Let the Technology Fit the Need

Oct. 27, 2008
The success of your networked digital video system hinges on your choice of networking technology.
By Shane McClelland and Scott Mansfield

The need for more effective security and surveillance is driving an unprecedented increase in the deployment of digital video technology in electronic security networks. Traditional CCTV solutions are difficult to implement and maintain in large-scale installations, and they become increasingly ineffective as they grow. Moreover, changing operational requirements are driving security professionals to implement integrated security systems over a large geographic area, and this requires digitized video that can be transported virtually anywhere using standard data networks based on networking technologies such as IP, ethernet and asynchronous transfer mode (ATM).

What Are Your Options?
In typical corporate data networks, IP-based data applications are transported over ethernet as the preferred local-area network technology because it offers low equipment cost and ease of implementation.

Facilities such as airports, seaports, rail systems and highways typically cover large geographic areas and can have many hundreds or thousands of surveillance cameras that must work with voice applications and sometimes low-speed, serial data applications. Networks that support these types of security systems cannot afford to fail. These networks need to be bulletproof, need to support multiple applications, and need to provide military-grade performance, reliability and security. In addition to video surveillance, other applications can run over the network, including

  • access control (card readers, biometrics);
  • sensor monitoring (motion, perimeter, temperature, explosives, radio frequency identification);
  • audio alert (public address, alarm);
  • video information displays;
  • voice communications (public phones, emergency phones, courtesy phones, two-way intercom);
  • command-and-control communications (private line, public Wi-Fi, frame relay).

ATM is the most-used technology in networks that carry several of these applications. ATM can protect various traffic classes and flows, and it leverages expensive WAN bandwidth that is normally paid for per-bit.

Unfortunately, security networks don't necessarily fit neatly into these categories, so it's not always easy to choose the right networking technology. It is important to keep in mind that the application and associated network performance requirements ultimately dictate the proper network technology to use for video surveillance and security networks.

Technology Issues to Consider
To date, video transmission (synchronous video with audio) for security purposes has been accomplished using traditional analog communication techniques over coaxial cable. For most companies this means supporting overlay networks. It's costly to design, implement and maintain separate networks, especially considering that video networks become increasingly complex as they grow.

Once the video signal has been digitized, transporting it over a communications network based on standard networking technologies like IP, ATM and Ethernet becomes much easier and more cost effective for most businesses. However, security professionals must consider several factors before making a technology selection. Cameras, viewing devices, compression methods, and appropriate network infrastructure must be designed and selected to ensure that your video will meet expectations.

Since video is bandwidth-intensive, implementing real-time, video-based solutions carries significant implications for the underlying communications infrastructure. Adding real-time video can affect bandwidth quantity and quality, network security, flexibility and reliability.

Consider the following issues before deciding on a networking technology.

Delay (latency). In video communications, delay is defined as the time it takes the packets that comprise the video image to exit the camera and reach the video monitor. Delay is caused mainly by propagation, process and queuing. ITU-T standards define 30 frames per second as the one-way limit for quality video.

In video-based security systems, delay will interfere with the viewing of the video signal, sometimes to the point that the imagery is unusable for security purposes. Care must be taken in order to minimize latency by using intelligent switching/routing equipment, low-delay transmission paths, and quality-of-service features in the network design. Care must also be taken to minimize network congestion. If mission requirements dictate the use of video communications, delay becomes an even bigger issue.

Jitter. Jitter-the variation in the arrival time of packets to their destination-may cause problems in video transmission and completely disrupt video communications, making them useless. Care must be taken in security networks to eliminate or minimize jitter. Proven solutions to jitter problems include the use of buffers that dynamically increase or decrease based on the inter-arrival delay variation of the packets.

Bandwidth limitations. Video-based security system networks, especially in their WAN connections, suffer from bandwidth constraints. As more and more real-time applications access the network, this puts an increasing amount of strain on the transport network. Bandwidth requirements per application may vary significantly. Therefore, special care must be taken to minimize bandwidth consumption through the use of intelligent traffic engineering (TE) and QoS/CoS technology.

Another way of reducing bandwidth is to compress the packet headers, especially for non-mission-critical data traffic and "interchatter" introduced by the growing numbers of automated end devices being attached to the network. This makes intelligent use of scarce bandwidth, especially to remote locations, of paramount importance.

Bit error rate (BER). Video-based applications are much more vulnerable to BER than typical data applications. When packet loss is greater than about five percent of the packets (for voice communications, less for video communications), other strategies may be required, such as forward error correction (FEC). It should be noted that introducing FEC to the system will introduce a delay to the link.

High availability and reliability. Video-based security system networks must offer the same high availability and high reliability that are inherent in today's commercial voice networks. Fully redundant hardware platforms, leading security features, and software redundancy features are an absolute requirement to meet the stringent availability requirements for real-time applications.

Security professionals rely on timely and accurate information to execute their missions. Real-time, video-based applications are creating a whole new range of options for security professionals to protect people and property. Networking technology is at the center of the deployment of digital video technology and solutions. To take full advantage of what these new solutions can bring to the table, security professionals must be able to couple video applications with a military-grade network to achieve the desired levels of bandwidth quantity and quality, as well as the required levels of security, reliability, flexibility, and interoperability.

Shane McClelland is director of marketing and market management for Marconi Systems. Scott Mansfield is principal engineer for Marconi Systems. For information, visit www.marconi.com/SIS.