who is EMP ( expanded out?)...Ericsson Mobile Platforms....Same group that did US TDMA.
Be carefull of leads and hype....
that this group is still working within Ericson at all is saying alot, as they misfired TWICE in a GIANT way and still survive.
On US TDMA, they ran out way too fast, taking the lead (along with Motorola), but they pushed the filter technology way too fast. There is a very complex front end filter called an equalizer, which helps find signals in random noise, and in this case the clock signal in which to find the data. The one that was specified for US TDMA (and was almost there in GSM at the time), was way out on the leading edge of the technology and was owned by EMP. It was a way of pushing MOT to the back burner, but of course was licensed to the 3GPP (or whatever existed like that then) group. But when they sold this to the US carriers, they pushed way too hard and were not able to get the channel characteristics right or THE Equalizer for about 4 years. By then they had rushed tons of chips out to market and made wild claims about how it worked and what it would do. They were advertising 3 slots of TDMA data and each one could carry a voice channel, so the promise was to have 3x the capacity for voice calls. But they designed the channel to have EXACTLY 3x the voice capacity at Analong phone quality levels. At the time (1990-1993) this was the state of the art. So when the equalizer failed, they had to keep putting in syncronization and timing bits on guardbands so that it got to about 1.7x the data rate and then with analog phone level voice quality. It would be VERY cool if someone could play the analog voice samples of the day to show just how lousy the channels were and how much artifacts the public was willing to setle for and still make calls. (PJ - take notice for CTIA type demo??)
Anyhow, the US-TDMA chipsets, handsets, systems were all out and growing with upstarts like Bell south, Criag McCaw, etc. But at the same time, a little upstart in San Diego started making calls on a technology which had the stuff finished, working, and offered voice quality much closer to the land lines of the time. At the start, the little engine who could, set the voice quality to the same as analog AND US TDMA, giving 20x capacity on day one, with ample range to expand to higher data rates if better quality was needed. When they demo'd the stuff, the Pactel guy (Bill Lee) asked what it would sound like if they dialed in the SAME as landline quality, and hence the 7x analog capacity was born and one could never go back... But better still, the system was still 3-4x more capacity than the US TDMA and the analog system out there. So from 1994 to 1996, the TDMA systems tried to come out (AT&T bought that division from McCaw), but failed to compete. They pushed their systems guys hard to find the right stuff, but it did not happen. Now tons of money was spent on defeating politically the standard, but technically there was no way.
FAST FORWARD to 2000.....same group pushes out EU 3G, after a few trials of that little upstart stuff in Bristol with Lucent/HP/others of the US CMA standard, but again, the same group misfired once again. This time, due to fears that the US might withdraw the GPS standard, they wanted to drop the dependence on synchronous GPS basestations and go with an asynchronous standard (What ever happened to Galileo anyway??). Once again, guess what part of the design was being pushed to the limit? yep....the same stupid Equalizer. At least most scientists learn from getting burned and accept that they do not touch fire twice.....but this group did it again. So they rushed out the WCDMA standard, but forgot that at the very time when the signals are the worst, you cannot be searching for a way to find them in the low level noise. So no way to find the new timing to lock onto the signal. So once again, the 3G group got burned.....but wait...US CDMA in Europe????Nope, they turned to an even better smoke and mirrors game this time. (At the time, there was about a 30-40% chance that Qualcomm could have gotten US CDMA into Europe....thus making hte EU standard a much differnt thing..)
the net effect of the bad equalizer is to turn up the signal to noise so that handoff can be done while you still have a bigger signal. Doign that meant that the cell radius had to get smaller or power amps larger. Since they had handsets out there and power amps meant larger batteries or shorter lives, then they had to figure out a new route. Either that or CDMA goes to EU...(Hence IMJ pounding on thier collective heads in 2001 that WCDMA would be 5 years off)....PANIC!!!!
Having fixed the US TDMA equalizer problem about 3 years late (Circa 1997-1998), the put GPRS onto the systems which immediately caused the system to overload. So they had this cool thing to work with now. Why you ask, what for? Oh, if your system is overloaded, then you need more basestations....so if you put in more (like about 3-5x more) then your GPRS loading will turn out ok. But if you knew what they wanted, you would know that at 3-5x more basestations, WCDMA had the right density to actually not drop calls at every cell boundary....So the great NOKIA/ERICY Hoax started out. Selling billions (+10-15B) of 3G bts to people who had already bougth $11b worth of the stuff to make it work right. This was the mess with the hot handset in the pocket, the island testing site in a dictatorship that actually abandoned the city they tested in, etc.
In 2003, Zyray came along and made a circuit which improved DSP systems and improved the problem by about 2.3 dB, but not the entire differnce.
At the same time, Qualcomm came up with their own circuit to fix it up and improved it a little, but the BTS were growing by then.
So, EMP you ask? Lead? for real?
If they were car mfgs they would have been out of business...
Be careful of leads and leads.
BTW - does the need for spectrum sound alot like the same cry from AT&T? Oh yea, old McCaw guys there...almost forgot...when the real issue is just put in more small cells and get on with it. but they got distracted for 9 months on trying to put more smoke and mirrors on the DOJ. maybe now they will just put up cells and get on with it. Physics is a hard thing to overcome......Politics and technology do NOT mix.
Sorry, you asked at a time when I am sitting around with family just goofing on SI....
(and no I do NOT expect that the senior VP guys from Ericson will appolgise for shouting me down at an IEEE confernce on CDMA systems design by simply yelling "It will never work" like they were occupy wall street guys or such...and i was simply showing RESULTS, not theory.)
Take care..Have a happy Holidays!! |
