In pursuit of devices that mirror real neural characteristics, this paper illustrates a leaky-integrate-fire (LIF) neuron implementated using Si doped HfO \(_2\) ferroelectric junction-less Tunnel FET (FE-JLTFET). The leaky-integration process is thought to be mimicked using BTBT phenomena. Furthermore, back gate biasing is not necessary to cause the floating body to restrict the excess charge carrier. Avoiding metallurgical doping, the problems and the manufacturing complexity can be solved. Using the ferroelectric insulator with normal oxide in the gate stack acts as a step-up voltage transformer. Thus, it only needs drain biased at 0.3V, which is significantly less than its corresponding DGJL FET, PD-SOI MOSFET, LBIMOS, bulk FinFETs and Si NIPIN based silicon neurons. The proposed device requires only 7fJ of energy per spike, exhibiting markedly superior energy efficiency compared to existing neuromorphic devices such as Phase Change CMOS, PD-SOI MOSFET, PCMO RRAM, FBFET, LBIMOS, DGJL FET respectively. Additionally, it exhibits spiking frequencies in the GHz range (8 times higher than biological neurons) at threshold voltage of 0.42V and threshold current of 0.5mA/ \(\mu \) m. As a result, it is assumed that the proposed device based leaky-integrated fire neuron is better suitable for implementing the SNN at large-scale due to less energy consumption, area efficient and CMOS adoptable.
