“The rapid explosion of artificial intelligence is mainly due to the rapid development in the fields of machine vision and image recognition. As people have increasingly stringent requirements for bandwidth, latency, and power consumption, edge AI and video capture and analysis technologies have begun to emerge.
“
The rapid explosion of artificial intelligence is mainly due to the rapid development in the fields of machine vision and image recognition. As people have increasingly stringent requirements for bandwidth, latency, and power consumption, edge AI and video capture and analysis technologies have begun to emerge.
In order to ensure the true accuracy of data, CMOS image sensors are becoming more and more high-definition, and the development of edge processor technology has also made it possible to process these data quickly, helping to improve the accuracy and flexibility of real-time analysis, fast anomaly detection, etc. .
Machine vision technology is being widely used in applications including robot-assisted surgery, security and surveillance platforms, industrial robotics, and more.
With the miniaturization, light weight and high speed, the data transmission method from the sensor to the processing system is changing.
Many existing high-speed interface technologies can help enable reliable transmission of high-resolution video data, but vision-based control systems also have shortcomings. For example, standard technologies such as Ethernet introduce additional latency due to protocol-related overhead. Since Ethernet PHYs cannot connect directly to the sensor’s native video interface, they also require additional wires and components such as crystal oscillators. For example, a 4-MP, 30-fps high-resolution imager can generate approximately 3.2 Gbps of video data. A single gigabit-based Ethernet link (such as 1000-BaseT) cannot provide enough throughput to carry this uncompressed high-resolution data, which introduces artifacts in the image stream that may Introduce errors in visual video processing.
In addition to data, in applications such as robot-assisted surgery, command transmission control is also required to make the front-end sensors act accordingly.
V3Link Serializer/Deserializer
The use of dedicated serializer/deserializer (SerDes) technology is a means of realizing high-speed data transmission. TI’s latest V3Link TSER953 serializer and TDES954 and TDES960 deserializers work together to transmit high-resolution video, control signals and power simultaneously over a single ultra-thin coaxial cable. These devices link sensors and processors to aggregate clock, uncompressed video, control, power, and general-purpose input/output signals over distances of up to 15 meters, as shown.
Block diagram showing data and power transfer schematics using V3 Link devices
The sensor signal can be transmitted from the serializer to the deserializer in one direction, and the power transmission is transmitted from the SoC side to the sensor. Others, including control, clock, GPIO, etc., all support bidirectional paths.
Embedded clocks enable video synchronization across multiple sensors, which allows for video stitching, image mixing, and stereo vision for 3D reconstruction and depth sensing, among others. Using the internally generated frame sync signal from the TDES960, multiple cameras can be synchronized with an accuracy of 600 ns, which enables multiple time-triggered applications in machine vision. The master clock synchronization of the extracted reference clock eliminates synchronization errors caused by the relative drift of the different oscillators clocking multiple imagers.
Adaptive equalizer technology reduces signal loss and power consumption
Additionally, to facilitate transmission of video data, control signals, and power over a single cable, V3Link devices include adaptive equalizer technology that compensates for up to 21 dB of loss at 2.1 GHz, enabling the use of very fine 28 to 32 AWG ( AWG) cable. The higher the AWG number, the thinner the cable and the higher the signal loss.
Thinner cables are more flexible and can support applications where sensors need to be positioned in tight, space-constrained applications such as endoscopes. The ability to transmit power and control signals on the same thin cable also minimizes the number of conductors.
At a typical power consumption of 250 mW on the sensor side, the V3Link serializer is extremely low power, enabling the integration of the sensor and serializer into a very compact area without additional cooling facilities. V3Link products’ proprietary master clock synchronization technology eliminates the need for a crystal or any oscillator on the sensor side, further reducing cost and overall space requirements.
in conclusion
From medical imaging applications to machine vision cameras, edge AI is driving the need for real-time video capture, transmission, and analysis. V3Link SerDes ICs help engineers meet these needs while reducing cable count, power consumption and overall system cost. V3Link devices provide universal link technology ideal for most applications requiring real-time capture, transmission and analysis of high-resolution video data. These devices support various cabling configurations such as coaxial, unshielded twisted pair, and shielded twisted pair, as well as multiple clocking modes such as synchronous and asynchronous.
TI has a long-standing investment in high-speed serializers/deserializers, with its FPD-Link™ serializers and deserializers for various video interfaces in automotive systems, including cameras for advanced driver assistance systems (ADAS). and infotainment system displays) optimize the design and transmission of high-resolution signals. Transport high-resolution, uncompressed data in automotive systems over fewer wires. Today, with the increasing demand for high-resolution original high-definition video data in industrial, medical and other fields, the V3Link high-speed serializer/deserializer can be used in medical endoscopes, various robots, video surveillance, and home appliances. More flexible and compact design.
The Links: MG75Q2YS40 NL8060BC31-41D