IC Design

design and verification

September, 2018

now browsing by month

 

[BOOK] High-Performance Computing Using FPGAs

155

http://www.mediafire.com/file/u9cj4680vole3xi/%5BHigh+Performance+Computing+Using+FPGAs.pdf

Free RTL and Verilog Testbench of SPI Wishbone Controller

24

http://www.mediafire.com/file/71ste4a9v5skzlb/spi_wb_master_lattice.rar

Advanced UVM Courses

17

http://www.mediafire.com/file/om4iiki8lh9k20z/UVM_Advanced_Course.rar

AXI4-based CAN controller

Features :

-Conforms to the CAN 2.0A, and CAN2.0B standards.

-Support both standard (11-bit identifier) and extended (29-bit identifier) frames.

-Up to 1 Mbps.

-Transmit message FIFO supports up to 64 messages.

-Receive message FIFO supports up to 64 messages.

-Support up to 4 ID filters.

-Sleep Mode with automatic wake-up.

-Loop Back Mode for diagnostic applications.

-Interrupts, status, error counters.

-Support 32 bits APB interface.

[Course] ELECOS 475 Computer Architecture. Princeton University 2016

http://www.mediafire.com/file/9cpi32mk5913d6i/ELECOS_475_Computer_Architecture._Princeton_University_2016.rar

 

[Course] 18-600 Foundations of Computer Systems CMU Univesity 2016

http://www.mediafire.com/file/66ivv422tta9otl/18-600_Foundations_of_Computer_Systems_CMU_2016.rar

 

SystemVerilog UVM 1.1 Workshop Synopsis

http://www.mediafire.com/file/63nvfumayfggm78/Synopsis_UVM_1.1_Training.rar/file

 

[Slide] ISSCC 2017 Deep learning processor

4.1 A 2.9 TOPS/W Deep Convolutional Neural Network SoC in FD-SOI 28nm for Intelligent Embedded Systems

14

In Paper 14.1, STMicroelectronics presents a deep convolutional neural network SoC in 28nm FD-SOI with energy efficiency of 2.9TOPS/W and peak performance of more than 676GOPS, operating at 200MHz with supply voltage of 0.575V.

14.2 DNPU: An 8.1TOPS/W Reconfigurable CNN-RNN Processor for General-Purpose Deep Neural Networks

19

In Paper 14.2, KAIST presents a reconfigurable CNN-RNN processor SoC in 65nm CMOS with energy efficiency of 8.1TOPS/W, operating at 50MHz with supply voltage of 0.77V.

14.3 A 28nm SoC with a 1.2GHz 568nJ/Prediction Sparse Deep-Neural-Network Engine with >0.1 Timing Error Rate Tolerance for IoT Applications

12

In Paper 14.3, Harvard University presents a fully connected (FC)-DNN accelerator SoC in 28nm CMOS, which achieves 98.5% accuracy for MNIST inference with 0.36μJ/prediction at 667MHz and 0.57μJ/pred at 1.2GHz.

 

14.4 A Scalable Speech Recognizer with Deep-Neural-Network Acoustic Models and Voice-Activated Power Gating

19

In Paper 14.4, MIT presents an IC designed in a 65nm LP process for DNN-based automatic speech recognition (ASR) and voice-activity detection (VAD). Real-time ASR capability scales from 11 words (172μW) to 145k words (7.78mW) and the noise-robust VAD has power consumption of 22.3μW.

14.5 ENVISION: A 0.26-to-10TOPS/W Subword-Parallel Dynamic-Voltage-Accuracy-Frequency-Scalable Convolutional Neural Network Processor in 28nm FDSOI

21

In Paper 14.5, KU Leuven presents an energy-scalable CNN processor in 28nm FDSOI achieving efficiencies up to 10TOPS/W by modulating computational accuracy, voltage and frequency, while maintaining recognition rate and throughput.

14.6 A 0.62mW Ultra-Low-Power Convolutional-Neural-Network Face-Recognition Processor and a CIS Integrated with Always-On Haar-Like Face Detector

19

In Paper 14.6, KAIST presents an ultra-low-power CNN-based face recognition (FR) processor and a CMOS image sensor integrated with an always-on Haar-like face detector in 65nm CMOS. The analog-digital hybrid Haar-like face detector improves the energy efficiency of face detection by 39% and the FR system dissipates 0.62mW at 1fps.

 

14.7 A 288μW Programmable Deep-Learning Processor with 270KB On-Chip Weight Storage Using Non-Uniform Memory Hierarchy for Mobile Intelligence

19

In Paper 14.7, the University of Michigan presents a programmable fully connected (FC)-DNN accelerator in 40nm CMOS. It achieves 374GOPS/W at 0.65V (288μW) and 3.9MHz, with configurable data precision, strategic data assignment in NUMA memory (270KB) and dynamic drowsy memory operation.

14.8 A 135mW Fully Integrated Data Processor for Next-Generation Sequencing

19

In Paper 14.8, National Taiwan University presents a data processor for Next-Generation Sequencing (NGS) in 40nm CMOS, which realizes DNA mapping, including suffix-array (SA) sorting and backward searching. With 135mW at 200MHz, it achieves significantly higher energy efficiency over CPU/GPU-based implementations.

http://www.mediafire.com/file/sx6rarstr3naxgt/ISSCC2017-14_Visuals_Deep-Learning_Processors.pdf/file

Installing QuestaSim SE 10.2C Win64

1.Download link :

http://www.mediafire.com/file/38o33r07r87800p/QuestaSim10.2c.rar

2.Run questasim-win64-10.2c.exe

Read More

[Book] Computer Architecture_ A Quantitative Approach. 5th -Elsevier (2011)

http://www.mediafire.com/file/41ny38x72tq76ph/%255BComputer_Architecture__A_Quantitative_Approach._5-Elsevier_%25282011%2529.pdf/file