This article originally appeared in the December 2024 issue of Security Business magazine. Don’t forget to mention Security Business magazine on LinkedIn and @SecBusinessMag on Twitter if you share it.
What do smart cities, construction sites and the oil and gas industry have in common? Among potentially other answers, the answer is that these verticals all struggle with bandwidth as a major concern.
Security executives in all these verticals (and others) want the image of a 4K (8MP) camera (or better) for general forensic use of a wide scene, or the digital zoom capabilities offered by 8MP cameras over that of a 2MP or 4MP camera. The enhanced solutions provided by 8MP imagers allow camera manufacturers the ability to provide color images in starlight. The onboard processing now allows camera manufacturers to offer edge analytics and even storage at the edge.
The problem has been that these high-megapixel cameras come at a cost to the network and storage. Storage is cheap and the price is decreasing, making it less of a concern, but not gone. However, bandwidth is a major limiting factor.
Smart cities, construction sites, and the oil and gas industry all suffer from locations with limited bandwidth. Fiber connectivity may not be possible. Point-to-point wireless may not be feasible due to trees, buildings, etc. Connectivity may be limited to a cellular connection – something that may not be able to support an 8MP camera even with H.265 compression and a manufacturer codec where only pixel change is sent in the stream.
As a perfect example, an 8MP camera sends about 16 MB of data per second. When you add a smart codec, let's be generous and say that number is reduced between 1/3 and 1/2. On a cellular connection in real-time, that may still cost $10,000-plus per camera every month if it needs to be kept from throttling. And that is only if the cellular provider’s network can even handle the impact.
For those instances where connectivity is not possible, but the camera need and viewing is still required, what is the answer? Security practitioners know that the answer cannot be no, not possible, or something else in the less-than-doable category. So now what?
A New Approach: Digital Barriers
Traditional H.264 and H.265 compression works for most applications, and manufacturer-specific codecs for additional compression help even the most normal applications; however, for situations that still need real-time live viewing where more bandwidth is not possible, solutions that can compress the compression provide the best of both worlds.
This is where new technology from Digital Barriers (www.digitalbarriers.com) thrives. Their customer references include the phrase “game-changing”, simply because the technology allowed the customer to monitor video when they needed it – from providing real-time video of terrorist takedowns, to remote camera solutions for smart cities and VIP events.
“5G and Edge computing are poised to transform security and operations, with video as a sensor driving efficiency and creating new possibilities,” says Kunal Shukla, CTO at Digital Barriers. “But what happens when we scale up to hundreds or thousands of bandwidth-hungry video sensors? How much additional fiber and 5G spectrum do we need to use? We’ve made the adoption of real-time video and analytics over cellular a technical and affordable reality by reducing bandwidth requirements by up to 90%.”
The difference Digital Barriers brings has been its ability to stream live video in low-latency environments without losing quality – yes, without creating grainy and pixelated video, streaming in real-time an 8MP camera at under 1 Mbps.
While this seems extreme, and possibly impossible, compression is all about the math. Where Digital Barriers has excelled is using its patented AI-based video codec to dynamically adjust to a scene to send only the needed pixels. This is different from most smart codecs, because it compresses the video as well as the pixels in the scene with movement that matters to security. Another nice feature is that they can set the threshold for compression, so in an extreme case, they can cap video transmission at under 1 Mbps and still stream live 8MP video without losing quality (at least to the naked eye).
Compression of the compressed video is designed to allow the Real Time Crime Center (RTCC) officer or the Global Security Operations Center (GSOC) operator the needed tools to see the video without dropping on bandwidth-constrained networks. The problem is this video is no longer admissible in court because of the video alteration.
When video needs to be kept immutable for evidentiary purposes, edge recording – either on the camera or with a co-located appliance – must be implemented. The primary stream would be sent to this storage device with all the proper watermarking. A secondary stream would be compressed and sent for live view. That said, most of the scenarios that require this already require a version without the benefit of real-time video streaming today.
