Working with Real-Time signals in matlab

46 次查看（过去 30 天）
Jaime Estarellas Cifre
评论： Jaime Estarellas Cifre 2024-10-3
I am trying to work with a real time signal witht the DSP Tool box but I am finding difficult to acomplish. I have seen the demos and read the documents related with the DSP Tool box but it doesn't really seem Matlab can work/process real time signals. My objective was to do a DAC to create a signal, an ADC to read this same signal, process it, and then do another DAC to see the processed signal. I have been able to work out an ADC and a DAC but none of the options available (at least for the 2022a version) are actually real time but trying to simulate it.
I will share the code in case anyone has curiostity, but not only it fails when approaching the frequency too nyquist's but also seems to have a hard time when the processing is just multiplying by two, can't imagine what would happen if I modulated the signal...
%============================================================%
%============== Continuous Real Time DAC & ADC===============%
%============================================================%
%========= Processes Data and Displays it Real Time =========%
%========= DisplaySignalV5: More advanced processes ========%
%============================================================%
% REQUIRED: Signal Processing Toolbox & DSP System Toolbox (NI Extension)
clear;
close all;
clc;
% === DAC and ADC === %
% Create DataAcquisition object for NI device
dq = daq("ni");
% Add Analog Output Channels to generate the signal on "ao0" and "ao1"
addoutput(dq, "Dev1", "ao0", "Voltage");
addoutput(dq, "Dev1", "ao1", "Voltage");
ch_out = dq.Channels(1:2);
ch_out(1).Name = "ao0_out";
ch_out(2).Name = "ao1_out";
% Add an Analog Input Channel to acquire the signal on "ai1"
addinput(dq, "Dev1", "ai1", "Voltage");
ch_in = dq.Channels(3);
ch_in.Name = "ai1_in";
% Set the sample rate
rate = 10000; % Hz
dq.Rate = rate;
% Specify the output signal - A square wave with a DC offset
f = 50; % Frequency of 10 Hz
t = (0:rate-1)/rate; % Time vector for 1 second
output = 1.25 + 0.25*square(2*pi*f*t); % Generate square wave with DC offset
% Create timescope to visualize the signals in real time
scope = timescope('SampleRate', rate, 'TimeSpanSource', 'Property', 'TimeSpan', 0.1, ...
'YLimits', [-2 2], 'Title', 'Output vs Input Signals', ...
'ShowLegend', true, 'ChannelNames', {'Inintal Signal DAC_1', 'Processed Signal DAC_2'});
% Set the callback function to execute when data is available
batchSize = rate; % Read data for 1 second at a time for more frequent updates
dq.ScansAvailableFcnCount = batchSize;
dq.ScansAvailableFcn = @(src, evt) plotAndWriteSignals(src, output, scope, rate);
% Preload the output data for "ao0"
preloadData = [output', zeros(length(output), 1)]; % Preload output for ao0, set ao1 to zero initially
preload(dq, preloadData);
% Start the output and input simultaneously in continuous mode
start(dq, "repeatoutput");
% Stop the generation and acquisition after key press
disp('Press any key to stop the continuous generation and acquisition.');
pause();
stop(dq);
disp("Generation and acquisition stopped.");
% Callback function to read data, plot input and output using timescope, and write to ao1 in real-time
function plotAndWriteSignals(src, output, scope, rate)
% Read available data in chunks as defined by ScansAvailableFcnCount
[data, ~] = read(src, src.ScansAvailableFcnCount, "OutputFormat", "Matrix");
processedSignal = 2.*data;
% Ensure that both 'data' and 'outputToPlot' are column vectors and have the same length
len = length(processedSignal);
outputToPlot = output(1:len)';
if isrow(processedSignal)
processedSignal = processedSignal';
end
% Update the timescope with the output and input signals
% The input data and output data must have consistent row sizes
scope([outputToPlot, processedSignal]);
% Write the acquired input data to the output channel "ao1"
% Create a matrix where each row contains values for both output channels
outputData = [outputToPlot, processedSignal]; % First column for ao0, second column for ao1
write(src, outputData);
end
  2 个评论
Walter Roberson
Walter Roberson 2024-10-2
addoutput(dq, "Dev1", "ao0", "Voltage");
addoutput(dq, "Dev1", "ao1", "Voltage");
ch_out = dq.Channels(1:2);
ch_out(1).Name = "ao0_out";
ch_out(2).Name = "ao1_out";
% Add an Analog Input Channel to acquire the signal on "ai1"
addinput(dq, "Dev1", "ai1", "Voltage");
You have not created any connection between the input and output signals. What you emit is being sent to the DAQ output terminals, but is not being looped back as inputs.
Jaime Estarellas Cifre
Jaime Estarellas Cifre 2024-10-3
Hi, the code I posted works, is just that is not actually real time. The output (of the first DAC) I am generating is actually a repeating pattern, so in essence I am just making the square wave periodic. If you increase the frequency near Nyquist (f = 5000Hz), the problems that come with this "trick" show clearly. What I needed was a continuos wave to work on. And, If I use
start(dq,"continuous")
instead of,
start(dq,"RepeatOuput")
I would need to preload a certain amount of signal, and that would only let me run the code for a set amount of time, and not until I make it stop. Errors like this show when using the "Continuous" approach:
Output underflow event: Last valid output was scan number 9878.
My objective was to modulate, filter, etc the signal inputed to the ADC, to then observe the processing in real time and continuosly through the second DAC.
I welcome any ideas on how could I achieve this, as I'd rather use Matlab than LabView.

请先登录，再进行评论。

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Measurements and Statistics 的更多信息

标签

产品


版本

R2022a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by