Hi All,

If I do attempt this, I really should get a copy of Skolnik's book. I
took a course on radar while getting my BS in EE, and a lot of the
references were to Skolnik (we even got some photocopies of chapters out
of the book) but I don't recall seeing that particular equation. Thanks
for the info!

I did some "back of the envelope" calculations which showed that in a
1MHz bandwidth, with a dish which gives about 1m^2 effective area (a
little more than 2 feet diameter,if your feed is very good in delivering
a unifrom dish illumination), running about 10 watts peak power (just
sending simple pulses about 2 microseconds long or so), with a mediocre
receiver (NF=3dB), and a large target (50m^2 RCS, equivalent to a 737 or
larger), you could get out to 24 miles. Better NF or higher power
increase range according to the 4th root of power.

Since I was suggesting only 100mW output power (since I haven't been able
to find a convenient way to cheaply produce a lot more at 5GHz), that
would give a range of 24 divided by the 4th root of 100 (10/0.1) or about
7 miles.

Using a better coding scheme (other than just simple pulses) would help
a fair bit. Linear FM chirps (as mentioned by a few posters) or another
form of pulse compression (phase codes, where you send a long burst of
binary or trinary or higher-order phase modulation and then use a proper
matched filter on the other end) could probably double the range without
having to do anything extremely fancy.

As for ID, I was thinking of just stopping the radar-type emission every
10 minutes and just turning on the transmitter in CW mode and sending CW
id, then resuming the scan.

Finally, on the subject of weather radar, I'd like to ask the experts
here something: How does weather radar obtain the component of wind (or
droplet) velocity perpendicular to the direction of radar beam
propagation? In other words, the doppler freq only tells you how fast the
object is moving toward or away from you. When dealing with aircraft, you
have to track the individual plane in order to get the other component
(that is, subtract its last position form the current one). Its seems to
me as  though this isn't possible with  WX radar since there is no target
to track, you are  getting an echo from millions of water dropplets
throughout a whole area.

Sean


On Mon, 6 Aug 2001, Matt Bennett wrote:

> "J.Feldhaar" wrote:
> >
> > Hi all,
> >
> > another tack on the issue: what about the bandwidth required for good
> > distance resolution, assuming a pulse or a chirp is used, considering the
> > mandatory use of multi kilowatts of peak power needed for the 30 miles
> > range earlier mentioned in this thread...??
> >
> > I personally have considered doing just this and I have not found an
> > amateur frequency range where this would be permitted...even the 5 GHz
> > band
> >
> > Comments?
> >
> > Greets,
> > Jochen Feldhaar DH6FAZ
>
> Ummm... What's "good resolution?"
>
> From _Intoduction to Radar Systems_ by Merrill I. Skolnik- the error in
> the time delay of a pulse (using rising and falling edges of the pulse)
> can be written (Eq. 11.5):
>
> Rise time error=(t/(4BE/No))^(1/2), where B is the bandwidth, E is the
> signal energy, No is the noise power per unit bandwidth and t is the
> pulse width.
>
> The range error would be 1/2*time error*c) (1/2 since radar is 2-way,
> but that also means that your power recieved goes down as a factor of
> 1/r^4)
>
> A chirp would get you a greater effective power on the target-
> increasing E.
>
> As to peak power required and so forth, I direct the curious to chapter
> 2 of the same book, where the radar equation is discussed in some detail
> (50+ pages).
>
> That's your range resolution, but azimuth resolution is a factor of your
> antenna and range.  I'm particularly fascinated by synthetic aperture
> radar- while the range resolution is still a factor of system bandwidth,
> azimuth resolution is *independent* of range with a focused SAR.
> Depends on a moving radar (or target) and you can't make an image in
> "real time" but it is fascinating anyway.
>
> Matt Bennett
>
> --
> http://www.piclist.com hint: The list server can filter out subtopics
> (like ads or off topics) for you. See http://www.piclist.com/#topics
>
>
>

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics