Print Page 9/4/2012

Two new Ericsson receivers, racked above the existing Sencore receivers, await deployment at Iowa Public Television.

JOHNSTON, IOWA—In a recent article on the changing face of routing within a broadcast facility, I speculated that future equipment might only have a National Electrical Manufacturers Association (NEMA) power connection and an RJ45 jack on the back panel. There will probably be a power indicator on the front panel and maybe a power switch and of course, the manufacturer’s logo, although I’d actually incorporate the power indicator into the logo (I could never figure out why Omneon didn’t spell out their name with those cool blue lights).

Whether or not we actually end up with hardware that has no front panel user interface or not remains to be seen but we certainly see networked control interfaces that allow us to access, configure and troubleshoot systems from a networked user’s computer running a web browser. In the digital world, the essence—the metadata, control instructions— everything is data and can be transported over a shared IP connection.

NO MORE CANDYCOATING
So what has this to do with station automation? Well, for better or worse, most stations approached automation from the standpoint of reducing the number of people it takes to run a facility. I know we have candycoated it and talked about repurposing people to do more important jobs, but look around most facilities and there are fewer people and automation is one of the primary drivers for that reduction. I don’t think we can continue to look at automation or operational efficiency as a way to cut costs.

I am also of the opinion that we cannot save our way to profitability and success; we have to create more compelling content for our audiences at the national and local levels. Therefore new automation projects need to actually allow the people that are displaced to contribute to creating content that will spur growth instead of downsizing them out the door to save a few dollars.

Rudimentary automation in broadcast stations began long before digital technology. I worked at radio stations in the 1970’s that used theoretically inaudible tones on audio tapes to trigger the next event. The RCA TCR-100 and Ampex ACR- 25 spot players combined robotics using quad tape cartridges to sequence commercial breaks. Traditional automation has essentially been focused on accurately running a sequence of events based on the traffic log and even in today’s world of video servers, that still seems to be the primary goal.

But what if we took a step back from thinking about automation in terms of running a sequence of events and thought more about it in terms of facility management? I have been involved for awhile in the SMPTE 34CS AdHoc Group on media device control over IP, and that is where I see us heading. One of the tasks we have been working on is creating a list of device categories, types and capabilities in order to develop open control protocols. I have been walking around Iowa Public Television, looking at the racks of equipment and making notes of what types of devices we have, how we use them and how we control them. It has been an interesting exercise for me because I have started to think about a more “heuristic” facility management architecture.

As the MDC group has discussed the required interchange needed between devices and control systems it becomes clear to see that much more than automation is possible. If the media devices and the control system can actually communicate not just simple commands like play and stop but can actually report capabilities and availability of the media devices, then the operation becomes more of a thinking organism. We see the basics for this in plugn- play devices that we install on our home computers. As devices are installed, they announce themselves to the operating system, negotiate interfaces and control and report when ready for use. Devices do this via USB, SATA and Ethernet, although the latter still requires more end user input to set up then is really desirable. In an IP over Ethernet environment, new or replacement equipment needs to be able to be installed, plugged in, turned on and then integrated into the environment based on the capabilities inherent in the system.

IMPERFECT EXAMPLE
Here is an imperfect example that we are working through at IPTV and my idealistic vision of what could be. As part of the ongoing next generation interconnect system (NGIS) project, the PBS network is converting all of its real-time satellite delivered programming from MPEG-2 to MPEG-4 over the next few months (see related story, p. 1). IPTV currently uses 12 Sencore MRD 3187A integrated receiverdecoders (IRD) to downlink HD and SD content from PBS and other sources for our primary and two secondary streams. Because of the nature of our operation and our predisposition in how we automate and limitations in the control architecture, we essentially park receivers on services within a transponder and leave them there. That means an IRD spends most of its time looking at a single service and outputting the SD or HD content from that service only. We do make changes when needed but this requires an operator to manually interact with the IRD. Since the 3187A is not upgradeable to receive DVB-S2 satellite service or MPEG-4 data streams, the existing 12 IRDs are being replaced with eight Ericsson RX8200 IRDs. The reduction in the number of IRDs used is a function of the soon to be deployed non-real time (NRT) component of the NGIS project (which will be the subject of an upcoming article).

In order to make the transition happen, we are taking advantage of the enhanced capabilities of the RX8200s by cloning the profiles from the existing 8127As and configuring them as DVB-S1 IRDs outputting MPEG-2 streams. Satellite services will then migrate around and run in parallel MPEG-2 and MPEG-4 for a few weeks and then MPEG-2 services will cease. During this process it will be necessary to go through and load profiles into the new IRDs manually. In my idealized vision of how this should actually work, I would plug my new RX8200s into the network architecture. The IRDs would announce themselves to the system complete with all of the capabilities included such as the capability of receiving DVB-S1 or DVB-S2, working with MPEG-2 and MPEG-4, outputting HD or SD content, etc. Then in the idealized facility management architecture, the system intelligence configures and manages the media devices and uses the appropriate tools to do the job.

As I said, this is an imperfect example since the devices that I am talking about are not currently on the list of media devices that SMPTE AdHoc is creating, but I may suggest that they be added, or at least considered. Be that as it may, the goal of creating an open set of control instruction sets that can be implemented across a wide array of devices is one step in moving towards a more intelligent facility management system that will free up time and people to do the really important work of creating compelling content. I would encourage my fellow station engineers to consider joining this AdHoc group so that the document that is produced is what the industry really needs.

Bill Hayes is the director of engineering for Iowa Public Television. He can be reached via TV Technology.

  Print Page