Page 112 - Photodetection and Measurement - Maximizing Performance in Optical Systems
P. 112

System Noise and Synchronous Detection

                                                         System Noise and Synchronous Detection  105

                         Output voltage  (log scale)  Desired  Harmonic responses
                                response





                                                                 7 f
                           0Hz  f mod   Signal mod    5 f mod  Frequency
                                           3 f
                                                                   mod
                         Signal + noise  power (log scale)  modulation




                                          Noise


                           0Hz                  f mod        Frequency
                       Figure 5.11 Synchronous detection using a square-wave modulator
                       results in harmonic responses at odd multiples of the modulation
                       frequency.


                       sis and synthesis to construct approximations to other functions. As the Walsh
                       functions are inherently digital, they can be very efficient at approximating
                       functions containing steps. (Fourier series are generally poor at this job, gen-
                       erating large errors at such steps: the Gibbs phenomenon.) However, they also
                       do a pretty good job with the continuous sine functions.
                         The 16 Walsh functions of order four are shown in Fig. 5.12. They include
                       both regular square waves of different periods and some nonperiodic wave-
                       forms. To make an efficient procedure, we first perform a Walsh function analy-
                       sis of a sine wave. Once the coefficients of each of the Walsh harmonics are
                       available, we will discard all but the three with the greatest amplitudes. Trans-
                       forming back again gives the approximated sine wave.
                         Figure 5.13 shows the Walsh transform of a ±1V sine wave, that is, the ampli-
                       tude coefficients of each Walsh function term. This was performed in Mathcad,
                                                                        th
                       which provides a Walsh transform algorithm. The zero coefficient of this analy-
                       sis is just the average value of the input sine wave, and therefore is itself zero
                       for this zero-mean sine wave. Also, all the even coefficients are zero. Further,
                       many of the remaining coefficients are very small and can be neglected without
                       imposing great errors. Here we take just the 1st, 3rd, and 7th coefficients, with
                       amplitudes respectively of 0.628 -0.125 and -0.260. Adding these together we
                       obtain a respectable, stepped approximation to the original sine wave. Figure
                       5.14 shows two cycles of the original sine wave, the three Walsh functions used
                       in the approximation, and the approximated sine wave made up from the sum
                       of the functions.
                         To make a Walsh function demodulator, we need only form the products of


                   Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
                              Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
                               Any use is subject to the Terms of Use as given at the website.
   107   108   109   110   111   112   113   114   115   116   117