By Mika Skarp
Just 3 short months ago, AT&T was awarded an historic bid to build out the U.S.’s public safety communications infrastructure. Under the FirstNet flag the omnibus project promises to deliver broadband services to public safety operatives from coast-to-coast by the end of 2017. 5 years ahead of schedule and only 9 months after the contract’s signing, FirstNet’s drive and leadership in this effort is as inspiring as it is exciting.
To put this into perspective the UK’s EE and ESN (Emergency Services Network), bearing a similar mandate for the last three years and amid much publicized delays, will only be completing the project two years from now. Still, let’s bear in mind that other countries that have announced such plans, spare South Korea, have yet to even begin. This only galvanizes FirstNet’s shining example.
But delivering assured mobile broadband for public safety on such a short time scale begs the question, how is AT&T able to do this? Aside from the added band 14 frequency, which is certainly nice to have in the AT&T frequency farm, it is not essential. To the more challenging mandate of delivering assured, life-critical connectivity, they are in a sense iterating the solution and doing so at this point using static or fixed network slices. This is not surprising to us as Cloudstreet has been offering static slicing solutions to our broadcasting customers for several years now and with excellent results. For this relatively simple capability, AT&T has a set of tools already at its disposal within their LTE network. As such, and in this capacity nothing new is needed to get started. Still, a core requirement of FirstNet’s system (written into the RFP as “Local Control”) will not be possible using static slicing. Rather, its next generation - dynamic network slicing - is the end goal, and Cloudstreet’s Dynamic Profile Controller™ (DPC™) is at the very heart of it.
At the recent PSCR Conference in San Antonio Texas, we touched on many topics about the AT&T/FirstNet roll-out, and in those discussions many questions came up about static slicing; Can you differentiate network performance between Apps? The answer is no, you can’t. And what happens if a police officer is having a coffee break and decides to stream or download a Netflix movie, will that get the special, static profile? To that, the answer is yes it will. And with this Q&A stakeholders at the conference could clearly see the limitations of this first, static slice-based iteration of FirstNet.
One question that was not asked, but was likely on the minds of many at the gathering was how other AT&T customers would be impacted by the provisioning of static slices for public safety that would permanently reduce capacity in the cells they occupy. Would they negatively impact customer experience? Based on our field tests and analysis the answer is no. While this would not be the case were we deploying massive IoT or M2M static slices on an industrial scale, the total number of public safety operatives represents but 1% of the population and is thus too small to make any noticeable difference. Additionally, in the case of a major incident, like say, a terrorist attack, the network would be congested anyway, but this also points to the limitations of static slices which wouldn’t satisfy the real-time requirements of field operatives. In short, static slicing does provide some of the requirement, but does not go nearly far enough to deliver on the promise and mandate of FirstNet. To make Local Control and a truly orchestrated system for public safety bandwidth allocation possible, Dynamic Network Slicing is the name of the game.
No surprise, this is already on the FirstNet roadmap for the coming years and indeed it is already opening up some exciting opportunities. While providing the ability to differentiate between applications in real-time within the same network, dynamic slicing directly addresses a multitude of use cases, particularly major events or situations. This represents a veritable sea-change in the industry that for many years focused the lion’s share of its attention on voice communications in multiple flavours. Recall that back in the day PSCR was granted some $300 Million to develop tools for public safety that were nothing but voice. But now the tide has truly turned. Today, PSCR is awarding over 30 organizations tens of million to explore the gamut of new technologies from indoor positioning, situation analysis and AI-assisted public safety to facial and object recognition systems and more. Suffice to say, the current discussions are on everything but voice. And this doesn’t just play into the requirements of FirstNet but opens up a burgeoning global market for these applications in which the U.S. will be a clear leader.
Certainly, and based on the fascinating discussions we took part in at PSCR 2017 we have moved beyond the voice era when it comes to public safety communications. And now, with the advent of capabilities like Dynamic Network Slicing, we are able to look into near and long term future of public safety where new technologies in areas of imaging, AI and ultra-fast edge network data bases drive real-time decision-making on the front lines of critical communications.
Indeed, and as exciting as this is, I think we may be only scratching the surface.