Jean ,
Your posting of this site is probably the only comments that I have seen from a company on technology issues. I compliment you for posting
it. If Ericsson can make these comments and name names such as QCOM and why their is95 will not work ,I say why don't Qcom say why it works . I have decided to place their FAQ on this board as some may not be able to get to this site. Jean thanks again for providing not your view or an analyst but facts from a leading provider.
Jim.
Update: The TDMA vs. IS 95 Technology Debate
As operators around the world begin to deploy digital wireless systems, they are faced with a
critical decision as to which digital technology to employ. There are two fundamental
choices: technology based either on time division multiple access (TDMA) or code division
multiple access (CDMA).
Until now, most digital systems have used one of the three TDMA-based standards: GSM,
D-AMPS or PDC. Recently, however, some operators have begun deploying the first systems
using the CDMA standard, IS 95.
With these new deployments, it's possible for the first time to begin assessing the relative merits of
the technologies. The debate -- which has raged for years -- is far from over. But some initial
conclusions are possible. Here are some common questions that are being asked today about
TDMA and IS 95.
Q: Why hasn't it been possible to resolve some of these fundamental technology issues?
A: Fair comparisons of the technology haven't been possible because of delays in deploying
CDMA-based systems. The first IS 95 systems were to have been deployed in 1991, but only
reached the marketplace in late 1995.
All comparisons up till now have therefore been based upon the actual performance of TDMA
systems and the theoretical performance of IS 95 systems. There has been no way to put the
theoretical claims of IS 95 proponents to the test.
Q: Why doesn't Ericsson support CDMA?
A: CDMA technology was first developed for military applications and has been around for many
years. Ericsson, in fact, did much of the early developmental work in CDMA and holds a number
of important patents for the technology.
The issue, therefore, is not CDMA per se, but a particular CDMA standard -- IS 95 -- developed
by Qualcomm. Ericsson maintains that this standard will not deliver benefits above and beyond that
which is already available with TDMA technology. More importantly, Ericsson doesn't believe that
its customers should have to experience the delays, risks and higher costs associated with an
unproven technology.
Q: One of the claims made for IS 95 is that it provides a 10 - 20 times capacity improvement
over analog cellular. Is this correct?
A: This has now been proven to be false. Operators deploying IS 95 systems are consistently
reporting capacity gains averaging six times analog, not 10 to 20. We are aware of no real systems
achieving capacity improvements greater than 10 times analog.
Q: Isn't this still superior to what's possible with TDMA systems?
A: No. TDMA-based systems can achieve increases of between five and 10 times analog capacity
through techniques such as frequency hopping, adaptive channel allocation and microcellular
approaches. The exact increase depends upon the specific TDMA standard and the technologies
employed.
In fact, capacity increases of greater than 10 times analog are possible in certain instances. In
particular, a microcellular technique called hierarchical cell structure allows for a virtually unlimited
capacity increases in specific areas. Hierarchical cell structure transforms a cellular network into a
layered system with macrocells handling fast-moving mobiles, microcells handling slow-moving
traffic and picocells supporting indoor users.
These dedicated cell layers can be installed virtually anywhere to provide extra capacity in densely
trafficked areas.
Q: Isn't hierarchical cell structure possible with IS 95?
A: No. Because of the need to balance the signal power within an IS 95 cell, the configuration of
these cells must be established right from the start. This significantly limits the ability of operators to
accommodate future capacity needs as the population in a given areas shifts or grows.
Q: CDMA proponents claim that IS 95 systems are easier and less expensive to engineer
than TDMA systems. Is this correct?
A: This has now been proven to be false. It was once believed that IS 95 systems, using a 1:1
frequency re-use pattern, would be easier to engineer than TDMA systems, which require
frequency coordination.. Instead, power balancing has emerged as a critical problem in IS 95
deployment. IS 95 cells change in size based on the real-time subscriber load. This makes cell
planning extremely complex and subject to multiple trial-and-error efforts before a system can be
brought on line. It is one of the principal reasons for continuing delays in the launch of systems like
the one in Los Angeles. In comparison, the tools and techniques required for frequency
coordination in TDMA systems are stable and well known.
Q: What exactly is "power balancing," and why is it so important?
A: IS 95 systems suffer from what is known as the "near-far" problem, that is, users that are close
to the cell site can drown out signals that are far from the cell site. This problem is exacerbated by
rapid fading associated with mobility. Hence IS 95 systems require very complex dynamic power
control.
All of these variables -- along with the variables of terrain (e.g. hills and valleys, tall buildings,
foliage, etc.) have to be taken into account when doing IS 95 cell planning. The resulting process is
very unpredictable. Tools do not yet exist to model the problem. Operators must go through an
extensive process of trial and error to get an IS 95 system to work under real world conditions.
This is what happened in Los Angeles, where the operator was forced to back away from the
original deployment plan. In Trenton, NJ, the operator chose not to deal with this complexity and
deployed the IS 95 system on a 1:1 basis, i.e. one IS 95 base station for each existing analog
(AMPS) cell site. In the end, they obtained no practical coverage advantages.
Q: So what of IS 95 proponents' claims that IS 95 base stations have superior coverage area
compared to analog or TDMA base stations. Is this correct?
A: There is in IS 95 a negative correlation between coverage and capacity, i.e. the coverage of the
basestation is reduced when the basestation is loaded with traffic. In practic, IS 95 offers no
coverage advantage over TDMA systems.
Q: So it will be a wash?
A: Well, not exactly. IS 95 base stations are more expensive than TDMA base stations and the
systems engineering is lengthier and more complex. This means that the cost per subscriber --
which is the ultimate measure of performance -- is likely to be much higher for IS 95-based
systems.
Q: What about voice quality? Won't IS 95 deliver better voice quality?
A: Voice quality is a function of the voice coder in the handset. Like TDMA operators, IS 95
operators will be offering service with both 8 kbps and 13 kbps voice coders. Voice quality should
be roughly equivalent to TDMA service.
Q: IS 95 proponents claim that with the large number of vendors planning to produce
subscriber units, handset supply is not an issue. Is this correct?
A: Handset availability is always an issue when new cellular technologies are launched. This is
especially true for IS 95, which employs handsets that are far more complex than TDMA phones
and require up to 90 MIPs of processing power. It will take several production generations before
all of the bugs can be worked out of these new products. If they are rushed to market, return rates
will be high and customer satisfaction will be low.
Q: Are there any indications that there will be problems with IS 95 handsets?
A: The only market where there is a sufficient track record in this area is Hong Kong, where the
first IS 95 system was launched in late 1995. Initial reports have been mixed at best. On April 29,
for example, the Hong Kong Apple Daily carried a full length feature story on problems that
subscribers were having with their phones. User complaints published in other Hong Kong
newspapers include: dropped calls, unstable signal transmission and poor voice reception.
Q: What about the future? Doesn't IS 95's soft hand-off capability make it better for the
kinds of broadband data applications that people will want in the future?
A: As things appear right now, soft hand-off is a liability, as 70% of the active phones in an IS 95
system seem to be in soft-hand-off at any one time. This has negative consequences on system
capacity. What consumers really care about is not whether hand-off is soft or hard but whether
they experience dropped calls. Actual results in Hong Kong suggests that IS 95 has a much higher
dropped call problem than TDMA.
As for broadband data applications, Ericsson is working with standards bodies on two new
standards that will enable data transfer rates of more than 100 kilobits per second (kbps) with
GSM. These are called the High Speed Circuit Switched Data (HSCSD) standard and the General
Packet Relay Services (GRPS) standard.
These standards will make possible a whole range of broadband applications, including access to
the World Wide Web and on-line services, video conferencing and database access services. In
fact, Ericsson, along with Telia Mobitel and Telia Research, recently demonstrated the first wireless
access of the Internet at 64 kbps. This demonstration included the high speed transmission of
sound, graphics and video.
Q: IS 95 proponents claim that now that IS 95 commercial systems have been launched in
the U. S., the issues of technical viability have been put to rest. Is this correct?
A: It is not correct to state that IS 95 systems are in commercial operational in the United States.
The Trenton, NJ test involves just 11 base stations supporting 300 users and users are not being
charged for airtime. IS 95 service in Los Angeles is being offered to selected customers. In neither
case has the service been made available to the general public nor has commercial pricing been
announced.
Even the system in Hong Kong has just 20,000 users -- far too few to truly test the system. Many
fundamental operational problems with IS 95 may not be evidenced until the systems are placed
under significant load (several hundred thousand subscribers). Once this happens, the superior
performance claims made by the IS 95 community are likely to evaporate. In the end, there will be
few if any advantages to compensate operators for the added risks, costs, and delays associated
with adopting this unproven technology.
Q: But don't all technologies go through a "shake down" period? Why is the IS 95
experience any different from what GSM operators experienced in the early stages of
deployment?
A: All new technologies do require a shake down period. That's why it's important to be careful
about accepting at face value the aggressive deployment schedules that are being made for IS 95
systems. Prudence dictates that all claims made about a new technology be scrutinized carefully.
That hasn't always been the case with IS 95.
Q: There are a lot of smart companies building IS 95 systems. What's the bottom line in the
debate?
A: There's no question that many companies are devoting significant resources to deploying IS 95.
And some of these systems will be deployed. In the end, however, these systems are not likely to
deliver the advantages that IS 95 proponents claim. The benefits will be no greater than for TDMA
technology. What will be greater, however, are the costs, risks and delays of deploying the
systems.
Ericsson's 85,000 employees are active in more than 100 countries. Their combined expertise in
switching, radio and networking makes Ericsson a world leader in telecommunications.
For further information, please contact
ke Persson, Vice President, Marketing and Sales
Ericsson Business Area Radio Communications
Tel: +46 8 404 4095
Per Bengtsson, Communications Director
Ericsson Business Area Radio Communications
Tel: +46 8 757 2159 |
