Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

[SOLVED] analog signal matching technique

Status
Not open for further replies.
J

jeffrey samuel

Advanced Member level 4
Advanced Member level 4
Joined
Jul 23, 2012
Messages
1,092
Helped
107
Reputation
214
Reaction score
107
Trophy points
1,363
Location
chennai,india
Activity points
6,373
is there a way to compare two analog signals(standard signals with time and amplitude scaling) and say whether they are identical using circuits

can you also post the same circuit

need some clarifications in it as well
 

The term time scaling needs explanation, I think. It could imply time shifting or even time compression/expansion.

As soon as time base manipulation is involved, you are looking for recorded waveforms in contrast to instantaneous values. Usually digital signal processing systems will be required to handle recorded waveforms.

The "indentity" criterion must be clarified, too.
 

Actually your question is not quite clear. there are many ways to compare to analog signals based on requirements. For example comparator is used to compare between two analog signals.

I think for your purpose you need to use analog to digital converter (ADC).
 

both are taken into account sir both shifting and expansion

sir i need to compare them physical that is their
1 voltage
2 form
3 type
4 delays etc

and show that the reference given as an ip of the circuit is equal to the value given to the other ip that is all

- - - Updated - - -

rongo024 said:
Actually your question is not quite clear. there are many ways to compare to analog signals based on requirements. For example comparator is used to compare between two analog signals.

I think for your purpose you need to use analog to digital converter (ADC).
Click to expand...

the only prob i encountered with comparator is that it gave op as max when the voltages match itself independent of the type of analog signal

i want to know how to compare two wave forms based on the criteria given
 
Last edited:

The best way is to use ADC and convert these two signals and apply DSP.
But ADC you have to consider quantization noise.
 
have you looked at any of the DSP libraries, e.g. using Microchip's dsPIC microprocessors
**broken link removed**

e.g. acquire the two signals via ADCs and then do an FFT on them?
 
We need to go through the table of content of a Signal Analysis lecture or text book to cover all signal parameters of interest and applicable methods to extract it. The question is boundless somehow.

From the present sight, digital signal processing seems to be essential. Most likeley you would want to evaluate suitable algorithms with a tool like Matlab before trying to implement it in a dedicated target hardware (if the latter is intended at all).
 
FvM said:
From the present sight, digital signal processing seems to be essential. Most likeley you would want to evaluate suitable algorithms with a tool like Matlab before trying to implement it in a dedicated target hardware (if the latter is intended at all).
Click to expand...
using Matlab is a very good idea - there are devices that enable a PC to acquire data using ADCs direct into Matlab or GNU Octave or Labview for analysis. An alternative is to use an embedded system to acquire the data and then transmit it to a PC for analysis (via Ethernet or WiFi or USB). Once you have the algorithms you can implement on a device such as a Microchip dsPIC or a Texas C6000 (depending on complexity).
 
rongo024 said:
The best way is to use ADC and convert these two signals and apply DSP.
But ADC you have to consider quantization noise.
Click to expand...

noise is always a problem

then also can a adc be effective enough to ensure whether the analog wave forms are same

- - - Updated - - -

FvM said:
The question is boundless somehow.
Click to expand...


so you mean that creating an hardware is really tedious


FvM said:
From the present sight, digital signal processing seems to be essential. Most likeley you would want to evaluate suitable algorithms with a tool like Matlab before trying to implement it in a dedicated target hardware (if the latter is intended at all).
Click to expand...


will writing a matlab program to compare the signals be really possible

- - - Updated - - -

horace1 said:
using Matlab is a very good idea - there are devices that enable a PC to acquire data using ADCs direct into Matlab or GNU Octave or Labview for analysis. An alternative is to use an embedded system to acquire the data and then transmit it to a PC for analysis (via Ethernet or WiFi or USB). Once you have the algorithms you can implement on a device such as a Microchip dsPIC or a Texas C6000 (depending on complexity).
Click to expand...

so you mean with out computer analysis this is not practically possible

- - - Updated - - -

horace1 said:
have you looked at any of the DSP libraries, e.g. using Microchip's dsPIC microprocessors
**broken link removed**

e.g. acquire the two signals via ADCs and then do an FFT on them?
Click to expand...

this is simple to say is designing the circuit to ensure it feasible
 

so you mean with out computer analysis this is not practically possible
Click to expand...
you can write your own code (in C++, c#, VB.NET, etc) to do the analysis.
It is just that tools such as Matlab and Octave come with DSP functions to do FFTs, filtering, etc which make life simpler.
You can spend time solving your problem rather than writing an FFT
 

anything you can do in matlab you can, with sufficient work, do in a programming language such as C++.

the advantage of using Matlab or Octave is to quickly sort out suitable algorithms then you implement them in a target system, e.g. a dsPIC or Texas C5000
 

I have attached a brief block diagram.
So this is the idea. Use 4 comparators(c1,c2,c3,c4), A delay element(d). S1 is analog signal 1 and S2 is analog signal 2. Output of comparator 1 is used to enable/disable comparator 2. And the same follows with the next comparators. Concept is like say at time T1 if both the signal voltages are equal then comparator 1 will enable comparator 2 now at time T2 again if both the voltages are equal then comparator 3 is enabled. and so on. Once you get a high at the output then we can say like over a period of time both the signal are similar if output is low then input signals do not match with each other. So the delay values should be computed based on the frequency ranges that you use.
signal1.JPG

- - - Updated - - -
 
If u want to measure full cycle the for every sample we cant use comparator. So by using delay element we can reach the full cycle with less number of comparators.
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top