[Hidden-tech] Any RF experts out there?

Tom Novelli tnovelli at gmail.com
Thu Oct 20 10:15:11 EDT 2011


Thanks for sorting that out, Jim (an actual RF expert, whereas I'm an
amateur who never quite got into ham radio).
I did a little more reading on GSM, TDMA, CDMA, etc (on Wikipedia, for
what it's worth) and all I can say is, what a struggle!  Timing,
interference, power levels, cell hopping, money, politics, standards
committees...

Then it occurred to me that the industry is going to all this trouble
to handle full-duplex realtime voice communication... and video.
Simple low-bandwidth store-and-forward text communication is ideal for
most purposes.  That includes texting, email, and web browsing.  Voice
is nice sometimes, but why realtime?  We probably spend 75% of our
time in phone calls saying "what? can you repeat that?" because of
choppy signals, talking at the same time, mumbling, accents,
background noises, distractions, etc.  I'd rather it work more like
texting: you hold a talk button (or use voice actuation) to record a
snippet of audio; it arrives at the listener's ear possibly delayed a
few seconds, but *intact*; and they can replay it if they weren't
paying attention the first time.  Heck, they could wait minutes or
hours to respond - no need to be fully engaged in a remote
conversation (I'd rather be fully engaged with my work and with people
in the same room!)  But we're stuck with the status quo because the
cellphone networks had to connect with the landline phone system in
order to be relevant.

Hmmm... we're kinda changing the subject here, maybe we should start a
new thread if this discussion keeps going...

Cheers,
Tom


On Thu, Oct 20, 2011 at 8:29 AM, ussailis at shaysnet.com
<ussailis at shaysnet.com> wrote:
> CDMA (code division multiple access) is not a real spread-spectrum method.
> Only frequency hopping and direct-sequence are. Also TDMA (time division
> multiple access) falls into this more hype than fact category.
>
> I think you will find the interference problem may have more to do with
> operational frequency band, and power output.
>
> I believe AT&T's GSM is a digital form of their old analog signal. I could
> be very wrong, but just a couple days ago I had this discussion.
>
> In a week or two I am going to be looking at an iPhone on the spectrum
> analyser to understand which band (800 MHz or 1800 MHz) it is working on. I
> got into this issue because I know of a "jail-broken" iPhone that works on
> T-mobile. T-mobile is an 1800 MHz system.
>
> LTE, 4G, and WiMAX are not there yet, despite all the hype. Here I think
> what is happened is "marketing" has called (or pushed) 2.5G to 3G, which
> pushes 3G to 4G.
>
> 2.5G was / is supposed to be an "Enhanced GSM," or EDGE. 3G is WCDMA or
> wideband CDMA. Somewhere the speeds have been doubled in the WCDMA spec, to
> 2 X 384 Kbps.
>
> 4G is to be another whole different thing, on a new frequency band around
> 3500 MHz. Here the Gov't hopes to reap (read rape) another 20B from the
> cell phone industry for a "frequency auction." And we know who ultimately
> pays for that.
>
> CDMA only means your signal is coded so many can exist at the same time,
> with others, on a channel. Each separated by a code, so the other end
> processor (tower or phone) responds to only one of the many that may be
> heard. Note that reception of two simultaneous signals of the same power
> level are a 'problem.'
>
> Time division makes more sense to me. Here every signal is chopped into
> small time segments, each has it's own slot for transmission. The processor
> re-assembles the lot. Just like email sent thru the web.
>
> It works because there are buckets of 'down-time' in a typical phone call.
>
> A better choice that is only seen in those DECT (digital European cordless
> technology) is a combination CDMA & TDMA. Alas, those have only caught on
> in campus-like settings, and in China. (I did a DECT design several years
> ago)
>
> Yes, I have seen a DECT for-the-home phone here, with a change in the
> meaning of the acronym. I haven't a clue what is inside, tho.
>
>
> Jim Ussailis


Google

More information about the Hidden-discuss mailing list