Process-Voltage-Temperature Variability Estimation of Tunneling Current for Band-to-Band-Tunneling-Based Neuron

• Published in: IEEE transactions on electron devices Vol. 71; no. 1; pp. 752 - 758
• Main Authors: Patil, Shubham, Sharma, Anand, Gaurav, R., Kadam, Abhishek, Singh, Ajay Kumar, Lashkare, Sandip, Mohapatra, Nihar Ranjan, Ganguly, Udayan
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-01-2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Band-to-band-tunneling (BTBT); Current measurement; Electric potential; MOSFET; neuron; Neurons; on regime (ION); Power management; Power sources; process variability; silicon-on-insulator (SOI); State machines; subthreshold regime (SS); Temperature measurement; temperature variability; Threshold voltage; Tunneling; Voltage; Voltage measurement; voltage variability
• ISSN: 0018-9383; 1557-9646
• DOI: 10.1109/TED.2023.3331660

Compact and energy-efficient synapse and neurons are essential to realize the full potential of neuromorphic computing. In addition, a low variability is indeed needed for neurons in deep neural networks for higher accuracy. Further, process (P), voltage (V), and temperature (T) (PVT) variation are essential considerations for low-power circuits as performance impact and compensation complexities are added costs. Recently, band-to-band tunneling (BTBT) neuron has been demonstrated to operate successfully in a network to enable a liquid state machine (LSM). A comparison of the PVT with competing modes of operation (e.g., BTBT versus subthreshold and above threshold) of the same transistor is a critical factor in assessing performance. In this work, we demonstrate the PVT variation impact on the BTBT regime and benchmark the operation against the subthreshold regime (SS) and ON-regime (<inline-formula> <tex-math notation="LaTeX">\text{I}_{ \mathrm{\scriptscriptstyle ON}} </tex-math></inline-formula>) of partially depleted silicon-on-insulator MOSFET. It is shown that the ON-state regime offers the lowest variability but dissipates higher power, hence not usable for low-power sources. Among the BTBT and SS regimes, which can enable the low-power neuron, the BTBT regime has shown <inline-formula> <tex-math notation="LaTeX">\sim 3\,\,{\times } </tex-math></inline-formula> variability reduction (<inline-formula> <tex-math notation="LaTeX">{ \boldsymbol {\sigma }_{I{_{D}}}{/} \boldsymbol {\mu }_{I{_{D}}} </tex-math></inline-formula>) compared to the SS regime, considering the cumulative PVT variability. The improvement is due to the well-known weaker P, V, and T dependence of BTBT versus SS. We show that the BTBT variation is uncorrelated with mutually correlated SS and <inline-formula> <tex-math notation="LaTeX">\text{I}_{{ \mathrm{\scriptscriptstyle ON}}} </tex-math></inline-formula> operation-indicating its different origin from the mechanism and location perspectives. Hence, the BTBT regime is promising for low-current, low-power, and low device-to-device (D2D) variability neuron operation.