|
From: | Marcus Müller |
Subject: | Re: [Discuss-gnuradio] GNU radio implementation for measuring frequency response |
Date: | Sat, 21 Oct 2017 15:10:57 +0200 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1 |
Hi Suman, be careful with "use instantly": Well, what you're doing is
channel sounding, and as such, the results are up to your
interpretation – for example, in the 100 MHz (!) wide channel
you're testing, what coherency times do you expect from your
environment? That is the time a single channel impulse response is
valid. Basically, how fast do the paths the signal can take
change, e.g. by movement of reflecting objects, or absorbing
(human, for example) bodies, or the transmitter or receiver? With
a 100 MHz overall observation bandwidth, you'd get a spatial
resolution of ~1.5m; so, if this is, for example, a room where
people move, this is the time where noone moves out of their
position by any significant amount of distance; that might be in
the order of seconds. If this is a channel between to moving cars,
more like milliseconds. Another aspect: I assume you're doing this with COTS SDR devices.
These typically are *not* calibrated devices, so you'll have to
figure out a way to remove the effect that your devices (from the
DAC over the filters, mixers, amplifiers, cables, antennas, to the
ADC and DSP) from your measurement. So, if you get data, don't forget that it's not a sufficient
description of the world, unless you understand the model of the
world you're using. Best regards, Marcus [1] https://github.com/gbaier/gr_channelsounder On 20.10.2017 23:48, Suman Bhunia
wrote:
|
smime.p7s
Description: S/MIME Cryptographic Signature
[Prev in Thread] | Current Thread | [Next in Thread] |