Reply to post: Re: hopping along between masts

Re: hopping along between masts

"I thought GSM (2G) was designed to work well at speeds of up to 150mph and that later standards have improved on that?"

Citation(s) welcome

That's why I asked the question.

Ok, so I've just spent my lunchtime looking. There doesn't seem to be an easy answer, mainly because GSM (2G) and later protocols are adaptive - they can alter various transmission parameters depending on the state of the radio link. There are several issues; there's the absolute distance from the base station - this applies to all mobile terminals and the standards have defined parameters which limit the physical size of a cell.

There's doppler shift - if the distance between the mobile unit and the base station is varying (the ultimate case, of course, being a mobile heading directly towards or away from a base), again standards have defined parameters which allow a certain range of frequency shifts before everything fails.

These two seem (if I've read the standards correctly) to be at least partly adaptive by varying the transmission parameters, trading off robustness for capacity of both the individual link and of the cell as a whole.

There is also the "network overhead"; that is the time it takes to handover a call from one cell to the next and the amount of data that must be passed around the network in order to do so. If the cells are close together and if the mobile station is moving tangentially relative to the cells there are some circumstances where the overhead of handoff becomes too great for the network to handle and calls might be dropped.

However, finding a simple reference which puts simple numbers to these parameters has been somewhat difficult. Anecdotally, most people have little or no trouble making calls from trains in areas with good coverage. Trains in the UK operate at up to 190mph(ish) though more normally at 140mph or less. Anecdotally there are also reports of passengers making successful calls from aircraft, below a certain altitude anyway.

One slightly-relevant reference I've found is this PDF which explains GSM-R. GSM-R is effectively standard GSM with some parameters "tweaked" (and some security enhancements) to make it more usable for rail. Some of those parameters affect the maximum speed at which the system works, and GSM-R is specified for speeds of up to 500km/h (310mph). That document is linked from this web page which explains the rationale behind GSM-R. There is also this overview (PDF) of the situation regarding a successor technology, which claims that with some tweaking LTE could be made to operate at 500km/h (see, for example, footnotes on p44).

Without reading (and understanding) all the specs it's a bit difficult to be more specific. Some more information might be gleaned from: A comparison of GSM-R and TETRA though this appears to have been written by someone with an axe to grind.

Phew! That was a busy Friday lunch break.

M.