re: Error Correction in HSCSD
What do you have against (lack of or minimized) error correction? When, why, how much is important? If reduced what is the consequence (in wireless mobile or stationary use)?
c2: "Intellectually honest? I try to be as rigorous as possible in my thinking. The HSCSD "cure" you suggest involving backwards de-evolution from GPRS is flawed because of the lack of error correction and cost. Not a mass-market tool, thus unlikely to make money. My opinion."
c2: GPRSux so de-evolve and congest the networks by ripping out error-correction and giving GSM another name. Not a great way to evolve. Just my opinion.
c2: As far as I know, error correction has not yet been incorporated into HSCSD.
EL: Well, has it or hasn't it? Perhaps. Can you verify that? I can't (or haven't taken the time to).
c2: My apologies to Illmarinen for characterizing his statement as "pimping." Uninformed? I'm not so sure. In the ultimate analysis, HSCSD is nothing but GSM without error correction.
Well, Ilmarinen being the class citizen of this thread that he is, I'm sure he will accept the apology.
As for me. You aroused my curiosity
If this sounds funny it is because it is translated from German:
nokia.at
Probably available in English if you care to look.
>> A New Coding Pattern
HSCSD is a new coding pattern for GSM data communication. With GSM altogether 22.8 is kBit/s at bit rate for order, with Datenuebertragungungen of it however only to 9.6 kBit/s is used - the remaining 13,3 kBit/s for error correction bits are used, so that the data arrive complete and safe.
With speech transmissions (thus when normal telephoning) it is already used since beginning of the GSM standard 13 kBit/s which can be explained thereby that the human ear does not register small errors, these during data communication is however inexcusable. What is situated thus more near to use as for data communication more of the max. possible 22.8 kBit/s?
The response to it gives the new channel coding, with which 14.4 kBit/s for the actual data and only the remainder for the error correction bits are used.
The gain of 50% at rate has also a small disadvantage: the cell cover reduces easily, because the further the base station is distant, the error correction bits become the more important. And if fewer corrections can be executed, then the distance of the base station may be to large not, if no problems are to occur. <<
... so now you know. Now I know. It's been like the blind leading the bloody blind. <g>
BTW: As you know GPRS may speed up with new coding schemes - based on reduced error correction - but now that EDGE is nigh, they may nebver be used, but if they are, same consequence:
The first option carriers will have to enhance GPRS is adding new coding schemes that increase rates to 15Kbits/sec or 21Kbits/sec per time slot (up from the current data transfer rate of 13Kbits/sec). These new coding schemes reduce the amount of forward error correction, and so require a strong signal. No carriers have committed to these coding schemes, however, because the next upgrade step, EDGE, offers far better performance.
... of course there are also new vocoders under consideration. GPRS will be around for a long time, and one way or another will get faster.
- Eric - |
