Basic Operations on DSP Processor

In this experiment some basic mathematical operations like convolution were performed  on Texas Instruments C2000 TMS320F28335 DSP processor. The code was written on Code Composer Studio 4 in C language.It is capable of handling large mathematical operations and also real time data can be processed with higher speed.For emulation XDS100v2 USB Emulator was selected.
 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: METHOD AND APPARATUS FOR FREQUENCY MODULATION SYNTHESIS

Application No.: 08/949,574

Patent No.: 6011448A

Date of patent: Jan. 4, 2000

Inventors: Daniel H. McCabe,Daniel H. McCabe,

Summary: The analysis of the Frequency Modulation involves multiplication and the feedback factor.Feedback factor can be implemented by using the bit-wise logical shift.This patent gives the excellent solution for large discrete increments involved in multiplication and also the sine look-up table hence the hardware required can be minimized. The invention provides a method of frequency modulation synthesis that can be carried out with shifts and additions,and without a large sine table.The sines are calculated using the coordinate rotation digital computer (CORDIC) algorithm. The use of this algorithm improves the efficiency.acquire sine values as opposed to an extensive sine look-up table. The method can be implemented with a programmed special purpose processor such as a digital signal processor. 
Link: http://www.google.co.in/patents/US6011448


IEEE Paper Review:A Pipelined CORDIC Architecture and Its Implementation in All-Digital                                      FM Modulator-Demodulator
Authors: Trio Adiono, Nur Ahmadi
Publisher :- IEEE
Published in: 6th Asia Symposium on Quality Electronic Design
Date of Conference: 4-5 Aug. 2015

Summary: — COordinate Rotation DIgital Computer (CORDIC), is an algorithm that is used to perform trigonometric-related calculations. CORDIC is often utilized in the absence of hardware multiplier since this algorithm requires only addition, subtraction, bit shifting, and lookup table.This paper provides an implementation of CORDIC algorithm using pipelined architecture.The proposed solution has greater clock speeds compared to other FM modulator techniques as seen from the comparison graph of paper.


Application 2:

 IEEE Paper Review:Implementation and Performance Analysis of Real-Time Digital Filter for Audio Signals
Authors: Pranab Kumar Dhar, Ui-Pil Chong, Jong-Myon Kim
Publisher :- IEEE
Published in: 2008 Third International Forum on Strategic Technologies

Summary: A real-time digital filter can perform filtering operation on real-time signals.In this paper effect of the window on FIR filter has been discussed. Real time audio processing allows modified audio to be “judged by hearing” while it is processed.The filter output using DSP block in MATLAB sets shows that smooth ripples appear in stop band but no ripples appear in pass band.The outputs of these filters are observed both by hearing the sound and analyzing the frequency response curves. Finally we analyzed the performance of these filters. The filters can be used to remove noise.Also for processing and transmitting of audio signal for specific purpose. 


Link: http://ieeexplore.ieee.org/document/4602912/

Patent Review: Hearing aid digital filter 
Application No.: US 09/324,128
Patent No.: US6292571 B1
Date of patent:18 Sep 2001
Inventors: Walter P. Sjursen


Summary: This patent mainly focuses on reducing the hardware required for digital filter with the use of single general purpose multiplier. It uses a DSP algorithm portion of a digital hearing aid by using signal processing algorithms that can be implemented in a minimum amount of dedicated circuitry. Circuitry is minimized in three Ways: First, only one general purpose multiplier is used to change the frequency response as a function of signal level, second, the filter coefficients are values represented by 2 raised to the nth power, and third, a non-recursive (finite impulse response) filter is used.Stability is ensured as FIR filters are used.


Link: https://www.google.ch/patents/US6292571

FIR Filter Design using FSM

Frequency sampling is the another method of FIR filter design.In this method desired frequency response Hd(w) is sampled at discrete values of frequency to obtain H(k).The final output sequence h[n] is obtained by Inverse Discrete Fourier Transform (IDFT) method.The final output sequence is always symmetric about the point N/2 which called Point of Symmetry.Computational efficiency is increased as signal is decomposed which makes calculation more faster.

FIR Filter Design using Windowing Method

The second type of the Digital filter design is FIR i.e. Finite Impulse Response in h[n] has finite length.There are different types of Window functions which includes Rectangular,Bartlet, Hanning, Hamming and Blackman. 

The selection of which depends upon the attenuation in stop band (As) which is given as input.Along with this fp,fs,Ap and Fsamp are also taken as input.

The h[n] which is obtained from inverse DTFT of ideal filter design,is multiplied with appropriate window function to give final output.

The phase response of the filter varies linearly with the frequency and no distortions are observed in the output.The output is same as delayed by some constant.

Design of Chebyshev Filter

The experiment was performed on Scilab with same procedure required to that of the Butterworth filter design.The difference in the code of Chebyshev is the formula for calculating the parameters of Analog filter.
Design of the digital filter remains the same either by Bilinear Transform Method or Impulse Invariant Method.Order of the chebyshev filter was found less than Butterworth filter which simplifies the hardware required for designing this filter.
From the magnitude spectrum it was observed that there are ripples in pass band and it is monotonic in stop band.The number of ripples represent the order of the filter.

Design of Butterworth Filter

This experiment was performed using Scilab which is the open source alternative to MATLAB.We designed both filters i.e. HPF and LPF using Scilab coding.
Here the parameters like passband attenuation (Ap),stopband attenuation (As),analog passband frequency (fp),analog stopband frequency (fs) and sampling frequency (Fs).were taken as inputs. Sampling frequency selected is 4 times higher than fp or fs.
The observed value of 'Ap' is less than input value whereas for 'As', it was found as greater. The order of the filter obtained was higher than 10 for each of the filter.
The frequency bands are complex.Also,magnitude spectrum is monotonic i.e varying in one condition.

OSM and OAM

Overlap Add Method (OAM) and Overlap Save Method (OSM) are the techniques used for filtering of long input data sequence x[n] with finite impulse response h[n].In OAM the output sequence is found in which the input sequence is divided in 'L' point blocks. Whereas in OSM block length ='N'. As the input signal is processed block by block , they are also called as Block Processing techniques. This also reduces the delay of processing.Both the techniques require same computational time