It will only meet the needs of part of the users. The software is
machine intensive and not suitable for hand helds at this point. It
works for Cell phones because the major portion is done at the cell
tower and then sent to the unit.
Andreas van Hooijdonk wrote:
Quote:
CSR has acquired CPS and NordNav, both providers of software GPS solutions
that will be added to CSR's Bluetooth chip.
The following article has more details and background information:
"Dale DePriest" <Dale@gpsinformation.het> wrote in message
news:12qvc289nucih5f@corp.supernews.com...
Quote:
It will only meet the needs of part of the users. The software is
machine intensive and not suitable for hand helds at this point. It
works for Cell phones because the major portion is done at the cell
tower and then sent to the unit.
Andreas van Hooijdonk wrote:
CSR has acquired CPS and NordNav, both providers of software GPS
solutions
that will be added to CSR's Bluetooth chip.
The following article has more details and background information:
When I read those articles I see that most of the technology still only
works for cell phones and not fully autonomous handheld devices. Saying
that the BT computer can do both jobs is just not realistic if the
computer has to do the full calculation on its own unless you have a
power hungry processor in there. I guess time will tell on what they are
able to achieve but the original unit shipped by ALK needed a hungry PC
to process the incoming data.
Dale
Andreas van Hooijdonk wrote:
Quote:
"Dale DePriest" <Dale@gpsinformation.het> wrote in message
news:12qvc289nucih5f@corp.supernews.com...
It will only meet the needs of part of the users. The software is
machine intensive and not suitable for hand helds at this point. It
works for Cell phones because the major portion is done at the cell
tower and then sent to the unit.
Andreas van Hooijdonk wrote:
CSR has acquired CPS and NordNav, both providers of software GPS
solutions
that will be added to CSR's Bluetooth chip.
The following article has more details and background information:
It's a step forwards for the price and a step backwards
for almost everything else.
In general, the 'classic' approach of analog frontend,
dedicated digital baseband (correlators) and small
controller CPU is the most energy efficient.
If you start to move the hardware-software boundary
towards the antenna, you start incurring a power penalty.
This is because you are using general-purpose hardware to
do a specific task it isn't really suited for.
Software GPS makes sense if you have to make do, cheaply,
with existing hardware. Then you can claim that with
the addition of a $1 GPS frontend you add GPS were there
was none before.
If all you need is an occasional position fix (emergency call),
Software GPS is currently the cheapest way to do it.
But it's not really suitable for navigation. If you read
the link above carefully, there is a mention of an improved
algorithm yielding a position fix in 3 seconds. That is
a mobile phone processor running flat out (at 1W or so)
for 3 seconds. Compare that with the less than 100mW of
a SirfStar III, doing a position fix every second.
There is also a tradeoff between sensitivity and processing
requirement. A simple 12 channel correlator (10 year old
GPS) does about 200 operations per clock, at 6MHz or so.
This is more than 1 billion operations per second.
On most mobile devices, software GPS must reduce that to
be able to work at all on current mobile processors. This
leads to lower sensitivity.
On something the size of a modern laptop, high sensitivity is
possible, at a significant reduction in battery endurance.
Then there is the aspect of flexibility - in theory it is
possible to do a software upgrade for Galileo. However
Galileo requires more processing power than GPS - it may not
be practical.