When deciding on an embedded vision system, there are various factors to think about during its construction. One such factor is ensuring the right camera interface is selected to complete your desired applications.
In the past, one of the standard-bearers for camera interfaces was MIPI. Yet, as new age applications continue to evolve and set new demands, MIPI interfaces have struggled to keep up with the pace. Instead, new camera interfaces are required to step up to the plate for modern embedded vision applications.
That’s where GMSL enters the picture.
To demonstrate why GMSL should be chosen over MIPI for camera interfaces, here are some of the key benefits of using this technology with your next embedded vision project:
Distance of Transmission
Keeping latency down while maintaining a high frame rate is necessary when using a camera interface. Yet when distance is introduced between the camera module and the computing unit, it can jeopardize that goal. That’s the case when not using GMSL.
GMSL technology is designed for long-distance transmissions of up to 15 metres. That means the host processor can be 15 metres away from the camera module without any dips in performance. This distance of transmission cannot be matched to the same extent by MIPI.
Automatic Repeat Request
Automatic Repeat Request (ARQ) is used by GMSL as a method for error control and packet recovery to supply reliable data transmission. With this method in place, it is possible for the side transmitting to re-transmit packets automatically, even when the packet from the receiver isn’t acknowledged. The result is that it helps applications remain error-free, providing the system with robust critical control functions.
EMI Performance
Cameras face a serious challenge in the form of electromagnetic interference (EMI). This is because harmful interference can be detrimental to the connector, cable, PCB combination, and relevant electronic devices. For performance to not be negatively impacted, it is essential they are protected from EMI. This is why GMSL incorporates EMI testing to prevent harmful interference. The GMSL serialiser and deserialiser ICs both have this programmable output built in, which improves safety-based camera applications.
Compatible Platforms
When creating bespoke systems and applications, computer engineers like things-embedded.com love working with GMSL cameras due to the number of compatible platforms available. Off-the-shelf support is available with the likes of Connect Tech’s Rogue and Rudi platforms, as well as the NVIDIA Jetson development kit. Because of this, rapid prototyping and swift deployment are possible with vision-enabled products.
Video Duplication
The GMSL Splitter mode can be of benefit when it comes to display applications. When in this mode, each GMSL PHY can receive the masked output for each video pipe. This process is known as video duplication, which can be beneficial for a range of applications and their performance levels.
Conclusion
Due to the added power and features offered, it is easy to see why GMSL is now viewed as the top option for camera interfaces rather than MIPI. As embedded vision advances, it needs technology that can keep up – and that’s exactly what GMSL does.
