At IPTV we have been going through our strategic planning for the last few months. As we were distilling our visions, one of the items that bubbled to the top was Mobile DTV. The strategic goal is not to turn on Mobile DTV by some fixed date but rather to develop a business plan for deploying the service. There is a not too subtle difference between the two concepts that is not lost on me.
The first thing we look at is the budget. In our case the expense is not just what it will cost to purchase the hardware necessary to deploy mobile but also the impact to our existing services. Currently we broadcast a full-time HD service as our primary channel and two standard definition services. Since the State of Iowa specifies that we provide free service to all Iowans, we have to carefully examine any reallocation of bandwidth that would either degrade or reduce our services. We have to be able to demonstrate that any service reduction has a corresponding service improvement. This is one of the primary reasons we didn't jump on the bandwagon for datacasting when it first came out. The services that we evaluated didn't really offer any value to our audience that was in line with our mission; nor did they appear to have a business plan that offered any real indication of sustainability. The fact that most of them have since failed is evidence that we were correct in our initial evaluation and further strengthens our resolve to make sure that if and when we move into Mobile DTV we'll have a sound strategy that meets our objectives.
One recent event has had some impact on our plans for a Mobile DTV service. While working with Harmonic to upgrade our encoders and statistical multiplexer, we discovered that the statmux still had our third and fourth standard definition encoders in the pool—and even though they were disabled, they were still taking up almost 2 MB of data space. Some additional changes to the filtering and preprocessing of the active encoders generated another 300 to 500 kilobits of bandwidth. These changes coupled with the overhead we already were seeing prior to the modifications means that we now have about 3 MB of bandwidth that we could use without requiring us to give up any of our current services or make any significant sacrifice in the quality of those services. We are definitely going to do some experimentation to verify these numbers and confirm this assumption. The rule of thumb for planning bandwidth requirements for the most robust service is about 4:1. Using that as a planning factor and assuming a total of 3 MB being available, IPTV could conceivably broadcast a 750-kilobit service and at that rate, the service would look quite good on virtually all of the mobile and portable screens coming on the market.
Hardware costs are also a factor that need to be fully developed. According to a recent presentation on Mobile DTV given by Marcus Alexander, NAB's Executive Vice President of Television, I learned that the estimated average cost for a station to upgrade to mobile was in the $50,000-$100,000 range.
While I agree this estimate is probably a good ballpark figure for a single station to get started in Mobile DTV, it doesn't present a real sustainable planning figure. For this amount, a station could add the necessary hardware to incorporate Mobile DTV data into their data stream and upgrade a single DTV exciter in a transmitter to broadcast the Mobile DTV data. Many stations have two exciters within their DTV transmitters to provide backup in the event of exciter problems as well as accommodating software upgrades. Not too many of us are comfortable with upgrading an exciter and having problems develop and not having a fallback. Upgrading the primary exciter to incorporate Mobile DTV capabilities without upgrading the backup exciter would mean that a failure in the primary exciter would result in the absence of mobile service despite the restoration of primary broadcast service. This assumes that the transport stream that has mobile data multiplexed into it can be used to drive a non Mobile DTV-capable exciter. This is a question that I have not yet received an answer to. If this is not the case, then the station will have to have the facilities to revert the stream back to a traditional non-Mobile DTV state. This might be acceptable for an operation that is experimenting and trying to develop a Mobile DTV strategy but it would be risky for an operation that depends on reliable mobile services.
Many stations have a primary and a backup transmitter, which presents even further challenges. At six of IPTV's nine transmitter sites our primary transmitter is an Axcera Visionary transmitter and the backup is a Harris Diamond. Both Axcera and Harris incorporate into their exciters dynamic control of the output RF signal to correct for linearity and power. Both use different methodologies and although we can and have used one manufacturer's exciter to drive the other manufacturer's transmitter, it required that we disable the auto-correction circuitry.
Depending on the amount of correction that typically is applied and the stability of the passive RF system and the environment, the absence of auto-correction may be problematic and would definitely not be a good long term solution. Conceivably we could upgrade one of each type of exciter at each site and just do a wholesale transmitter switch in the event of a primary exciter failure. But that assumes that the backup transmitter system has enough output power to provide adequate market coverage. At our sites this is the case but at many commercial broadcast facilities, their backup transmitter is a lower-powered DTV system that was used to provide coverage to their city of license pending the end of analog and the installation of a full-power primary DTV system. Dropping 7 or 8 dB would have a noticeable effect on the primary DTV service and might make Mobile DTV unreceivable.
None of the figures take into account the addition of circular or elliptical transmission facilities. How much additional vertical power is needed is debatable but a station running its DTV transmitter at its maximum capability will either have to add additional transmitter power or sacrifice some horizontal power to drive the vertical transmission system. This may mean risking audience at the fringe to try and gain coverage for mobile services close in. With IPTV's mission being to serve all Iowans, this could mean putting viewers at risk for whom we are the only terrestrial service available. That would be a tough position to justify.
These are some of the fundamental elements that we are just now beginning to analyze so that when we do develop our Mobile DTV plan, we have as full an understanding of all of the costs. I have been around a number of people and projects for whom deploying the technology is the strategic goal but the real success comes when the technology is deployed to either augment or create a successful business. There is no doubt about the fact that the technology works and that once it is deployed the performance will improve—just look at what has happened with terrestrial DTV. Technology is a manageable challenge, but I think the real substantive challenge confronting the fledgling Mobile DTV industry is finding a working and sustainable business model.
Bill Hayes is the director of engineering for Iowa Public Television.