No abstract

No abstract

In this paper, a voltage-scaled SRAM for both error-free and error-tolerant applications is presented that dynamically manages the energy/quality trade-off based on application need. Two variation-resilient techniques, write assist and Error Correcting Code, are selectively applied to bit positions having larger impact on the overall quality, while jointly performing voltage scaling to improve overall energy efficiency. The impact of process variations, voltage and temperature on the energy-quality tradeoff is investigated. A 28 nm CMOS 32 kb SRAM shows 35% energy savings at iso-quality and operates at a supply 220 mV below a baseline voltage-scaled SRAM, at the cost of 1.5% area penalty. The impact of the SRAM quality at the system level is evaluated by adopting a H.264 video decoder as case study.Index Terms-Error-tolerant, error-free, energy-quality tradeoff, ultra-low power processing, near-threshold, SRAM, approximate computing, resiliency.

A new methodology to realize efficient multiplexers using quantum-dot cellular automata (QCA) is presented in this paper. The novel designs here demonstrated fully exploit the intrinsic logic capabilities of the basic building block in the QCA domain: the Majority Gate. An efficient logic formulation is derived for the 4:1 multiplexing function that can be recursively applied to the realization of multiplexers with any fan-in, by adding in the worst-case path only one level of Majority Gate for each input doubling. A 16:1 multiplexer designed by applying the proposed recursive approach requires less than 1600 cells and consumes only 12 clock phases to complete the operation. INTRODUCTIONRecently the emerging technology based on quantum-dot cellular automata (QCA) has received a great deal of attention, and several solutions have been proposed to design efficient logic circuits, such as multiplexers [1-3], adders [4][5][6][7][8][9][10][11], multipliers [12-18] and comparators [19][20][21][22][23][24][25][26].The QCA nanostructure originally demonstrated in [27] uses a square cell with four quantum dots and two free electrons as the basic element to realize both logic structures and interconnections. The free electrons can tunnel through the dots within the cell, but, owing to Coulombic repulsion, they can lead to only two possible stable states, also named polarizations, that are associated with the binary states 1 and 0. Even though adjacent cells interact through electrostatic forces and tend to align their polarizations, QCA cells do not have intrinsic data flow directionality. The latter is provided by means of four clock signals clk x (with x ranging from 0 to 3), shifted by 90°from each other, that are progressively associated to the cells within a QCA design [28,29]. This clock scheme, named the zone clocking scheme, makes the QCA designs intrinsically pipelined, because each clock zone behaves like a D-latch.The fundamental logic gates inherently available within the QCA technology are the inverter and the majority gate (MG); non-elementary digital modules are designed combining instances of these basic gates and exploiting an efficient zone clocking scheme . The design of complex gates, such as multiplexers (MUXes), is not actually straightforward, and proper strategies have to be adopted at both logic and layout levels to improve performance and area behaviors.Large fan-in multiplexers (i.e. larger than 4:1) are crucial as fundamental blocks in several popular applications; few examples are the realization of lookup tables for special function units and the decoders in memory circuits [30][31][32][33][34][35][36]. This paper is focused on the design of efficient QCA-based large fan-in MUXes and proposes a novel approach that smartly exploits the majority logic function

No abstract