adaptive_filter.cpp

Output

Source

#include "sigpack.h"

using namespace std;
using namespace sp;

int main()
{
  // Number of samples
  int N = 400;

  // Create a FIR filter and adaptive
  FIR_filt<double, double, double> G;
  FIR_filt<double, double, double> Ghat;

  // Filter coeffs.
  arma::vec b = "-0.2 -0.1  0.1 0.3 0.7";
  arma::vec c = "-0.4  0.0  0.2 0.1 0.4";
  G.set_coeffs(b);
  int M = b.size();

  // Setup adaptive filter
  //    Ghat.setup_rls(M, 0.8, 5);
  //    Ghat.setup_nlms(M,0.6,0.001);
  //    Ghat.setup_lms(M,0.05);
  Ghat.setup_kalman(M, 10, 1, 5);
  //    Ghat.setup_newt(M,0.99,0.01,15);

  // Signal vectors
  arma::vec x(N, arma::fill::randn); // Input sig
  arma::vec y(N, arma::fill::zeros); // Model sig
  arma::vec d(N, arma::fill::zeros); // Output sig
  arma::vec z(N, arma::fill::randn); // Measurement noise
  arma::vec e(N, arma::fill::zeros); // Err sig

  // Log matrix
  arma::mat Wlog(M, N);

  // Apply G to input signal and add some measurement noise
  for (int n = 0; n < N; n++)
  {
    if (n == 200)
      G.update_coeffs(c);
    d(n) = G(x(n)) + 0.0001 * z(n);
  }

  // Filter - sample loop
  for (int n = 0; n < N; n++)
  {
    // Apply adaptiv filter
    y(n) = Ghat(x(n));

    // Calc error
    e(n) = d(n) - y(n);

    // Update filter
    //      Ghat.rls_adapt(e(n));
    //      Ghat.nlms_adapt(e(n));
    //      Ghat.lms_adapt(e(n));
    Ghat.kalman_adapt(e(n));
    //      Ghat.newt_adapt(e(n));
    // Save to log
    Wlog.col(n) = Ghat.get_coeffs();
  }

  // Display result
  cout << "Filter coeffs:    " << c.t() << endl;
  cout << "Estimated coeffs: " << Ghat.get_coeffs().t() << endl;
  gplot gp0, gp1;
  gp0.window("Plot", 10, 10, 500, 500);
  //    gp0.set_output("Wlog.png");
  gp0.set_term("qt");
  gp0.plot_add_mat(Wlog, "b");
  gp0.plot_show();

  gp1.window("Plot2", 600, 10, 500, 500);
  //    gp1.set_output("LMS_err.png");
  arma::vec J = 10 * log10(e % e);
  gp1.set_term("qt");
  gp1.plot_add(J, "|Error|^2");
  gp1.plot_show();

  return 0;
}