A novel method for high-level synthesis of datapaths in digital filters using a moth-flame optimization algorithm

被引：0
作者：
Mohammad Reza Esmaeili
Seyed Hamid Zahiri
Seyed Mohammad Razavi
机构：
[1] University of Birjand,Department of Electrical Engineering
来源：
Evolutionary Intelligence | 2020年 / 13卷
关键词：
Digital VLSI circuits; Datapath; Digital filters; High-level synthesis; Moth-flame algorithm; Optimization;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
High-level synthesis (HLS) is one of the most important processes in digital VLSI circuit design. Owing to complexity and enormity of the design space in HLS problems, employing meta-heuristic methods and swarm intelligence has been considered as a highly favorable option when solving such problems. This research work proposes a moth-flame optimization (MFO) algorithm-based method for HLS of datapaths in digital filters, where scheduling, allocating, and binding steps were performed simultaneously. It was observed that the efficiency of the proposed method enjoyed an improved efficiency thanks to the mentioned simultaneous steps while being combined with the MFO algorithm. By comparing the performance of the proposed method with Genetic algorithm based method and particle swarm optimization based method for HLS of digital filters benchmarks, it can be inferred that the proposed method outperforms the other two methods in HLS of digital filters. This is evidently approved by a maximum improvement observed in the rates of the delay, the occupied area of the chip, and the power consumption for 2.99%, 6.58%, and 6.48%, respectively. In addition to the mentioned improvement, another striking characteristic of the proposed method is its fast runtime in reaching a response. This could significantly lower the costs while increasing the design speed of circuits having large dimensions. As well, an averagely 20% rise was also discerned in the algorithm runtime compared to the other two methods.
引用
收藏
页码：399 / 414
页数：15
相关论文
共 55 条
[1]  
Kim NS(2017)heterogeneous computing meets near-memory acceleration and high-level synthesis in the post-moore era IEEE Micro 37 10-18
[2]  
Xiong J(2018)Securing hardware accelerators: a new challenge for high-level synthesis IEEE Embed Syst Lett 10 77-80
[3]  
Hwu WW(2017)TL-HLS: methodology for low cost hardware trojan security aware scheduling with optimal loop unrolling factor during high level synthesis IEEE Trans Comput Aided Des Integr Circuits Syst 36 655-668
[4]  
Pilato C(2019)VeriIntel2C: abstracting RTL to C maximize high-level synthesis design space exploration Integration 64 1-12
[5]  
Garg S(2018)VLSI-based pipeline architecture for reversible image watermarking by difference expansion with high-level synthesis approach Circuits Syst Signal Process 37 1575-1593
[6]  
Wu K(2018)Enabling automated bug detection for IP-based design using high-level synthesis IEEE Des Test 35 54-62
[7]  
Karri R(2005)Unification of scheduling, binding, and retiming to reduce power consumption under timings and resources constraints IEEE Trans VLSI Syst 13 1113-1126
[8]  
Regazzoni F(2003)A game theoretic approach for power optimization during behavioral synthesis IEEE Trans Very Large Scale Integr (VLSI) Syst 11 1031-1043
[9]  
Sengupta A(1989)Algorithms for hardware allocation in data path synthesis IEEE Trans Comput Aided Des Integr Circuits Syst 8 768-781
[10]  
Bhadauria S(1993)SALSA: a new approach to scheduling with timing constraints IEEE Trans Comput Aided Des Integr Circuits Syst 12 1107-1122
123456