On The Move


A Processor with Vision

I just came back from the MIPI alliance face-to-face meeting last week that was engaging and interesting.
MIPI D-PHY v2.0 key contributors

At the beginning of the week, as part of the plenary session the MIPI alliance recognized several specifications with the key contributors.
Here’s a picture of main contributors for D-PHY v2.0 spec including Intel Henrik and Synopsys Raj Kumar.

Paul VishnyThe MIPI alliance also presented Lifetime Achievement award to Paul Vishny, the Legal council who helped shape the MIPI Alliance from the first inception days. I had the opportunity and honor to work with Paul on multiple engagements helping non-members be comfortable with the MIPI alliance policies and become members and adopters of MIPI IP.

There were a lot of discussions around the applications that can be powered (better said low powered) by MIPI specifications, around 5G, Sensors, and continuous adoption of display and camera technologies in both mobile and beyond mobile applications such as automotive, IoT and wearables.

MIPI strategic focus areas March16 (source: MIPI alliance twitter)
MIPI strategic focus areas March16 (source: MIPI alliance twitter)

While the traditional mobile market is reaching maturity and limited growth is expected in the coming years, there are promising applications that continue to use the strengths of MIPI based specifications which are used in high volume mobile electronics driving lower cost and lower power to service the needs of adjacent markets.

One application that I find very interesting is Machine learning, it combines many hardware and software capabilities and bring a new kind of application to the mainstream. Bringing intelligence to machines requires capabilities in multiple levels, ultra low power, performance-optimized and cost efficient interfaces that can deliver the required information to the processor plus the ability to process the information quickly and derive the next actions.
One example of a of a Vision Processor Unit (from Movidius) that is used for machine learning applications is described in this cool video:

So what’s next?
The industry supporting MIPI interfaces should continue to invest in improving the low power and efficiency of the interfaces used for applications beyond mobile. The success of MIPI interfaces in mobile helps substantiate a wide and robust eco-system, which in my mind requires few refinements to address application specific needs such as higher-reliability automotive applications. In the next couple of years we will see MIPI interfaces used in beyond mobile electronics such as Virtual reality, Machine learning, Wearables and Automotive ADAS.

If you like to read this or other previous posts, send this URL to your friends and tell them to Subscribe to this Blog.
One option to subscribe is as follows:
• Go into Outlook
• Right click on “RSS Feeds”
• Click on “Add a new RSS Feed”
• Paste in the following “http://feeds.feedburner.com/synopsysoc/sqLy?format=xml”
• Click on “Accept” or “Yes” or whatever the dialogue box says.

  • del.icio.us
  • Digg
  • Facebook
  • Google Bookmarks
  • Print
  • Twitter
  • StumbleUpon