At the IMS (IP Multiedia Subsystem) World Forum I gave a pre-conference workshop that summarized some of the results gathered in the IMS Status Report. The conference backed up the findings in the Report, and highlighted the many technology and business challenges we face as IMS rolls out. An interesting discussion point raised on both VoLTE (Voice over Long Term Evolution) roll-out and Rogers’s One Number (RON) deployment is will roaming voice revenues disappear over the coming decade as voice becomes just an app? We’ll discuss this later in the article, this is going to be a critical issue for many operators’ profitability. After years of circumspection, IMS is well on its way to ubiquity, even given the technology and business challenges, as shown below, a result from the operator survey contained in the IMS Report.
The conference had 200+ attendees, 66% were operators. Telefonica and Vodafone Spain were just a few subway stops up the road in Madrid. If only operators in other countries could be equally as conveniently located, well perhaps in time around the Paddington area in London as Vodafone, Everything Everywhere and Telefonica consolidate into the city from places like deepest, darkest Newbury.
Some of the issues discussed through the conference included:
- Cloud: granted a degree of cloud-washing is taking place; that is using the cloud label when it doesn’t technically apply. There is a recognition that even though IMS is shown as a set of boxes and wires, it’s mostly software and can take advantage of cloud economics given the processor intensity of most IMS implementations. Another aspect is on how IMS can work with cloud-communication services, we’ll discuss this in more detail later in this weblog article.
- Impact of Web RTC (Real Time Communications) on IMS. That is, over the coming 18 months all browsers become SIP end-points, this is both an opportunity and threat for IMS. But it appears it’s not on the ‘IMS radar.’ I’ll do a weblog article on its importance later, as an industry we need to be more responsive to emerging threats and opportunities.
- Lack of web integration remains an issue limiting IMS to real-time communication sessions, missing the richer communication experience enabled by most OTT (Over The Top) communication services.
- Lack of experience in RCS/VoLTE/IMS (Rich Communications Suite / Voice over Long Term Evolution) roaming. This is a critical component of an operators advantage compared to OTT, but it remains largely untested with a lack of urgency in getting it resolved.
- Belief that QoS (Quality of Service) matters to customers. It does, but not all the time, and that’s the problem, especially as the pipes get fatter.
- Opinion was it will take >2 years before RCSe becomes mass market or at least pervasive on handsets.
- VoLTE, the beginning of the end of roaming voice revenues?
- 2G switch off. I was only 2G connected throughout the event, using WiFi due to the unreasonable mobile broadband roaming rates. If I have 3G and WiFi on my smartphone doesn’t last the day with normal use. So 2G switch off will frustrate lots of customers, not just laggards. Markets like Japan are unique and should not be used as a proof-point for other markets, didn’t i-mode teach us anything? The voice of the consumer and its political impact is far stronger in other markets.
- Weird situation that if RCSe is not about new revenue then let’s take a softly-softly approach to trials. There’s a lack of urgency, it’s a fashionable topic, but the intensity of commitment to making it work just doesn’t feel like its there. A question often asked was, “Is it too late?” My reaction is, its only too late once you’re buried underground!
- OTT appears to be the top concern, operators are finally realizing Vonage, Viber, Skype, Whatsapp, Pingme, Google, Facebook, and the thousands of other similar services are reducing their relevance as CSPs (Communication Service Providers) to customers.
Reviewing just a few of the presentations. I’ll provide more highlights in a later article.
Noriaki Hagiya (NTT): Deploying IMS network and migration strategies
Their decision on IMS was strategic, this is a common theme for many operator deployments, and discussed in the IMS Report. The initial application of IMS was managing the ATM to IP migration for 3G, it was then reused for their VoLTE deployment. Their drive to LTE was simply the 3* spectrum improvement (note they did not have HSPA+) which appeared from their slides to the open door the door to 4G, with its much greater spectrum efficiency compared to LTE. They described their SEN (Service Enabler Network) using a SCIM (Service Capability Interaction Manager) and a composition engine to create operator services. They do not appear to be focused on the use of APIs for innovation across internal, partner and third parties for IMS at present.
Christophe Coutelle (Huawei): Maximizing the value of IMS
Christophe gave an excellent presentation setting out 3 drivers for IMS: 1) do voice well, 2) multimedia and multi-screen, and 3) expose and enable. On driver (1) he explained the 2 main approaches in the market of softswitch and AGCF (Access Gateway Control Function) for small deployments, e.g. Telecom Malaysia; and P-CSCF and SIP-MSAN (Proxy Call Session Control Function and Session Initiation Protocol Multi-Service Access Network) for larger deployments, e.g. T-com. He highlighted the aggressive move to VoLTE through 2012 and 2013 by a number of operators. Recommended to start RCSe on 3G as early as possible to start learning and solving the roaming and user interface / experience issues as soon as possible. Mentioned the iBasis LTE roaming service to help get up the learning curve fast. On multimedia he discussed the challenges in video, especially on interoperability. He finished on the the importance of ‘expose and enable’, with CaaS (Communication as a service) being a $700M market in 2012, growing to $3B in 2015. And gave the example of China mobile which has 120 services enabled by 3rd parties such as insurance apps and video door bell redirect to mobile.
Larry Baziw (Rogers): Rogers One Number
Larry gave the highlight presentation of the conference that reviewed RON (Rogers One Number), a consumer UC (Unified Communications) service between the mobile and the PC, using the mobile number as the unifying identity. It was the first service they launched on IMS, and has exceeded take-up targets. Its available for free to postpaid customers. It targets youth, trans-youth, and families with youth, but as with any service its reach goes beyond the initial target segments. It enables features such as call pull, where a conversation can be transferred mid-call from the mobile to the PC. Larry gave a frank review of the feature migration issues, such as MMTel (MultiMedia Telephony) only supporting 3-way calling (original fixed bias of the standard) while GSM supports 6-way calling which is popular in the Canadian youth segment. One of the many examples of how IMS history limits it on such obvious requirements. An important point is Rogers led the project management activities across 10 vendors (including Ericsson, Broadsoft, Counterpath, etc.). Larry also discussed the user experience issues about using CDR (Call Detail Record) history in populating the NAB (Network Address Book). the reason this was such a good presentation is it focused on the commercial, organizational, and implementation challenges.
Dr Chen Dan (ZTE): IMS based PSTN/ISDN Emulation
Provided case studies from China Telecom and Telefonica Peru on use of IMS to replace legacy PSTN with savings from: power consumption 34%; O&M 53%; maintenance 66%; and space 37%. She reviewed a number of IMS deployment options, e.g. hosted IMS at a group level across a number of OBs (Operating Businesses); IMS in a box for small deployments; as well as a full cloud implementation with software/hardware decoupling running in a private, hybrid, or public cloud. She explained several tools they’ve created to manage the IT barriers in IMS deployment including a data migration tool, and a service provisioning node, which is similar to the mediation layer Verizon created in their IMS deployment and described in the IMS Report
Jimmy Erhbar (COLT)
Jimmy provided a frank review of where many operators are, that is IMS-ready and watching and waiting.
Cloud panel discussion Micaela Giuhat (Genband), Eduardo Alonso (Acme Packet) and Michael Blauer (Broadsoft)
Some of the points raised in the discussion included:
- Lack of open API implementations to enable IMS implementations to work with cloud-based services, really the broader category of web-based services was more appropriate as they’re all API enabled.
- Lack of definition on what cloud means in the context of IMS and IMS applications
- Dr Chen from ZTE highlight a good practical summary of the IMS deployment options and Jimmy Erhbar from COLT highlighted cloud implementations will start with applications.
- There was a discussion around cloud-based UC services such as Cisco’s HUCS (Hosted Unified Communications Service) and the lack of interoperability with IMS capabilities for QoS.
- Who is responsible when things go wrong in a service between the cloud provider and the network operator?
- Are vendor’s IMS solutions ready for a clear separation between the software and hardware with an efficient implementation in a virtualized environment?
- A hosted ‘group IMS’ with each OB sharing the capabilities was seen as a valid model, provided its from a single vendor. Most operators consider a single vendor IMS approach the only practical deployment option, which calls into question most of the effort spent on the IMS standard.
Michele Zarri (Deutsche Telekom): VoLTE Roaming
Michele explained the TRF (Transit Roaming Function) to keep the existing roaming model in place for VoLTE. There was strong push-back from the audience on whether the model can be maintained. The earlier RON deployment was used as an example where customers will increasingly use OTT to avoid roaming fees. Some comments made highlighted a fundamental challenge in that operators’ employees never pay for their telecommunications service, if they did they’d quickly start using OTT and understand the threat. VoLTE will need to adapt to this situation else OTT will continue to gain a greater foothold. Operators must remember they are a unique niche in the market, their employer covers their bill no questioned asked. Virtually all other customers have to pay their bills, or justify any expense they claim.
Pieter Veenstra (KPN): Standardization national IP interconnect
Pieter again gave an excellent presentation I’ll discuss in more detail in a later weblog on IMS simplifications. He recommended using SIP not ENUM for national interconnect; the creation of a routing engine to simplify interconnect; and interesting self-learning approach to HD Voice.
Yachen Wang (China Mobile): Overview of CM-IMS (China Mobile’s IMS)
This case study is discussed in the IMS Report. Yachen Wang gave a good review of their current situation. 600M mobile subs, 20M PSTN subs across enterprise and home. They have a 3 step IMS strategy:
- Step 1: Unifed Comms for enterprise (2009-2011) – converged IP Centrex
- Step 2: VoIP / PSTN replacement (2011-2013)
- Step 3: VoLTE and FMC (need full regional / national coverage of LTE – like Verizon) 2013+
They currently have 10M IMS subscribers. Each province has its own IMS. All main IMS vendors are involved. They are currently focused on solving problems created through BYOA (Bring Your Own Access) such as QoS, security, firewall blocking. They are also investigating how to support IMS services over WiFi, as China Mobile has a national WiFi strategy. Here the Wireless Broadband Alliance will play an important role.
In Summary: We’ve got to realize that its not about incremental revenue, its about maintaining revenues. In my case, between AT&T, Verizon and Vonage; they’re taking $4k a year from me, and most people on the planet are an operators’ customer. Its a good business, we’ve got to focus on managing the transition in the business to focusing upon relevancy, and the changes technology will have on the revenue mix, such as roaming voice revenues disappearing over this decade. For next year at the conference I hope we have many more presentations like Larry Baziw’s (Rogers) which focused on the commercial, organizational, and implementation challenges much more than the technology, as technology is increasingly the easier part of the problem.