Litcius/Paper detail

Energy-Efficient Single-Ended Read/Write 10T Near-Threshold SRAM

Erfan Abbasian, Sobhan Sofimowloodi

2023IEEE Transactions on Circuits and Systems I Regular Papers40 citationsDOI

Abstract

Modern system-on-chip-based applications require low-power/energy SRAMs for long-term operation. To deal with this issue, near-threshold SRAM design is an effective approach. In this regard, this paper presents an energy-efficient single-ended 10T (SE10T) near-threshold SRAM. The proposed SE10T improves read stability and writability with the help of a built-in read-assist scheme and a power-gating technique, respectively, and reduces power/energy consumption by using single-ended read/write operation and stacking of transistors in the cell core. Simulation results in 32-nm CMOS technology at a 0.6 V show that the proposed design improves read stability/writability by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.03\times /1.35\times $ </tex-math></inline-formula> , reduces leakage power by 46.09%, and offers improvements of 86.09%/88.81% and 73.82%/62.72% in read/write power and read/write energy, respectively, in comparison with the conventional 6T SRAM. The minimum operation voltage of the proposed design is the lowest ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$V_{min} =590$ </tex-math></inline-formula> mV), which is reduced by 41% compared to the conventional 6T. However, read/write delay in the proposed design is increased by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2.48\times /5.40\times $ </tex-math></inline-formula> due to being single-ended, and the layout area of the proposed design is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.893~\mu \text{m}^{2}$ </tex-math></inline-formula> , which is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1.82\times $ </tex-math></inline-formula> larger than that of the conventional 6T.

Topics & Concepts

CMOSStatic random-access memoryComputer scienceEnergy (signal processing)ChipTransistorNotationStability (learning theory)Power (physics)Computer hardwareEmbedded systemAlgorithmElectrical engineeringArithmeticMathematicsVoltageEngineeringPhysicsTelecommunicationsStatisticsQuantum mechanicsMachine learningLow-power high-performance VLSI designAdvanced Memory and Neural ComputingVLSI and FPGA Design Techniques