Nowadays Interest

I'm currently investigating on Algebraic Pattern Recognition for FPGA Mappers. So far I have been able to extract and reconstruct the algebraic operations from the FPGA netlist; I'm also recognizing some DSP aplications suchs as FIR filters (althought it is not yet 100% reliable). At this moment I'm looking for New Data Structures that can help me achieve this. It is a very hard problem, and probably unsolvable, but hey I'm a PhD right? - Nothing to lose but my own time. lol.

Topics of interest

Algebraic Pattern Recognition for DSPs

High Level Synthesis

Functional Verification

Security and Encryption

Routing on FPGAs

New CMOS technologies

Simulation and Layout of VLSI designs

Taylor Expansion Diagrams (TED)

The TED is a canonical data structure capable of representating multivariate polynomials subject to the imposition of a total ordering on the variables. The size of the TED graph depends on the order of its variables; therefore the complexity of any manipulation subject to the size of the TED can be improved reordering the variables to find the best order that minimize the size of the TED.  

TEDs can be used in conjuction with GAUT Synthesis Tool to extract the Data Flow Graph and perform high level optimizations on it.

I'm the main developper of the TED data structure, and you can download the program from the TEDify Webpage which I also mantain. If you expirience any problem or find any bug, I'm the one to blame. So, please let me know and I will take care of it.