Chapter 11
Nonlinear Dynamics of Circuits with
Mem-Elements
In recent years, the use of memristors as nonlinear dynamical elements for real-time
analog signal processing has been a topic of ever increasing interest. Memristors
are widely employed in neuromorphic architectures and cellular neural networks,
where they behave as nonlinear dynamic devices within neurons (Chap. 9) or
they implement synaptic connections in nanotechnology with adaptation capabilities and reduced area consumption and power dissipation, see, e.g., [1–13], and
references therein. The article [14] stresses that scalable electronic devices, as
Mott memristors, implementing a source of controllable chaotic behavior, and
that can be incorporated into a neuromorphic network, are expected to be an
essential component of future computational systems. Memristors have also been
found effective for implementing rich nonlinear dynamics at the core of reservoir
computing [15]. A new paradigm has been developed in [16], where memcomputing
machines use memristors both as dynamical nonlinear devices for analog computing
and as devices that memorize the result of computation in the same physical
location, with the goal to overcome some bottlenecks of Von Neumann machines
where the computing and memory phases are performed at different locations.
Memcapacitors and meminductors are also currently receiving an increasing
attention. The recent article [17] reports on a voltage-controlled memcapacitor
where capacitive memory arises from reversible and hysteretic geometrical changes
in a lipid bilayer mimicking the composition and structure of biomembranes (cf.
Example 2.28) in Chap. 2. It is pointed out that such mem-elements are capable of
co-locating signal processing and memory via history-dependent reconfigurability
at the nanoscale and, as such, they are expected to be vital for next generation computing materials striving to match the brain’s efficiency and flexible
cognitive capabilities. This should in turn pave the way to develop low-energy,
biomolecular neuromorphic mem-elements as models to study capacitive memory
and signal processing in neuronal membranes. From an application viewpoint,
memcapacitive properties observed in certain metal-oxides nanostructures (e.g.,
VO 2 ) have suggested their use for frequency tuning of some metamaterials [18].
© Springer Nature Switzerland AG 2021
F. Corinto et al., Nonlinear Circuits and Systems with Memristors,
https://doi.org/10.1007/978-3-030-55651-8_11
387
Nonlinear Dynamics of Circuits with
Mem-Elements
In recent years, the use of memristors as nonlinear dynamical elements for real-time
analog signal processing has been a topic of ever increasing interest. Memristors
are widely employed in neuromorphic architectures and cellular neural networks,
where they behave as nonlinear dynamic devices within neurons (Chap. 9) or
they implement synaptic connections in nanotechnology with adaptation capabilities and reduced area consumption and power dissipation, see, e.g., [1–13], and
references therein. The article [14] stresses that scalable electronic devices, as
Mott memristors, implementing a source of controllable chaotic behavior, and
that can be incorporated into a neuromorphic network, are expected to be an
essential component of future computational systems. Memristors have also been
found effective for implementing rich nonlinear dynamics at the core of reservoir
computing [15]. A new paradigm has been developed in [16], where memcomputing
machines use memristors both as dynamical nonlinear devices for analog computing
and as devices that memorize the result of computation in the same physical
location, with the goal to overcome some bottlenecks of Von Neumann machines
where the computing and memory phases are performed at different locations.
Memcapacitors and meminductors are also currently receiving an increasing
attention. The recent article [17] reports on a voltage-controlled memcapacitor
where capacitive memory arises from reversible and hysteretic geometrical changes
in a lipid bilayer mimicking the composition and structure of biomembranes (cf.
Example 2.28) in Chap. 2. It is pointed out that such mem-elements are capable of
co-locating signal processing and memory via history-dependent reconfigurability
at the nanoscale and, as such, they are expected to be vital for next generation computing materials striving to match the brain’s efficiency and flexible
cognitive capabilities. This should in turn pave the way to develop low-energy,
biomolecular neuromorphic mem-elements as models to study capacitive memory
and signal processing in neuronal membranes. From an application viewpoint,
memcapacitive properties observed in certain metal-oxides nanostructures (e.g.,
VO 2 ) have suggested their use for frequency tuning of some metamaterials [18].
© Springer Nature Switzerland AG 2021
F. Corinto et al., Nonlinear Circuits and Systems with Memristors,
https://doi.org/10.1007/978-3-030-55651-8_11
387