Radar is an excellent addition to perimeter security, but you will have to overcome these challenges first.
Radar integrated with AI-enabled surveillance cameras offers an excellent addition to perimeter security. But as with every technology, this comes with some challenges for customers and integrators.
Speaking to asmag.com recently, Amit Isseroff, VP for R&D at Magos Systems, listed six significant challenges radars pose. From a technology perspective, most of them revolve around the need for large open spaces, low resolution, and reflections and ghost targets generated by metal objects. While AI can solve most of these concerns, there are also certain business challenges that integrators can expect.
1. Need for open spaces
You would need open spaces to allow the radar time to track the suspected target, verify real object movement, and filter out random noise from vegetation. This becomes even more crucial when the approach path is tangent (roughly perpendicular) to the radar, making the tracking process more difficult as the doppler effect is less dominant in these scenarios.
“Relying on AI technology for fused target detection can reduce the required open space, as the second layer of video detection augments the initial radar detection and shortens the required tracking time for high confidence detection,” Isseroff explained. “To demonstrate – without AI, the required open “sterile” space next to a fence for reliable radar operation is at least 10 meters in a typical installation where the radar is positioned to monitor along with the fence. With AI, this distance is cut down to roughly three meters.”
Also read: What's new in post-pandemic perimeter security?
2. Reflections and ghost targets
Reflections and ghost targets are two other issues that radars pose but can easily be dealt with AI. As radar relies on transmitted radio waves reflecting off-targets and received at the radar again, it is inherently prone to suffer from reflections. A typical scenario is when a radar is set to monitor movement outside the perimeter fence but also has a line of sight to inside areas.
In such a scenario, the fence acts as a mirror for RF, and the radar cannot differentiate between a person walking inside the fence from his “reflection” outside. AI can help here, filtering out ghost targets by verifying that there are no real threats where the reflections are observed.
3. Low resolution
The inherent low resolution in radars, especially “angular” resolution, hampers the differentiation of objects that are at the same distance from the radar. In scenarios where the radar is set to monitor an area next to a busy road, traffic can cause a lot of constant “noise” or clutter that will mask targets walking next to the road and render the radar blind to these targets.
“Increased resolution can be achieved by using multiple receiving antennae, but this tends to increase the radar unit size and price,” Isseroff said. “Multiple Input Multiple Output (MIMO) has been implemented in communication technology in the past two decades and has recently been introduced to radar technology. MIMO uses multiple transmit and receiving antenna, and when correctly applied can keep the unit size minimal, increase resolution and thus improve detection performance even in cluttered environments.”
4. Challenges in deployment
The most common challenge faced by integrators is proper deployment. Deploying Radar involves a somewhat different logic from camera/fence deployment. Not only do they need line-of-site, but also require careful positioning to provide the best detection results.
“You can easily overcome this challenge by properly training the integrators in advance, performing an initial site survey, and using site-design tools while consulting with the radar experts to validate proper deployment,” Isseroff explained.
Also read: Top perimeter security trends 2022
5. Integration with legacy systems
The next challenge would be integrating radar into an existing security system. Most customers already have an eco-system with their preferred security software suite and other systems, such as access control. Depending on the model and brand of these solutions, integration could be complex.
“Those are overcome on the software side by integrating Magos Area Surveillance Software (MASS) with all major security systems, providing generic standard solutions, and offering quick support for implementing new integrations on demand or allowing the integrator to implement the integration himself based on the software API/SDK,” Isseroff added.
6. Meeting customer expectations
The final challenge integrators often face is managing expectations. While radars are almost always a superior solution for perimeter security, they are not “magic” boxes that can detect in any environment, through walls and fauna, and provide zero false alarms in any situation.
To overcome this challenge, both the customer and the integrator must undergo basic training on radar technology to understand its advantages and where it is suitable for efficient and effective deployment.