Prev: Zero Padding for Non-Causal Filters
Next: DSK 6713 - Simulink/Code Composer
From: Ikaro on 23 Mar 2010 08:32
On Mar 23, 2:29 am, "sanam1" <sanamsingh(a)n_o_s_p_a_m.hotmail.com>
wrote:
> Hi,
> Thanks for your reply. Suppose if I have two signals and I correlate them
> then I get degree of similarity between them. But my question is that I
> want to achieve same result but with an operation other than correlation.
> Hope I have made my question clear.
> Regards,
> Sanam
>
> >sanam1 wrote:
>
> >> Hi. What are the alternatives we have to cross-correllation ?
> >> If I want to find degree to similarity between two sampled signals what
> >> options do I have besides doing the obvious correlation??
>

What are you trying to accomplish and why is correlation not
sufficient ??
You want something that gives the same results as the correlation but
by some other operation, so I am just trying to understand why.


You could also use any metric distance if you are dealing with a
hilbert space at treat your signals as vectors. The dot product
(correlation at 0 shift) is a classic example. Distance metrics are
monotonic functions of similarities you can use Manhantan, Mahalanobis
(if dealing with distributions), or other any of the p-vector sum
distances ....

There is also coherence analysis were similarity is computed in
spectral domain (used frequently in eeg analysis).
From: Vladimir Vassilevsky on 23 Mar 2010 09:54




sanam1 wrote:

> Hi,
> Thanks for your reply. Suppose if I have two signals and I correlate them
> then I get degree of similarity between them.

Define what is "degree of simularity".
Correlation gives you squared distance between the signals minus bias.

> But my question is that I
> want to achieve same result but with an operation other than correlation.

If you want to achieve the same result as the correlation, you should
use the correlation.

> Hope I have made my question clear.

None at all.

> Regards,
> Sanam


>
>
>
>>sanam1 wrote:
>>
>>
>>>Hi. What are the alternatives we have to cross-correllation ?
>>>If I want to find degree to similarity between two sampled signals what
>>>options do I have besides doing the obvious correlation??
>>
>>Define "similarity".
>>Depending on this, there could be infinitely many ways to measure it.
>>
>>VLV
>>
From: Vladimir Vassilevsky on 23 Mar 2010 12:41


sanam1 wrote:

> Hi,
> How can I use affine transformation here??

Let's say you have X(t)

Then define:

X'(t) = A*X(t) + B
t' = C*t + D

A,B,C,D - constants

Is X'(t') similar to X(t) ?

VLV




From: Clay on 24 Mar 2010 10:50
On Mar 23, 9:54 am, Vladimir Vassilevsky <nos...(a)nowhere.com> wrote:
> sanam1 wrote:
> > Hi,
> > Thanks for your reply. Suppose if I have two signals and I correlate them
> > then I get degree of similarity between them.
>
> Define what is "degree of simularity".
> Correlation gives you squared distance between the signals minus bias.
>
> > But my question is that I
> > want to achieve same result but with an operation other than correlation.
>
> If you want to achieve the same result as the correlation, you should
> use the correlation.
>
> > Hope I have made my question clear.
>
> None at all.
>
>
>
> > Regards,
> > Sanam
>
> >>sanam1 wrote:
>
> >>>Hi. What are the alternatives we have to cross-correllation ?
> >>>If I want to find degree to similarity between two sampled signals what
> >>>options do I have besides doing the obvious correlation??
>
> >>Define "similarity".
> >>Depending on this, there could be infinitely many ways to measure it.
>
> >>VLV- Hide quoted text -
>
> - Show quoted text -

How about time reversal, followed with conjugation and finally
convolution ;-)

Clay

From: fatalist on 24 Mar 2010 15:47
On Mar 22, 3:44 pm, "sanam1" <sanamsingh(a)n_o_s_p_a_m.hotmail.com>
wrote:
> Hi. What are the alternatives we have to cross-correllation ?
> If I want to find degree to similarity between two sampled signals what
> options do I have besides doing the obvious correlation??
>
> Thanks in advance

State-space signal embedding followed by nearest-neigbor search in m-
dimensional state space

Can replace your cross-correlation as well as auto-correlation for
many purposes

Read US Patent 7124075 to get a better idea

http://www.google.com/patents/about?id=dB97AAAAEBAJ&dq=7124075

First  |  Prev  | 
Pages: 1 2 3
Prev: Zero Padding for Non-Causal Filters
Next: DSK 6713 - Simulink/Code Composer