The millimeter-wave sensor, in essence, is a sensor that detects signals in the millimeter-wave band. Generally, millimeter-wave refers to electromagnetic waves with frequencies ranging from 30 to 300 GHz and wavelengths between 1mm and 10mm. Due to its propagation speed being equivalent to that of light in air, it is considered to be highly accurate, particularly when detecting objects at long distances. They are particularly useful in applications such as radar, communication systems, and autonomous vehicles, where they can be utilized to precisely determine the position, velocity, and movement of objects in the environment.
Figure: the millimeter-wave sensor
How does millimeter-wave ensure accuracy? Next, let's take a look at the millimeter-wave radar and how it achieves high-precision detection.
The accuracy of wireless communication systems operating within the millimeter-wave frequency range (30 GHz to 300 GHz), also known as mmWave, depends on several factors, including:
The design of the antenna used for millimeter-wave communication is crucial for achieving high accuracy. The antenna must be able to transmit and receive signals in a highly directional manner, resulting in higher data rates and better signal quality. Common types of antennas used in millimeter-wave sensors include patch antennas, horn antennas, and phased-array antennas.
Figure: Millimeter-wave sensor antenna layout
Signal processing plays a crucial role in the accuracy and performance of millimeter-wave sensors. By utilizing advanced signal processing techniques, such as beamforming, Doppler processing, distance processing, and clutter removal, millimeter-wave sensors can provide highly accurate and reliable measurements in various applications, including the automotive, industrial, and security fields. The following are some key signal processing techniques used in millimeter-wave sensors:
Each individual small antenna in the antenna array emits signals in different directions, and the purpose of beamforming is to combine the signals they emit into a more directional pattern, transmitting the signal towards a specific direction to enhance the accuracy of the sensor.
Figure: Digital Signal Beamforming Processing
Doppler processing is used to measure the velocity of moving objects. Millimeter-wave sensors emit signals that are reflected off objects and received by the sensor. By analyzing the frequency shift of the reflected signals, Doppler processing can determine the velocity of the moving object
Figure: Detecting Object Motion
Based on the Doppler principle, by obtaining the time delay between the transmitted signal and the received signal, processing and analyzing the beat signal, the distance can be calculated; by calculating the frequency shift of the reflected signal, the speed of the object can be determined. You can refer to the Wikipedia page for an introduction to what the Doppler effect is.https://en.wikipedia.org/wiki/Doppler_effect
Noise reduction and other data processing techniques:
Noise refers to unwanted signals that may interfere with the detection of the object of interest. Noise reduction techniques are used to filter out unwanted signals and improve the accuracy of sensors. Different algorithms for noise reduction have a significant impact on the accuracy of detection scenarios for millimeter-wave sensors.
Figure: Diagram illustrating noise reduction
Furthermore, advanced signal processing techniques such as Fourier transform, wavelet transform, and Kalman filter are employed as fundamental methods to extract useful information from the raw signals received from sensors.
The accuracy of millimeter-wave systems can be affected by the operating environment. For instance, obstacles such as walls, buildings, and trees can attenuate or reflect signals, resulting in decreased accuracy. Additionally, the accuracy of detection can also be affected by the installation angle.
Figure: Experimental Results of Material Penetration with Millimeter-wave Sensors
TI conducted a comparative experiment at 60 GHz, which demonstrated the ability of millimeter-wave sensors to detect and count people through various materials such as plastic, drywall, and wood.Results for this experiment were generally favorable with minimal amounts of interference due to the obstruction. However, the plywood proved the most difficult to detect objects through. We tuned the minimum required SNR before the experiment to account for that. The double drywall scenario created some reflections noticeable in the video as well.
In summary, the development of millimeter-wave sensors in the market has become mature, especially in hardware design such as antennas, which has become mature. Therefore, the core factor affecting the accuracy of millimeter-wave sensors is mainly the algorithms used in signal processing methods. Additionally, the installation environment should also be considered in order to obtain ideal detection results. Next, let's take a look at the parameters of millimeter-wave radar to see which specific parameters can help us judge the accuracy.
The parameter table of millimeter-wave sensors will vary depending on the specific sensor model and manufacturer. However, understanding a few main parameters can help you understand how to choose precision.
Millimeter-wave radar refers to radar operating in the millimeter-wave band. The wavelength range of millimeter waves is 1-10mm, corresponding to a frequency of 30-300GHz. The commonly used millimeter-wave radar frequency bands are 24GHz, 60GHz, and 77GHz. Among them, the wavelength of 24GHz is 1.25cm, but it is still called millimeter wave in the industry. The wavelength of 60GHz is 5mm, and the wavelength of 77GHz is 3.9mm. Higher frequency corresponds to shorter wavelength, which results in higher measurement resolution and accuracy.
Millimeter-wave radar at 77GHz with higher precision is mainly used in the automotive field, while 24GHz millimeter-wave radar has slightly lower performance than 77GHz millimeter-wave radar. However, its various performances are more mature, and its cost is lower, making it more widely applicable in areas such as smart homes, smart appliances, smart bathrooms, security, and agriculture.
The range resolution defined by radar is the ability to distinguish different targets placed at the same angle direction (azimuth) but at different distances from the radar.
That is, the accuracy of measuring the velocity of a single target, and the data depends on the signal-to-noise ratio (the unit that measures the quality of the received signal of the radar). Whether the signal-to-noise ratio is high or not is a fundamental parameter for measuring the target detection performance of millimeter-wave radar.
It is also important to determine the direction of the target (azimuth) in a clear manner. This can only be achieved within the field of view (FOV) of the radar, which defines the angle coverage range of the radar in the azimuth (horizontal plane) and elevation (vertical plane) angles.
As with range resolution, it is also important to distinguish two independent targets placed at different angles (azimuth angles) but at the same distance. Since the frequency shift caused by signal delay cannot be used to identify signals from each target, multiple antennas need to be used at different locations, in conjunction with angular resolution parameters, to better distinguish them.
|Model||24GHz millimeter wave sensor||60GHz millimeter wave sensor||77GHz millimeter wave sensor|
|Distance resolution(cm)||Minimum 60||Minimum 3.75||Minimum 3.75|
|Speed resolution||Low||High (2.5 times that of 24G radar)||High (2.5 times that of 24G radar)|
|Antenna size||big||Smaller (about 6 times smaller than 24G antenna size)||Smaller (about 6 times smaller than 24G antenna size)|
|Application Scenario||General scenarios that do not require high parameters, such as presence detection, proximity detection||Can be used in industrial, security, automotive and other fields||It is mostly used in transportation and automobile fields|
In general, the higher the operating frequency, the shorter the detection range of the sensor, but it has strong anti-interference ability, and higher resolution and accuracy, some can even achieve millimeter-level detection accuracy. However, overall, millimeter-wave sensors can achieve high precision in various applications, from tracking personnel in crowded environments to measuring fluid flow velocity.
Of course, in some application scenarios, very high precision data may not be necessary. For example, if you only need to detect the presence of a human body, different sensors can achieve this, such as infrared sensors, ultrasonic sensors, and PIR sensors, which are common sensors for detecting the presence of a human body.
An infrared sensor is an optoelectronic switch sensor that integrates both emission and reception functions. It consists of a pair of infrared emission and reception tubes. The emission tube emits infrared light at a certain frequency, which is reflected back by an obstacle (a reflecting surface) in the detection direction and received by the reception tube. After processing by the comparator circuit, the green indicator light will light up, and at the same time, the signal output interface will output a digital signal. It can be widely used in robot obstacle avoidance, interactive media, industrial automation production lines, and many other applications.
Regarding the precision of infrared sensors:
Since infrared sensors mainly rely on infrared light for detection, there may be significant precision issues in outdoor environments with highly variable lighting conditions. Infrared sensors are susceptible to environmental light interference and may produce erroneous readings, whereas millimeter-wave sensors are less susceptible to such interference.
Figure: Infrared Sensor
The ultrasonic sensor is another commonly used sensor for human detection and can also be used for fast ranging or obstacle avoidance applications. Ultrasonic waves are a part of sound waves that are inaudible to the human ear and have a frequency higher than 20kHz. Like sound waves, they are generated by the vibration of matter and can only propagate in a medium. Based on this principle, ultrasonic sensors are designed to emit and receive ultrasonic waves.
Regarding the precision of ultrasonic sensors:
Regarding the accuracy of ultrasonic sensors, their principle of operation imposes some limitations in detecting scenes. Ultrasonic sensors are susceptible to environmental noise, and their accuracy can be influenced by the shape and size of the detected object, as well as the temperature, humidity, and air pressure of the environment.
Figure: Ultrasonic Sensor
Passive infrared (PIR) sensors possess the ability to gauge the thermal energy of objects as well as detect their motion. This particular kind of sensor solely measures infrared radiation and does not emit it, hence it is referred to as a passive infrared sensor. Generally, all objects emit some form of thermal radiation in the infrared spectrum which is invisible to our eyes but can be detected through an infrared sensor, thereby enabling the detection of human presence.
Regarding the precision of PIR sensors:
Regarding the precision of PIR sensors, their principle dictates that they are susceptible to temperature changes which may result in erroneous readings, particularly in unstable temperature conditions. For instance, when utilized for automatic doors, PIR technology is only capable of detecting motion, signifying that a PIR sensor cannot detect stationary objects as there is no change in their thermal radiation, making it impossible to discern whether a person has entered a building or differentiate between a human or animal presence.
Figure: Passive infrared (PIR) sensor
In general, while each type of sensor has its own strengths and weaknesses that are related to its design principles, there are certain applications where considering cost and low precision requirements, ultrasonic sensors, infrared sensors, and passive infrared sensors can all meet the demand. Therefore, one can opt for different types of sensors to achieve the desired detection function.