ITEXPO begins in:   New Coverage :  Asterisk  |  Fax Software  |  SIP Phones  |  Small Cells

November 2006, Volume 9/ Number 11

Return to IT Table of Contents

Real-World IPTV Deployment Concerns

The Top 10 Things You Need to Know to Successfully Deploy IPTV

By Kevin Kosche


Internet protocol television (IPTV) is creating new business opportunities for broadband service providers. You, the operator, now can introduce exciting revenue-enhancing triple play (voice/video/data) services. Your subscribers will be able to get all the services from you, conveniently bundled into a compelling package.

IPTV is a hot growth sector of the telecom equipment industry and has suddenly attracted many companies competing in this space. You face a dizzying array of solution architectures, equipment combinations, and applications.

You have no doubt given considerable thought to the new services you will offer � basic and premium digital television (DTV) channels, video on demand (VoD), personal video recording (PVR), gaming, Web access via TV, and targeted advertising. You have analyzed required capital investments and have projected revenues. You have seen vendors� proposals, roadmaps, and their visions into further communications and digital home services. You have developed vendor evaluation criteria, checked references, and perhaps even visited other deployed providers.

Have you overlooked anything? If you have not deployed video services before, you may not know where the �land mines� lie � the �unknown unknowns�. Some vendors� short track record leaves even them learning. However, doing 80+ IPTV deployments over six years has taught Siemens (News - Alert) where these issues are, and how to either eliminate them or minimize their impact. Consider the following IPTV real-world fundamentals. You are well advised to investigate these before you select an IPTV solution.

1. Disruptive Innovation
IPTV is still new and has the potential to be disruptive with respect to established TV and communications usage modes, content distribution and control, regulatory restriction, and more. Changing the establishment is a daunting and risky prospect for the operator. A more certain approach might be to first replicate existing TV viewing fundamentals.

Your consumers range from small children all the way to the very elderly � some technically savvy and some not at all. Don�t assume that all your subscribers are computer literate. There is a large demographic of television viewers � roughly, the 39- to 50-year-old age bracket � the massive passives � that is happy to view broadcast television without interaction and without controlling what flows over the screen. These viewers� habits will not change overnight. What would happen if the basic TV interface suddenly changed? Can you afford that many service calls? You should first �do no harm� to basic viewing, while introducing enhanced services in an intuitive, easy-to-navigate manner.

There are companies that control the production, licensing, and distribution of content. They have established rules about content distribution and advertising, and these rules will not change overnight. You can�t circumvent them, and you are not likely to single-handedly change their business models. What you can do is work within their business models, expanding their distribution, and creating more revenue for all.

2. Content Acquisition
�Content is King!� Abundant premium content is key to your rollout. Incumbent video services providers, your competition, will take defensive measures, including enhancing their own content, so your content offering has to instigate the churn of many of their customers.

In your pursuit of content, you can quickly become mired in an onerous process of establishing contacts, negotiating licensing, and responding to multiple technical questionnaires and subsequent clarifications to convince content owners of your IPTV implementation�s level of security. This could introduce many months� delay into your rollout.

Can your IPTV solution vendor facilitate this? Do they have a proven process to guide you through these steps? Have they been �certified� by the major studios? Do they have relationships with content owners and aggregators? Most importantly, do they have multi-vendor encryption and digital rights management solutions? You should expect the answers to these questions to all be �YES.�

3. Branding
The user interface (UI) displayed by the client software can offer opportunities to present and promote your brand and create loyalty for your service. You should expect that the look and feel of the UI can be customized for this purpose. There should be flexibility on all screens and at all menu levels to accommodate your trademark colors, fonts, layouts, logos, etc.

The displayed brand should be yours, not your vendor�s. Why would your vendor�s brand need to be seen by your subscribers? Is your brand stronger than your vendor�s? Are they intending to reach your subscribers directly?

4. Architecture Scalability
Many IPTV solutions are quite new and have not yet been commercially deployed. Rather, they have been demonstrated in controlled laboratory scenarios or perhaps been used in limited trials involving friendly, non-paying users. These vendors are still waiting to �land the big one.� Do you want to be their first?

System scalability troubles will not surface in a lab or trial. You might encounter scalability limitations when your IPTV subscriber base numbers in the tens of thousands, and then again at hundreds of thousands. Poor scalability will manifest itself in the form of excessive network demands, (necessitating very expensive bandwidth upgrades), compromised quality of service, and your architecture expanding into hundreds or thousands of servers. It will certainly invalidate the capital expenditure projections that you did as part of your business case. Are you prepared, financially, and operationally, to build and maintain a large server farm and numerous facilities?

How do you mitigate this risk? Careful analysis of the scalability of solutions you are considering. Proof, as evidenced by large-scale success of your vendors� reference customers. Visit those customers and ask them about scalability.

5. Multicast vs. Unicast
Many viewers will be watching the same popular channels at the same time. Multicasting is the most efficient technology for video delivery, reducing required bandwidth in your network. If your solution does not multicast, each channel has to be replicated in the transport network for each viewer watching it, significantly wasting network resources. With multicast, each video channel is transported once through your network.

With the efficient networks afforded by multicast, basic services such as broadcast TV can be introduced at a very reasonable cost. This efficiency then leaves precious bandwidth available for other applications that must necessarily be delivered unicast, such as video on demand and online gaming.

6. IPTV Applications vs. Web Browser-based Client on the Set-top Box (STB):
Some IPTV solutions are based on a Web browser client/server architecture, wherein the processing for rendering menu screens, guide navigation, etc., is performed by servers installed at your central office. With a browser-based client, the STB�s job is simple � it simply renders pages sent from the server, much like an ordinary computer rendering a Web page, and, of course, decodes the incoming video stream. Browser-based clients offer the advantage of a smaller SW footprint requirement in the STB. A less powerful, less expensive STB might better operate with a browser client. However, this architecture incurs two distinct disadvantages. The first is that a browser-based client requires unicast communication with the server in response to virtually any action performed by the user. For example, if the user opens a program guide, that displayed guide page is presented only after a unicast request to the server followed by a unicast page sent from the server to that STB. This sequence may result in sluggish response, depending upon how many people are accessing the application, to user key clicks. Think about your favorite website. Is the responsiveness of that site always predictable?

Note also the impact to service if your server should go down or otherwise lose connection to the network � your users� TV experience will be interrupted. If your network is not 100% available, you risk a compromised user experience. If you intend to deploy a browser-based client, consider carefully the bandwidth that you must allocate to its unicast communication. The second major disadvantage to this type of client is the amount of back office hardware required to support it. If you add servers and allocate their capital and operational cost over your subscriber base, then your STB savings may evaporate.

Conversely, an IPTV application on the STB puts the processing for these operations into the STB. The result is that the required STB hardware resources may need to be slightly greater than that needed with a browser-based client. An IPTV application on the STB, however, can operate mostly autonomously from the central servers. Application execution is done without server participation. User interface navigation is performed locally on the STB. Guide data is resident on the STB. Unicast communication with the server is minimal, required only for such things as VoD rental transactions. The user can still watch TV even if the server is off line. Because of superior communication/scaling and service level performance, the IPTV application on the STB better fits into telco network environments.

7. Integration to 3rd Party Systems:
IPTV deployments require integration between the IPTV solution itself and numerous other essential legacy systems, including systems for OSS/BSS, Caller ID, emergency alert system (in the U.S.), ad insertion systems, etc. These systems all likely came from different sources, and you cannot just assume that they will naturally play together.

They come to work together through a deliberate integration effort. But, you can benefit if your solution vendor has, through prior deployments, already encountered this specific third-party equipment and already done this specific integration. If so, great! If not, your solution must accommodate these other systems, most likely through external application programming interfaces (APIs). You need to understand what data needs to be passed and in what format. These APIs should be present, well-documented, and conform to applicable open standards. You should also ascertain the flexibility and cost of all parties to tailor these interfaces as needed.

8. Set-top Box Choice
You will need choices � more than one � in set-top boxes (STBs). Some of your subscribers will need a PVR box, others won�t. Some with premium services may need a more powerful box than others. Some will take high-definition programming. Over time, boxes may be discontinued and new ones introduced. Thus, you may very well have a mixture of boxes in your deployment.

You are buying the features and services touted by your solution vendor. Do all features run on all their compatible STBs? The economic reality of a large deployment is that the boxes dominate your capital expenditures; thus you may be economically compelled to deploy low- and middle-range boxes. Don�t be fooled by demos of very jazzy functionality that in fact won�t run on real-world boxes � the ones that work for your business model.

An STB model doesn�t just by chance work with your IPTV solution. It has to be integrated to the solution, either by the box manufacturer or the IPTV solution vendor, or both. Depending on the client software design, its portability, and its maturity, this may be a relatively straightforward or very grueling process. So how is this performed, by whom, and how long does it take? This will pace the timeline between a new box introduction and its deployment. Does it require custom code development, or is the client portable from box to box? Do they have a defined process to certify new boxes? How many new boxes do they tend to introduce in a given year?

Once you make a substantial investment fielding boxes, you want to keep those in usage for years. Each box model that your vendor has certified and continues to support represents a potential support and backward compatibility burden. Over time, older boxes may lack the processing power, etc. needed to run ever more complex, newer applications. How long will your solution vendor commit to support old boxes? Does their history confirm that?

9. Operator Usability
Much is said about the end user experience, and rightly so, but consider also the daily operation of the system from your perspective. Your personnel will be using the system�s back office components. What about the solution�s administrative features?

Consider the tasks associated with defining service packages, pricing, and assignment of packages to subscribers. How do you define DTV channel line-ups? How do you define various types of packages that can be offered? Does the system provide for regional variances in channel line-up and package definitions? Does it permit subscriber attributes to be defined so that you can manage subscribers on a groups/classes basis? The sophistication of these features will impact the efficiency of your support staff.

How do you manage content and its associated business rules? Your on-demand content will have defined distribution windows, and outside of these you will be prohibited from distributing it. You may receive content well in advance of the dates in which you are legally authorized to offer it to your subscribers. Then, some time later, you must cease offering that content. Failure to properly manage content could result in loss of current and future content rights, or worse. How does the system manage the receipt, encryption, storage, pricing, activation, deactivation, and deletion of content? With potential libraries containing thousands of content items, you cannot manage this manually.

How are new subscribers provisioned? Do you intend a truck roll to install the service to each new subscriber, or will you mail an STB, cables, and instructions to them? Will you have to manually manage the STB/subscriber association, or will the system auto-provision that box when it is connected to your network? Wouldn�t it be nice to brand, market, and sell your video and other services in existing retail outlets? Make sure your deployment can support auto-provisioning.

10. Maintenance
Maintenance difficulties and expense incurred by poor system scalability has already been mentioned, and there are more operational realities that can become burdensome with a poor design or if key operational features are lacking.

Is the solution vulnerable to viruses? Your business case did not include the expense and maintenance for virus protection to each of hundreds of thousands of STBs. Some operating systems have a history of virus attacks. What operating systems will your servers and STBs rely on?

How will you troubleshoot a user problem? Does the system provide for remote diagnostics, logging, and status query, or will you have to do a truck roll to the subscriber�s home � and, thereby, also delay satisfying that customer? Does the design provide friendly error messages to the subscriber so they don�t call you over simple errors, such as unplugged network cables?

How does the solution perform client software upgrades? Are upgrades administered centrally by the operator or do they require that the user do the upgrade? Are STBs all upgraded in unison, or one-by-one? How long would it take to upgrade a quarter-million boxes? Are services interrupted during that time? How does the upgrade effect network load? Can you query the loaded software revision for a specific box?

What happens after a widespread power outage? When power is restored, and all your STBs are simultaneously trying to boot, what will your network load look like? What documents/files do the STBs need to retrieve from your central servers? Are these files delivered unicast, duplicated for each of your hundreds of thousands of boxes? How long would that take? Your system will recover more elegantly if the client software resides on the STB, and the STB needs only minimal data from the server. Additionally, those files should be multicast to the STBs.

Regardless of the solution you ultimately select, the system integration demands that your technical staff have or acquire new skills. Build-up, maintenance, and expansion of IP-based services are not trivial and may be a significant departure from the technologies you have �grown up with.� You need IP experts. Your network also needs to be rock-solid and 100% compatible with your IPTV solution. If there is any weakness in your network, video will find it! Does your planned solution provider offer advanced training for technical personnel? If not, how do you intend to get these employees up to speed?

Your triple play plans are based on solid reasons. You need to get it right the first time. Choosing the wrong solution or vendor could cost you significantly in terms of competitiveness, delay, investment, and perhaps, career. Think it can�t happen? Search the Internet for the major announcements of your potential IPTV vendor�s account wins. How many of those are deployed?

Nonetheless, there are solutions to help in getting your solution deployed as planned and on schedule. Choose well, and get a commanding head start! IT

Kevin Kosche is Vice President, Operations, Myrio Corporation, a Siemens company. For more information please visit

If you are interested in purchasing reprints of this article (in either print or PDF format), please visit Reprint Management Services online at or contact a representative via e-mail at [email protected] or by phone at 800-290-5460.


Today @ TMC
Upcoming Events
ITEXPO West 2012
October 2- 5, 2012
The Austin Convention Center
Austin, Texas
The World's Premier Managed Services and Cloud Computing Event
Click for Dates and Locations
Mobility Tech Conference & Expo
October 3- 5, 2012
The Austin Convention Center
Austin, Texas
Cloud Communications Summit
October 3- 5, 2012
The Austin Convention Center
Austin, Texas