Oh-no, another Qudiot..
Look up some old Q-specification and at the bottom of the
page you will see "not yet implemented" for the 48 tap
filter.
Tough job for the RF guys, but easy for the DAC-DSP days.
But like you, I am sure they fixed it.
There is actually on valid point in the 4,096Mcs debate,
but no Qudiot have understood to mention it (well, except
Clark, but he was no idiot)
I have no idea what you are dancing with, my GSM phone
burps 4 times a day and opens it ears once every other
second, synchronization and power control,AGC setup
is instant, battery standby times are great, doesn't
participete nor waste battery in gaussian noise cocktail
parties, basic functions runs on less than 10MIPS.
With well isolated channels and timeslots QoS is
predictable, great for businees, although I still
remember that one call Sonera dropped on me 2 years ago
when I passed a bus on the other side of the base station.
In this business I have no need for transmitting
live pictures from my mobile sauna at 2MBps, although
my buddies at Hki Univ of Tech still have the world
world championship in that.
If in urge and need I send a SMS message to get company.
Interesting that QCOM wihdrew, coitus interruptus,
from the Finnish sauna beauty licensing contest, their
better half is now providing WLAN access.
Small tap coefficients is said to be the reason.
--
"Help me out here. I am not seeing radiation outside of its operating channel."
Your eyes, as your ears, have dynamics of around 100dB,
should be extremely easy for even an almost blind
to see the 30dB within the first 1MHz and and the
60dB for the next.
-30dB is the real interest rate your bank used to give you,
3% or american beer, -60dB is like the alcohol in some
blood, 0.1%, -100dB is what it takes magic to produce
and understand., 0.003%, only 50 ohm RF guys need to
apply.
Why is the spectrum down 30dB (peanuts, 3%) at a
bandwidth of 3 grid lines, one being 0.5 Mhz, giving
1.5Mhz, not the 1.25Mhz mentioned earlier??
Bt there seem to be 16 taps used for making the
passband equiripple 4 bumps, leaves the rest 32 for
the stopband, which explaines the lousy transition
band.
But clearly they have not really optimized the
FIR filter for the lousy RF filter, making that
passband some 2-3 dB sloppy, losing half of the
capacity.
Gaussian background noise and finnish regulations,
me thinks you should read your Wozencraft,
skip the Jacobs and try to figure out what Gauss wrote.
Ask Viterbi what is the difference between filtered
gaussian noise and a filtered CDMA signal, he might know..
--
And finally, digital filters run at sampling rates,
charge and discharge gates smaller than fP, when
they need to be awake and running.
When burping, or speaking, the PA is the only thing to
drain the battery.
Except for that color display, tough stuff to
reload a totally new set of pixels 10 times
a second, but the japanese are taking care of
that.
Ilmarinen
Anything else??
Btw, m(illi)Hz is something submarines and geophysics
use, M(ega)Hz is something for regular mobile guys,
800kHz for 4,000kHz is 1/5 = 25% alpha, less than a
v32bis modem.
Do you expect 100 hours talktime from your QCDMA
handset?? or do you have a problem like most
qudiots with talk time, burp time and wake up
time for a couple of timeslots every 1-4 second??
Or even the difference between speaking and listning??
Like overheating GPRS PA-amplifiers when receiving
4-7 time slots??
Foot in the mouth or ear??
(I'm sure Ruffian would suggest a third place) |
