Problems ahead?

Read the ads, listen to the hype and it all seems so easy: IP-based security and surveillance systems are easier to install, offer superior video quality and quicker access to stored data, and save big money, all while sharing space on an existing network. Cameras can be added easily without dedicated cabling and, if you believe everything you read and hear, the whole matter has become plug-and-play simple. It's true you can do more than ever with IP-based security systems, and you will gain better quality, storage and retrieval functions without a lot of installation fuss. However, IP-based security systems risk overloading your shared network. The truth, after all, is that everything made by man has a breaking point. With IP-based security systems, if you're not careful, you just may find yours. So what should you watch out for when implementing this new technology? Latency, for one thing. If you tax your network beyond its capabilities, suddenly latency could affect some of your networked video systems. Remember that it takes only 400 to 500 milliseconds for the brain to perceive latency. Beyond what can happen to your video, other tasks performed on your shared network could also be negatively affected if your network doesn't have the strength to manage everything you require. PCs may slow to a crawl, or worse yet, crash. Successful security systems integrators should take several steps to avoid these pitfalls. The first step is to realize that good network design isn't just about coaxing a bunch of PCs and devices into talking to one another. Today's system designers must be capable of developing design parameters that include speed policies, bandwidth limits, quality of the service desired and more. The key to minimizing these types of bandwidth limitation problems is distributed storage of video at the subnet level. With storage properly distributed across the network, the only video that will be pushed to a centralized storage medium is tied to events. The data at the distributed storage device can be searched and retrieved at will, however. Determining what an "event" is opens up the whole world of video analytics and integration with other systems. Designers should know when IP-based security networks make the most sense and when they don't. For example, with so many facilities upgrading their network infrastructure to support Voice over Internet Protocol (VoIP), it's easier to piggyback video and even access control upgrades on top of many VoIP programs. Take full advantage of these situations where a robust network is already in place, but be careful if the budget only allows for older, less capable networks. Finally -- and there is much more to be said on this topic -- system designers need to stay current with the rapid changes taking place in the marketplace. But don't assume that newer equals better. Many people, aware that video compression is about to change, are ready to take the latest step up the ladder. We have, after all, moved from JPEG to MPEG-4, and the world has improved. It follows that as we get ready to make the jump to H.264, technology will improve again. However, H.264 will take off only when we have the processing power and the new required DSP chips. That may happen sooner than we realize. As with everything else, the best advice is to not jump into the pool until we know just how deep the water really is. Chuck Wilson is the executive director of the National Systems Contractors Association. NSCA is the leading not-for-profit association representing the commercial electronic systems industry.