Coherence Bandwidth

[Nik_Victor]

I have a small problem... I am trying to find coherence bandwidth with th
help of corelation coefficients (and not with the formulas 1/5*RMS Dela
Spread or 1/50*RMS Delay Spread)...

I have some data in dB and their phase and from these i calculat
frequency response, impulse response, i export a Power Delay Profile plu
some spread values (mean excess and rms delay).

In a book i found a formula in order to calculate corelation coefficient
which is CC=P(tau)/integration(P(tau)) ---- (where P(tau) is the powe
delay profile)...

Ok. How do i find the integration of P(tau) when i have discrete values o
data. (for example for tau = 200ns P(t)= -15dB e.t.c. )

I use Matlab, and i have stuck.

I have to thanks in advance those who will read this post and share wit
me any ideas.

[USENET]

Hi,

For the case of discrete distribution of power delay profile, the
integral becomes a summation.

Hope that helps
Ramesh


Nik_Victor wrote:
> I have a small problem... I am trying to find coherence bandwidth with the
> help of corelation coefficients (and not with the formulas 1/5*RMS Delay
> Spread or 1/50*RMS Delay Spread)...
>
> I have some data in dB and their phase and from these i calculate
> frequency response, impulse response, i export a Power Delay Profile plus
> some spread values (mean excess and rms delay).
>
> In a book i found a formula in order to calculate corelation coefficients
> which is CC=P(tau)/integration(P(tau)) ---- (where P(tau) is the power
> delay profile)...
>
> Ok. How do i find the integration of P(tau) when i have discrete values of
> data. (for example for tau = 200ns P(t)= -15dB e.t.c. )
>
> I use Matlab, and i have stuck.
>
> I have to thanks in advance those who will read this post and share with
> me any ideas.