Selected MemoCIS Achievements
1. Researchers from the Czech Republic, Switzerland, Germany, Italy, Cyprus, Spain and the UK, bring together four distinct scientific communities to work together on the common vision of creating Memristor based systems, by pooling together skills from material science, semiconductor devices, circuit theory, device modeling and simulation, sensor design, and circuit and system design.
2. German Action partners demonstrates BiFeO3 memristor-based encryption of medical data
3. UK and Swiss partners from the MemoCIS Action emulate short-term synaptic devices with memristive devices.
4. UK and Swiss partners from the MemoCIS Action implement a spike-timing based learning rule using TiO2-x memristors
5. Italian and UK partners demonstrate gradual set dynamics in HFO2 Memristors
6. German based partners demonstrate the use of tunnel junction based memristors as artificial synapses
7. Italian and UK partners demonstrate HFO2 based Memristors are suitable for neuromorphic applications
8. Italian partners demonstrate neuromorphic learning and recognition with one-transistor-one resistor synapse
9. Italian partners demonstrate unsupervised learning by spike timing dependent plasticity in phase change memory (PCM) synapses
10. Italian partners demonstrate physical unbiased generation of random numbers with coupled resistive switching devices
11. German and US partners develop Robust simulation of a TaO memristor model
12. Italian partners investigated the role of metal-oxide interfaces in the multiple resistance switching regimes of Pt/HfO2/TiN devices and showed that both metal oxide interfaces are active in the formation/dissolution of conductive filaments
13. UK partners develop the first instrument to enable en-masse RRAM characterisation with ns pulsing resolution
14. UK partners develop a memristor biasing parameter optimizer that can also be used for increasing device reliability
15. UK partners implement a spike-based perceptron learning rule using TiO2−x memristors
16. UK partners develop a conductive AFM investigation methodology for determining switching thresholds in titanium dioxide thin films to give new insights for device programmability, reliability and failure mechanisms
17. UK and China partners develop a memristor SPICE model accounting for synaptic activity dependence
18. UK partners perform X-ray absorption spectroscopy (XAS) to characterize the atomic-scale structures of a nonstoichiometric TiO2−x thin films
19. UK and Swiss partners demonstrate emulation of short-term synaptic dynamics with memristive devices
20. Swiss partners develop the worlds first electrochemical biosensor base of the memristive effect and DNA aptamers, for the detection of prostate specific antigens for early detection of cancer
21. UK partners emulate the electrical activity of the neuron using a silicon oxide RRAM cell
22. UK and Dutch partners investigation the Switching Mechanism in TiO2 Based RRAM by using a two-dimensional EDX approach to show that the switching mechanism involves both the redistribution of both Ti and O ions within the active layer
23. UK and Austrian partners demonstrate unsupervised learning in probabilistic neural networks with multi-state metal-oxide memristive synapses, showcasing successful learning in the presence of corrupted input data and probabilistic neurons
24. German and Czech partners investigate convergence and numerical issues related to accurate memristor modeling
25. German and Czech partners develop a generalized rule of homothety of ideal memristors and their siblings
26. German and Italian partners develop a Generalized Boundary Condition Model for memristors
27. German and Italian partners develop a class of memristor circuits consisting of a cascade connection between a static nonlinear two-port and a dynamic one port
28. German partners get insights on underlying mechanisms of NbOx memristors through application of theory of nonlinear dynamics
29. Greek partners (industrial and academia) demonstrate memristor-based architecture for High-Radix arithmetic systems
30. German partners (industrial and academia) develop a physics-based Spice model for the Nb2O5 memristor
31. Italian and UK partners demonstrate HfO2-based Memristors are suitable for Neuromorphic Applications
32. Czech partners develop a behavioral model for simplified identification of memristor parameters
33. Czech partners develop an improved models of the TiO2 memristor
34. Serbian partners design memristor based circuits for microwave applications
35. German and USA partners discover memory loss in memristors and develop models
36. Greek and Italian partners develop a two transistor non-ideal memristor emulator
37. Spanish and Greek partners develop the experimental model of a non-ideal memristor
38. Dutch, Belgium and Swiss (IBM Research Laboratory) partners demonstrate memristor-based, computation-in-memory architecture for Data-Intensive Applications.
39. Greek partners develop multi-state memristive nanocrossbar for high-radix computer arithmetic systems
40. UK and Swiss partners demonstrate emulation of short-term synaptic dynamics with memristive devices
41. Swiss and UK partners implement a spike-based perceptron learning rule using TiO2−x memristors
42. Portuguese partners explore Memristor-based Willshaw network and analyze capacity and robustness to noise in the presence of defects, thus exploring Memristor reliability from an architectural point of view
43. Greek and Dutch partners explore five alternative architectures towards developing reliable memristive crossbar memories
44. Italian and German partners identify a class of versatile circuits, made up of standard electrical components, to be memristors
45. Norwegian and Swiss partners develop a novel readout circuit for memristive biosensors
46. Spanish and Greek partners develop an analytical energy model for the reset transition in unipolar resistive-switching RAMs
47. Czech partners develop an analog emulator of genuinely floating memcapacitor
48. Czech partners develop memristor models for SPICE simulation of extremely large memristive networks
49. Czech partners identify “The Simplest Memristor in the World,”
50. Italian partners develop a novel Flux—Charge Analysis Method for memristors
51. Italian and German partners write/edit a book on “Resistive Switching: From Fundamentals of Nanoionic Redox Processes to Memristive Device Applications”, with (but not limited to) contributions from MemoCIS Action members
52. UK partner develops Gradient-descent-based learning in memristive crossbar arrays
53. Italian and German partners investigate “Synchronization conditions in simple memristor neural networks
54. Dutch partners develop Computation-in-memory based parallel adder
55. US and Spanish partners model and demonstrate a Hopfield Network Analog-to-Digital Converter with Hybrid CMOS/Memristor Circuits
56. Spanish and US partners develop a generalized reconfigurable memristive dynamical system for neuromorphic applications
57. Swiss partners develop SiNW-FET in-Air Biosensors for High Sensitive and Specific Detection in Breast Tumor Extract
58. Greek partners develop Oscillation-Based Slime Mould Electronic Circuit Model for Maze-Solving Computations
59. Swiss partners develop Event-based softcore processor in a biohybrid setup applied to structural plasticity
60. Swiss partners examine resistance impact by long connections on electrical behavior of integrated Memristive Biosensor
61. UK and Greek partners developed a novel versatile Memristor model that emulates realistically several metal-oxide technologies.
62. Italian and German partners explore the development of novel RRAM-based recurrent network for associative memories, unsupervised learning of patterns, and error correction.
63. Italian and Swiss partners demonstrated the development of a hybrid CMOS-RRAM hardware system.
64. Cypriot and UK partners are collaborating for the development of mem-capacitors and mem-inductors that can operate in the radio frequency/microwave bands.
65. Cypriot and UK partners are collaborating for the development of optically controlled metasurfaces using mem-capacitors.
66. UK partners have pushed metal-oxide memristor technologies to a new level of performance after experimenting with several metal-oxide bi-layer combinations.
67. UK partners have demonstrated a new state-of-art in scaling densities of RRAM memristors.
68. Italian partners have developed a novel RRAM technology based on SiOx which shows superior resistance window, multilevel operation, low variability, and good retention at high temperature (1 hour at 260°C).
69. Italian partners have developed a novel SiOx-based technology for volatile RRAM with Ag and Cu top electrodes.
70. Italian partners characterized HfO2 based memristive devices in view of their use as synapses in spiking neural networks.
71. UK partners have recently demonstrated a new type of optical memristor device that can be controlled by wavelength and polarization-specific light.
72. German and Greek partners derived the analytical solution to the state equation of a real-world memristor based upon tantalum oxide under any positive DC voltage input and for all initial conditions.
73. German partners reported fading memory effects in a SrTiO3-based VCM cell fabricated in house, and gained some new insights into the mechanisms at the origin of this peculiar dynamic phenomenon.
74. Czeck and German partners reported a deep analysis of memristors with nonlinear exponential memristance versus state maps, including the derivation of a closed-form solution for the Abel differential equation governing the current flowing through devices of this kind in response to arbitrary voltage stimuli.
75. Czeck and Serbian partners have achieved several promising results in the numerical investigation of extremely-large memristive networks and of memristor-based microwave filters.
76. German partners introduced a robust continuous and differentiable approximation to the Strachan's model.
77. German and Italian partners employed a circuit, consisting of purely-passive two-terminal elements from Circuit Theory and falling into the class of extended memristors, in place for Chua’s diode to provide nonlinearity to the Chua's oscillator.
78. German partners introduced the theory behind a novel memristor-centered strategy to control the motion of the limb of a humanoid robot called Myon.
79. HRL Laboratories LLC, Hewlett Packard Laboratoriesand German partners demonstrated that a thyristor-based CMOS circuit is not a genuine memristor.
80. Swiss and German partners used the extraordinary sensing capability of silicon nano-wires to detect the Ebola matrix protein.
81. Italian partners developed different types of models of HfO2–based memristive devices.
82. Czeck partners extended the class of two-terminal memristor devices to include memristive multi-ports in order build up circuit-theoretic foundations on more complex memristive structures.
83. Czeck partners revealed that the spin-transfer torque effect, caused by the current flowing through the magnetic element of spin torque nano-oscillators, modifies the natural magnetic damping torque by a mechanism that can be considered as the operation of a hidden locally-active memristor.
84. Czeck partners developed a method for deriving the Lagrange equations of circuits, containing memristors, memcapacitors, meminductors and other higher-order elements from Chua’s periodic table, on the basis of the knowledge of the potential functions of the circuit elements.
85. Czeck and Serbian partners have conducted a large number of numerical simulations of extremely-large networks with tens of millions of memristors.
86. UK and Greek partners implemented (a) a general empirical model for analog ReRAM devices, (b) a physical device characterization routine and (c) a corresponding parameter extraction algorithm for the developed memristor model.
87. Spanish and Swiss partners managed to design chips with a mixed CMOS+memristor process provided by St-Microelectronics and CEA-LETI.
88. Dutch partners succeeded to initiate and get funding for an EU project (H2020-ICT-2017-1) on memristive devices for computing (MNEMOSNE).
89. Dutch partners got a patent on “Computing device for “big data” applications using memristors”, US 9824753, Nov 21, 2017.
90. Swiss and Italian partners have demonstrated a spike-driven threshold-based learning rule that goes beyond the conventional Spike-Timing Dependent Plasticity mechanisms.
91. Cypriot and UK partners have developed a generic adaptive reactive element that is a memcapacitor/meminductor used for adaptive metasurfaces.
92. UK partners have provided a useful database for selecting appropriate electrode materials in TiO2-based devices.
93. Spanish partners have developed and tested a piecewise model for the reset and set transitions of a bipolar ReRAM memristive device in the flux-charge domain.
94. UK partners have developed a method to reduce the formation of crystallites in ZnO nanorod thin-films by the use of high microwave power.
95. Swiss and Italian partners have developed a novel hybrid 4-transistor/1-resistor synapse capable of spike-rate-dependent plasticity.
96. Italian partners have introduced the concept of stateful neural networks consisting of resistive switches that can perform all logic functions with the same network topology.
97. Greek and German partners have developed a complete analytical model for the time evolution of the state of a real-world memristor under any DC stimulus and for all initial conditions.
98. Serbian partners have presented the use of memristive switches for the implementation of main-line mounted loaded-line phase shifters.
99. Serbian partners have developed a novel memristor-based multilayer dual-mode resonator suitable for the design of reconfigurable multi-band filters.
100. Swiss partners have developed a novel, portable, bio-detection system based on memristive biosensors to provide a fast, automatic and simulatneous sensing parameters from a single chip.
101. Dutch partners have presented two computation-in-memory based accelerators making use of memristive devices, Memristive Vector Processor and RRAM Automata Processor.
102. Italian and UK partners have developed a new concept where the inherent volatile properties of a nano-scale memristive device are used to detect and compress information on neural spikes as recorded by a multielectrode array.
103. UK partners have managed to seamlessly fuse digital and analog design for the development of reconfigurable computing using memristors
104. UK partners have developed a memristor-based threshold logic gate design toward the optimisation of computer workflows.
105. UK partners have presented a high sensitivity threshold detection circuit concept that used its offset voltage as the reference threshold and thus directly transforms differential input signal samples into digital values.