My research interests are primarily in the field of VLSI-CAD. They include FPGA physical design, reconfigurable computing, and ASIC floorplanning/placement. I do some FPGA designs as well. In the past couple of years, I have focused on Stochastic Computing, which uses an analog-like encoding scheme: numbers are represented by streams of 1’s and 0’s. The ratio of 1’s to the length of the stream shows the value as a “probability”. Simple logic can perform complex operations on stochastic streams (e.g., an AND gate can multiply two probabilities).
Advances in the FPGA technology have made them increasingly more powerful in the past decade, and their market share is increasing every year. The main reason for their popularity is their flexibility in changing designs without having to go through more fabrication cycles. However, the performance, power, and area efficiency of FPGAs lags behind their ASIC counterparts. My objective is to bridge the gap between ASICs and FPGAs by devising better architectures, CAD algorithms, and programming paradigms.
Sayed Abdolrasoul Faraji, M. Hassan Najafi, Bingzhe Li, Kia Bazargan and David Lilja, "Energy-Efficient Convolutional Neural Networks with Deterministic Bit-Stream", in Hardware Architectures for Deep Learning, Masoud Daneshtalab and Mehdi Modarressi, Institution of Engineering and Technology (IET), 2019.
Weikang Qian, Marc D. Riedel, Kia Bazargan, and David J. Lilja, "Synthesizing combinational logic to generate probabilities: theories and algorithms," in Advanced Techniques in Logic Synthesis, Optimizations and Applications, Sunil Khatri and Kanupriya Gulati editors, Springer Publishing, 2011.
Kia Bazargan, "Chapter 10.2: FPGA Technology Mapping, Placement, and Routing", in The Handbook of Algorithms for VLSI Physical Design Automation, Charles J. Alpert, Dinesh P. Mehta, and Sachin S. Sapatnekar, CRC Press
Sachin Sapatnekar, Kia Bazargan, "Chapter 10.4: 3D Design", in The Handbook of Algorithms for VLSI Physical Design Automation, Charles J. Alpert, Dinesh P. Mehta, and Sachin S. Sapatnekar, CRC Press.
Soheil Mohajer, Zhiheng Wang, Kia Bazargan and Yuyang Li, “Parallel Unary Computing Based on Function Derivatives”, ACM Transactions on Reconfigurable Technology and Systems (TRETS), 2020.
Zhiheng Wang, Devan Larson, Morgen Barker, Soheil Mohajer, and Kia Bazargan, "Deterministic Shuffling Networks to Implement Stochastic Circuits in Parallel", in IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2020, doi: 10.1109/TVLSI.2020.2984731
M. H. Najafi, Shiva Jamali-Zavareh, D. J. Lilja, M. D. Riedel, K. Bazargan, and R. Harjani, "An Overview of Time-based Computing with Stochastic Constructs," in IEEE Micro, 2017.
M. H. Najafi, D. J. Lilja, M. D. Riedel and K. Bazargan, "Polysynchronous Clocking: Exploiting the Skew Tolerance of Stochastic Circuits," in IEEE Transactions on Computers, vol. 66, no. 10, pp. 1734-1746, Oct. 1 2017.[Selected as IEEE Transaction on Computers' Feature Paper of the Month]
M. Hassan Najafi, P. Li, D. J. Lilja, W. Qian, K. Bazargan, M. Riedel, "A Reconfigurable Architecture with Sequential Logic-based Stochastic Computing," ACM Journal on Emerging Technologies in Computing Systems, Vol. 3, No. 4, 2017.
M. Hassan Najafi, S Jamali-Zavareh, D. J. Lilja, M. Riedel, K. Bazargan, and R. Harjani, "Time-Encoded Values for Highly Efficient Stochastic Circuits," IEEE Transaction on Very Large Scale Integration Systems, Vol. 25, No 5, 2017.