14 November 2022 in IEEE Transactions on Vehicular Technology
For the direction of arrival (DOA) in array signal processing, eigenvalue decomposition (EVD) is one key issue in hardware implementation of the multiple signal classification (MUSIC) algorithm. Therefore, we introduce the look-ahead sim- plified one-sided Jacobi's method to efficiently decompose those symmetric matrices in this article and prove that the new method has the best orthogonality of eigenvector and locates eigenvectors closest to the true solution in theory. Both the numerical perform- ance and real-time are important in engineering, so we present the novel flexible hardware architecture in single floating point arithmetic for EVD on field-programmable gate arrays (FPGAs). Finally, the simulated and raw data are used to investigate the performance of some different approaches in the context of both the EVD and MUSIC algorithm. The experimental results show that our proposed method has the best performance.
11 February 2021 in IEEE Transactions on Circuits and Systems II: Express Briefs
Partial sorting, selecting M largest/smallest numbers from N inputs, is of interest in many applications. Thus, this brief presents a novel scheme for real-time max/min-set-selection sorters on field-programmable gate arrays (FPGAs). Our scheme has good expansibility but no close relation with the vector length, whose basic idea is reducing the programming complexity through the essence of sorting networks for Bitonic sequences. In addition, two modified forms of our scheme are proposed to solve the high data rate sequence and a multiple of 2 max/min-set-selection problem respectively. Finally, the effectiveness of our scheme is demonstrated on one Xilinx XC7VX690T FPGA by performing comparisons with state-of-the-arts.
27 September 2022 in Circuits, Systems, and Signal Processing
28 October 2020 in IEEE Transactions on Circuits and Systems II: Express Briefs
Due to the robustness, the matrix inversion methods based on matrix factorization are often adopted in engineering while the triangular matrix inversion is one key part of those methods. This brief reviews the adjoint matrix and presents a novel scheme for field-programmable gate arrays (FPGAs) calculating the triangular matrix inversion. Employing the more characteristics of both the triangular matrix and its inversion, the proposed diagonal-wise algorithm for the triangular matrix inver-sion has the high parallelism and extensibility in the hardware implementation and is suitable for the different matrix triangular factorization (QR, LDL, Cholesky and LU). Meanwhile, the recursive diagonal-wise algorithm is designed for the large scale triangular matrices. Compared with the traditional row-/column-wise methods, our algorithm has a good performance at the low computation load. Finally, the experiments are conducted on one Xilinx Virtex-7 FPGA to illustrate the performances of the four different methods.