Search for: All records

To accurately determine the reliability of SRAMs, we propose a method to estimate the wearout parameters of FEOL TDDB using on-line data collected during operations. Errors in estimating lifetime model parameters are determined as a function of time, which are based on the available failure sample size. Systematic errors are also computed due to uncertainty in estimation of temperature and supply voltage during operations, as well as uncertainty in process parameters and use conditions. 

A convenient method based on deep neural networks and an evolutionary algorithm is proposed for the inverse design of FinFET SRAM cells. Inverse design helps designers who have less device physics knowledge obtain cell configurations that provide the desired performance metrics under selected wearout conditions, such as a set specific stress time and use scenario that creates a specific activity level (duty cycle and transition rate). The cell configurations being considered consists of various process parameters, such as gate length and fin height, in the presence of variations due to process and wearout. The front-end mechanisms related to wearout include negative bias temperature instability (NBTI), hot carrier injection (HCI), and random telegraph noise (RTN). The process of inverse design is achieved quickly and at good accuracy. 

Effective assessment of degradation induced by electromigration (EM) is necessary for the design of reliable circuits based on FinFET technology. In this paper, a new methodology is proposed where FinFET SRAM cell array activity is used to evaluate the resistance degradation due to EM. The implementation of this methodology consists of analysis of stress evolution, a time-dependent resistance model, cell array activity extraction, and a customized algorithm for cell array reliability evaluation. The stress model is derived from the material transport equation which contains the driving forces due to the gradient of vacancy concentration,temperature, hydrostatic stress, and EM itself. The time-dependent resistance shift describes the effect of stress evolution. The customized algorithm is applied to calculate the resistance degradation while considering the characteristics of metal wire arrays in SRAMs. The statistical degradation in a FinFET SRAM cell array reveals that, for the tested case, in addition to the percentage of the workload in various operating modes, the cell array activity distribution also affects EM degradation. More evenly distributed cell activity results in better EM reliability. 

Accelerated lifetime tests are necessary for reliability evaluation of circuits and systems, but the parameters for choosing the test conditions are often unknown. Furthermore, reliability testing is generally performed on test structures that have different properties than actual circuits and systems, which may create inconsistencies in how circuits and systems work in reality. To combat this problem, we use ring oscillators, which are similar to circuits, based on the 14nm FinFET node as the circuit vehicle to extract wearout data. We explore the effects of testing time, sample size, and number of stages on the ability to detect failures for various test conditions, focusing on front-end time dependent dielectric breakdown, which is one of the most dominant wearout mechanisms. 

We build a modelling and simulation flow to study how the front-end wearout mechanisms affect the FinFET SRAM soft error rate. This flow incorporates process variation, such as device dimensions, and degradation parameters. We first checked the impact of process parameters on critical charge and soft error rate. It is found that a larger gate length and higher temperature help us obtain better reliability for a FinFET SRAM cell under radiation, with a higher Qcrit and lower SER. Then, the time-dependent shift of Qcrit and SER is displayed. Within its range between 0% and 50%, a lower duty ratio leads to worse reliability due to soft errors. Moreover, a higher transition rate causes worse reliability. 

FinFET SRAM cells suffer from front-end wearout mechanisms, such as bias temperature instability and hot carrier injection. In this paper, we built a library based on deep neural networks (DNNs) to speed up the process of simulating FinFET SRAM cells' degradation. This library consists of two parts. The first part calculates circuit configuration parameters, wearout parameters, and the other input variables for the DNN. The second part calls for the DNN to determine the shifted circuit performance metrics. A DNN with more than 99% accuracy is achieved with training data from standard Hspice simulations. The correctness of the DNN is also validated in the presence of input variations. With this library, the simulation speed is one hundred times faster than Hspice simulations. We can display the cell's degradation under various configurations easily and quickly. Also, the DNN-based library can help protect intellectual property without showing users the circuit's details. 

Circuits may fail in the field due to a wide variety of failure modes. If there are frequent failures in the field, circuits are returned to the manufacturer, and the causes of failure must be identified. The challenge is that wearout mechanisms are confounded in circuit and system-level failure data. Using such failure data, it is often hard to separate the underlying failure causes without time-consuming and expensive physical failure analysis. To distinguish the wearout mechanisms for each failure sample, we have developed a quick and low-cost methodology using maximum likelihood estimation and probability analysis to determine the origin of the failure distributions, region of error, and sorting accuracy. We apply our methodology to analyze the competing wearout mechanisms in 14nm FinFET ring oscillators, as an example, using simulation. We also consider the problem of Trojan detection. 

Because data from a variety of wearout mechanisms is confounded in circuits, we apply machine learning techniques to detect the parameters of competing failure mechanisms in ring oscillators, which more closely mimic circuit behavior than test structures. This is the first known application using data analysis to distinguish competing wearout mechanisms in circuit-level data. To quickly and efficiently analyze failure data, we propose to use maximum likelihood estimation to separately determine the parameters of each underlying distribution by only observing the time-to-failure of samples. The quasi-Newton method 

Advanced FinFET SRAMs undergo reliability degradation due to various front-end and back-end wearout mechanisms. The design of reliable SRAMs benefits from accurate wearout models that are calibrated by accelerated test. With respect to testing, the accelerated conditions which can help separate the dominant wearout mechanisms related to circuit failure is crucial for model calibration and reliability prediction. In this paper, the estimation of optimal accelerated test regions for a 14nm FinFET SRAM under various wearout mechanisms is presented. The dominant regions for specific mechanisms are compared and analyzed for effective testing. It is observed that for our SRAM example circuit only bias temperature instability (BTI) and middle-of-line time-dependent dielectric breakdown (MTDDB) have test regions where their failures can be isolated, while the other mechanisms can’t be extracted individually due to acceptable regions’ overlap. Meanwhile, the SRAM cell activity distribution has a small influence on test regions and selectivity.