Multiple-Object Optimization of Design in Error-Compensation Circuit of Fixed-Width Booth Multiplier

In this proposal, fine-tuning the accuracy of fixed-width two's-complement modified Booth multiplier using multiple-object intelligent evolution algorithm is proposed. Based on the trade-off of truncation error, circuit area, speed, and power consumption, we choose the optimal solution for specific application. This makes fine-tuning of accuracy instead of column-wise coarse-tuning in fixed-width multipliers possible. We propose two methods based on the BSB to estimate and compensate the fixed-width multiplier for minimization of truncation error. Two primary methods can improve the signal to noise (SNR) with a small area penalty in fixed-width multiplier. Statistics on the SNR were analyzed, and the hardware performance results validate the effectiveness of the proposed methods. In real-world implementation, compared with the discrete cosine transform (DCT) core using direct-truncated (D-T) multipliers, the DCT cores using the proposed BSB with Auxiliary Element (BSB AE) and BSB Hierarchy Control (BSB HC) multipliers increase average peak signal-to-noise ratio (PSNR) by 27% and 31% with only a 6% and 7% area penalty, respectively. In this proposal, based on these primary results, we are developing an efficient multiple-object optimization method for the minimization of MAE and the hardware performance will be evaluated, such area, accuracy, and power.

Project ID:PB10708-1953
External Project ID:MOST107-2221-E182-062