When dealing with security systems you will often find yourself working with large amounts of data that has to be transported across networks from camera to server to viewing station. All of this data has to be managed and accounted for or viewing and recording quality will drop.
A number of tech support calls result in us checking the bandwidth usage of the cameras and their routing. 90% of cases where cameras aren’t working with new installations can easily be fixed by checking the network settings – firewalls and managed switches – and making sure there is enough bandwidth.
So what is the bandwidth?
The bandwidth is the amount of network traffic that your network has been setup to deal with. It is also the amount of network that will be generated by your cameras. The bandwidth consumption can vary by manufacturer but as a worse case (max amount of traffic) a 5 megapixel camera using MJPEG can generate 20 Mbps of traffic. This would mean a standard recorder with a 1Gbps network connection will start having trouble when there is 20 or more cameras. This is assuming all switches between the camera and recorder also have a 1Gbps capability.
Understanding bandwidth is critical because it impacts network load and storage use, both causing an increase in costs. There are few factors that you can consider to optimise the bandwidth usage without compromising the security system efficiency.
- Choose the right video compression
There is an inverse relationship between compression and bandwidth: the more compressed a video or image is, the lower the bandwidth will be. Some video compression format, like H.264, compress similar pixels in a frame, compare previous frames and transmit only the pixels changed between frames. Sending only the changed information instead of each frame of video saves significant bandwidth. If we switch the 5MP camera to H.264 the bandwidth will go down to about 5 Mbps which will allow a lot more cameras on the network (closer to 100).
- Reduce the camera resolution
When setting your camera resolution you should consider the complexity of the surveillance scenario. While 2MP cameras could be extremely useful where clear good quality images are required to provide post event analysis and identification, such as in large public areas, retail environments, parking facilities and transport systems or when live viewing is not available or essential, cameras with a lower resolution could be enough to monitor small and mid-sized sites with limited details and movements.
- Decrease the frame rate
As image quality and frame rate increase, bandwidth requirements increase as well. The frame rate should meet business requirements, but it does not need to be higher than is necessary. Motion pictures are captured at 24FPS. Human visual capabilities normally register images captured at 24FPS as fluid motion. Regular television sets use 25FPS (PAL) or 30FPS (NTSC). These rates are often too high for some video surveillance applications and in most applications less than 12-15 FPS is normally sufficient.
- Set your camera on motion detection
Motion detection is used to trigger an alarm only when cameras detect a movement or a change of scene; this means less data to be sent through the network.
For bandwidth requirements it is useful answer the following questions:
- How much bandwidth does each camera use?
- How many simultaneous users are connected to the VMS?
- How many simultaneous video and audio streams will users access?
- What is the minimum acceptable video resolution and quality required by users?
- What is the maximum amount of bandwidth required by users?