Limits...
Implementation of olfactory bulb glomerular-layer computations in a digital neurosynaptic core.

Imam N, Cleland TA, Manohar R, Merolla PA, Arthur JV, Akopyan F, Modha DS - Front Neurosci (2012)

Bottom Line: Our system is based on a digital neuromorphic chip consisting of 256 leaky-integrate-and-fire neurons, 1024 × 256 crossbar synapses, and address-event representation communication circuits.The neural circuits configured in the chip reflect established connections among mitral cells, periglomerular cells, external tufted cells, and superficial short-axon cells within the olfactory bulb, and accept input from convergent sets of sensors configured as olfactory sensory neurons.Our circuits, consuming only 45 pJ of active power per spike with a power supply of 0.85 V, can be used as the first stage of processing in low-power artificial chemical sensing devices inspired by natural olfactory systems.

View Article: PubMed Central - PubMed

Affiliation: Computer Systems Lab, Department of Electrical and Computer Engineering, Cornell University Ithaca, NY, USA.

ABSTRACT
We present a biomimetic system that captures essential functional properties of the glomerular layer of the mammalian olfactory bulb, specifically including its capacity to decorrelate similar odor representations without foreknowledge of the statistical distributions of analyte features. Our system is based on a digital neuromorphic chip consisting of 256 leaky-integrate-and-fire neurons, 1024 × 256 crossbar synapses, and address-event representation communication circuits. The neural circuits configured in the chip reflect established connections among mitral cells, periglomerular cells, external tufted cells, and superficial short-axon cells within the olfactory bulb, and accept input from convergent sets of sensors configured as olfactory sensory neurons. This configuration generates functional transformations comparable to those observed in the glomerular layer of the mammalian olfactory bulb. Our circuits, consuming only 45 pJ of active power per spike with a power supply of 0.85 V, can be used as the first stage of processing in low-power artificial chemical sensing devices inspired by natural olfactory systems.

No MeSH data available.


Related in: MedlinePlus

Crossbar connectivity depicting the convergence of OSNs expressing the same receptor type to the same glomerulus. A total of 48 distinct glomerular columns are implemented in one core.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3368244&req=5

Figure 4: Crossbar connectivity depicting the convergence of OSNs expressing the same receptor type to the same glomerulus. A total of 48 distinct glomerular columns are implemented in one core.

Mentions: In lieu of actual sensors, we simulated the presentation of odors with 48-dimensional combinatorial activity patterns, each dimension representing the level of activation of one sensor type. Each of these input dimensions was mediated by 10 OSNs, each capable of an arbitrary number of activation levels represented by stochastic spike rates. The presence of a particular odorant was simulated with a characteristic pattern of relative activation levels of the sensors. The sensors were assumed to be broadly tuned and sensitive to concentration, each responding with a specific intensity to an odorant and each showing monotonic rise in activity as the concentration of the odorant increased. The OSNs were assumed to reside outside of the core and to provide spiking input to the core through the rows of the crossbar. The 10 inputs associated with the same sensor type were implemented homogeneously and projected convergently to the same glomerular column as depicted in Figure 4. The activation of a sensor was simulated by sending AER packets (“spikes”) into the core (to address the crossbar rows associated with the sensor) at a frequency corresponding to the activation level. At each time step, sensors that were highly responsive to a presented odorant had a higher probability of sending in AER packets into the core, compared to sensors that were less responsive. Consequently, any sensor that can be configured to output AER packets at a frequency proportional to its activation level can be used effectively with this chip.


Implementation of olfactory bulb glomerular-layer computations in a digital neurosynaptic core.

Imam N, Cleland TA, Manohar R, Merolla PA, Arthur JV, Akopyan F, Modha DS - Front Neurosci (2012)

Crossbar connectivity depicting the convergence of OSNs expressing the same receptor type to the same glomerulus. A total of 48 distinct glomerular columns are implemented in one core.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3368244&req=5

Figure 4: Crossbar connectivity depicting the convergence of OSNs expressing the same receptor type to the same glomerulus. A total of 48 distinct glomerular columns are implemented in one core.
Mentions: In lieu of actual sensors, we simulated the presentation of odors with 48-dimensional combinatorial activity patterns, each dimension representing the level of activation of one sensor type. Each of these input dimensions was mediated by 10 OSNs, each capable of an arbitrary number of activation levels represented by stochastic spike rates. The presence of a particular odorant was simulated with a characteristic pattern of relative activation levels of the sensors. The sensors were assumed to be broadly tuned and sensitive to concentration, each responding with a specific intensity to an odorant and each showing monotonic rise in activity as the concentration of the odorant increased. The OSNs were assumed to reside outside of the core and to provide spiking input to the core through the rows of the crossbar. The 10 inputs associated with the same sensor type were implemented homogeneously and projected convergently to the same glomerular column as depicted in Figure 4. The activation of a sensor was simulated by sending AER packets (“spikes”) into the core (to address the crossbar rows associated with the sensor) at a frequency corresponding to the activation level. At each time step, sensors that were highly responsive to a presented odorant had a higher probability of sending in AER packets into the core, compared to sensors that were less responsive. Consequently, any sensor that can be configured to output AER packets at a frequency proportional to its activation level can be used effectively with this chip.

Bottom Line: Our system is based on a digital neuromorphic chip consisting of 256 leaky-integrate-and-fire neurons, 1024 × 256 crossbar synapses, and address-event representation communication circuits.The neural circuits configured in the chip reflect established connections among mitral cells, periglomerular cells, external tufted cells, and superficial short-axon cells within the olfactory bulb, and accept input from convergent sets of sensors configured as olfactory sensory neurons.Our circuits, consuming only 45 pJ of active power per spike with a power supply of 0.85 V, can be used as the first stage of processing in low-power artificial chemical sensing devices inspired by natural olfactory systems.

View Article: PubMed Central - PubMed

Affiliation: Computer Systems Lab, Department of Electrical and Computer Engineering, Cornell University Ithaca, NY, USA.

ABSTRACT
We present a biomimetic system that captures essential functional properties of the glomerular layer of the mammalian olfactory bulb, specifically including its capacity to decorrelate similar odor representations without foreknowledge of the statistical distributions of analyte features. Our system is based on a digital neuromorphic chip consisting of 256 leaky-integrate-and-fire neurons, 1024 × 256 crossbar synapses, and address-event representation communication circuits. The neural circuits configured in the chip reflect established connections among mitral cells, periglomerular cells, external tufted cells, and superficial short-axon cells within the olfactory bulb, and accept input from convergent sets of sensors configured as olfactory sensory neurons. This configuration generates functional transformations comparable to those observed in the glomerular layer of the mammalian olfactory bulb. Our circuits, consuming only 45 pJ of active power per spike with a power supply of 0.85 V, can be used as the first stage of processing in low-power artificial chemical sensing devices inspired by natural olfactory systems.

No MeSH data available.


Related in: MedlinePlus