Neuron-like Bursting in a Negative Differential Resistance System
Jitendra Kumar a
a Instituto de Tecnología Química (ITQ), Consejo Superior de Investigaciones Científicas-Universitat Politècnica de València, 46022, Valencia, Spain
Proceedings of MATSUS Fall 2026 Conference (MATSUSFall26)
D4 Iontronics
Palma, Spain, 2026 October 26th - 30th
Organizers: Roberto Fenollosa Esteve and Francesco Rossella
Oral, Jitendra Kumar, presentation 164
Publication date: 22nd July 2026

Bursting responses play a key role in spiking neural networks and neuromorphic computing, carrying richer temporal structure than isolated spikes and improving the reliability of neural communication. Since synaptic transmission is probabilistic, clustered spikes naturally increase the probability of successful downstream activation. Unlike conventional clock-driven computing, neuromorphic systems operate asynchronously in an event-driven manner,[1,2] enabling low-latency, energy-efficient processing while facilitating the integration of heterogeneous components without requiring clock synchronization.[3,4] These characteristics make them particularly well suited for real-time edge intelligence in dynamic environments.

Controlled bursting emerges in a negative differential resistance (NDR) system when it is biased near the folding point and driven by time-varying inputs. The nonlinear device response, together with external modulation, generates stable spike clusters whose temporal structure can be tuned through the input amplitude, frequency, and bias conditions. Consequently, both the burst rate and the number of spikes per burst are controllable.

This dependence on multiple control parameters enables flexible temporal encoding, making the system suitable for asynchronous signal classification based on dynamical patterns rather than fixed-rate sampling. Burst length and firing activity can be continuously adjusted through the operating conditions.

When the device is driven in the oscillatory regime near a Hopf bifurcation, it effectively suppresses high-frequency noise while amplifying weak coherent signals, enabling the detection of signals buried in noise at signal-to-noise ratios as low as 1/1000. This noise-filtering capability arises from relaxation toward stable attractors, which attenuate stochastic fluctuations while reinforcing coherent input-driven transitions.

© FUNDACIO DE LA COMUNITAT VALENCIANA SCITO
We use our own and third party cookies for analysing and measuring usage of our website to improve our services. If you continue browsing, we consider accepting its use. You can check our Cookies Policy in which you will also find how to configure your web browser for the use of cookies. More info