Basic Operations on DSP Processor

In this experiment some basic mathematical operations like convolution were performed  on Texas Instruments C2000 TMS320F28335 DSP processor.The theoretical aspects of DSP technology are not too difficult. We just have different algorithms for different operations. But, the real world doesn't work on just mathematics and algorithms. We need to have some physical hardware that will implement these operations. This is where the DSP processor comes in. The code was written on Code Composer Studio 3 in C language.The output of the instructions was stored in the registers and these values were obtained in real-time using the debugging functionality of the software.

DSP Application on One Dimensional Signal

Patent Review: MOTOR DRIVE CONTROL USING PULSE WIDTH MODULATION PULSE SKIPPING

Application No.: 13/963,317

Patent No.:US 9,240,749 B2

Date of patent: Jan. 19, 2016

Inventors:Charles E.Green,Fenton,MO(US);

                 Joseph G.Marcinkiewicz,St.PETENS,mo(US)

Summary: A conterol system for a motor includes a pulse width modulation module,a pulse skip determination module,and a duty cycle adjusment module.the pulse width modulation module generates three duty cycle values based on three voltage requests,respectively.A plurality of solid stste switches control three phase of motor in response to the three duty cycle values,respectively.The pulse skip determination module generates a pulse skip signal.The duty cycle adjusment module selectively prevents the plurality of solid state switches from switching during intervals specifiedc by the pulse skip signal.

Link: http://www.google.co.in/patents/US9,240,749 B2 

IEEE Paper Review:Sinusoidal PWM signal generation using TMS320C6711 DSPfor power control in mobile phones Authors: K.Karthikeyan, J.Jaisheela, Dinesh Kumar M and Dr.K.Senthil Kumar
Publisher :- IEEE
Published in:Process Automation, Control and Computing (PACC), 2011 International Conference on
Date of Conference: 20-22 July 2011

Summary: Pulse Width Modulation is a method of controlling the amount of power to a load without having to dissipate any power in the load driver. PWM signals are widely used in telecommunications, power control systems, voltage regulation and in audio signal processing. This paper deals with the optimized power control in mobile applications. PWM power control systems became reality with the advent of modern semiconductor switches like MOSFETs and Insulated Gate bipolar transistor (IGBT). PWM signals can be easily generated using microcontrollers rather than Digital signal processors (DSP), but when the sensor data from light intensity sensors are processed in real time, then microcontrollers can no longer be used because of less accuracy and speed and this is where DSP comes into play. Programming a DSP to generate PWM signal in accordance with the sensor data is not an easy task since the interfacing codes are not as simple as for microcontrollers. Hence the simulation of the PWM signal has been done to study the efficiency of the C code. This work is a part of a project to develop DSP-FPGA based BLDC motor control.

 

Design of FIR filter using Frequency Sampling Method

Frequency Sampling Method:

In this experiment, the various parameters like pass band attenuation, stop band attenuation, pass band frequency, stop band frequency and sampling frequency are passed as input and the order of the filter is calculated. The flow is shown below :

Hd(w)----------------H(k)------------------h(n)

         (By sampling)             (by IDFT)

I obtained the magnitude response of this by the same way I obtained it for FIR filter by Windowing method.

Design of FIR filter using window method

                     Here we design Digital Filter using windowing technique and study spectrum of filter. We  have to choose to cases for i/p specification parameter for LPF and HPF respectively.

     Then design filterby selecting appropriate window function are as follow :-

          Hd(w) = Mod(Hd(w) ) → By IDFT (on integrating Hd(w)) → hd(n)  then, h(n) = hd(n).w(n)

         After this plot magnitude spectrum and phase spectrum and verify values of Ap and As in Pass Band and Stop Band from magnitude spectrum.

          Observe the Phase Spectrum plot.

Design of Chebyshev Filter

Chebyshev filter design :

This experiment was similar to the Butterworth Filter design which we had done earlier.We had hassled a lot in that experiment to find out in built scilab functions to plot the magnitude and phase response of the desired filter. We learnt from that and searched for more such functions earlier only so that implementation would be made easy. 

Design of Butterworth Filter

               This was the first experiment where we used Scilab for implementing the code. 

In this experiment, the various parameters like pass band attenuation, stop band attenuation, pass band frequency, stop band frequency and sampling frequency are passed as input and the order of the filter is calculated. Similarly, the cutoff frequency is calculated.

                       The normalized transfer function is evaluated according to the filter type,i.e LPF or HPF(replacing s by 1/s). From the normalized transfer function, the denormalized function is calculated by substituting the value of cut off frequency. The response in z -domain is equivalently calculated by IIM or BLT transformations. 

OAM & OSM

 The fourth experiment of the course was on Linear FIR Filtering Methods -Overlap-add method and Overlap save method are algorithms to compute DFT of input signals of very large lengths. In OAM, the input signal is divided into small chunks of equal length and the convolution of these small chunks with the second signal is carried out. This can result into overlapping of values for the same indexes depending on the length of one chunk and the second input signal. Thus the output signal to be obtained should be addition of all the overlapped values.

In OSM, instead of dividing the input, we divide the output in small chunks and use only those input values which are required to calculate this output. Both the OAM and OSM require same amount of computations and hence memory.