Upgrade Your Security Camera by Replacing PIR Sensors with 60-GHz Radar

March 25, 2024

Sponsored Story

Consumer security cameras are growing in popularity when it comes to smart home devices, whether they’re attached to porches, eaves, or as part of a doorbell.

Significant technology advancements have been made in recent years, offering features like high-definition video, remote access, motion detection, and integration with smart home systems. However, they still have several limitations. Those shortcomings include:

• Limited field of view (FoV), which may not effectively cover large areas or multiple angles.
• False alarms, lead to unnecessary alerts and notifications.
• High power consumption, resulting in repeated battery charging or changes, potentially the need for wired AC, can limit placement significantly.

Current Motion-Detection Technology: PIR Sensors

The most common motion-detection technology employed in today’s consumer security cameras is passive infrared (PIR), which detects movement by sensing changes in infrared radiation within its FoV. When an object moves within the sensor’s detection area, the amount of infrared radiation reaching different parts of the sensor changes.

A typical PIR sensor consists of a pyroelectric sensor, which is divided into segments. Each segment is covered by a filter, allowing only infrared radiation within a certain wavelength range to pass through. These segments are arranged so that they can detect motion across the sensor’s FoV.

The sensor then analyzes the signals from the various segments. When a significant change in infrared radiation is detected across multiple segments within a short period, it indicates motion within the sensor’s range. This triggers the sensor, resulting in an action, such as starting the camera, activating lights, etc. Because PIR is a passive technology, it is relatively energy efficient. The limitations and deficiencies of PIR technology include limited range, a line-of-sight requirement, a limited detection angle, and reduced effectiveness at night or in challenging environmental conditions.

Alternative Technology: 60-GHz Radar

An alternative to the PIR sensors is 60-GHz radar technology, also known as millimeter-wave radar. It operates by emitting short electromagnetic pulses/waves at a frequency of 60 GHz, enabling the analysis of the reflected signals to detect objects and gather information about their properties (e.g., speed, distance, angle, etc.).

The radar system’s emitted 60-GHz waves are part of the millimeter-wave spectrum, which lies between microwave and infrared wavelengths. The emitted waves propagate through the air and travel until they encounter objects in their path. When the waves hit an object, they scatter in various directions.

A portion of the emitted waves reflects off the object's surface, back toward the radar system where the signals are “received” and processed to extract information about the detected objects. Processing involves analyzing factors such as time delay, Doppler shift, amplitude, and phase of the reflected signals. Through that analysis, the system can make approximations of the person’s/object’s distance, speed, size, etc. The accuracy achieved by this technology is far greater than can be achieved by PIR, and it is not affected by temperature, weather, or darkness conditions.

Consumers may not know whether or not the technology inside their cameras uses 60-GHz radar, but they will realize its benefits, with higher reliability (e.g., reduction in false alarms), fewer battery charges or changes, and increased accuracy. By reducing the number of false alarms, battery life is significantly prolonged because the camera does not activate every time there’s a false alarm. The radar can be programmed to distinguish between a human and a small animal. Hence, the camera needn’t be turned on every time a squirrel enters the FoV.

In addition, the amount of Cloud storage is reduced because you’re not storing the false alarms. The design aesthetics for a radar-based design are far more pleasing because you can simply hide the sensor behind a piece of glass, whereas the PIR sensor typically sits behind a plastic dome.

At the same time, the security-camera original equipment manufacturers (OEMs) benefit by having a competitive edge with superior products and happy customers. And thanks to Infineon, they can also take advantage of a reference design and ready-to-use proprietary algorithms. Note, that the OEMs can increase their processing capability and add AI or machine-learning (ML) capabilities to their designs. That would increase the accuracy to the point where it could distinguish between children and adults. TinyML is a good solution for a product of this nature. For anything more complex, processing would need to be handled in the Cloud.

Ready-To-Go Reference Design

Infineon offers a radar-based reference design that's implemented with the company’s latest 60-GHz radar sensor, the BGT60TR13C. The part integrates with an antenna (antenna-in-package or AiP), which simplifies the design process. The reference design is built on a Raspberry Pi (RPi), which provides convenient access to the camera module through an HDMI interface, on a platform that developers are familiar and comfortable with.

Using one transmitter (Tx) and three receivers (Rx), Infineon has been able to achieve a FoV of 130 degrees, with a range of about 8 m. It is implemented with a PSoC6 MCU (the CY8C6245) to run a proprietary algorithm for presence detection, which works in tandem with the radar sensor. That particular PSoC offers multiple, granular ultra-low power modes, which are enabled when nothing is being sensed.

The RPi handles the housekeeping functions, like turning on the camera when a presence is detected, and then potentially sending images to the Cloud. The RPi also serves as the hub between the camera and an Android phone running the GUI for low-power presence detection. The RPi can turn on a flash when an up-close presence is detected for more positive identification.

Infineon has also removed one of the barriers to entry with radar by providing the necessary algorithms, which can be quite challenging when dealing with three-dimensional objects. Designated the “camera sensor fusion algorithms,” they provide reliable, low-power wake-up upon presence detection in the security camera’s FoV.

Typical Use Case

In a typical deployment, the camera would be combined with the BGT60TR13 60-GHz radar sensor for presence detection and a CY8C6245 PSoC6 MCU to run the presence-detection algorithm. This should result in a 130-degree field of vision (FoV) with a presence detection of around 8 m. The Raspberry Pi board handles the Bluetooth, Wi-Fi, and camera connectivity.

In the deployment, Infineon’s Wi-Fi solutions offer “Network Offload” capabilities, which enable the security camera to stay connected to the cloud at all times, even when the video processor is in standby mode. This configuration allows for low-latency remote wakeup and faster video streaming. Options for developers include various Wi-Fi and Bluetooth combo products from Infineon, such as the ultra-low-power Wi-Fi 5 1x1 CYW43022 for extended battery life or the higher performing Wi-Fi 6/6E 2x2 CYW5557x or the 1x1 CYW55513 that can provide extended range and a larger coverage area. In addition, designers can opt to have the MCU and Wi-Fi in a single chip, providing most of the components integrated into one small package. Note that Infineon currently offers a development board with this configuration running on an Android app.

As stated earlier, the benefits of radar-enabled security cameras for end users include higher levels of accuracy and reliability, less power being consumed, meaning fewer battery changes or recharge cycles, and fewer false-alarm images to decipher thanks to the increased accuracy. For security camera manufacturers, they’ll be assured of happy customers thanks to these same benefits, resulting in a competitive edge. Contact Infineon for more information.

---

Tanja Hofner has 30 years experience of running hardware, software, and application engineering teams. She currently leads a system development team for sensor solutions at Infineon. Tanja previously worked in various engineering and management positions at InvenSense and Maxim Integrated.