| [sf, sf2, ctime, spa, mpa, error_cond]=B_Duration_second_flucts(p, Fs, Q, Lplus, Lminus, make_plot)
|
function [sf, sf2, ctime, spa, mpa, error_cond]=B_Duration_second_flucts(p, Fs, Q, Lplus, Lminus, make_plot)
% % B_Duration_second_flucts: Calculates the secondary fluctuations for the B-duration of impulsive noise
% %
% % Syntax:
% %
% % [sf, sf2, error_cond]=B_Duration_second_flucts(p, Fs, Q, Lplus, Lminus, make_plot);
% %
% % **********************************************************************
% %
% % Description
% %
% % This program calculates the duration of the subsequent fluctuations
% % according to MIL-STD-1474D Requirement 4.
% %
% % Two methods are used to calculate the secondary fluctuation durations.
% % Mtehod 1 uses interpolation which is generaly more accurate but may
% % have bugs. Method 2 is the sum of the points above the threchshold
% % which is less accurate but a quick check for the interpolation method.
% %
% % Using two methods provides for an error condition to calcualte whether
% % there is a mistake.
% %
% % **********************************************************************
% %
% % Input variables
% %
% % p=radnn(1, 50000); % Sound pressure time record in Pa for a single
% % % channel data array the
% % % default value is randn(1, 50000);
% %
% % Fs=50000; % Sampling rate in Hz.
% % % default value is 100000 Hz.
% %
% % Q=1; % Data point that the subsequent fluctuations
% % % proceed.
% % % default value is [buf Q]=max(abs(p));
% %
% % Lplus=1; %
% % % default value is Lplus=0.1*max(abs(p));
% %
% % Lminus=-1; %
% % % default value is Lminus=-0.1*max(abs(p));
% %
% % make_plot=1; % 1 Plots the time records and places a circle at
% % % each chosen peak. Otherwise no plots are made.
% % % default value is 0. (No plotting)
% %
% % **********************************************************************
% %
% % Output variables
% %
% % sf is the interpolated subsequent fluctuation duration in seconds
% %
% % sf2 is the rough estimate of the subsequent fluctuation duration by
% % counting points without interpolation.
% %
% % ctime array of indices of all points above Lplus or below Lminus.
% %
% % spa array of indices of all single consecutive points above Lplus
% % or below Lminus.
% %
% % mpa array of indices of the first and last points of multiple
% % consecutive points above Lplus or below Lminus.
% %
% % error_cond is 1 if the absolute value of the difference betweeen
% % c and c2 is greater than the sum of the multiple points and the
% % single points above the threshold.
% %
% % **********************************************************************
%
%
% Example='1';
%
% % Example impulsive data with background noise
%
% Fs=50000; fc=200; td=1; tau=0.1; delay=0.1; A1=3; A2=20;
% [p, t]=analytic_impulse(Fs, fc, td, tau, delay, A1, A2);
% % p % Pa sound pressure, single channel data array.
% % p should have only one impulse.
% Fs=50000; % Hz sampling rate, 50000 Hz is the default
% make_plot=0; % 0 for no plot
% % 1 for make the plot
% % Default is to not make any plots.
% % The plot is the absolute value of the data array.
% % blue line for the 10 dB threshold.
% % green circle for each point above the threshold.
% % red circle for each single point above the
% % threshold.
% % black circle for each first and last point of a
% % multiple point series above the threshold.
%
% [sf, sf2, ctime, spa, mpa, error_cond]=B_Duration_second_flucts(p, Fs, Q, Lplus, Lminus, make_plot);
%
%
% % **********************************************************************
% %
% % References:
% %
% % Mil Standard 1474D, DEPARTMENT OF DEFENSE
% % DESIGN CRITERIA STANDARDm, NOISE LIMITS, 12 February 1997
% % www.silencertests.com/docs/mil-std-1474d.pdf
% %
% % Guido F. Smoorenburg, "Damage Risk Criteria for Impulsive
% % Noise," New Perspectives on Noise Induced Hearing Loss,
% % Raven Press, New York, pages(471-490) 1982
% %
% %
% % **********************************************************************
% %
% % B_Duration_second_flucts is written by Edward L. Zechmann
% %
% % date 27 January 2009
% %
% % modified 31 January 2009 Added plotting of B-duration
% %
% % **********************************************************************
% %
% % Please feel free to modify this code.
% %
% % See also: B_mil_1474D_duration, A_Duration, B_Duration, C_Duration, D_Duration
% %
if (nargin < 1 || isempty(p)) || ~isnumeric(p)
p=randn(1, 50000);
end
% If sampling rate is not specified, then
% assume a data acquisition rate of 50 KHz;
if (nargin < 2 || isempty(Fs)) || ~isnumeric(Fs)
Fs=100000;
end
if (nargin < 3 || isempty(Q)) || ~isnumeric(Q)
[buf Q]=max(abs(p));
end
Q=round(Q(1));
Q(Q<1)=1;
if (nargin < 4 || isempty(Lplus)) || ~isnumeric(Lplus)
Lplus=0.1*max(abs(p));
end
if (nargin < 5 || isempty(Lminus)) || ~isnumeric(Lminus)
Lminus=-0.1*max(abs(p));
end
% The default is to not make any plots
if (nargin < 6 || isempty(make_plot)) || ~isnumeric(make_plot)
make_plot=0;
end
% new_zero is the midpoint
new_zero=0.5*(Lplus+Lminus);
% new_thres is the half_length
new_thres=0.5*(Lplus-Lminus);
p3=p((Q+1):end);
p2=abs(p3-new_zero);
% B_Duration_second_flucts
ctime=find(p2 > new_thres);
c=0;
ct1=0;
ct2=0;
c2=length(ctime);
dct=diff(ctime);
flag1=0;
mpa=[];
spa=[];
if isempty(ctime) || logical(length(p2) < 10)
sf=0;
sf2=0;
ctime=0;
spa=0;
mpa=0;
error_cond=0;
t_array=0;
t_array=0;
else
if isempty(dct)
% C-duration for a single very large impulse
thres=new_thres/max(max(p3));
[maxp maxp_index]=max(p2);
maxp_thres=maxp*thres;
x_thres_index=find(p2 > maxp_thres);
if ~isempty(x_thres_index)
% find start time for C-duration, which is the first 10 dB
% threshold-crossing before the peak pressure
e1=x_thres_index(1)+1;
if (x_thres_index(1)+1) > 1 && (x_thres_index(1)+1) < length(p2)
e1=x_thres_index(1)+1;
else
e1=1;
end
flag1=0;
while (flag1 == 0) && (e1 > 1)
e1 = e1-1;
if p2(e1) <= maxp_thres
flag1=1;
end
end
% interpolate to find better begin time
if abs(p2(e1)-p2(e1+1)) >= 10^-12
ct1 = (maxp_thres-p2(e1))/(p2(e1+1)-p2(e1))+e1;
else
ct1=e1;
end
% find end time for C-duration, i.e. the last 10 dB threshold-crossing
% after peak
if (x_thres_index(end)-1) > 1 && ((x_thres_index(end)-1) <= length(p2))
e1=(x_thres_index(end)-1);
else
e1=2;
end
flag1=0;
while (flag1 == 0) && (e1 < length(p2))
e1 = e1+1;
if p2(e1) <= maxp_thres
flag1=1;
end
end
% interpolate to find better end time
if abs(p2(e1)-p2(e1-1)) >= 10^-12
ct2 = (maxp_thres-p2(e1-1))/(p2(e1)-p2(e1-1))+e1-1;
else
ct2=e1;
end
else
ct2=1;
ct1=1;
end
% C-duration in indices
c = (ct2-ct1);
c2=c;
error_cond=0;
t_array=(ct2-ct1);
else
% mpa multiple points above threshold
% spa single point above threshold
% C-duration for a single very large impulse
thres=new_thres/max(max(p2));
[maxp maxp_index]=max(p2);
maxp_thres=maxp*thres;
mpa=zeros(c2, 1);
spa=zeros(c2, 1);
mpc=0; % multiple point counter
spc=0; % single point counter
for e1=1:(length(dct));
if dct(e1) > 1
if flag1 == 1
mpc=mpc+1;
mpa(mpc)=ctime(e1);
else
spc=spc+1;
spa(spc)=ctime(e1);
end
flag1=0;
else
if flag1 == 0
mpc=mpc+1;
mpa(mpc)=ctime(e1);
end
flag1=1;
end
end
mpa=mpa(1:mpc);
spa=spa(1:spc);
t_array=[];
% Every pair of multiple points above threshold
% must have an end point
% if no last end point, then the last
% point becomes the last end point
if mod(length(mpa), 2) > 0
mpc=mpc+1;
mpa(mpc)=ctime(end);
end
nmpa=floor(length(mpa)/2);
t_array=zeros(nmpa+spc, 1);
for e1=1:nmpa;
% get height of first data point before crossing threshold
if (mpa(2*e1-1)-1) > 1
pt1=p2(mpa(2*e1-1)-1);
else
pt1=p2(1);
end
% get height of data point after crossing threshold
pt2=p2(mpa(2*e1-1));
% interpolate for data point before crossing threshold
if abs(pt2-pt1) >= 10^-12
t3L=1*(pt2-maxp_thres)/(pt2-pt1)+0;
else
t3L=0;
end
% get height of data point
pt22=p2(mpa(2*e1));
% get height of next data point
if (mpa(2*e1)+1) < length(p2)
pt3=p2(mpa(2*e1)+1);
else
pt3=p2(end);
end
% interpolate for data point after crossing threshold
if abs(pt22-pt3) >= 10^-12
t3U=1*(pt22-maxp_thres)/(pt22-pt3)+0;
else
t3U=0;
end
t_array(e1)=t3U+t3L+mpa(2*e1)-mpa(2*e1-1);
end
for e1=1:spc;
% get height of previous data point
if (spa(e1)-1) > 1
pt1=p2(spa(e1)-1);
else
pt1=p2(1);
end
% get height of data point
pt2=p2(spa(e1));
% interpolate for data point before crossing threshold
if abs(pt2-pt1) >= 10^-12
t3L=1*(pt2-maxp_thres)/(pt2-pt1)+0;
else
t3L=0;
end
% get height of next data point
if (spa(e1)+1) < length(p2)
pt3=p2(spa(e1)+1);
else
pt3=p2(end);
end
% interpolate for data point after crossing threshold
if abs(pt2-pt3) >= 10^-12
t3U=1*(pt2-maxp_thres)/(pt2-pt3)+0;
else
t3U=0;
end
t_array(e1+nmpa)=t3U+t3L;
end
% calculate the interpolated sum
c=sum(t_array);
% calculate the rough estimated sum
c2=length(ctime);
% Calculate if the maximum error is satisfied
error_cond=0;
max_error=spc + nmpa;
if abs(c2 - c) > max_error
error_cond=1;
end
if make_plot == 1
close all;
figure(2);
% plot the absolute value of the data array
plot(1/Fs*(0:length(p)-1), abs(p));
hold on;
% draw the threshold line
plot( 1/Fs*[0 length(p)-1], maxp_thres*[1 1], 'k');
% put a green circle for each point above the threshold
for e1=1:length(ctime);
plot( (Q-1+ctime(e1))/Fs, p2(ctime(e1)), 'og', 'linestyle', 'none', 'markersize', 7);
end
% put a red circle for each single point above the threshold
for e1=1:length(spa);
plot( (Q-1+spa(e1))/Fs, p2(spa(e1)), 'or', 'linestyle', 'none', 'markersize', 7);
end
% put a black circle for the beginning and end point of each series of multiple points above the threshold
for e1=1:length(mpa);
plot( (Q-1+mpa(e1))/Fs, p2(mpa(e1)), 'ok', 'linestyle', 'none', 'markersize', 7);
end
end
end
end
% convert c and c2 to seconds
sf=1/Fs*c;
sf2=1/Fs*c2;
if isempty(sf)
sf=-1;
end
if isempty(sf2)
sf2=-1;
end
|
|