MAX: Synopsys' two major propositions focus on intelligent technology, and the 2024 R&D Competition is waiting for you to join!

”

With the vigorous development of intelligent technology, the 19th China Graduate Electronic Design Competition (hereinafter referred to as "R&D Competition") has been grandly opened in March 2024. As a leader in technological innovation, Synopsys has been working with Yandian Competition for 28 consecutive years to contribute to China's technological innovation and talent development.

In the era of Pervasive Intelligence, thanks to the rapid development of smart home, smart city, smart healthcare, autonomous driving, intelligent control and other fields, each of our lifestyles is being changed by intelligent technology.

This year's Synopsys competition will focus on smart applications and focus on radar non-contact sensing. This technology is currently a hot technology in medical monitoring, intelligent transportation and other fields. Through this competition, Synopsys hopes to stimulate students' enthusiasm and innovative thinking about cutting-edge technologies, and work with us to promote the application and advancement of intelligent technologies in a wider range of fields in the future.

In addition, Synopsys has set up a "Green Channel for the National Competition", where the top 10% of outstanding teams will advance to the national finals and participate in the final awards. We warmly welcome every student who is passionate about science and technology and bravely pursues challenges to join this competition of scientific and technological innovation.

Awards

First prize team (1 team): 10,000 yuan

Second prize team (2 teams): 5,000 yuan

Third prize team (3 teams): 3,000 yuan

Additional Benefits:

*Internship opportunities: Participants will be given priority to receive internship opportunities at Synopsys and related partner companies, and send their resumes to [email protected].

*Invited to participate: Outstanding winners will be invited to participate in the Synopsys ARC Processor Summit, and the final plan will be subject to the official announcement of the company.

Timeline

Official start: March 2024

Proposition Expert

Cheng Songbo

Senior R&D Manager, ARC Processors, Synopsys

He has more than 15 years of experience in processor IP and processor chip development. He is currently working at Synopsys and is responsible for the development and prototyping of Synopsys' ARC processor family of IP (ARC Instruction Set Architecture and RSIC-V Instruction Set Architecture) for high-performance application processors, AI computing processors, real-time processors, and ultra-low-power embedded processors.

Tournament propositions

01 AIoT electronic system design based on ARC processor

Teams are required to use an ARC processor development board provided by Synopsys to collect data from a variety of sensors (e.g., microphones, cameras, motion sensors, etc.). Leverage the unique hardware and software resources of the ARC processor to accelerate data processing or motion control at the edge to solve specific real-life problems, such as:

■ Human-computer interaction: noise reduction, voice recognition, vocal recognition, facial behavior recognition, etc.

■ Personal health: exercise detection, situational recognition, early disease prediction, health monitoring, etc.

■ Industrial Internet: multi-sensor data fusion, behavior prediction, acoustic fault detection, etc.

The fields involved in the participating systems include but are not limited to smart home, smart city, wearable devices, intelligent driving, intelligent control, industrial Internet and other fields.

Interpretation of the competition questions

This challenge requires participants to design a solution based on Synopsys' ARC processor IoTDK or AIoTDK development board. Data is captured through on-board sensors such as microphones, cameras, motion sensors, etc., and efficiently processed and responded to with the help of the software and hardware capabilities of the ARC processor. Furthermore, machine learning (ML) technology is used to achieve high-accuracy detection and recognition functions to solve real-world problems, such as human-computer interaction and motion detection.

Output requirements

■ PPT introduction of system design scheme

■ Functional demonstration video to show the actual effect of the solution

■ Technical documents, detailing the scheme design and algorithm implementation

■ Complete project source code, including comments, and ensure that it can be compiled and run

■ The source code needs to be uploaded to Github for open source sharing

Judging Criteria

■ Creativity and innovation: Evaluate the creativity of the work and the breakthrough of the design idea, as well as the advancement of the technical methods adopted

■ ARC Processor Application: Investigate how the team can make full use of the software and hardware characteristics of ARC processor to solve key problems, especially focusing on the in-depth utilization and performance optimization of processor features (such as DSP/MLI), and the effective integration of machine learning methods

■ Presentation and documentation: PPT should have a clear theme and clear logic, the video should fully show the function implementation, the technical document should have a clear structure and sufficient argumentation, and the code should be annotated in detail and can be compiled smoothly.

02 Multi-person vital sign detection based on 60G millimeter-wave radar

This competition focuses on the innovative technology in the field of smart home - 60G millimeter-wave radar. In the field of smart home, short-range radar systems based on millimeter-wave technology are becoming a new research hotspot. Compared with traditional 24G or 5.8G centimeter-wave radar, vision and infrared technology, 60G millimeter-wave radar can more accurately capture human behavior, such as movement trajectory, posture, heart rate and breathing, while protecting user privacy, providing richer data support for health monitoring. This competition requires participants to use a 60G millimeter-wave radar system based on ARC processor to achieve non-contact monitoring of the vital signs of multiple people (more than two people).

Interpretation of the competition questions

Participants are encouraged to combine 60G millimeter-wave radar technology with ARC processors to enable non-contact detection of multiple users' vital signs, such as respiratory rate and heart rate, in an indoor environment.

Output requirements

■ Use the Calterah radar development platform (RDP-60S244-IB-A-AIP) to detect breathing and heartbeat, and the detection results should be output to the PC in real time through the serial port and displayed (using the serial port debugging tool or GUI software)

■The device should be able to detect the breathing and heartbeat of more than 2 people at the same time

■The submitted materials should include source code, design PPT, technical documents, test videos and test reports

Judging Criteria

■Accuracy and stability: In the case of radar facing the radar, 100 times of detection of multiple people or data output for 3 minutes (whichever is reached first) are carried out, and the mean and variance of the measurement results are calculated. Priority will be given to entries with a mean value close to the true value and a small variance

■ Bonus points:

(1) Able to detect the breathing and heartbeat of 2 people at the same distance but in different directions

(2) Output breathing and heartbeat waveforms at the same time, and display them in real time through the host computer

(3) Automatically identify the presence or absence of a human body, and automatically start or stop the detection accordingly

(4) Support multiple orientations of the tester, including facing, sideways, and back

■ Presentation and documentation: The content of PPT should highlight the theme and have clear logic, the video should fully demonstrate the functional implementation, the technical document should have a clear structure and detailed arguments, and the code comments should be complete and can be compiled smoothly