original script uses random number generation as its input - to fit my needs requires input of lab data I have collected instead of random number; I get 1-by-0 errors

% DFM Allan Deviation

% 3-16-22

% Provides Allan Deviation Curve for inputted Data file in .TXT format

% .TXT file sample time is manually inputted as SpS

clc

SpS = 25; % samples per Second

Fs = double(SpS);

% count the number of lines in the text file then deposit in LC

lines = -inf;[fid,msg] = fopen('11.txt', 'r');

linecount = 0;

lines_in_file = 0;

while ~feof(fid)

thisline = fgetl(fid);

if ~ischar(thisline); break; end

linecount = linecount +1;

num_nonspace = nnz(~isspace(thisline));

if num_nonspace >0

lines_in_file = linecount;

end

end

disp(linecount);

LC = linecount-1;

omega = double(LC);

%omega = LC

load ('11'); % input the data

t0x = 1/SpS; % determine tau from samples per second, SpS (entered at top)

t0 = double(t0x);

thetax = cumsum(omega, 1)*t0;

theta = double(thetax);

%display(theta);

%display(t0);

%display(omega);

maxNumM = 1000000;

L = size(theta, 1);

display(theta);

maxM = 2.^floor(log2(L/2));

m = logspace(log10(1), log10(maxM), maxNumM).';

m = ceil(m); % must be Integer Value

m = unique(m); % remove duplicates

display(L);

tau = m*t0;

%disp(m);

avar = NaN(numel(m), 1); % create a matrix of zeros to speed up memory access

for i = 1:numel(m)

mi = m(i);

avar(i,:) = sum((theta(1+2*mi:L) - 2*theta(1+mi:L-mi) + theta(1:L-2*mi)).^2, 1);

end

avar = avar ./ (2*tau.^2 .* (L - 2*m));

%disp(avar);

adev = sqrt(avar);

figure()

%loglog(tau, adev)

title('Allan Deviation')

xlabel('\tau');

ylabel('\sigma(\tau)')

grid on

axis equal

Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 1-by-0.

Error in DFM_Allan_Deviation (line 57)

avar(i,:) = sum((theta(1+2*mi:L) - 2*theta(1+mi:L-mi) + theta(1:L-2*mi)).^2, 1);

(Here is the original file found in the help examples)

% Inertial Sensor Noise Indication using Allan Deviation

% this is the Allan Deviation part

% This script came from the Mathworks Help Section; search "Inertial Noise"

% The following is this script unchanged from the Mathworks example

% Load logged data from one axis of a three-axis gyroscope. This recording

% was done over a six hour period with a 100 Hz sampling rate.

load('LoggedSingleAxisGyroscope', 'omega', 'Fs')

t0 = 1/Fs;

theta = cumsum(omega, 1)*t0;

maxNumM = 100;

L = size(theta, 1);

maxM = 2.^floor(log2(L/2));

m = logspace(log10(1), log10(maxM), maxNumM).';

m = ceil(m); % m must be an integer.

m = unique(m); % Remove duplicates.

tau = m*t0;

avar = zeros(numel(m), 1);

for i = 1:numel(m)

mi = m(i);

avar(i,:) = sum( ...

(theta(1+2*mi:L) - 2*theta(1+mi:L-mi) + theta(1:L-2*mi)).^2, 1);

end

avar = avar ./ (2*tau.^2 .* (L - 2*m));

adev = sqrt(avar); % Allan Deviation is the Square root of Allan Variance

figure % Plot The Output

loglog(tau, adev)

title('Allan Deviation')

xlabel('\tau');

ylabel('\sigma(\tau)')

grid on

axis equal

% this script will not run because "omega" is not supplied ????

% I am trying to produce omega from my data to plug into this script.

% all I get is the "1-by-1 leftside does not match 1-by-0 right side"

[Voss]

What does display(L); show? Does that value of L seem right to you?

Unless load('DFM3') overwrites omega and/or SpS, it looks like theta is a scalar, so that L is 1 and the indexes 1+2*mi:L, etc., are all empty, on the line that gives the error.

I guess to know for sure it will be useful to have the files 11.txt and DFM3.mat; can you please attach them?

[Duane Melvin]

thanks for your response! tis is driving me wacko (a little).

here are the files (originally DFM but I changed it to "11")

11.txt is like this (there are 112000 lines but I show just the first 10)

0.99679
0.99638
0.9972
0.99658
0.997
0.997
0.99639
0.99659
0.99639
0.99659

[Duane Melvin]

display(L) shows 1 actually I dont know if this seems right

[Duane Melvin]

what could I do to theta to make it not scalar???? I copied the script below for your convienience

theta = cumsum(omega, 1)*t0;

maxNumM = 100;

L = size(theta, 1);

maxM = 2.^floor(log2(L/2));

m = logspace(log10(1), log10(maxM), maxNumM).';

m = ceil(m); % m must be an integer.

m = unique(m); % Remove duplicates.

tau = m*t0;

avar = zeros(numel(m), 1);

for i = 1:numel(m)

mi = m(i);

avar(i,:) = sum( ...

(theta(1+2*mi:L) - 2*theta(1+mi:L-mi) + theta(1:L-2*mi)).^2, 1);

end

Answer 1

[Voss]

It seems likely that omega should not be a scalar, because theta = cumsum(omega,1)*t0 with scalar omega is the same as theta = omega*t0 (i.e., doing cumsum on a scalar just returns the original scalar).

Should omega be the contents of the file 11.txt? If so, then you don't need to read (fopen/fgetl) and also load the file; you can do one or the other.

(By the way, don't forget that every fopen() needs an fclose()!)

% DFM Allan Deviation
% 3-16-22
% Provides Allan Deviation Curve for inputted Data file in .TXT format
% .TXT file sample time is manually inputted as SpS
clc
SpS = 25;    % samples per Second
Fs = double(SpS);
% count the number of lines in the text file then deposit in LC
% lines = -inf;[fid,msg] = fopen('11.txt', 'r');
% linecount = 0;
% lines_in_file = 0;
% while ~feof(fid)
%     thisline = fgetl(fid);
%     if ~ischar(thisline); break; end
%     linecount = linecount +1;
%     num_nonspace = nnz(~isspace(thisline));
%     if num_nonspace >0
%         lines_in_file = linecount;
%     end
% end
% disp(linecount);
% LC = linecount-1;
% omega = double(LC);
%omega = LC
omega = load('11.txt')   % input the data

omega = 10×1

0.9968 0.9964 0.9972 0.9966 0.9970 0.9970 0.9964 0.9966 0.9964 0.9966

t0x = 1/SpS;  % determine tau from samples per second, SpS (entered at top)
t0 = double(t0x);
thetax = cumsum(omega, 1)*t0;
theta = double(thetax);
%display(theta);
%display(t0);
%display(omega);
maxNumM = 1000000;
L = size(theta, 1);
display(theta);

theta = 10×1

0.0399 0.0797 0.1196 0.1595 0.1994 0.2392 0.2791 0.3190 0.3588 0.3987

maxM = 2.^floor(log2(L/2));
m = logspace(log10(1), log10(maxM), maxNumM).';
m = ceil(m);   % must be Integer Value
m = unique(m); % remove duplicates
display(L);

L = 10

tau = m*t0;
%disp(m);
avar = NaN(numel(m), 1);  % create a matrix of zeros to speed up memory access
for i = 1:numel(m)
   mi = m(i);
   avar(i,:) = sum((theta(1+2*mi:L) - 2*theta(1+mi:L-mi) + theta(1:L-2*mi)).^2, 1);
end
avar = avar ./ (2*tau.^2 .* (L - 2*m));
disp(avar);

   1.0e-06 *

    0.1078
    0.0269
    0.0393
    0.0615

adev = sqrt(avar);
disp(adev);

   1.0e-03 *

    0.3284
    0.1641
    0.1982
    0.2480