Great Western Coffee Shop

January has not been a good month for punctuality in Cornwall, in particular evening down trains leaving Plymouth between 1630 and 1800. Repeated problems with signals at Par and this week, failure of track circuits and the new signals at Menheniot. Evening trains yesterday up to 90 minute late at Liskeard, and again today 20-30 minutes late. Yesterday no one seemed to have a clue what was going on, my reliable local contacts said that track circuit failures meant they didnt know where trains were!
 It does seem a pity after all the investment in extra signals that we now have a rather unreliable service.

In my simplistic view the use of track circuits as a means of detecting the presence/location of a train in this age of advanced telecomms/GPRS/RFID/Beacon technology, etc, does seem rather archaic and proven to be prone to failure.

If each train was fitted with a device that transmits a unique RFID pulse and each signal (and/or other trackside objects) fitted with devices to receive these pulses, isn't that a more resilient and accurate way of tracking trains?

ETCS does that, but differently. It relies on the train knowing where it is along the track, and communicating with network control. It uses several sources of position data, because using any single system is a risk. In the case of ETCS it's GNSS (GPS etc.), balises in the track, and a high integrity odometer.

Of course ETCS doesn't use the latest technology - this  is the railway! And one known gap in all such systems to date is being unable to directly check which track you are on, except at a balise. I think that leads to extra balises being needed - and they are expensive to install. I think adding new sensors to ETCS is in theory quite possible, though the bureaucracy involved might make it a lot less so. Though note that anything installed in large numbers on the track, even if you think it is inherently cheap, has a high cost both initially and in maintenance (including checking its presence and operation).