摘要
针对ASIC芯片设计中时钟树综合效率和时序收敛的问题,提出了一种高效的时钟树综合方法,特别适用于现代先进深亚微米工艺中的高集成度、高复杂度的设计中。改进了传统时钟树综合方法,通过采用由下至上逐级分步综合的方法实现。该设计方法在SMIC 0.18μm eflash工艺下的一款电力线载波通信芯片中成功流片验证,结果表明分步综合能够在实现传统设计功能的前提下,在完成时序收敛时有效减少不必要的器件插入,从而减小芯片面积,降低整体功耗,有效改善绕线拥塞度。
A new method was proposed to solve the problem of clock tree synthesis (CTS) efficiency and timing closure. This method was suitable for modern highly sophisticated and integrated deep sub-micron VLSI design. The traditional CTS was optimized and it was realized by a bottom-up step- by-step synthesis approach. A power-line-communication chip design was used to verify the proposed method in a 0. 18 μm eflash CMOS technology. The result shows that based on this method, the functionality was successfully realized, excess buffers were avoided during timing closure. As a result, the chip area and power consumption can be reduced and the congestion improved.
出处
《半导体技术》
CAS
CSCD
北大核心
2012年第3期169-171,179,共4页
Semiconductor Technology
基金
北京工业大学博士科研启动基金(X0002019201101
X0002019201102)