Random telegraph signal analysis with a recurrent neural network

Lambert, Nicholas J.; Esmail, A. A.; Edwards, Megan; Ferguson, A. J.; Schwefel, Harald G. L.

doi:10.1103/physreve.102.012312

Cited by 5 publications

(5 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, in a submicron-sized metal-oxide-semiconductor field-effect transistor (MOSFET), if there is a trap to capture or emit a single charge carrier, device currents in the transport channel exhibit arbitrary patterns of two-level RTSs over time. Similar RTS phenomena are discerned in superconducting qubits 14 , 15 , single photon avalanche diodes 16 , ultrasensitive biosensors 17 , and single-cell activities in ion channels 18 . Real-time dynamics in quantum electrical devices also reveal rapid jumping sequences 19 , 20 .…”

Section: Introductionmentioning

confidence: 58%

“…Recently, two interesting studies attempted to build neural network models for examining RTSs. One study applied a clustering neural network to assess a long-time series of currents in a resistive random access memory device with improved weighted TLP 40 , and the other work constructed a long short-term memory (LSTM) recurrent neural network (RNN) to evaluate phase RTSs in superconducting double dots 15 . Hence, systematic analysis models of complex RTSs are sought-after, and to the best of our knowledge, a step-by-step protocol for the quantitative RTS analysis and its explicit descriptions are still needed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep neural network analysis models for complex random telegraph signals

Robitaille

Yang

Wang

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Time-fluctuating signals are ubiquitous and diverse in many physical, chemical, and biological systems, among which random telegraph signals (RTSs) refer to a series of instantaneous switching events between two discrete levels from single-particle movements. A reliable RTS analysis is a crucial prerequisite to identify underlying mechanisms related to device performance and sensitivity. When numerous levels are involved, complex patterns of multilevel RTSs occur and make their quantitative analysis exponentially difficult, hereby systematic approaches are often elusive. In this work, we present a three-step analysis protocol via progressive knowledge-transfer, where the outputs of the early step are passed onto a subsequent step. Especially, to quantify complex RTSs, we resort to three deep neural network architectures whose trained models can process raw temporal data directly. We furthermore demonstrate the model accuracy extensively with a large dataset of different RTS types in terms of additional background noise types and amplitude size. Our protocol offers structured schemes to extract the parameter values of complex RTSs as imperative information with which researchers can draw meaningful and relevant interpretations and inferences of given devices and systems.

show abstract

Section: Introductionmentioning

confidence: 58%

Section: Introductionmentioning

confidence: 99%

Deep neural network analysis models for complex random telegraph signals

Robitaille

Yang

Wang

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…For example, in a submicron-sized MOSFET, if there is a trap to capture or emit a single charge carrier, device currents in the transport channel exhibit arbitrary patterns of two-level RTSs over time. Similar RTS phenomena are discerned in superconducting qubits [14], single photon avalanche diodes [15], ultrasensitive biosensors [16], and single-cell activities in ion channels [17]. Real-time dynamics in quantum electrical devices also reveal rapid jumping sequences [18,19].…”

mentioning

confidence: 64%

Extensive Study of Multiple Deep Neural Networks for Complex Random Telegraph Signals

Robitaille¹,

Yang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Time-fluctuating signals are ubiquitous and diverse in many physical, chemical, and biological systems, among which random telegraph signals (RTSs) refer to a series of instantaneous switching events between two discrete levels from single-particle movements. Reliable RTS analyses are crucial prerequisite to identify underlying mechanisms related to performance sensitivity. When numerous levels partake, complex patterns of multilevel RTSs occur, making their quantitative analysis exponentially difficult, hereby systematic approaches are found elusive. Here, we present a three-step analysis protocol via progressive knowledge-transfer, where the outputs of early step are passed onto a subsequent step. Especially, to quantify complex RTSs, we build three deep neural network architectures that can process temporal data well and demonstrate the model accuracy extensively with a large dataset of different RTS types affected by controlling background noise size. Our protocol offers structured schemes to quantify complex RTSs from which meaningful interpretation and inference can ensue.

show abstract

“…For instance, short-channel MOSFETs display RTSs in drain currents, where two distinctive currents are observed when a single charge trap independently captures or emits an electron [ 43 , 44 , 45 ]. As another example, a two-level fluctuator at the superconducting qubit systems is considered to be responsible for the current RTS [ 46 , 47 ]. Each RTS is determined by three quantities of the amplitude between two distinct values of a measurement variable denoted as

and two dwell time constants

and

at the high and low levels, as indicated in Figure 2 e. Statistically, the trapping and detrapping events obey a simple Poisson process from a series of independent abrupt switches whose distributions are quantified by average characteristic times,

, and

[ 48 ], and the corresponding PSD of the Poisson distribution exhibits a Lorentzian spectrum with

= 2.…”

Section: Noise Processesmentioning

confidence: 99%

“…It is natural that the properties and behavior of noise are highly idiosyncratic in each quantum processor architecture, where both classical and quantum fluctuations coexist under Gaussian and non-Gaussian statistics [ 66 ]. Low-frequency behaviors aggravating decoherence are clearly observed in the form of RTSs associated with two-level or quantum fluctuators in materials [ 46 , 47 ], and the noise spectral density in qubit parameters exhibits a power-law trend [ 46 , 66 , 67 ]. The next subsection is devoted to explaining our step-by-step analysis of RTS time tracing method adopting neural network models, which is essential to quantify RTS parameters in complicated signals.…”

Section: Noise Processesmentioning

confidence: 99%

Material-Inherent Noise Sources in Quantum Information Architecture

Yang

Kim

2023

Materials

View full text Add to dashboard Cite

NISQ is a representative keyword at present as an acronym for “noisy intermediate-scale quantum”, which identifies the current era of quantum information processing (QIP) technologies. QIP science and technologies aim to accomplish unprecedented performance in computation, communications, simulations, and sensing by exploiting the infinite capacity of parallelism, coherence, and entanglement as governing quantum mechanical principles. For the last several decades, quantum computing has reached to the technology readiness level 5, where components are integrated to build mid-sized commercial products. While this is a celebrated and triumphant achievement, we are still a great distance away from quantum-superior, fault-tolerant architecture. To reach this goal, we need to harness technologies that recognize undesirable factors to lower fidelity and induce errors from various sources of noise with controllable correction capabilities. This review surveys noisy processes arising from materials upon which several quantum architectures have been constructed, and it summarizes leading research activities in searching for origins of noise and noise reduction methods to build advanced, large-scale quantum technologies in the near future.

show abstract

Random telegraph signal analysis with a recurrent neural network

Cited by 5 publications

References 32 publications

Deep neural network analysis models for complex random telegraph signals

Deep neural network analysis models for complex random telegraph signals

Extensive Study of Multiple Deep Neural Networks for Complex Random Telegraph Signals

Material-Inherent Noise Sources in Quantum Information Architecture

Contact Info

Product

Resources

About