Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
pattern recognition

Subject: pattern recognition

From: muzaffar

Date: 21 Apr, 2008 15:41:02

Message: 1 of 5

HI, every body,
i have data in the form columns.(signals). i want to
compare data A to that of B.
how i can do pattern recognition of signals?
kind regards.

Subject: pattern recognition

From: David

Date: 21 Apr, 2008 16:54:03

Message: 2 of 5

"muzaffar " <muzaffarbashir@yahoo.com> wrote in message
<fuicie$pjs$1@fred.mathworks.com>...
> HI, every body,
> i have data in the form columns.(signals). i want to
> compare data A to that of B.
> how i can do pattern recognition of signals?
> kind regards.
>

A==B

beyond that we'll need more information... like what kind
of signals? what kind of patterns are you looking for?
how close do they have to match? what tools do you have,
signal processing toolbox, blockset, etc?? etc, etc, etc.

Subject: pattern recognition

From: muzaffar

Date: 22 Apr, 2008 07:52:01

Message: 3 of 5

HI,
i am afraid that i dont mean A==B in that so simple sense.
i have data and then want to compare these with sample
(template) for biometrics.
i have
-----------------------------------------------------------
--------------------------
MATLAB
Version 7.4 (R2007a)
Simulink
Version 6.6 (R2007a)
Bioinformatics Toolbox
Version 2.5 (R2007a)
Communications Blockset
Version 3.5 (R2007a)
Communications Toolbox
Version 3.5 (R2007a)
Control System Toolbox
Version 8.0 (R2007a)
Curve Fitting Toolbox
Version 1.1.7 (R2007a)
Filter Design Toolbox
Version 4.1 (R2007a)
Financial Toolbox
Version 3.2 (R2007a)
Fixed-Point Toolbox
Version 2.0 (R2007a)
Fuzzy Logic Toolbox
Version 2.2.5 (R2007a)
Gauges Blockset
Version 2.0.5 (R2007a)
Genetic Algorithm and Direct Search Toolbox
Version 2.1 (R2007a)
Image Processing Toolbox
Version 5.4 (R2007a)
MATLAB Compiler
Version 4.6 (R2007a)
Mapping Toolbox
Version 2.5 (R2007a)
Neural Network Toolbox
Version 5.0.2 (R2007a)
Optimization Toolbox
Version 3.1.1 (R2007a)
Partial Differential Equation Toolbox
Version 1.0.10 (R2007a)
Signal Processing Blockset
Version 6.5 (R2007a)
Signal Processing Toolbox
Version 6.7 (R2007a)
Simulink Response Optimization
Version 3.1.1 (R2007a)
Spline Toolbox
Version 3.3.2 (R2007a)
Stateflow
Version 6.6 (R2007a)
Statistics Toolbox
Version 6.0 (R2007a)
System Identification Toolbox
Version 7.0 (R2007a)

Subject: pattern recognition

From: Gordon Cooper

Date: 22 Apr, 2008 10:22:02

Message: 4 of 5

"muzaffar " <muzaffarbashir@yahoo.com> wrote in message
<fuicie$pjs$1@fred.mathworks.com>...
> HI, every body,
> i have data in the form columns.(signals). i want to
> compare data A to that of B.
> how i can do pattern recognition of signals?
> kind regards.
>
You might find this program useful;

http://www.mathworks.com/matlabcentral/fileexchange/loadFile
.do?objectId=18409&objectType=file

Gordon Cooper

Subject: pattern recognition

From: Dave Robinson

Date: 24 Apr, 2008 10:07:02

Message: 5 of 5

"muzaffar " <muzaffarbashir@yahoo.com> wrote in message
<fuicie$pjs$1@fred.mathworks.com>...
> HI, every body,
> i have data in the form columns.(signals). i want to
> compare data A to that of B.
> how i can do pattern recognition of signals?
> kind regards.
>

Sounds like you are trying to do a single dimensional
'correlation' operation. There have been lots of work done
here. The method you need to implement depends on whether
the signals you are trying to match are the same signal,
perhaps drowned in noise, or can vary in the 'time domain'
(X axis) such as would occur in speech recognition, where
the template utterance may well last longer/shorter than the
target utterance.

In the first case you can apply straightforward correlation
techniques.

help corr()

In the second case you will need to investigate dynamic
programming techniques, or Hidden Markov models of your
template. Suggest you have a Google, I am sure you will be
inundated with hits.

Regards

Dave Robinson

Tags for this Thread

No tags are associated with this thread.

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us