Optimizing Bandwidth for IP Video Surveillance
This guide explains how to manage network bandwidth in video surveillance systems by utilizing sub-streams and proper configuration to prevent network congestion.
Why this matters
When you mismanage bandwidth in a video surveillance deployment, your customer faces a network that grinds to a halt, causing dropped voice calls and sluggish internet speeds. If you do not understand how to properly balance video data, you risk turning a high-end security installation into a liability that degrades the client's entire business infrastructure.
The core idea
Bandwidth represents the total volume of data that can be transmitted over your network connection within a specific timeframe, typically measured in Megabits per second. In an IP surveillance environment, the camera acts as a data generator, pushing a continuous stream of pixels to the Network Video Recorder or a remote viewing client. Resolution refers to the level of detail in the video, such as 4K or 1080p, while the frame rate, or frames per second, determines how fluid the motion appears. High resolution combined with a high frame rate creates a massive amount of data, which can easily saturate a local or wide area network.
A codec is the compression algorithm, such as H.265 or H.264, which shrinks that raw data for easier transport. The most vital concept for remote viewing is the sub-stream, also known as the secondary stream. This is a parallel, lower-resolution, and lower-frame-rate feed produced by the camera specifically for mobile apps or multi-view displays, keeping the main high-definition stream reserved for event recording.
How it works in practice
In our product line, such as Uniview, cameras are designed to output dual streams simultaneously. You configure the main stream for local recording at the maximum possible quality, typically using H.265+ for the best compression efficiency. However, you should always configure the sub-stream at a much lower resolution, such as D1 or CIF, and limit the frame rate to 10 or 15 frames per second. When a user connects via the EZStation software or a mobile app like EZView, the system should default to pulling this sub-stream.
By using the sub-stream, you reduce the throughput requirement per camera from 4 to 8 Megabits per second down to roughly 500 Kilobits to 1 Megabit per second. This ensures that even when an end-user monitors sixteen cameras on a single screen, the total bandwidth usage remains well within the limits of standard office networking hardware, preventing the common phenomenon where the entire office network lags whenever the security team opens their software.
Worked example
Imagine a customer calls stating that their reception staff cannot load their email or make VoIP calls whenever the security guard opens the NVR software to watch the front entrance. The customer assumes their internet service provider is at fault and wants to upgrade their monthly plan. If you simply tell them to use a lower quality codec for their primary recording, you are sacrificing image quality that the customer needs for evidence collection, which is a major professional error.
The correct approach is to log into the NVR remotely, navigate to the remote management settings, and confirm that the multi-screen view is set to request the sub-stream rather than the main stream. Once you switch the global view settings to use the secondary stream, the total bandwidth consumed by the surveillance software drops by over 80 percent, the network stabilizes instantly, and the reception staff regain their connectivity without you ever needing to compromise the actual recorded video quality stored on the hard drives.
Where people go wrong
Many technicians make the mistake of assuming that the codec is the primary lever for bandwidth control. While switching from H.264 to H.265 can provide better compression, it does not solve the underlying issue of trying to push twelve high-definition video feeds through a bottleneck. Another common error is failing to set a frame rate limit on the sub-stream, leaving it at 30 frames per second, which is unnecessarily high for a multi-window overview screen.
A third mistake is ignoring the network switch capabilities; if the switch is only 100 Megabits per second, even a few high-definition cameras can hit the physical ceiling of the hardware. Finally, failing to educate the customer on why they see a lower resolution image in the multi-view grid leads them to believe the equipment is faulty, when in reality, it is a strategic design choice to preserve the integrity of their network.
Key takeaways
Always prioritize the use of the sub-stream for multi-window viewing and mobile access. Keep the main stream optimized for evidence-quality recording, not for live grid viewing. Limit sub-stream frame rates to 10-15 frames per second to maximize network stability. Verify that the NVR and software settings are configured to request the sub-stream automatically. Explain the trade-off between live view performance and recording quality to the client to manage expectations.