Digital Mobile Wireless in the Americas
Hello Raymond,
<< If there where would have been any american customers on the 850 band then GSM would have been in service on that band much much faster than what IS-95 was capable of. >>
I'm not sure I totally understand that statement. Until November 1995 when APC Sprint Spectrum launched PCS-1900 (GSM) service, all US cellular subscribers used 850 MHz AMPS or AMPS/TDMA.
<< I respect your view but if you think that "American technology" had nothing to do with the choice I think you are wrong. >>
Stockholm born European technology was initially selected over "American technology" for the U.S. AMPS migration to digital, although San Diego born CDMA was subsequently cleared for standardization and AMPS migration.
After public demonstrations in California for the Advanced Radio Committee Subcommittee (ARTS) of CTIA in 1987 of Motorola's FDMA system (following AT&T's earlier FDMA demonstration) using a 7.5 kHz time-slot channel and a lab tested 6.2 kbps vocoder, and Ericsson's TDMA system using 15 kHz bandwidth and a 13 kbps vocoder (that when matured became the GSM-FR coder), Ericsson's TDMA system was chosen as the US digital migration path for AMPS and from the outset dual-mode 850 MHz handsets were a strict mandated requirement. U.S. TDMA was, of course initially standardized in TIA as IS-54 starting in 1989 and completed standardization in 1992 and was evolved later to the more workable IS-136 TDMA standard. IS-95 completed TIA standardization in 1993.
No effort was made in the prior decade to standardize GSM in 850 MHz. It would have been counter productive for Ericsson. In the last decade TDMA became the dominant digital mobile wireless standard in the Americas and Ericsson was the dominant TDMA infrastructure provider and they worked closely with the UWCC members to evolve IS-136 to ITU approved IMT-2000 3G TDMA EDGE as a migration path to WCDMA using the GPRS backbone.
Interestingly enough, in 2000, Ericsson (hoping to be as successful with CDMA technology acquired from QUALCOMM as they had with TDMA) was aligned with CITEL and CDG, QUALCOMM, Lucent, Nortel, and Motorola, and against GSMA, Nokia and Siemens, in lobbying Brazil's Anatel to preserve the 'Americas ANSI-41 Band Plan.' That GSMA lobbying effort was obviously the genesis of the initiative to standardize and commercialize GSM in 850 MHz that has allowed GSM to become the dominant mobile wireless technology in the Americas, and it didn't take long for Ericsson to aggressively back that initiative.
<< How difficult can it be to go from the 900 band to 850? >>
No more difficult than it would be for CDMA2000 to migrate its 850 MHz systems to 900 MHz, but we still haven't seen that happen.
Also considerably less difficult than CDG described it in Herschel Shostek's FUD laden June 2001 whitepaper published shortly after AT &T Wireless and Rogers announcements of their TDMA to GSM flips. It still took several years from the commencement of 3GPP standardization to initial commercial implementation.
<< Then when GSM got a chance with the 1900 auctions the Americans courts was used to try to stop it. >>
The American courts didn't stop it, and didn't come close to stopping it with the frivolous suits over SARS emissions. PCS 1900 (Modified DCN 1800 and Upbanded GSM Phase 2) standardization began in JTC and was completed in TSC T1P1 in the US and approved as a US ANSI standard in March 1995 and APC Sprint Spectrum launched all digital GSM service in the US 8 months later in the Baltimore/DC metros.
<< Independent studies has afterwards come to the conclusion that GSM with the EFR coder which was used in the GSM 1900 rollout had a better speech quality than IS-95. >>
The 'US-1 EFR' CELP based codec developed by Nokia and the University of Sherbrook in Canada to NPAG (North American PCS 1900 Action Group) requirements, selected by NPAG over 4 other candidates, standardized in T1P1 here in the US, and later standardized within ETSI SMG was a great voice coder. It was not, however, commercially available when Sprint Spectrum, Bell South Mobility DCN, Pac Bell, and Omnipoint launched GSM service in the US in 1995 and 1996 and initial GSM handsets in the US used the original GSM-FR 13K RPE-LTP coder. The EFR codec became commercially available in a (very) few devices in late 1996. Widespread in 1997. About the same timeframe as QUALCOMM's QCELP 13K vocoder.
While lab measurements are important, at the end of the day evaluating speech quality requires subjective analysis. CDG had several subjective studies posted that purported to show that the speech quality of QUALCOMM's QCELP 13K vocoder was preferred over the GSM EFR codec. It certainly wasn't to me and I much preferred the more natural speech quality of my EFR enabled Bosch worldphone that I used for overseas travel to the QCP800 I purchased from BAM in early 1997 and used for nationwide AMPS and CDMA domestic travel, but the digital speech quality of the QCP800 was pretty darned good for its time.
<< Another thing that was always mentioned was the superior battery times when using the IS-95.I think we all know the truth of that statement by now. >>
Yes. We do. No contesto.
Let me close with this thought. Despite the fact that GSM got off to a slow start in the U.S. and the upper and lower Americas, at the end of 2006 according to Informa T&M WCIS (formerly EMC) while CDMA boasts 48% of the subscriber connections in the U.S. & Canada, 39% of the subscriber connections were GSM, and in the Americas as a whole 56% of subscriber connections were GSM and 33% are CDMA. Today, CDMA is the dominant technology in (only) 3 global markets of any consequence -- the U.S, Canada, and Korea, and GSM has a major presence in two of those markets.
Best,
- Eric - |
