Re: Is 900 mHz legal in the USA for RC??

In article <mik8d11kccgd59k2j1k22h0raeo7fkmsto@xxxxxxx>,
Abel Pranger <abel_pranger@xxxxxxxxxxxxxx> wrote:

| >The problem with spread spectrum at low price points is the lack of
| >update resolution. This is critical when flying a model, but not
| >critical when used for telephony.
|
| I think you are on to it, Ed. It's also true that you have to trade
| off update rate for number of control channels, and that is one of the
| reasons why 2 channel R/C car radios aren't quite the problem that 6-9
| channel model airplane radios will be to design.

Wouldn't that just be a matter of assigning enough bandwidth?

PPM systems update each channel 50 times per second (it varies, but
let's use that as a goal.)

Assuming 16 channels with 16 bits per channel (which gives 65536
possible stick positions per channel, better than the Spektrum's
4096), a six byte `address' for the transmitter (like an ethernet MAC
address) and a four byte checksum (if the checksum doesn't match,
throw the entire packet out) and one more byte for things like setting
failsafe values, that works out to 43 bytes/packet. Sending 50
packets per second, you'd get about 16 Kb/s.

We could also shave much of this off by only broadcasting as many
channels as a given plane needs, or by going down to 10 bits/channel
rather than 16 bits/channel. The Futaba PCM1024 system only uses 10
bits/channel, for example. Doing only 10 bits/channel and 8 channels
it only works out to 21 bytes/packet, or 8 Kb/s.

802.11g can do 54 Mbps in in 30 mHz of bandwidth, so it seems
reasonable that we could put our 16 Kb/s into 9 kHz of bandwidth,
which is approximately the size of our channels now. Assume that we
want to be able to support 50 simultaneous planes at once, and we'd
need 450 kHz of bandwidth total, which is approximately what we have
now in the 72 mHz band.

802.11g cards can decode a packet in just a few miliseconds, and do it
cheaply, so it seems reasonable to believe that a R/C receiver could
do the same. Certainly, doing everything in 20 ms (1/50th of a
second) seems possible, so your latency would be around 1/25th of a
second at worst. How good are your reflexes? :)

If more planes than 50 went up at once, the transmission rate would
have to drop, so the latency would increase, but nobody would get
`locked out' if everything was designed properly.

(Of course, these are all `back of the napkin' type calculations, so
they should go with some grains of salt.)

Not all 900 mHz/2.4 gHz/5.8 gHz systems are spread spectrum, and so
not all would play nice with a system like this. Ideally another band
would be allocated strictly for spread spectrum R/C stuff, 500 - 1000
kHz, both air and ground (since they could play nicely together) but
at this point, it seems to just be a political and economic problem,
not a technical one. The technical parts needed are already out
there, and relatively cheap. But somebody would have to go and make
the equipment. Having the right chunk of bandwidth for it would
certainly help provide incentive for that, but bandwidth is money, so
....

--
Doug McLaren, dougmc@xxxxxxxxxx
"Bachelors should be heavily taxed. It is not fair that some men
should be happier than others." --Oscar Wilde
.

Relevant Pages

  • Re: DDR2 versus FBD
    ... DDR2-400/533 supports 8 DIMMs across two channels ... In general for DDR, capacity decreases as bandwidth ...
    (comp.sys.ibm.pc.hardware.chips)
  • Re: UK: incompatible change to Freeview is coming
    ... that there is far more bandwidth available on both satellite and cable. ... their market share by broadcasting competing channels on that bandwidth. ...
    (alt.tv.tech.hdtv)
  • Re: Camera Card Reader
    ... which I suspect will be even more rubbish crammed into even less bandwidth! ... Whilst some of the digital TV is quite good, other channels are not, particularly where bit-rate has been sacrificed at the expense of packing in more channels. ... prefer not to have to listen to the news on a digital connection. ... quality their best audio engineers could arrange to be broadcast. ...
    (rec.photo.digital)
  • Re: DDR2 versus FBD
    ... fewer pins means more channels. ... the address and command lanes are bypass routed to the next DIMM ... That's why the unloaded latency is higher, ... limit that decreases as the bandwidth goes up. ...
    (comp.sys.ibm.pc.hardware.chips)
  • Re: picture moevement with HDTV
    ... channels in SD they run out of bandwidth. ... it is), 'breakup" of the picture. ... OTA is a much better picture than you get with cable or sat. ...
    (alt.tv.tech.hdtv)