Synaptic Neural Network Core Based Sensor System

• US 20160132766A1
• Filed: 11/07/2014
• Published: 05/12/2016
• Est. Priority Date: 11/07/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adams, Samuel S., Belluomini, Wendy A., Friedlander, Robert R., Kraemer, James R.
• Granted Patent: US 9,881,253 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 1/26 Power supply means, e.g. regulation thereof for memories G11C
• CPC Class Codes: G06N 3/049 Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
• CPC Class Codes: G06N 3/063 using electronic means

Abstract

A sensor system comprises: an energy storage device; an intermittent energy release device electrically coupled to the energy storage device, wherein the intermittent energy release device causes the energy storage device to release stored energy intermittently; a sensor electrically coupled to the energy storage device; a register electrically coupled to the sensor, wherein the register stores readings from the sensor; a synaptic neural network core electrically coupled to the sensor, wherein the synaptic neural network core converts the readings from the sensor into a synthetic context-based object that is derived from the readings and a context object; a transponder electrically coupled to the synaptic neural network core; and a storage buffer within the transponder, wherein the storage buffer stores the synthetic context-based object for transmission by the transponder to a monitoring system.

Claims

1. A sensor system comprising:

an energy storage device;

an intermittent energy release device electrically coupled to the energy storage device, wherein the intermittent energy release device causes the energy storage device to release stored energy intermittently;

a sensor electrically coupled to the energy storage device;

a synaptic neural network core electrically coupled to the sensor, wherein the synaptic neural network core converts the readings from the sensor into a synthetic context-based object that is derived from the readings and a context object;

a transponder electrically coupled to the synaptic neural network core; and

a storage buffer within the transponder, wherein the storage buffer stores the synthetic context-based object for transmission by the transponder to a monitoring system.

2. The sensor system of claim 1, further comprising:

a register electrically coupled to the sensor, wherein the register stores readings from the sensor.

3. The sensor system of claim 1, further comprising:

synaptic connections that electrically connect electronic neurons within the synaptic neural network core, wherein the synaptic connections convert the sensor readings from the sensor into a non-contextual data object that is used, with the context object, to generate the synthetic context-based object.

4. The sensor system of claim 1, further comprising:

an ambient power collection device coupled to the energy storage device, wherein the ambient power collection device converts ambient forces into electricity.

5. The sensor system of claim 4, wherein the ambient forces are from a group consisting of radio frequency energy, heat, electrical induction forces, acceleration forces, and vibration.

6. The sensor system of claim 1, wherein the intermittent energy release device comprises a breakdown diode.

7. The sensor system of claim 1, further comprising:

a sensor identification register, wherein the sensor identification register stores an identifier of the sensor as the context object;

a set of one or more excitatory sensing units within the sensor, wherein excitatory sensing units produce a first type of signal in response to sensing a first type of physical event associated with a physical condition, and wherein the first type of signal is transmitted on a first type of electronic neuron;

a set of one or more inhibitory sensing units within the sensor, wherein inhibitory sensing units produce a second type of signal in response to sensing a second type of physical event associated with the physical condition, and wherein the second type of signal is transmitted on a second type of electronic neuron;

a set of synaptic connections within the synaptic neural network core that selectively couple the first type of electronic neuron and the second type of electronic neuron to a non-contextual object register, wherein the non-contextual object register stores non-contextual data received from the set of synaptic connections; and

a synthetic event synaptic neural network core electrically coupled to the synthetic event descriptor register and the sensor identification register, wherein the synthetic event synaptic neural network core generates a synthetic event descriptor from contents of the non-contextual object register and the sensor identification register.

8. The sensor system of claim 1, further comprising:

a resistor that electrically couples the intermittent energy release device to the sensor, wherein the resistor is sized to allow electrical power to be transmitted to the sensor for a predetermined amount of time.

9. A method of optimizing sensor operations, the method comprising:

storing electrical energy on an energy storage device;

intermittently releasing stored electrical energy from the energy storage device to a sensor, wherein intermittently released stored electrical energy from the energy storage device activates one or more sensing units in the sensor;

capturing sensor readings by the one or more sensing units in the sensor;

transmitting the readings to a register for storage;

loading the readings from the register onto a synaptic neural network core;

converting, by the synaptic neural network core, the readings into a synthetic event identifier, wherein the synthetic event identifier is generated from the readings and a context object;

loading the synthetic event identifier onto a register on a transponder device; and

transmitting the synthetic event identifier from the transponder device to a monitoring system.

10. The method of claim 9, wherein the context object is an identifier of a sensor type for the sensor.

11. The method of claim 9, further comprising:

generating power from ambient forces, wherein the power is generated by an ambient power collection device that is coupled to the energy storage device, wherein the ambient power collection device converts ambient forces into electricity, and wherein the ambient forces are from a group consisting of radiofrequency energy, heat, electrical induction forces, acceleration forces, and vibration.

12. The method of claim 9, wherein the intermittent energy release device comprises a breakdown diode.

13. The method of claim 9, further comprising:

storing, on a sensor identification register, an identifier of a sensor type for the sensor as the context object;

producing, by a set of one or more excitatory sensing units within the sensor, a first type of signal in response to sensing a first type of physical event associated with a physical condition, and wherein the first type of signal is transmitted on a first type of electronic neuron;

producing, by a set of one or more inhibitory sensing units within the sensor, a second type of signal in response to sensing a second type of physical event associated with the physical condition, wherein the second type of signal is transmitted on a second type of electronic neuron;

selectively coupling, by a set of synaptic connections within the synaptic neural network core, the first type of electronic neuron and the second type of electronic neuron to a non-contextual object register, and wherein the non-contextual object register stores non-contextual data received from the set of synaptic connections; and

generating, by a synthetic event synaptic neural network core that is electrically coupled to the synthetic event descriptor register and the sensor identification register, a synthetic event descriptor from contents of the non-contextual object register and the sensor identification register.

14. A sensor system comprising:

an energy storage device;

an intermittent energy release device electrically coupled to the energy storage device, wherein the intermittent energy release device causes the energy storage device to release stored energy intermittently;

a synaptic neural network core electrically coupled to the intermittent energy release device, wherein the synaptic neural network core converts the readings directly received from a sensor into a synthetic context-based object that is derived from the readings and a context object;

a transponder electrically coupled to the synaptic neural network core; and

a storage buffer within the transponder, wherein the storage buffer stores the synthetic context-based object for transmission by the transponder to a monitoring system.

15. The sensor system of claim 14, further comprising:

an ambient power collection device coupled to the energy storage device, wherein the ambient power collection device converts ambient forces into electricity.

16. The sensor system of claim 15, wherein the ambient forces are from a group consisting of radiofrequency energy, heat, electrical induction forces, acceleration forces, and vibration.

17. The sensor system of claim 14, wherein the intermittent energy release device comprises a breakdown diode.

18. The sensor system of claim 14, wherein the intermittent energy release device comprises a breakdown transistor.

19. The sensor system of claim 14, further comprising:

a sensor identification register, wherein the sensor identification register stores an identifier of the sensor as the context object;

a set of one or more excitatory sensing units within the sensor, wherein excitatory sensing units produce a first type of signal in response to sensing a first type of physical event associated with a physical condition, and wherein the first type of signal is transmitted on a first type of electronic neuron;

a set of one or more inhibitory sensing units within the sensor, wherein inhibitory sensing units produce a second type of signal in response to sensing a second type of physical event associated with the physical condition, and wherein the second type of signal is transmitted on a second type of electronic neuron;

a set of synaptic connections within the synaptic neural network core that selectively couple the first type of electronic neuron and the second type of electronic neuron to a non-contextual object register, wherein the non-contextual object register stores non-contextual data received from the set of synaptic connections; and

a synthetic event synaptic neural network core electrically coupled to the synthetic event descriptor register and the sensor identification register, wherein the synthetic event synaptic neural network core generates a synthetic event descriptor from contents of the non-contextual object register and the sensor identification register.

20. The sensor system of claim 14, further comprising:

a resistor that electrically couples the intermittent energy release device to the sensor, wherein the resistor is sized to allow electrical power to be transmitted to the sensor for a predetermined amount of time.