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

Chip data illustrating the enhanced signal-to-noise ratio (SNR) in a mitral cell compared to an OSN. A constant low-concentration odor was presented between 30 and 50 ms (solid black line). This increases the spiking probability of individual OSNs during one time step (1 ms) from 0.1 to 0.15. Background spiking is depicted by open circles; additional odor-evoked spikes are depicted by solid black dots. Since multiple OSNs converge on one mitral cell, mitral cell activity during the stimulus period exhibits a significant increase over its background level. The signal-to-noise ratio in an OSN is 0.3 while that in the mitral cell is 0.8.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 7: Chip data illustrating the enhanced signal-to-noise ratio (SNR) in a mitral cell compared to an OSN. A constant low-concentration odor was presented between 30 and 50 ms (solid black line). This increases the spiking probability of individual OSNs during one time step (1 ms) from 0.1 to 0.15. Background spiking is depicted by open circles; additional odor-evoked spikes are depicted by solid black dots. Since multiple OSNs converge on one mitral cell, mitral cell activity during the stimulus period exhibits a significant increase over its background level. The signal-to-noise ratio in an OSN is 0.3 while that in the mitral cell is 0.8.

Mentions: With 48 glomerular columns, each receiving input from a distinct type of chemosensor, combinatorial representation in principle can represent a large number of different odorants. For example, if the chemosensors each have m distinct activation levels, then m48 separate odors can be represented in a single core. Noise in the chemosensors can be partially compensated by the convergence of multiple sensors of the same type onto one glomerulus, increasing the signal-to-noise ratio as well as the maximum sensitivity to low-concentration odorants. Consequently, the output of glomerular circuitry can offer a considerably more robust and reliable representation of low-concentration odors than can individual chemosensors. An example measurement from the chip is shown in Figure 7. We here define the signal-to-noise ratio as the number of spikes induced by odorant presentation divided by the total number of spikes (i.e., activity due to the stimulus plus background activity) during the presentation interval.


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)

Chip data illustrating the enhanced signal-to-noise ratio (SNR) in a mitral cell compared to an OSN. A constant low-concentration odor was presented between 30 and 50 ms (solid black line). This increases the spiking probability of individual OSNs during one time step (1 ms) from 0.1 to 0.15. Background spiking is depicted by open circles; additional odor-evoked spikes are depicted by solid black dots. Since multiple OSNs converge on one mitral cell, mitral cell activity during the stimulus period exhibits a significant increase over its background level. The signal-to-noise ratio in an OSN is 0.3 while that in the mitral cell is 0.8.
© Copyright Policy - open-access
Related In: Results  -  Collection

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

Figure 7: Chip data illustrating the enhanced signal-to-noise ratio (SNR) in a mitral cell compared to an OSN. A constant low-concentration odor was presented between 30 and 50 ms (solid black line). This increases the spiking probability of individual OSNs during one time step (1 ms) from 0.1 to 0.15. Background spiking is depicted by open circles; additional odor-evoked spikes are depicted by solid black dots. Since multiple OSNs converge on one mitral cell, mitral cell activity during the stimulus period exhibits a significant increase over its background level. The signal-to-noise ratio in an OSN is 0.3 while that in the mitral cell is 0.8.
Mentions: With 48 glomerular columns, each receiving input from a distinct type of chemosensor, combinatorial representation in principle can represent a large number of different odorants. For example, if the chemosensors each have m distinct activation levels, then m48 separate odors can be represented in a single core. Noise in the chemosensors can be partially compensated by the convergence of multiple sensors of the same type onto one glomerulus, increasing the signal-to-noise ratio as well as the maximum sensitivity to low-concentration odorants. Consequently, the output of glomerular circuitry can offer a considerably more robust and reliable representation of low-concentration odors than can individual chemosensors. An example measurement from the chip is shown in Figure 7. We here define the signal-to-noise ratio as the number of spikes induced by odorant presentation divided by the total number of spikes (i.e., activity due to the stimulus plus background activity) during the presentation interval.

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