← 返回 JSSC 论文列表JSSC 2006第9期Memory0.25μm/0.5μm
Adaptive Algorithm Using Hot-Electron Injection for Programming Analog Computati
提出一种利用热电子注入编程模拟计算存储单元的自适应算法,精度达0.2%。
1024x16和96x16浮栅阵列,0.25μm和0.5μm n-well CMOS工艺
热电子注入Fowler-Nordheim隧穿自适应算法模拟计算浮栅阵列
▸创新点1:热电子注入精确编程(方法创新)。该方法利用热电子注入技术实现高精度编程,能够在3.5个数量级的电流范围内达到0.2%的编程精度,显著提升了编程的准确性。
▸创新点2:Fowler-Nordheim隧穿全局擦除(方法创新)。采用Fowler-Nordheim隧穿技术实现全局擦除,确保了大规模阵列的高效擦除操作,提升了系统的整体性能。
▸创新点3:自适应预测算法(系统创新)。该算法通过自适应调整编程脉冲数量,平均仅需7-8个脉冲即可完成编程,显著减少了编程时间和能耗,提高了系统的效率。
▸创新点4:大规模阵列验证(系统创新)。该算法在1024x16和96x16浮栅阵列上进行了验证,分别在0.25μm和0.5μm n-well CMOS工艺中表现出优异的编程性能,证明了其在实际应用中的可行性。
Abstract
This paper describes a new predictive algorithm that
can be used for programming large arrays of analog computational
memory elements within 0.2% of accuracy for 3.5 decades of cur-
rents. The average number of pulses required are 7–8 (20
s each).
This algorithm uses hot–electron injection for accurate program-
ming and Fowler–Nordheim tunneling for global erase. This al-
gorithm has been tested for programming 1024
16 and 96
16
floating-gate arrays in 0.25
m and 0.5
m n-well CMOS pro-
cesses,