We've looked at an overview of digital signage computing platforms and integrated graphics capability supporting signage displays, and we're ready to take a closer look at the physical characteristics of a system, including maintenance and security. These aspects will help identify what I think are the trends coming into play soon.


Digital signage systems are in public places, by definition - they are meant for information or advertising content. This means physical mounting and security are more of a concern than the average embedded system. The units can't dislodge easily for operational and safety reasons and they have to be tamper and theft proof. They are often mounted either out of reach or in some type of secure cabinet.




But at the same time, two other design challenges exist. Digital signage systems are usually networked in order to deliver regularly updated content to them for use, so they can't be totally isolated to provide physical security. Additionally, they have to be maintainable in case the display or the compute electronics fail or need upgrading, to avoid huge lifecycle costs.


How does a designer tackle this problem set, and what are some key trends to watch for? Let's break this down into the main parts of the hardware implementation.



The packaging of the computer complex (or media player as it is typically referred to in the industry) attached to a digital signage system is important. It takes into account three primary aspects related to maintenance and physical security:


  • cooling, which drives reliability,
  • cabling, which drives reliability, ease of installation, and maintenance,
  • mounting, which also drives installation, maintenance, and physical security,


Small digital signage systems can be constructed by OEMs using small computer boards (like COM Express modules or Mini-ITX boards) or custom small form factor systems, similar to what we showed in part 2 of this series . Larger systems can use larger form factors depending on performance and functionality needs, space, cooling, and power permitting.


As digital signage systems are moving to deployments with more systems involved, like the Beijing airport example  cited in part 1, most digital signage OEMs/VARs are looking for an integrated solution: an external box or signage "appliance" of some type that is self-contained, attached to the display, and can be installed and changed out quickly. By integrating into a single box, cooling and reliability can be tested, cabling can be simplified into a unified I/O area, and mounting can be made very secure as a box can be fastened by its flanges to a signage display.


As with many embedded computing applications, people are debating the pros and cons of a box approach versus a pluggable form factor approach. Pluggable form factors would have benefits of faster maintenance without potentially having to dismount the entire digital signage display for repair. For instance, Intel® Embedded Alliance Associate member Axiomtek is innovating with their slot in PC design for signage applications. Other Alliance members working on pluggable solutions are congatec and Winmate. The tradeoffs involved with compute boxes versus pluggable solutions is a trend to watch.



Obviously the choice for a networking protocol of almost any "edge" device, embedded or enterprise, today is TCP/IP. Digital signage systems operate like endpoints in an enterprise network, and need to connect quickly and easily. Operating systems which support TCP/IP are plentiful. The driving question: is the medium wired Ethernet, Wi-Fi, or WiMAX?


While there's something to be said for the ease of setting up a single wired Ethernet system, personally I'd be leaning toward a wireless implementation for the future. Pulling cables for 600 systems isn't cheap. Keeping cables secure, especially Ethernet cables with RJ-45 connectors, is a bit of a concern in a public area if the system is exposed. Digital signage systems don't usually move once installed, so any issues of signal coverage and integrity can be handled.


Until recently, content wasn't usually streamed on-demand for playback on a digital signage system. It's usually been delivered as a multimedia content file that is to be stored and played in some type of recurring schedule. This means the network bandwidth demands haven't been huge. Content files can be transferred prior to use, and as long as they get there in time for the intended use that's sufficient. If there was real-time content, it was usually smaller information types like weather, stock quotes, and the like that is relatively quick and easy to transfer.


That moderate bandwidth need made Wi-Fi a viable option for many installations, and in many cases it still is. If there's a Wi-Fi infrastructure already in place, connecting a point-of-sale or inventory control system in the vicinity, piggybacking a signage system on it makes sense. Wi-Fi scales relatively inexpensively and supports the necessary bandwidth. It also supports encryption which can keep content secure. 802.11n improves the available bandwidth and keeps Wi-Fi a viable option for many applications.


There is a growing demand for more streaming content, and corresponding higher bandwidth. While 802.11n is a big improvement over earlier implementations of Wi-Fi, WiMAX (802.16) brings 75 Mbps of bandwidth to the network.


WiMAX also provides much stronger security than Wi-Fi, supporting AES and 3DES today, and Intel is pushing for even stronger additions to 802.16 such as EAP and AES with up to 256bit keys. The bottom line is WiMAX will be very secure for distributed applications.


As WiMAX rollouts are beginning on a larger and larger scale, it's becoming a more attractive option because of the combination of bandwidth and coverage. WiMAX networks operated by service providers cover large geographic territories, and the infrastructure is set up and maintained by the service provider.


With all this in mind, WiMAX will probably become an easier and easier choice for distributed digital signage networks. Even without integrated chipset capability on a motherboard, WiMAX can be added today via Intel WiMAX modules or third-party ExpressCards easily. Integrated chipset capability isn't far off.





The "sign" part

Care should be given to selecting a screen that has the right reliability and environmental characteristics for the application - in most cases, a commercial grade LCD screen will be the best choice. The good news is that whether a commercial grade or consumer grade LCD is selected, driving the display is probably the easy part of a digital signage design. Most systems today are driven in a standard HD format, over an interface like HDMI or DVI.


Once the challenges of mounting the computer and networking the system are solved, it's really about mounting the display to keep it secure and safe from the elements. There are a wide range of wall, ceiling, and cabinet mounting options to look at - too many to explore here.


With a strong implementation accounting for physical security, reliability and maintenance, and secure and fast networking, a digital signage system can be implemented and kept in operation more cost effectively than ever before.


In the next installment of this series, we'll take a look at software side of a digital signage system, but we'd like to hear your thoughts on this post. What do you feel is the best approach to packaging a computer for digital signage? Which networking strategy wins: cabled Ethernet, Wi-Fi, or WiMAX? What considerations are there in selecting a display best suited for signage? Discussion welcomed.




Don Dingee
OpenSystems Media®, by special arrangement with Intel® Embedded Alliance