With the rapid shrinking of technology nodes in the deep submicron era, the increasing power density in SRAM circuits has become a significant concern due to increased average power and leakage current. The SRAM is a vital component of cache memory in system on system on chip (SOC) designs, where it consumes a significant portion of power. This paper presents novel low power 16 nm SRAM cells design using 3 different architectures namely architecture-1 (A-1), architecture-2 (A-2), and architecture-3 (A-3) in CMOS technology and results are compared with conventional 6 T (C-6 T) SRAM cell. All the results are verified with the LTSPICE simulator. All the proposed SRAM cells design show improvements in terms of average operational and leakage power reduction, while A-1 additionally shows enhanced WSNM, and reduced WRITE delay as compared to the C-6 T SRAM cell. The proposed cell shows an improvement as compared with the published results and is suitable for high speed, low power memory circuit design in the VLSI industry.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Novel Low Power SRAM Cells in 16 nm CMOS Technology

  • Prakhar Chaturvedi,
  • Abhijit Asati

摘要

With the rapid shrinking of technology nodes in the deep submicron era, the increasing power density in SRAM circuits has become a significant concern due to increased average power and leakage current. The SRAM is a vital component of cache memory in system on system on chip (SOC) designs, where it consumes a significant portion of power. This paper presents novel low power 16 nm SRAM cells design using 3 different architectures namely architecture-1 (A-1), architecture-2 (A-2), and architecture-3 (A-3) in CMOS technology and results are compared with conventional 6 T (C-6 T) SRAM cell. All the results are verified with the LTSPICE simulator. All the proposed SRAM cells design show improvements in terms of average operational and leakage power reduction, while A-1 additionally shows enhanced WSNM, and reduced WRITE delay as compared to the C-6 T SRAM cell. The proposed cell shows an improvement as compared with the published results and is suitable for high speed, low power memory circuit design in the VLSI industry.