A 16Gb 9.5Gb/S/pin LPDDR5X SDRAM With Low-Power Schemes Exploiting Dynamic Voltage-Frequency Scaling and Offset-Calibrated Readout Sense Amplifiers in a Fourth Generation 10nm DRAM Process.

International Solid-State Circuits Conference(2022)

引用 7|浏览31
暂无评分
摘要
Mobile systems for 5G communications and emerging technologies, such as advanced driver assistance system (ADAS), augmented reality (AR), and artificial intelligence (AI), demand high-density, high-speed, and low-power DRAM.LPDDR5 has played an important role in such systems, by offering continuous improvements in memory density (8, 12, and 16Gb) and speed (up to 8.5Gb/s) [1, 2, 3].LPDDR5X has new high-speed enabling features: per-pin DFE training, pre-emphasis for DQ driver, receiver offset calibration training, and a read duty-cycle adjuster. By employing these speed enhancement techniques, we have successfully implemented a 9.5Gb/s/pin 16Gb LPDDR5X using a fourth generation 10nm DRAM fabrication technology. This paper also presents a power-efficient LPDDR5X using (1) a non-stacking bank architecture, (2) an offset-calibrated sense amplifier for the global data lines, (3) an extended dynamic voltage and frequency scaling core (DVFSC), (4) a stacked output driver, and (5) a low-power data aligner.
更多
查看译文
关键词
speed enhancement techniques,fourth generation 10nm DRAM fabrication technology,power-efficient LPDDR5X,sense amplifier,extended dynamic voltage,frequency scaling core,low-power data aligner,low-power schemes exploiting dynamic voltage-frequency scaling,offset-calibrated readout sense amplifiers,fourth generation 10nm DRAM process,mobile systems,augmented reality,low-power DRAM.LPDDR,memory density,high-speed enabling features,per-pin DFE training,calibration training,size 10.0 nm
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要