Toolkit
Story: China developing domestic 3G phone
TD-SCDMA address the main problem with CDMA - how to cope with congestion.
CDMA has a time to transmitt function with an expotential O - main component of the "apparently insignificant part of the function". The O is a function of simultaneous users and packets about. GSM solved this by rejecting CDMA. (This is the air interface only!).
By introducing TDMA into CDMA, as GSM, you have the possibility to impose "one thing at a time" when the network is overloaded. The roaming problem is strange that they have not solved - and they may take the GSM sollution.
WCDMA will have this problem.
WLAN has this problem
CDMA2000 has this problem.
CDMA has this problem - all behave very poor once imposed with heavy load - just as the (queue) theory suggested.
DECT/DPRS will provide non-mobile IMT2000 today outside the USA.
It seems like we forget old theory, and ignore old findings - while the Chinese looked at the problem and came up with a diffrent solution - but similar one.
Full Talkback thread

