Memristors are the electronic components that are newly designed to represent some biologic systems such as the brain. Is there any explicit model to represent one neuron, several neurons and the neural network?
Memristor = Memory+ resistor . It can be used as a memory element since its resistance doesn't change after removing the applied electrical bias that means it is non-volatile and it also has non-scale and it can provide higher density and lower power consumption. Also Memristor can be in designing oscillators since it has a variable resistance and it increases/decreases according to the polarity of the applied voltage bias. Memristor also can be used in Neuroscience, since it behaves like synapse in connecting between two neurons.
I don't consider myself an expert (more a very interested observer), but I do recognise Memristors as a new building block for future technology and possible even a new microprocessor design.
You can follow this technology at http://www.memristor.org/ but there is plenty of material available for you to pursue. Personally I attended WCCI and spoke with many researchers first hand and chair there session on Memristors. I like the book Memristors and Memristive Systems and even reviewed the book on Advances in Neuromorphic Memristor Science and Applications.
Stan Williams is doing a great job at HP and Hynix has plans to build memory IC's (ReRAM).
There is a spice model and a lot of activity in many domains; both analog and digital.
I can only encourage people to support Mostafa's idea and help him the community to pursue cognitive modelling using Memristors. For those interested in this topic, the Memristor and Memristive Systems Symposium is a good place to begin as are the following resouces:
-World Congress on Computational Intelligence published in the Advances in Computational Intelligence http://www.springer.com/computer/database+management+%26+information+retrieval/book/978-3-642-30686-0
- Memristors and Memristive Systems http://www.springer.com/engineering/circuits+%26+systems/book/978-1-4614-9067-8
- Advances in Neuromorphic Memristor Science and Applications http://www.springer.com/biomed/book/978-94-007-4490-5
memristor could also be used in opamps and A/D systems, however, the main point of utilization of memristor is memories with low power dissipation. There are also great challenges on implementing such memories. and cope with its data integrity
I agree. Although I believe memristors have a greater role in our future. They become the new NAND gates and provide scientists or engineers with the ability to completely redesign the concept of transactional computation because they are able to simultaneously calculate and engage in low-powered long-term memory cells. Given you believe they are suitable for analogue applications, that would make them able to harmoniously unify all aspects of microelectronic engineering.
Matlab model - http://webee.technion.ac.il/people/skva/Memristor%20Models/Verilog-A/memristor%20model.txt However Jim Walker also suggests that ...
(http://www.nobeliefs.com/memristor.htm)
POSSIBLE INVENTIONS UTILIZING MEMRISTORS Include
1. memory for cameras, cell phones, iPods, iPads, etc. 1 to 5 years
2. universal memory replacing hard drives, RAM, flash, etc. in all computer devices 5 to 10 years
3. complex self learning neural networks and hybrid transistor/memristor circuits 5 to 15 years
4. memristic logic circuits on par with CPUs and other transistor circuits 15 to 20 years
5. advanced artificial thinking brains 20 to 30 years?
6. artificial conscious brains ?
7. memory and brains capable of living millions of years ?
8. duty-cycle artificial conscious beings capable of interstellar travel ?
9. creation of a real god I'm kidding!
Obviously if HP and Hynix don't start producing their promised mega memory and SSD products, scepticism will continue. I look forward to hearing more detail about the 'Machine'!
The utility of memristances is that they can implement self-learning in a simple form without extra circuitry. The conductance between memristor nodes is dependent on the timing of the voltage through the memristor. If SNN is wanted to be implemented with CMOS technology only, then the inclusion of STDP would imply an extra circuitry if compared with mixed CMOS-memristance systems. Also, memristors are two-terminal devices that are occupying a lower area than conventional MOS transistors and the packaging is higher. The main drawback that I see to memristance devices is their low endurance. This is a technology that is not as mature as CMOS but I hope that in the near future the technology would be enhanced and higher endurance times could be reached. Memristance can also be used to implement memories and Programmable logic.
Regarding the Memristance model, maybe this can help:
Mahvash, M., Parker, A.C. A memristor SPICE model for designing memristor circuits (2010) Midwest Symposium on Circuits and Systems, art. no. 5548803, pp. 989-992.
Another memristor model that is also quite popular
Kvatinsky, S., Friedman, E.G., Kolodny, A., Weiser, U.C. TEAM: Threshold adaptive memristor model (2013) IEEE Transactions on Circuits and Systems I: Regular Papers, 60 (1), art. no. 6353604, pp. 211-221.