Is your network PhaseReady?

attend the Phase Ready Seminar from Chronos

Phase Ready?

As you think about the evolution of your network, don't limit your thoughts to the frequency stability you need now. We want to help ensure that network roll-outs today will still be relevant come the explosion of HetNets and Small Cells.

Donut or Doughnut - the wrong question?

Back in March I blogged asking whether you (well your phase delivery network actually) were a donut or doughnut person. The donut would be a ring of PRTCs (Primary Reference Time Clocks) at the edge of your network access layer delivering phase to Gateway devices and eNodeBs.The doughnut would have resilient (perhaps Caesium supported) PRTCs in your network core. The terms of this question were really about a battle of technologies as well as network architecture. I've since realised that this question is really not valid for a reliable supportable phase delivery network for TDD and LTE-A services that you are now on the verge of rolling out.

For me most of the technology issues are resolved (microwave links delivering phase being the most challenging), although implementations will vary wildly and standards have yet to catch up with real world network issues.

I think the key requirement for a phase delivery network is not the timing technology that underlines it but what happens when things fail! For example GPS directly connected at the eNodeB or Cell Site Gateway should easily deliver 1.5us at the air interface, but for how long will it continue to meet this requirement if some kind soul cuts through the GPS cable? Will it be long enough for your field force to get to site and effect a repair before the timing requirement is breached at an economic cost?

With this in mind I believe that the ideal phase implementation may well be a donut supported by a doughnut! I also believe this will be best delivered by an phase system independent of the network equipment. With PRTCs at the donut layer supported by PRTCs in the doughnut; especially if supported by Caesium in the core and Automatic Path Asymmetry Compensation (as implemented in the TimeProvider 2700) at the donut layer; next day mobilisation of field force could still allow fixes before the network edge breaks the 1.5us barrier.

Continue reading
2409 Hits

Donut or Doughnut? Where will your Phase come from?

I've spent a lot of time since the beginning of the year talking to carriers and equipment vendors about the problems using PTP to get phase sync to the edge of networks. This includes talking to both the carrier and vendor about a network that is being installed this year and will need edge phase sync down the line. I've also spoken to two mobile carriers with very different philosophies of likely network rollout to achieve adequate phase sync, and another operator planning a network for rollout in a couple of years still to make such choices.

The fundamental choice seems to boil down to this - are you a donut or doughnut person? In much of the world a donut has a ring shape - hollow in the centre (center if we're following the spelling of donut!). Many believe they can deliver phase sync from "the edge of the core" or access layer. This would involve a layer of GPS PRTC / Boundary Clocks a minimal number of hops from the network edge - a phase sync "donut".

Here in the UK a doughnut is a disc shaped confection with no hole but a large amount of jam (jelly?) in the middle. In this case a smaller number of larger capacity and more resilient GPS PRTC devices would be deployed in the network core; disseminating phase sync through a network path that would probably demand full PTP awareness and on-path support.

Of course many factors in your network design could tip you from the donut to the doughnut - will you be deploying in largely legacy sites or is this a "green field" project; is it planned to have some or all of your traffic carried through a third party Managed Service; will you be using third party "dumb pipes" to transport your PTP packets for at least part of their journey to the edge?

Even if your LTE-A plans are a year or two away, you need to be thinking whether you are a donut or doughnut network. We at Chronos are developing "Phase Ready" tools to help carriers make these decisions, then implement and monitor them.

Continue reading
2361 Hits

Will it be Worth All the Pain?

Excellent piece from Graham Finnie, Chief Analyst, Heavy Reading in Light Reading - MWC14: What Are Telcos For? - giving a "from the outside" view of the Congress.

His conclusion, or nightmare scenario if you're a carrier, is that the network could just be a commodity that customers pay a bandwidth and "line rental" fee for. Services are already migrating away from the carriers (why SMS when you have WhatsApp and iMessage; why call with WhatsApp Voice and FaceTime Audio) and will continue to do so.

Of course most of the services running on your network are customer rich and income poor, and so the opportunity for you to sell underlying services is not a great prospect; and you're one step away from Internet Bubble 2.0 bursting, and this one won't be your fault! You'll still get the blame for teenage porn viewing and cyber-bullying though.

Even if charges for service providers on your network were viable, you have the spectre of Net Neutrality hanging over your head. This is described as "the founding principle of an open Internet" and without it would be "the end of the Internet". Net Neutrality though means all traffic must be treated equally - not a good start for revenue generation! And what politician is going to side with you against "the little guy"?

There will be no sympathy for your plight of course. You have a reputation (deserved or otherwise) of milking your customers over many years; exploiting the mobile cash cow in the way Independent Television companies did in the 60s and 70s. A license to broadcast was a license to print money.

So now you're rolling out scalable, flexible Carrier Ethernet backhaul; implementing the tight Traffic Engineering you'll need to deliver microsecond timing at the network edge for all the LTE-A services to come; paying for more and more spectrum and then having the cost of your "old" 2G and 3G spectrum increased; and all for the prospect of an affordable monthly fee!

And will you make anything from a worldwide simulcast of Paul McCartney live from the Cavern (or a TV set that looks like the Cavern)? And who's fault will it be if its not delivered perfectly to any device anywhere?

The future's bright......

Continue reading
2383 Hits

LTE-B: Terrible TLA but coming to a device near you soon

One of the key upcoming technologies in the 3GPP world for some years now has been MBMS "Multimedia Broadcast Multicast Service" and latterly eMBMS "Evolved Multimedia Broadcast Multicast Service". This allows adjacent sites to transmit the same broadcast data, allowing devices an effectively higher data rate for streaming content. Promised but as yet undelievered, eMBMS requires microsecond (1us to 32us dependent on implementation) phase synchronisation between sites.

Three stories have appeared in short order to show this technology is becoming a reality. Last week Samsung announced they are working with Korea Telecom to deliver eMBMS services to their LTE subscribers. This was quickly followed by Verizon trialing "LTE Multicast" tied to the Super Bowl, and Telstra conducting the first stadium trial of "LTE Broadcast" at a T20 Cricket match at the MCG (well they call T20 Cricket but that's a subject I'm not going to touch here!).

I feel the marketeers won't be able to resist using LTE-B for this. Its even better than LTE-A!!



Continue reading
2755 Hits

Is the "Phase Ready" message getting through?

A couple of contrasting situations in big European Mobile Carriers demonstrates that the message is starting to get through - but there's still a lot of work to be done!

One major carrier who work closely with us at Chronos on sync and timing are busily engineering their edge network to be as symmetrical as possible to make sure their PTP for phase performance is a good as it can be. They currently have NO projects that require phase, although they use both SyncE and PTP for edge frequency.

I also heard this week of a carrier that committed only last year to a large rollout of Ethernet Microwave radios for backhaul, but selected a platform that cannot support PTP on path support. This is a supported product and so no criticism on selecting this for the project they had at the time, but they are now considering a trial of LTE-A requiring phase at the edge via a backhaul pipe that will impair phase performance with no upgrade path to help!

It really does make sense to think of the phase supporting future when considering frequency only projects today. This future is not as far away as you may think, and is likely to be in the lifetime of equipment you are currently considering for use.

Continue reading
1970 Hits

And You Thought Phase Would Be Tough?

I must admit that over the last year or so I thought, as far as PTP is concerned, that "frequency is done". From the early development days of PTP for frequency we went from "you can't trust it over more than 3 hops" to moving frequency through long networks of switches and microwave links without batting an eyelid. In this context the crusade to ask "Is Your Network Phase Ready?" began.

Intriguing then that I know of a situation TODAY where a Gateway device from a very reputable manufacturer with embedded PTP client is incapable of delivering adequate FREQUENCY to Node Bs after only a few network hops with even moderate traffic levels. This leads me to reinforce a view I've spoken about before - just because something is PTP aware doesn't mean it works!

We're seeing the drive today for full on path support in sync networks delivering phase to the edge, and people wanting to play in the Small Cells space (like Microwave Backhaul providers)  realise that they will have to interact with these PTP packets and will all be implementing a variety of boundary and transparent clock type solutions to do so.

I am still very much of the opinion then that your sync network should be INDEPENDENT of your transport network, or at least independently monitored and qualified for time and timing performance.

If you can't move adequate phase performance to the edge of your network you WILL have a negative impact on your customers' experience. Can you trust this key performance element to the people who supplied and support your transport network?

Continue reading
10561 Hits

Tested Against What?

There was an interesting question from the floor during our Phase Ready Seminar on Wednesday; basically questioning the need for independent benchmarking and monitoring of phase in a network when manufacturers are building in an array of synchronisation performance statistics into their PTP Aware Switches and Routers.

From our perspective the response is simple - what are these statistics actually measuring and what are they measuring against? Of course in day to day operations the data available from within the end to end network management platform will be a key to deciding what actions if any are required should a particular set of circumstances arise. However, we think that, at the very least, these performance metrics should be calibrated against a set of independently verified and traceable measurements to put them in greater context.

This is why we are working with several equipment vendors to integrate SyncWatch into their EMS and OSS systems.

Continue reading
3924 Hits

Is Your Backhaul Sync Available to You?

PDH and SDH/SONET core networks made life pretty easy for the NodeB that needed 50ppb timing for the Air Interface. The transport technology inherently carried traceable time, and this was available via the T1/E1 interface that carried your traffic. This timing quality was available whether this core network was yours or provided by a third party!

Now we're looking to a future of microsecond sync at the edge, and planning for how we get that level of synchronisation out there. Some of the backhaul technologies being mooted for Small Cells are TDD based and therefore need good synchronisation themselves for optimal performance. FTTx and GPON for example in the fixed category, and several TDD radio systems whether point to point or point to multipoint.

The questions that strike me here are:

What level of synchronisation is needed for these transport technologies?

Is this sufficient for your phase synchronisation needs?

Can this technology carry your own timing (PTP packets for example) without compromising performance?

Can you trust your network performance to a third party?

Is this sync available to you as the user anyway?

Start to get answers to some of these questions,or just ideas on what sort of questions you need to ask, at our free "Is Your Network Phase Ready?" Seminar at Kings Place on Wednesday May 8th.

Continue reading
6722 Hits

Phase Ready - Why Now?

Why is it important to consider phase synchronisation at the network edge when, at least here in the UK, you're only just rolling out your LTE Macro layer?

A valid question, but you may well also be rolling out 3G Small Cells for indoor coverage in public spaces and enterprises,data offload, or Outdoor Event systems. Choices that make perfect technical and economic sense for these systems may well come back to haunt you if you don't take heed of the future (and not so far flung future) phase requirements your network will have.

Take backhaul technologies for example. You will of course find backhaul that's perfectly suited to your requirement now; perhaps getting SyncE right out to the Small Cells to give great frequency sync for the Air Interface. But is this backhaul phase ready? When you need to overlay this great frequency network with PTP packets to get microsecond sync to Small Cells, will this backhaul deliver the timing performance you need?

Decisions taken now for Small Cells need to take account of the requirement you WILL have for phase when you're delivering LTE-A services. Think smart and you won't need to rip out your two year old backhaul network because it wasn't future proof.

Our "Is Your Network Phase Ready" seminar at Kings Place on May 8th has a session "Small Cells Backhaul - the “Wild West” Explained". We'll try to give an impartial view of the options currently available, and their possible impact on delivering microsecond phase. Places at the Seminar are FREE - sign up HERE today.

Continue reading
7030 Hits

Its Not Just Small Cells!

A recent Light Reading piece - eMBMS Unleashes New Potential for MBB Business Growth - shows that it won't just be the deployment of Small Cells and LTE-A that drives the need for phase synchronisation at the network edge.

eMBMS (Evolved Multimedia Broadcast Multicast Service) is the LTE flavour of MBMS; described by 3GPP for many years but not as yet deployed in any network - until now! Plans from Verizon and others have been announced to roll eMBMS services out over the next year or so, primarily for streaming live video content and supplying video on demand.

As content becomes king, delivery platform is less and less a concern for users. If their Mobile Carrier gives them Cable TV like services - with Cable like performance of course - why have separate subscriptions? But beware the fickle user. The Light Reading piece also says "Studies have also shown that for every second a user has to wait before an online video begins playing, 5.8% of all users will leave before the video ever loads ."

eMBMS relies on the creation of a Single Frequency Network (SFN) to send data simultaneously to many users in a sector, or between sectors. This will require microsecond phase alignment of cells to deliver this content efficiently, and eMBMS is equally at home in LTE-FDD networks as it will be in LTE-TDD.

So if you think "my FDD network doesn't need phase" well think again!

Continue reading
3341 Hits

Why Phase Ready?

Many of us have been, and are still, involved in the migration from SDH to Carrier Ethernet in networks; moving from the “certainties” of SDH timing to PTP and Synchronous Ethernet. The crossing of the “Bellhead / Nethead” boundary has mostly got beyond “Ethernet doesn’t need sync” and, in a surprisingly short period of time, getting good frequency performance to the edge of networks over Carrier Ethernet is just about done.

Now we see the next timing challenge on the horizon – PHASE synchronisation at the network edge. And the key driver for this requirement is not that networks will be using Time Domain transport technologies like LTE-TDD, but the coming of LTE-A, Heterogeneous Networks (HetNets), Small Cells and Enhanced InterCell Interference Coordination (eICIC).
It’s a given now (or seems to be) that in next generation mobile networks the Macro layer will simply not have the coverage and spectrum to deliver data densities and rates customers now require. The latest Small Cells Market Data white paper from the Small Cells Forum looks at the scale of likely rollouts. For example “Mobile Experts published a new forecast claiming that 70 million small cells will be shipped by 2017, including Femtocells deployed by mobile operators and picocells used for high-capacity urban networks. LTE small cells are a major part of the forecast growth over the next five years, with more than two-thirds of small cells deployed in 2017 devoted to LTE-FDD or TD-LTE.”

A blog post from leading industry consultant Frank Rayal called “More and More Small Cells, But Where’s the Gain?” notes that “In the absence of advanced interference management solutions, most capacity benefit will be obtained from deploying small cells in a targeted way with intimate knowledge of the location of traffic hotspot and interference profile. In other words, forget about mass uniform deployments for now. Planning will remain essential.”

Effective eICIC requires stringent phase synchronisation between deployed Small Cells and the Macro cell(s) under which they operate. And effective eICIC will be the tool to enable “mass uniform deployments”. This requirement and the issues around it are why we’ve launched

Continue reading
6902 Hits