Question about reducing noise in a radar receiver

Hi,

I have to find a new radar receiver design to increase the SNR. The signal
bandwidth is 2 MHz (a CW Phase modulated signal). With a common receiver
design (receiver BW at 2 MHz too), the SNR is too low for detection. To
increase the SNR, I can reduce the noise power in my receiver by reducing
the receiver bandwidth. Can someone tell me if it's possible to do that
and how (filter bank maybe?).

Thank you

LPG1976 wrote:
> Hi,
> 
> I have to find a new radar receiver design to increase the SNR. The signal
> bandwidth is 2 MHz (a CW Phase modulated signal). With a common receiver
> design (receiver BW at 2 MHz too), the SNR is too low for detection. To
> increase the SNR, I can reduce the noise power in my receiver by reducing
> the receiver bandwidth. Can someone tell me if it's possible to do that
> and how (filter bank maybe?).

The receiver will have a broadband IF. Redesign that to make it narrower.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
�����������������������������������������������������������������������

If the desired signal BW is 2 MHz and the receiver predetection IF BW
is 2 MHz, then what more is there to filter?

You can lower the noise figure of the front end so there is less noise
in the receiver to start with...


Or you can reduce the BW of the transmitted signal and then also reduce
the BW of the receiver to match it... but then you cannot send as much
data....

Mark

"Mark" <makolber@yahoo.com> writes:
> [...]
> You can lower the noise figure of the front end so there is less noise
> in the receiver to start with...

Or you can increase the gain of the receive antenna...

Or the output power of the transmitter...

Or the gain of the transmit antenna...
-- 
%  Randy Yates                  % "Maybe one day I'll feel her cold embrace,
%% Fuquay-Varina, NC            %                    and kiss her interface, 
%%% 919-577-9882                %            til then, I'll leave her alone."
%%%% <yates@ieee.org>           %        'Yours Truly, 2095', *Time*, ELO   
http://home.earthlink.net/~yatescr

This is a radar so there's no data to transmit as such.

You can reduce the bandwidth but that will reduce the range resolution
(assuming that is important).

You can increase SNR in the ways Randy describes, or by *coherently*
integrating the received signal for longer. But if you have CW
operation and your power budget already has you continuously
integrating the *output of a matched filter*, then there's not much you
can do in terms of DSP or filter architectures - you need more power
one way or another (transmitter power amps, antenna directionality) or
a lower noise receiver.

Are you sure your calculations are correct, ie you are measuring the
matched filter output SNR?

ggroupsthomas@gmail.com wrote:
> This is a radar so there's no data to transmit as such.
> 
> You can reduce the bandwidth but that will reduce the range resolution
> (assuming that is important).
> 
> You can increase SNR in the ways Randy describes, or by *coherently*
> integrating the received signal for longer. But if you have CW
> operation and your power budget already has you continuously
> integrating the *output of a matched filter*, then there's not much you
> can do in terms of DSP or filter architectures - you need more power
> one way or another (transmitter power amps, antenna directionality) or
> a lower noise receiver.
> 
> Are you sure your calculations are correct, ie you are measuring the
> matched filter output SNR?
> 

you can also use multiple pulses, see Whalen's Detection Theory book.

> you can also use multiple pulses, see Whalen's Detection Theory book

not in this case, as the OP states it is a CW (continuous wave) radar.

<ggroupsthomas@gmail.com> wrote in message 
news:1139966530.535939.126650@o13g2000cwo.googlegroups.com...
>> you can also use multiple pulses, see Whalen's Detection Theory book
>
> not in this case, as the OP states it is a CW (continuous wave) radar.
>
Phase modulated

best of luck - Mike

>>> you can also use multiple pulses, see Whalen's Detection Theory book
>> not in this case, as the OP states it is a CW (continuous wave) radar.
>Phase modulated

I don't see that it being phase modulated makes the slightest bit of
difference. The phase modulation simply means the signal has
non-negligible bandwidth, so the receiver bandwidth must be adequate to
receive it (and hence its total thermal noise is higher). The radar
will have range resolution as a result of this bandwidth.  If there
were no modulation, the receiver bandwidth could be reduced to only
that of the expected doppler shift.

Either way, the signal is continuous, so signal and noise *powers* can
be inserted in the "radar equation". The OP may wish to integrate the
matched filter or doppler filter bank outputs for as long as he wishes
in order to increase the SNR of the signal used for detection
decisions, but there are no "pulses" (by definition amplitude
modulated) to contend with.

I suppose you might consider a phase modulated code repetition to be a
"pulse", but that would be rather unconventional as the duty cycle is
1! In the pulsed case, the radar equation "expects" unmodulated pulses
with bandwidth = 1/pulse_length if you simply stick the numbers in
without thinking, so if pulse compression is used (so the actual duty
cycle is higher) then you must factor this in.

But in this CW case, phase modulated or not, the transmitter output
power is essentially constant, so there are no more degrees of freedom
re the SNR *at the receiver*.
The SNR at the *detector* depends in this case simply on the
integration time, which might be chosen depending on the detector
update rate required, as well as the standard SNR-based criteria (prob
of false alarm, detection, etc) ...

The radar equation looks (and is) pretty simple, but it is possible to
make gross miscalculations without verifying the way that is was
derived and how that corresponds to a particular system, and without
considering the difference between SNR at the receiver (eg at RF or IF)
and at the detector. It's possible this is the cause of the OP's
problem in calculation (or else a very low tx power, very high range,
or an unusual detection statistic is required).

ggroupsthomas@gmail.com wrote:
>>you can also use multiple pulses, see Whalen's Detection Theory book
> 
> 
> not in this case, as the OP states it is a CW (continuous wave) radar.
> 
Indeed he did.  multiple CW?