Shouldn’t PPDR be TDD?

The Public Protection and Disaster Relief (PPDR) community has repeatedly made the case for dedicated spectrum for their use.  Most of the studies conducted on PPDR spectrum demands show a higher demand for uplink capacity than downlink (e.g. WIK for the BMWi and ECC Report 199).  This is in part due to the lower spectrum efficiency of mobile uplinks and the need for streaming lots of video from the ‘scene of a crime’ to the control centre. PPDR users are currently demanding 20 MHz (2 x 10 MHz) of spectrum in the (yet to be assigned) 700 MHz band.  But there are problems with these demands:

• The channelisation of the 700 MHz band (in Europe) has not yet been finalised
• The band is currently in use in many EU countries for television broadcasting
• The band will be in high demand from commercial mobile operators

All of this means that not only is there going to be a hotly fought battle over the 700 MHz band but that even if an assignment is agreed, it might be another 5 to 10 years before networks could begin to be rolled out.

But using FDD (paired) spectrum for PPDR networks is an odd way to go.  It would effectively mean that the networks would be designed for the uplink requirement and that the corresponding amount of downlink capacity that would be delivered would be in excess of that needed. This is inefficient!  Being inefficient with spectrum is a practice that is almost outlawed, and we wouldn’t want PPDR users to be breaking the law would we?

But maybe there is a simple solution – the use of TDD spectrum for PPDR services would seem to be a no-brainer:

• A TDD network could be configured to balance the uplink and downlink requirements efficiently
• It is easier to configure TDD networks to provide ‘direct mode’ (e.g. calls from one handset to another without the control of central infrastructure) – a common requirement for PPDR networks
• TDD networks only require a single piece of spectrum, not a pair of assignments and it might be easier to find such a block

It is this latter point that leads to an interesting opportunity.  It was mentioned at the recent joint EC/CEPT workshop on preparation for WRC-15, that the 2 GHz TDD IMT bands (1900 – 1920 MHz and 2010 – 2025 MHz) were likely to be re-allocated as there has been little to no use of them.  Isn’t this an ideal opportunity for a win-win outcome.

• PPDR users want dedicated spectrum
• PPDR requirements are uplink constrained
• TDD technology allows efficient allocation of uplink and downlink capacity
• There is 35 MHz of TDD spectrum that is largely unused at an EU level
• This TDD spectrum is already allocated for mobile and in many countries could be made available now

So why hasn’t one been added to one to produce two? The arguments against this from the PPDR community would probably go along the lines of:

• There is no ‘off the shelf’ equipment available for the band
• It might not be available in all countries (after all, much of it is actually assigned to specific operators)
• It is a high frequency (2 GHz) and would require lots more infrastructure than a 700 MHz solution

All of the above are true, but the result is only one of higher cost and not one of impracticality  (and thus one plus one still make two).  The PPDR community also recently conducted a number of studies (e.g. the work by the Alexander Grous at the LSE) which showed that a socio-economic benefit of several billion Euro of PPDR getting access to the 20 MHz of spectrum they are demanding.  These billions of Euro would easily pay for the difference in cost of 2 GHz TDD networks compared to 700 MHz FDD networks and what’s more the benefits could begin to be realised today, rather than in 5 to 10 years time.

It’s probably also worth pointing out that satellite operator Solaris Mobile have proposed the use of their 2 GHz spectrum (1980 – 2010 MHz paired with 2170 – 2200 MHz) to support PPDR activities, supplemented in rural areas by their satellite services.  Add this to the TDD spectrum possibilities at 2 GHz and there is a congealing of possible options in this range of frequencies.  Whether or not this solution has any traction, there are very few instances in spectrum allocation where there are such blindingly obvious ways to ‘kill two birds with one stone’ as the saying goes.  Maybe if the two birds sat in the same room at the same time, there would be a higher chance of a knock-out.