Electronic Enthusiasts Network reported (text / Huang Jingjing) Millimeter-wave radar technology has been well-known in the fields of autonomous driving, smart cars, and smart homes in recent years. In Google's millimeter-wave gesture radar project Soli, we saw that smart watches equipped with Soli chips can perceive The user's actions, its millimeter-wave gesture radar module has also been applied to Google's Pixel 4 mobile phone, which is the first application of millimeter-wave radar in consumer electronics products.

When it comes to millimeter-wave radar, most readers probably think of automotive radar first. In fact, the millimeter-wave radar used in automobiles is completely different from the millimeter-wave radar used in intelligent hardware in terms of scenarios and technical requirements. Based on the research and development of millimeter-wave radar technology based on intelligent hardware applications, domestic universities already have a strong scientific research team. Electronic Enthusiasts.com also had the honor to have an exclusive interview with the leader of this scientific research team, an authoritative expert in the research and development of millimeter-wave radar technology, namely Professor Gu Changzhan from the Department of Electronic Engineering of Shanghai Jiao Tong University and the Institute of Artificial Intelligence, and talked about his own research and development experience, The research progress of the team in ultra-short-range millimeter-wave radar and his views on the application prospects of this technology.

Participated in the Google Soli project, and then returned to China to teach, focusing on the exploration of millimeter-wave radar technology

Gu Changzhan received his bachelor's and master's degrees from the Department of Information and Electronic Engineering of Zhejiang University in 2006 and 2008, respectively, a master's degree in electrical and computer engineering from the University of Florida in 2010, and a doctorate in electrical engineering from Texas Tech University in 2013. . Gu Changzhan, who graduated with a Ph.D. in 2013, came to Silicon Valley and initially engaged in the research and development of wireless connection chips, and worked at Marvell's Silicon Valley headquarters.


Professor Gu Changzhan, Department of Electronic Engineering and Artificial Intelligence Research Institute, Shanghai Jiao Tong University

Later, since 2016, Gu Changzhan participated in the industrialization of millimeter-wave gesture recognition technology, and the millimeter-wave gesture radar module he designed was applied to Google Pixel4 mobile phone, realizing the first millimeter-wave radar sensor in consumer hardware products. Mass production landed.

This is his project experience at Google's US Silicon Valley headquarters before returning to China. After he returned to China, he joined Shanghai Jiaotong University in 2019. He said that on the one hand, he wanted to train students and let more young people participate in the field of millimeter-wave radar; on the other hand, he hoped to take advantage of China's mature and complete hardware industry chain , to help achieve better and faster implementation of millimeter-wave radar intelligent hardware products.

The technical difficulties of ultra-short-range millimeter-wave radar, the chip solution needs to be optimized

After decades of development, radar technology has moved from military, large gas detection and other professional fields to civilian fields. Professor Gu Changzhan said that traditional radars are large in size and high in power consumption, and the target is far away from the radar, such as a few kilometers or even hundreds of kilometers away from an aircraft. With the advancement of semiconductor technology and the rise of autonomous driving, the size of millimeter-wave radar has gradually become smaller, and it is increasingly used in automobiles. Compared with traditional military and atmospheric radars, automotive radar is a medium-range application, that is, a target vehicle. The distance from the radar is tens to hundreds of meters.
In the field of vehicle-mounted millimeter-wave radar, it is currently in the golden period of "domestic substitution". In recent years, a number of outstanding companies such as Catland Microelectronics and Anda Technology have emerged. The CMOS route adopted is also an inevitable way for the development of millimeter-wave chips. I believe Domestic chips will play an important role in the field of autonomous driving in the future. In addition to in-vehicle scenarios, millimeter-wave radar, as a technology that can achieve high-precision, non-sensing detection, and does not violate privacy, also has many application opportunities in daily life scenarios. Not long ago, Huawei officially announced the addition of millimeter wave sensing technology to its whole-house smart solution.

Encouraged by the huge demand scenarios, Professor Gu started to lead the team to develop intelligent perception radar technology for a wide range of daily life scenarios, such as gesture interaction, closed space monitoring, human posture recognition, and vital sign detection. Professor Gu analyzed that these emerging intelligent perception scenarios are mostly <>

Professor Gu pointed out that in the emerging field of intelligent perception, there is currently no better chip-level solution, and most chips still use the architecture and algorithm system of vehicle radar, and there are many bottlenecks in practical applications. In-vehicle millimeter-wave radar technology is more suitable for medium and long-distance detection in terms of hardware system architecture, algorithm process, etc. In addition, there are essential differences in the physical characteristics of electromagnetic wave scattering and target movement corresponding to vehicle and life scenes, which makes the millimeter-wave perception based on automotive radar. If the technical solution is to be used in smart hardware in daily life, there are still natural shortcomings such as short-range blind spots, insufficient recognition accuracy, and high power consumption, which makes the vehicle-mounted solution unable to meet the requirements of the millimeter-wave radar in emerging fields such as smart home and wearables. Due to the requirements of power consumption and accuracy, the application dimension is limited.

Led the Shanghai Jiaotong University team to achieve a breakthrough in the "non-contact vital signs detection" of millimeter-wave radar

Professor Gu Changzhan said that our Shanghai Jiaotong University research group has focused on system-level research on hardware architecture and algorithms for short-range millimeter-wave radar perception, solved the technical problem of simultaneous co-frequency and full-duplex, realized the integration of transmitting and receiving signals, and greatly reduced the size of the sensor. The size of the millimeter-wave radar makes the development of millimeter-wave radar further in miniaturization, and it is possible to load it into more small household smart devices and medical health equipment. In addition, the key technical problem of micron-level displacement motion detection is realized, which helps to collect extremely small motion information that cannot be recognized by human eyes. This series of technological breakthroughs has made the application of millimeter-wave radar in many fields such as human-computer interaction, health monitoring, and smart medical care to play a greater role, showing a good prospect.


Especially in the non-contact vital sign detection, the Shanghai Jiaotong University team has made further breakthroughs. Professor Gu introduced that millimeter wave sensing technology can use the Doppler principle to realize non-contact detection of vital signs information such as heartbeat, respiration, and pulse of the human body. The precise displacement trajectory information called "Doppler Cardiogram DCG", combined with the self-developed artificial intelligence medical prediction model, can accurately reflect the subtle changes in the whole heartbeat, respiration, and pulse movement process, so as to deduce the correlation between cardiovascular and cerebrovascular diseases. A series of health-related parameter indicators to achieve monitoring and early screening of various diseases.

For example, based on our Doppler cardiogram DCG, the heartbeat RR interval, QRS complex and P/T wave can be obtained in a non-contact way that is comparable to traditional contact ECG. Our team has carried out a series of clinical experiments with several affiliated hospitals of Shanghai Jiaotong University, and clinically verified the millimeter wave sensing technology to screen for apnea, arrhythmia, premature beats, atrial fibrillation, heart sounds, hypertension and other cardiopulmonary and cardiovascular Feasibility of related diseases.

It can be seen that the scientific research results of the non-contact vital signs detection based on millimeter-wave radar technology by the Shanghai Jiaotong University team will be able to assist in obtaining key vital signs data and help to apply this emerging technology to smart medical care. Looking forward to the future, this technology can help people to prevent and treat diseases in a non-sensing way. It may be just a matchbox-sized device on the desk to help people with a history of respiratory and cardiovascular diseases, or high-intensity overtime work. of people monitor their health status in advance.

Academia and industry work together to promote the development and implementation of millimeter-wave radar technology

At present, the transformation of scientific research results of short-range millimeter-wave radar is mainly in the aspects of presence perception, gesture interaction and non-contact vital sign detection. It has also achieved scientific research results on short-range millimeter-wave radar.

Professor Gu said that in the field of presence perception, foreign MIT, Stanford and domestic Peking University, Zhejiang University, Shanghai Jiaotong University, etc. all have teams engaged in the research of presence perception technology. A good alternative to traditional infrared technology. The next technological trend is how to control chip costs while improving indicators such as distance, angle, and coverage.

In addition, in terms of gesture interaction applications, the “non-contact” demand during the epidemic has brought new opportunities for millimeter wave gesture interaction technology, which will play an important role in specific scenarios such as elevator door control and gesture somatosensory games.

Professor Gu also mentioned that there are relatively sufficient talents for automotive millimeter-wave applications, which can also be seen from the rapid growth of China's automotive millimeter-wave radar industry in the past few years. However, millimeter-wave radar is still in a state of talent shortage in off-vehicle applications. At present, these talents are mainly concentrated in Europe and the United States. However, in recent years, the returnees from the chip industry have improved the problem of insufficient talent pool to a certain extent.

In China, Peking University, Zhejiang University, Shanghai Jiaotong University and other universities have laboratories to promote the research of millimeter wave sensing in off-board applications. Professor Gu's team includes more than ten doctors and masters, and their research interests include the application of millimeter-wave radar in human-computer interaction, medical health, smart home, elderly care and other fields. Professor Gu very much recognizes the word "one core for life", which represents focus and devotion. He said that no matter whether the student's topic is chip, system or algorithm, we are committed to allowing each student to be exposed to practical hands-on research content and understand the significance of their own research work. I often tell students that we pursue "impactful useful research" and hope that the research work done during school can benefit the society, contribute to the progress of national science and technology, make millimeter wave sensing technology serve the people, and improve the quality of life of the masses.

summary:

Short-range millimeter-wave radar is an emerging application in the non-vehicle field, but its chip development and application process are accelerating. Professor Gu Changzhan pointed out that in the automotive field, millimeter-wave radar will help the further evolution of autonomous driving technology. In the industrial and consumer fields, simple applications such as presence perception and barrier gates have achieved initial market results. The next stage will see higher-level applications such as smart medical care, health monitoring, human-computer interaction, and gesture recognition.

With the development of millimeter-wave radar technology towards miniaturization and low cost, we have now seen a huge demand for millimeter-wave radar modules in non-vehicle application scenarios, and it is expected that it will increase in the next 3-5 years.

In addition, he also said that with the development of digitization and artificial intelligence, intelligent perception technology will inevitably attract more attention. Domestic manufacturers do not have the legacy problems of international manufacturers, and they have the opportunity to take advantage of latecomers to achieve overtaking.

Leave a Reply

Your email address will not be published.