Homec4science
Diffusion Instrument Control (TekRSA-Freq-vs-Time) Browse

rINSTCONTROL/@MyAvgTrace/MyAvgTrace.m
47ca0ecc4598TekRSA-Freq-vs-Time

Last Change
Raw File

MyAvgTrace.m
View Options

% Adds averaging capabilities to MyTrace
%
% The averaging type is 'lin' (or 'linear')/ 'exp' (or 'exponential').
% Linear averaging is a simple mean
% x=\sum_{n=0}^N x_n,
% exponential is an unlimited weighted sum
% x=(1-exp(-1/n_avg))*\sum_{n=0}^\inf x_n exp(-n/n_avg).
classdef MyAvgTrace < MyTrace
properties (Access=public)
% Target number of averages, when it is reached or exceeded
% AveragingDone event is triggered
n_avg=1
avg_type='lin'
end
properties (GetAccess=public, SetAccess=protected)
% Counter for the averaging function, can be reset by clearData
avg_count=0
end
methods (Access=public)
% Adds data to the average accumulator. When the averaging counter
% reaches n_avg (or exceeds it in the exponential case), completed
% is set to 'true', otherwise 'false'.
% In exponential regime the averaging proceeds indefinitely so that
% avg_count can exceed n_avg.
% In linear regime when the averaging counter exceeds n_avg, new
% data is discarded.
function completed = addAverage(this, b)
assert(isa(b,'MyTrace'), ['Second argument must be a ' ...
'MyTrace object']);
if isempty(this)
% Initialize new data and return
this.x=b.x;
this.y=b.y;
this.name_x=b.name_x;
this.unit_x=b.unit_x;
this.name_y=b.name_y;
this.unit_y=b.unit_y;
this.avg_count=1;
completed=(this.avg_count>=this.n_avg);
return
end
assert(length(b.y)==length(this.y), ...
['New vector of y values must be of the same', ...
'length as the exisiting y data of MyTrace in ', ...
'order to perform averanging'])
switch this.avg_type
case 'lin'
if this.avg_count<this.n_avg
% Increase the counter and update the data
this.avg_count=this.avg_count+1;
this.y = (this.y*(this.avg_count-1)+b.y)/...
this.avg_count;
end
case 'exp'
% In the exponential case averaging proceeds
% indefinitely, so do not check if avg_count<n_avg
this.avg_count=this.avg_count+1;
this.y = b.y*(1-exp(-1/this.n_avg))+ ...
this.y*exp(-1/this.n_avg);
otherwise
error('Averaging type %s is not supported', ...
this.avg_type)
end
completed=(this.avg_count>=this.n_avg);
end
% Provide restricted access to the trace averaging counter
function resetCounter(this)
this.avg_count=0;
end
% Overload clearData so that it reset the averaging counter in
% addition to clearing the x and y values
function clearData(this)
this.x=[];
this.y=[];
resetCounter(this);
end
% Extend the info stored in trace metadata compare to MyTrace
function Mdt=makeMetadata(this)
Mdt=makeMetadata@MyTrace(this);
addParam(Mdt,'Info','avg_type',this.avg_type, ...
'comment','Averaging type, linear or exponential');
addParam(Mdt,'Info','avg_count',this.avg_count, 'comment', ...
'Number of accomplished averages');
addParam(Mdt,'Info','n_avg',this.n_avg, 'comment', ...
['Target number of averages (lin) or exponential ' ...
'averaging constant (exp)']);
end
function setFromMetadata(this, Mdt)
setFromMetadata@MyTrace(this, Mdt)
if isfield(Mdt.Info, 'avg_type')
this.avg_type=Mdt.Info.avg_type.value;
end
if isfield(Mdt.Info, 'n_avg')
this.n_avg=Mdt.Info.n_avg.value;
end
if isfield(Mdt.Info, 'avg_count')
this.avg_count=Mdt.Info.avg_count.value;
end
end
end
%% Set and get methods
methods
% Ensure the supplied value for averaging mode is assigned in its
% standard form - lowercase and abbreviated
function set.avg_type(this, val)
switch lower(val)
case {'lin', 'linear'}
this.avg_type='lin';
case {'exp', 'exponential'}
this.avg_type='exp';
otherwise
error(['Averaging type must be ''lin'' ' ...
'(''linear'') or ''exp'' (''exponential'')'])
end
end
function set.n_avg(this, val)
% The number of averages should be integer not smaller than one
this.n_avg=max(1, round(val));
end
end
end
c4science · Help