, length of output signal N=L+M-1. For circular convolution, length N=max(L,M).

We studied auto-correlation and cross-correlation. For auto-correlation, the observation was that the output signal is a palindrome sequence and the output of correlation is both-sided.

Cross correlation is the correlation between two signals to measure the degree of similarity.

Link for code:https://drive.google.com/drive/folders/0B076kcYDr0DCbzZndW5SMGdjMmM