2. What is Neuromorphic Engineering ?
Neuromorphic engineering is an interdisciplinary amalgam of neuroscience,
biology, computer science and a number of other fields that attempts first to
understand how the brain manipulates information, and then to supplant or
supplement the input and/or output of the nervous system.
4. Neurons
Neurons are the most
fundamental unit of the
human brain. There are
about 1011 neurons in a
human brain each one of
which is furthur connected to
about 1000-10000 neurons.
Synapse is a connection site
between two neurons. They
usually connect axon and
dendrites.
5. Many VLSI models for neurons have been
developed till far just focussing on their
functionality instead of classifying their parts. The
simplest of all is where electrical nodes are
considered as neurons whereas silicon neurons
are the most detailed ones.
Various models are as follows :
VLSI Model of Neurons
1. Neurons as electrical nodes
2. Perceptrons (Mc Culloch Pitts neurons)
3. Integrate and fire Neurons
4. Silicon Neurons
6. Silicon Neuron Circuit Blocks
● Conductance Dynamics
● Spike Event Generation
● Spiking Thresholds and Refractory periods
● Spike-frequency adaptation and adaptive thresholds
● Axons and dendritic trees
12. Application of Neuromorphic circuits
• To enable the reanimation of paralyzed limbs
• Brain or nerve signals can be used to control computer cursor movements and
robotic arms
• Developing cochlear implants and functional retinal prostheses
• Used in non-invasive brain stimulators to improve normal people’s attention while
gaming
Therefore, it is necessary to study these circuits and the effects of temperature and
process changes on their speed and power consumption.
14. Mapping of Circuit to the Application
PVT variation
Circuit simulation result
mapped to MCR graph
Neuromorphic circuit analysis
Variation in device
dimensions
HSpice Simulation Error Analysis
Design a Neuromorphic Circuit
Approach to problem
15. References
➢ Lecture Notes, Neuromorphic Electronics, Department of Informatics, University of Oslo.
➢ Bipin Rajendran, Yong Liu, Jae-sun Seo, Kailash Gopalakrishnan, Leland Chang, Daniel J.
Friedman, Mark B. Ritter, Specifications of Nanoscale Devices and Circuits for
Neuromorphic Computational Systems, IEEE TRANSACTIONS ON ELECTRON DEVICES,
VOL. 60, NO. 1, JANUARY 2013.
➢ Neuromorphic silicon neuron circuits, 31 May 2011 doi: 10.3389/fnins.2011.00073
➢ Ahmad Abdel Majid Suleiman Bashaireh, Design Robustness Analysis of Neuromorphic
Circuits