Abstract:As the feature size of CMOS technology rapidly shrinks into the very deep submicrometer, the power density is getting higher and the circuit's reliability is seriously impacted by thermal noise. In order to reduce power dissipation, decreasing the supply voltage is a promising way. However, soft errors and decreased supply voltage make circuits behave probabilistic. Fortunately, some applications can tolerate a certain level of operation errors. The tradeoff between energy and reliability provides an opportunity to save the power consumption in the circuit. An asymptotically linear analysis (ALA) strategy is proposed to model the reliability evaluation of combinational logic circuits. Under the assumption that the gate error probability is not more than 0.05, an asymptotic linear matrix A is constructed by using the first-order term of the output error probability. When propagating the gate error probability from the input to the output ports, the ALA strategy only requires one multiplication between the matrix A and the gate error probability vector. There are significant differences in the complexity of computation of the ALA scheme and the Bayesian network (BN) and the probability transfer matrix (PTM) schemes, while its relative error (RE) is no more than 0.08 on the benchmark of LGSynth91. Furthermore, three schemes are proposed to adjust the gate error probability to meet the constraint on the error probability of the probabilistic CMOS (PCMOS) circuit's outputs. The proposed schemes can be performed to search the gate error probability combinations in the logic and circuit design stage of VLSI. During the power optimization, the switching activity factor in the asymptotically linear region is approximately invariant. It makes the power optimal model convex. The energy-saving ratio of the proposed power optimization scheme is 5%-6% higher than the existing schemes. The time consumption of the proposed three schemes is one-thousandth of the time consumed by the BN and PTM schemes. All the three schemes are verified by conducting experiments on the ISCAS'85 and IWLS'15 benchmark circuits. The simulation results establish the feasibility of the ALA and the performance of the proposed schemes.

The Probability Logics for Nanoscale Inverters Cascade

Design of Nano-Scale Noise Tolerant CMOS Logic Circuits Based on Probabilistic Markov Random Field Approach

Design of probabilistic-based Markov Random Field logic gates in 65nm CMOS technology

Probabilistic Modeling Of Nanoscale Adder

A Probability-Density Function Approach to Capture the Stochastic Dynamics of the Nanomagnet and Impact on Circuit Performance

Synthesizing Combinational Logic To Generate Probabilities: Theories And Algorithms

The Synthesis of Stochastic Circuits for Nanoscale Computation.

The Synthesis of Combinational Logic to Generate Probabilities

Ground-State Probabilistic Logic with the Simplest Binary Energy Landscape for Probabilistic Computing

Soft Error Reliability Evaluation of Nanoscale Logic Circuits in the Presence of Multiple Transient Faults.

Gate Leakage Analysis of Nano-MOSFETs Using Ensemble Full Band Monte Carlo with Quantum Correction

Asymptotically Linear Analysis and Gate Probability Allocation Schemes in Probabilistic Circuits.

Critical nonlinear aspects of hopping transport for reconfigurable logic in disordered dopant networks

A General Scheme for Noise-Tolerant Logic Design Based on Probabilistic and DCVS Approaches.

Complementary Logic with Voltage Zero‐Loss and Nano‐Watt Power Via Configurable MoS2/WSe2 Gate

Modeling and Optimizing of Carbon Nanotube Based CMOS Logic Circuit

Design Trends and Challenges of Logic Soft Errors in Future Nanotechnologies Circuits Reliability

Simple Noise Margin Model for Optimal Design of Unipolar Thin-Film Transistor Logic Circuits

Self-similar Magneto-electric Nanocircuit Technology for Probabilistic Inference Engines

Fault-tolerant Analysis for Feedback Based Digital Circuit via Probabilistic CMOS Model

Re-investigation of gate oxide breakdown on logic circuit reliability