Message-ID: <3AC08874.8322911@usa.net>
Date: Tue, 27 Mar 2001 14:32:52 +0200
From: "Alberto di Bene" <dibene@usa.net>
Organization: Undisclosed
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
Subject: LF: [TECH] A question
References: <3.0.1.16.20010320155747.2b1f164c@mail.cc.kuleuven.ac.be>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

I have a question for you all.

I am implementing both in Argo and Spectran a function that will
display the relative level (in dB) of the spectral lines when you hover the
mouse cursor over them. So far so good.
What I am uncertain about, is that word, 'relative'. Relative to what ?

One possibility is to first click on a given point of the display, take this
bin amplitude as a reference, and the values subsequently displayed will
be relative to this reference level.  But frankly I don't like this solution
very much, even if this will be perhaps the correct way to measure
the S/N ratio, but this is another topic.

Another solution which has been suggested is to take as reference
level the saturation level of the ADC. This is quite valid is case of
a single signal, but on a real case what will happen with this method
is that the measured level will vary depending on the total input to the
ADC, even if the signal being measured is constant in amplitude.

I gave a thought about using the Parseval equation, but before pursuing
this further, I decided to ask the question here, as perhaps this problem
has been already encountered and solved by others.
I am open to any suggestions and advices.   Thanks.

73   Alberto   I2PHD