The Forum was hosted by Taiwan Semiconductor Research Institute ("TSRI"), Interuniversity Microelectonics Centre ("imec") and Europractice. One of the main themes throughout the Forum was the need for us all to collaborate, especially between the semiconductor industry and academia, and something we activity promote here at SoC Labs. The presentations and attendees from both communities was very impressive and hopefully will lead to a lot of innovative collaborations.

Monday 16th September at 2024 IEEE 37th International System-on-Chip Conference saw the announcement of a new design contest for creation of an academic Chiplet based disaggregated SOC using the ARM ecosystem. 

Rewards/ Prizes: 

SoC Labs will arrange for the winning design:

funding toward die fabrication costs for custom chiplets

fabrication of a custom interposer/package

design support during the year

Tomorrow sees the start of the 37th IEEE International System-on-Chip Conference (SOCC) in Dresden. Last night saw the legendary Kraftwerk play live in the city. I am not sure if the organisers of the conference arranged that but they have certainly been very helpful in supporting a a Special Session for SoC Labs on Thursday. 

In association with CMC Microsystems we are holding an on-line workshop. The workshop will outline how the broad range of IP made freely by Arm Academic Access combined with academic-focused reusable reference designs and tape out flows (being developed by the SoC Labs academic community) are making System on Chip projects easier to undertake by individual students and larger academic teams. 

Five projects from SoC Labs presented their work in a Special Session at the 36th IEEE International System-on-Chip Conference (SOCC) in Santa Clara in the heart of Silicon Valley. It was nice to see teams from across the world come together and share not only their own work but collaborate on shared interest in Arm based SoC designs. Both the Hardware track, with a focus on novel hardware design solutions, and the Education track, where development of skills and experience in SoC design and collaboration where the more important aspect, were represented.

SoC Labs with support from Arm are pleased to announce a SoC Design Contest: Bridging the Skills Gap along with a new NanoSoC reference design and example competition design flow to help engage.

Background:

A k-Nearest-Neighbour (kNN) classifier is a non-parametric type of classifier, which predicts the class of unseen examples based on their Euclidean distance from known examples stored in non-volatile memory. Because of its non-parametric nature, its training phase consists of just a mere storage operation. In contrast to other classifiers, there is no need for a computationally expensive parameter update process.