Myth No.2 ~ 'IP CCTV uses up all my bandwidth'

A couple of years ago we wrote...
'The amount of bandwidth that a camera uses is determined by the image quality and frame rate set for the camera, and the compression algorithm used. The compression technology is predetermined by the manufacturer, but good quality cameras will allow the image quality and the number of images per second to be set for each camera. For corporate security applications, cameras with MPEG-4 compression set to high image quality and 12-15 ips give good quality images in almost real time, and will use about 1.5 Mbps of bandwidth.'

We have found the 1.5 Mbps bandwith scenario to work very well, but have things moved on? What about H.264 compression and mega-pixel cameras? Mega-pixel cameras required more bandwidth, but H.264 is a more efficient compression algorithm that requires less bandwidth than MPEG-4. Is this an indication of how things will develop...increasingly high definition images, and continually improving compression algorithms to allow high definition images to be transmitted?

The report went on to say:
'High quality transmission of data for several CCTV cameras can be comfortably handled by typical office networks. Where more cameras are needed, it’s easy to establish separate networks. Today there are easy and affordable data-carrying solutions for all IP surveillance applications.
 A typical single network camera will have a video feed ranging from 0.2 to 2.0 Mbps. So a typical
office network of 100 Mbit will cope easily with the transmission demands of several cameras. When you have more than a few cameras, put in a separate camera network. It’s easy to separate networks with readily available network switches and routers. Existing structured cable, where available, can be used, but all the cameras should feed into Layer 2 Managed Switches.

For enterprise solutions using large numbers of cameras, an Ethernet backbone is recommended. When setting up a network it’s important to ensure that it has enough capacity to cope with the worstcase-scenario. So, plan for the maximum demand occasions, when all cameras are recording.

There are some useful techniques that enable installers to integrate CCTV into an organisation’s network without overloading it. For example, with ‘schedule recording’ users can choose not to have all cameras recording all of the time. This means that less data will need to be transferred.'


What are your experiences? We'd love to hear from you.