Re: [Discuss-gnuradio] GPS tracking on a map over mobile app software using RF

[Marcus Müller]

Hi Antti,

Well, it wouldn't really help you at all, because, what you're going to do?

Legally, you could only use the ISM bands (which are already congested). 
Or you'd need to buy a licence for licensed spectrum so that people can 
use that spectrum; you'd need some central station to arbitrate access 
to the expensive spectrum you've bought, because in a dense situation a 
"oh, everyone just uses this whenever they please" is not a solution. 
People that buy licenses for spectrum, offer infrastructure to control 
medium access and route data around have a name:

They are network operators. And it's unlikely you'd build a network that 
is better at being efficient in using spectrum to transport many 
people's data than LTE.

So, whilst adding another piece of (universal) radio equipment to your 
(fixed-purpose) radio equipment is an interesting thing, it won't solve 
the problem of spectral congestion. You could argue that your GPS data 
is "really not much data at all", and you'd be absolutely right. To 
transport "not much data", you can use "not much bandwidth", or a very 
high spreading factor, and get your data across a long distance at 
limited transmit power (which is a legal necessity in ISM spectrum). 
Now, the problem with that is, that there's really really many people 
that are trying to do the same, at the same time, in an uncoordinated 
manner. It's really hard to find sufficiently orthogonal spreading 
sequences so that this works out positively. Simply because "not much 
data" times "really many people within reach" becomes "pretty much total 
data", after all.

In short: if you have cake, you can distribute it. If the cake is 
already distributed, you either need to buy more cake (buy an LTE 
license and operate an LTE network) or distribute the crumbs (trying to 
find unused spectrum in the free-to-use ISM bands). Either way, there's 
no free cake.

So, if you ask me, this (sharing a few bits of position information) 
sounds a lot like the emerging problem field of interference modelling 
of IoT sensor node networks, only made harder by the fact that the 
network operator might not have reserved spectrum. As far as I know, 
there's not been a great solution to that problem.

But, to be brutally honest, sounds like you're solving a "non-problem" 
to me. It's been quite a while since I wasn't able to even log on to an 
LTE network at an event because of too many people, and not because it 
was in an abandoned subway station, well shielded from the LTE network. 
In that situation, well, you're in that hole, and whoever gets your RF 
signal is there, anyways. Many apps share data in the background – but 
they do so in blatantly inefficient ways, generating kilobytes and 
kilobytes of JSON or XML traffic where a simple exchange of two 32 bit 
numbers would have sufficed. So, I'd rather try to use the 
infrastructure for data transport that is given to you in a clever 
manner (namely, minimum amount of transferred data, make the network 
stack on your mobile device do actual quality of service prioritizing 
instead of what handwaving s*** Android does these days, disable data 
communication for all but important applications when getting little 
chance to send data – these are all software modifications to be made to 
a smartphone rather than additional hardware PLUS the software) than to 
try and build up a parallel infrastructure.

Cheers,
Marcus


On 08/14/2017 06:31 PM, Antti Ruohonen wrote:
> Thank you for your reply, Marcus. Your suggestion of an attachable 
> device is totally fine, actually very good idea.
>
> So, this SDR device would be the medium to the software radio (app in 
> the mobile phone), which controls the device. How would this scenario 
> be possible to achieve the goal? Goal is to send (GPS) data over the 
> radio frequencies to update the app's map. :)
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio