Firing Dynamics and Offset Regulation of a Meminductive Rulkov Neuron with Application to IoMT Data Security

Discrete neuron models efficiently reproduce complex firing behaviors, while meminductive coupling introduces history-dependent regulation without substantially increasing model complexity. A discrete meminductor is coupled to the Rulkov neuron to construct a meminductive-feedback Rulkov neuron model (MF-Rulkov). The results show that the system can exhibit periodic bursting and chaotic bursting. By regulating the effective feedback strength, the meminductive state variable modifies the oscillation range, spike amplitude, interspike interval, and number of spikes per burst, thereby controlling firing amplitude and frequency. Changes in the offset-boosting term also shift the oscillatory trajectories within the state space, producing adjustable positive and negative offset states. The MF-Rulkov model is implemented on an STM32 platform, and its chaotic sequences are incorporated into the tiny encryption algorithm (TEA) to develop a ChaosTEA encryption scheme. Experiments on medical image encryption demonstrate favorable statistical performance and computational efficiency, supporting lightweight secure data transmission in the Internet of Medical Things (IoMT).