Finite element emulation-based solver for electromagnetic computations

Electromagnetic (EM) computations are the cornerstone in the design process of several real-world applications, such as radar systems, satellites, and cell-phones. Unfortunately, these computations are mainly based on numerical techniques that require solving millions of linear equations simultaneously. Software-based solvers do not scale well as the number of equations-to-solve increases. FPGA solver implementations were used to speed up the process. However, using emulation technology is more appealing as emulators overcome the FPGA memory and area constraints. In this paper, we present a scalable design to accelerate the finite element solver of an EM simulator on a hardware emulation platform. Experimental results show that our optimized solver achieves 101.05x speed-up over the same pure software implementation on MATLAB and 35.29x over the best iterative software solver from ALGLIB C++ package in case of solving 2,002,000 equations.

[1]  Khalid H. Abed,et al.  Mapping a Jacobi Iterative Solver onto a High-Performance Heterogeneous Computer , 2013, IEEE Transactions on Parallel and Distributed Systems.

[2]  Y. El-Kurdi,et al.  Hardware Acceleration for Finite-Element Electromagnetics: Efficient Sparse Matrix Floating-Point Computations With FPGAs , 2007, IEEE Transactions on Magnetics.

[3]  Richard Barrett,et al.  Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods , 1994, Other Titles in Applied Mathematics.

[4]  Prawat Nagvajara,et al.  Sparse LU Decomposition using FPGA ⋆ , 2008 .

[5]  Viktor K. Prasanna,et al.  High-Performance and Parameterized Matrix Factorization on FPGAs , 2006, 2006 International Conference on Field Programmable Logic and Applications.

[6]  Matthew N. O. Sadiku Numerical Techniques in Electromagnetics, Second Edition , 2000 .

[7]  Youcef Saad,et al.  A Basic Tool Kit for Sparse Matrix Computations , 1990 .

[8]  Matthew N. O. Sadiku,et al.  Numerical Techniques in Electromagnetics , 2000 .

[9]  M. Sadiku Numerical Techniques in Electromagnetics , 2000 .

[10]  Aljoscha Smolic,et al.  Evaluation and FPGA Implementation of Sparse Linear Solvers for Video Processing Applications , 2013, IEEE Transactions on Circuits and Systems for Video Technology.

[11]  Dennis W. Prather,et al.  FPGA-based acceleration of the 3D finite-difference time-domain method , 2004, 12th Annual IEEE Symposium on Field-Programmable Custom Computing Machines.

[12]  Yunqing Huang,et al.  Time-Domain Finite Element Methods for Maxwell's Equations in Metamaterials , 2012 .

[13]  Anastasis C. Polycarpou,et al.  Introduction to the Finite Element Method in Electromagnetics , 2006, Introduction to the Finite Element Method in Electromagnetics.

[14]  Viktor K. Prasanna,et al.  Sparse Matrix Computations on Reconfigurable Hardware , 2007, Computer.