RF transceiver chips with low-kHz bandwidth?

for

Yes, and this used to matter for ad prices and such, but now that the majority of viewers are on cable/sat, the stations don't care as much anymore... :-(

Charlie

for

You'd be surprised how many people out here are not on sat/cable. And those tend to be the ones with well-paying jobs, meaning disposable income.

However, there is a trend that the media moguls seem not to grasp and this one is much more serious: The kids of those folks. Once they head off to college many forego TV altogether. Internet and Netflix is all the rage there. I know several who don't even have a TV anymore, they watch everything via their computers.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

Yeah... that part of the Friis equation is very misleading and a source of confusion.

Oh, I dunno, I doubt I was the only one in my college class who blanched a little at the end result!

Wikipedia mentions it as well:

.

Yeah, but the gain term for the antenna makes up for this and hence the overall result of the Friis equation is correct. When someone says, "if you double the frequency, the path loss is twice as large" the appropriate response is often, "well, no, because if I get to use the same-sized antenna, its gain will likely have doubled as well."

As you allude to, where Friis has his "free space path loss" there should really have a qualifier along the lines of "free space path loss using the aperature of an isotropic radiator which, at lamba^2/(4*pi), is not exactly intuitive and can lead to unexpected results." :-)

I don't know that I've seen a derivation of the aperature area of an isotropic radiator, though... my recollection is that it was taken as a given ("it's equal to a circle with a circumference of lamba") and from there you started deriving things like gain as 4*pi*A/lamba^2, so then once you have the fields and hence the Poynting vector for something like a dipole, you could calculate its gain relative to an isotroptic radiator (using gain=area of sphere/aera of antenna pattern) and then the effective aperature from there (using gain=4*pi*A/lambda^2). But perhaps I have this backwards, and there's some motivation for *defining* gain and from that determining an isotropic radiator's aperature.

---Joel

A single pair of feed terminals can couple only to a single mode of the EM field. The etendue (area-solid angle product) of a single mode is lambda**2/2, and a fully polarized isotropic radiator radiates into 2*pi steradians (4*pi steradians is a sphere, but there's a cosine term due to obliquity that makes it 2*pi instead).

Therefore the area of an isotropic radiator is lambda**2/4*pi.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net

Mmm.... ok!

I should probably page through some of my old EM texts tonight and see what I find...

Feel free to include this in the appenfix of the next edition of your book. You can probably work it into a discussion of the derivation of the Airy pattern for circular aperatures or somesuch? :-)

Thanks,

---Joel

f a

h_loss.

f you

nna,

e

ly

ropic

s

ted

elds

ulate

aera of

s some

This is great, (I'd never heard of the Friss equation.) Are dipole antenna's like electric dipole transitions in atoms? For Rb atoms you can measure the cross section (of a single atomic transition) as a function of density* and find that it's ~(wavelength/pi)^2? or something like that. It doesn't depend much on the size of the atom.

George H.

*density of Rb is a strong function of the temperature and it's hard to get a good measure of the temperature.

d

t- Hide quoted text -

"Therefore the area of an isotropic radiator is lambda**2/4*pi. "

Excellent, thanks Phil, It's the same for atomic dipoles, say Rb atoms at 795nm.?

George H.

It's touched on briefly in Section 8.4.2 of the second edition.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net

Hide quoted text -

Yup. It's a thermodynamic argument. I should have said, "effective area".

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net

Thread read. Just the same, can you add (modulate with) some nice predicable PRN thus using the extra bandwidth to your advantage? Regulatory issues may make hash of this idea.

A good idea and most likely agencies won't have any beef with that if you remain under the power limits. However, the chips are rather restricted in terms of modulation. The buffet is usually limited to on/off AM, bi-level AM, FSK and PSK. That's it :-(

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

the

got

Any

antennas

to

uC

actually

The radio chip should just see the (bi-level or on/off) AM, the rest done in a PIC or similar (at each end).

That's how people do correlation and it does improve range tremendously. Until you get interference of the non-pure noise kind ...

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

With AM, you really need a good AGC to feed the detector a proper signal level compared to the detector threshold. The signal strength can vary considerably due to multipath nulls, even when there is a minor movement.

In order to generate a goof AGC signal, the data should not contain long strings of 0's or 1's, thus, at least a scrambler or preferably a suitable code should be used, to generate a more even distribution of

0 or 1 sequences.