lsw00kor wrote:
> Hello, everyone.
> I would like to discuss OFDM's robustnesss againt freq-selective channel.
>
> 1) I'm using BPSK( for some reasons, even though mostly we use QPSK, or
> QAM..) on a very very wide channel.
> 2) I didn't employ a channel equalizer at the frontend before fft.
> 3) I, as a reference, followed Richard Van Nee's 'OFDM for wireless
> multimedia communications' to set the parameters:
>
> [1] worst case delay spread = 10 nsec.
> [2] 4 times [1] = 40 nsec.
> [3] 6 times [2] = 240 nsec is set as an OFDM symbol's duration
> [4] For Rs bits (BPSK) per an OFDM symbol, Rs * 240 nsec bits are
> assigned.
> [5] deltaf = 1/200nsec = 5MHz
>
> 4) But, weirdly, I always have very very bad error performance. I can't
> figure it out. I think I followed the guidelines exactly to setup the
> parameters.
>
> 5) But on the other hand, we usually don't test OFDM in this kind of wide
> bandwidth..  So, the guidelines in [1] - [5], I wonder if they can really
> work in my scheme, too.
>
> 6) Could anybody give me some advice please?
>
> Your advice is highly appreciated. Thank you very much.


1. lsw00kor
 Nov 8, 4:23 pm   show options

Newsgroups: comp.dsp
From: "lsw00kor" <lsw00...@hotmail.com> - Find messages by this author

Date: Tue, 08 Nov 2005 15:23:58 -0600
Local: Tues, Nov 8 2005 4:23 pm
Subject: BPSK + OFDM over a very wideband channel
Reply | Reply to Author | Forward | Print | Individual Message | Show
original | Report Abuse

Hello, everyone.
I would like to discuss OFDM's robustnesss againt freq-selective
channel.


1) I'm using BPSK( for some reasons, even though mostly we use QPSK, or

QAM..) on a very very wide channel.
2) I didn't employ a channel equalizer at the frontend before fft.
3) I, as a reference, followed Richard Van Nee's 'OFDM for wireless
multimedia communications' to set the parameters:


[1] worst case delay spread = 10 nsec.

really?

what kind of channel is it?

Mark

Hello, everyone.
I would like to discuss OFDM's robustnesss againt freq-selective channel.

1) I'm using BPSK( for some reasons, even though mostly we use QPSK, or
QAM..) on a very very wide channel.
2) I didn't employ a channel equalizer at the frontend before fft.
3) I, as a reference, followed Richard Van Nee's 'OFDM for wireless
multimedia communications' to set the parameters:

[1] worst case delay spread = 10 nsec.
[2] 4 times [1] = 40 nsec.
[3] 6 times [2] = 240 nsec is set as an OFDM symbol's duration
[4] For Rs bits (BPSK) per an OFDM symbol, Rs * 240 nsec bits are
assigned.
[5] deltaf = 1/200nsec = 5MHz

4) But, weirdly, I always have very very bad error performance. I can't
figure it out. I think I followed the guidelines exactly to setup the
parameters.

5) But on the other hand, we usually don't test OFDM in this kind of wide
bandwidth..  So, the guidelines in [1] - [5], I wonder if they can really
work in my scheme, too.

6) Could anybody give me some advice please?

Your advice is highly appreciated. Thank you very much.