Abstract: A reconfigurable neural network circuit is provided. The reconfigurable neural network circuit comprises an electronic synapse array including multiple synapses interconnecting a plurality of digital electronic neurons. Each neuron comprises an integrator that integrates input spikes and generates a signal when the integrated inputs exceed a threshold. The circuit further comprises a control module for reconfiguring the synapse array. The control module comprises a global final state machine that controls timing for operation of the circuit, and a priority encoder that allows spiking neurons to sequentially access the synapse array.

Abstract: A reconfigurable neural network circuit is provided. The reconfigurable neural network circuit comprises an electronic synapse array including multiple synapses interconnecting a plurality of digital electronic neurons. Each neuron comprises an integrator that integrates input spikes and generates a signal when the integrated inputs exceed a threshold. The circuit further comprises a control module for reconfiguring the synapse array. The control module comprises a global final state machine that controls timing for operation of the circuit, and a priority encoder that allows spiking neurons to sequentially access the synapse array.

Abstract: A reconfigurable neural network circuit is provided. The reconfigurable neural network circuit comprises an electronic synapse array including multiple synapses interconnecting a plurality of digital electronic neurons. Each neuron comprises an integrator that integrates input spikes and generates a signal when the integrated inputs exceed a threshold. The circuit further comprises a control module for reconfiguring the synapse array. The control module comprises a global final state machine that controls timing for operation of the circuit, and a priority encoder that allows spiking neurons to sequentially access the synapse array.

Abstract: Embodiments of the invention relate to a time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a universal substrate of adaptation. One embodiment comprises a neurosynaptic device including a memory device that maintains neuron attributes for multiple neurons. The module further includes multiple bit maps that maintain incoming firing events for different periods of delay and a multi-way processor. The processor includes a memory array that maintains a plurality of synaptic weights. The processor integrates incoming firing events in a time-division multiplexing manner. Incoming firing events are integrated based on the neuron attributes and the synaptic weights maintained.

Abstract: A reconfigurable neural network circuit is provided. The reconfigurable neural network circuit comprises an electronic synapse array including multiple synapses interconnecting a plurality of digital electronic neurons. Each neuron comprises an integrator that integrates input spikes and generates a signal when the integrated inputs exceed a threshold. The circuit further comprises a control module for reconfiguring the synapse array. The control module comprises a global final state machine that controls timing for operation of the circuit, and a priority encoder that allows spiking neurons to sequentially access the synapse array.

Abstract: Embodiments of the invention relate to a neuromorphic network for producing spike-timing dependent plasticity. The neuromorphic network includes a plurality of electronic neurons and an interconnect circuit coupled for interconnecting the plurality of electronic neurons. The interconnect circuit includes plural synaptic devices for interconnecting the electronic neurons via axon paths, dendrite paths and membrane paths. Each synaptic device includes a variable state resistor and a transistor device with a gate terminal, a source terminal and a drain terminal, wherein the drain terminal is connected in series with a first terminal of the variable state resistor.

Abstract: Embodiments of the invention relate to a neuromorphic network for producing spike-timing dependent plasticity. The neuromorphic network includes a plurality of electronic neurons and an interconnect circuit coupled for interconnecting the plurality of electronic neurons. The interconnect circuit includes plural synaptic devices for interconnecting the electronic neurons via axon paths, dendrite paths and membrane paths. Each synaptic device includes a variable state resistor and a transistor device with a gate terminal, a source terminal and a drain terminal, wherein the drain terminal is connected in series with a first terminal of the variable state resistor.

Abstract: Embodiments of the invention relate to a time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a universal substrate of adaptation. One embodiment comprises a neurosynaptic device including a memory device that maintains neuron attributes for multiple neurons. The module further includes multiple bit maps that maintain incoming firing events for different periods of delay and a multi-way processor. The processor includes a memory array that maintains a plurality of synaptic weights. The processor integrates incoming firing events in a time-division multiplexing manner. Incoming firing events are integrated based on the neuron attributes and the synaptic weights maintained.

Abstract: Embodiments of the invention relate to a time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a neural network. One embodiment comprises maintaining neuron attributes for multiple neurons and maintaining incoming firing events for different time steps. For each time step, incoming firing events for said time step are integrated in a time-division multiplexing manner. Incoming firing events are integrated based on the neuron attributes maintained. For each time step, the neuron attributes maintained are updated in parallel based on the integrated incoming firing events for said time step.

Abstract: Embodiments of the invention relate to a neuromorphic network for producing spike-timing dependent plasticity. The neuromorphic network includes a plurality of electronic neurons and an interconnect circuit coupled for interconnecting the plurality of electronic neurons. The interconnect circuit includes plural synaptic devices for interconnecting the electronic neurons via axon paths, dendrite paths and membrane paths. Each synaptic device includes a variable state resistor and a transistor device with a gate terminal, a source terminal and a drain terminal, wherein the drain terminal is connected in series with a first terminal of the variable state resistor.

Abstract: A method and system are disclosed for measuring a specified parameter in a phase-locked loop frequency synthesizer (PLL). In one embodiment, the method comprises introducing multiple phase errors in the PLL, measuring a specified aspect of the introduced phase errors, and determining a value for the specified parameter using the measured aspects of the introduced phase errors. In one embodiment, the phase errors are introduced repetitively in the PLL, and these phase errors produce a modified phase difference between the reference signal and the feedback signal in the PPL. In one embodiment, crossover times, when this modified phase difference crosses over a preset value, are determined, and these crossover times are used to determine the value for the specified parameter. In an embodiment, the parameter is calculated as a mathematical function of the crossover times. The parameter may be, for example, the bandwidth of the PLL.

Abstract: Embodiments of the invention relate to a time-division multiplexed neurosynaptic module with implicit memory addressing for implementing a neural network. One embodiment comprises maintaining neuron attributes for multiple neurons and maintaining incoming firing events for different time steps. For each time step, incoming firing events for said time step are integrated in a time-division multiplexing manner. Incoming firing events are integrated based on the neuron attributes maintained. For each time step, the neuron attributes maintained are updated in parallel based on the integrated incoming firing events for said time step.

Abstract: A receiver is adapted to receive an input signal having a first voltage swing and to generate an output signal having a second voltage swing, the output signal being indicative of the input signal, the second voltage swing being greater than the first voltage swing. The receiver includes a first sub-rate receiver block and at least a second sub-rate receiver block. A receiver clock is divided into a first sub-rate clock phase and at least a second sub-rate clock phase, the first sub-rate clock phase being used to drive the first sub-rate receiver block and the second sub-rate clock phase being used to drive the second sub-rate receiver block. Each of the first sub-rate receiver block and the second sub-rate receiver block includes at least one gated-diode sense amplifier.

Abstract: A method and system are disclosed for calibrating a mid-voltage node in an integrated circuit including an input-output circuit having charge-recycling stacked voltage domains including at least first and second voltage domains. In one embodiment, the method comprises transmitting data through the input-output circuit, including transmitting a first portion of the data across the first voltage domain, and transmitting a second portion of the data across the second voltage domain. The method further comprises measuring a specified characteristic of the data transmitted through the input-output circuit; and based on the measured specified characteristic, adjusting a voltage of said mid-voltage node to a defined value. The voltage of the mid-voltage node may be adjusted to accomplish a number of objectives, for example, to achieve a desired trade-off between power and performance, or so that the two voltage domains have the same performance.

Abstract: A receiver is adapted to receive an input signal having a first voltage swing and to generate an output signal having a second voltage swing, the output signal being indicative of the input signal, the second voltage swing being greater than the first voltage swing. The receiver includes a first sub-rate receiver block and at least a second sub-rate receiver block. A receiver clock is divided into a first sub-rate clock phase and at least a second sub-rate clock phase, the first sub-rate clock phase being used to drive the first sub-rate receiver block and the second sub-rate clock phase being used to drive the second sub-rate receiver block. Each of the first sub-rate receiver block and the second sub-rate receiver block includes at least one gated-diode sense amplifier.

Abstract: A method and system are disclosed for calibrating a mid-voltage node in an integrated circuit including an input-output circuit having charge-recycling stacked voltage domains including at least first and second voltage domains. In one embodiment, the method comprises transmitting data through the input-output circuit, including transmitting a first portion of the data across the first voltage domain, and transmitting a second portion of the data across the second voltage domain. The method further comprises measuring a specified characteristic of the data transmitted through the input-output circuit; and based on the measured specified characteristic, adjusting a voltage of said mid-voltage node to a defined value. The voltage of the mid-voltage node may be adjusted to accomplish a number of objectives, for example, to achieve a desired trade-off between power and performance, or so that the two voltage domains have the same performance.

Abstract: A method and system are disclosed for measuring a specified parameter in a phase-locked loop frequency synthesizer (PLL). In one embodiment, the method comprises introducing multiple phase errors in the PLL, measuring a specified aspect of the introduced phase errors, and determining a value for the specified parameter using the measured aspects of the introduced phase errors. In one embodiment, the phase errors are introduced repetitively in the PLL, and these phase errors produce a modified phase difference between the reference signal and the feedback signal in the PPL. In one embodiment, crossover times, when this modified phase difference crosses over a preset value, are determined, and these crossover times are used to determine the value for the specified parameter. In an embodiment, the parameter is calculated as a mathematical function of the crossover times. The parameter may be, for example, the bandwidth of the PLL.

Abstract: A method of forming a circuit includes forming a transimpedance amplifier having a first input node and a second input node. The method also includes forming a feedback circuit having a first transistor having a drain terminal connected to the first input node, a source terminal, and a gate terminal, a second transistor having a drain terminal connected to the second input node, a source terminal, and a gate terminal, and a third transistor having a drain terminal connected to the source terminal of the first transistor and the source terminal of the second terminal.

Abstract: Techniques are disclosed for minimizing the effects of soft errors associated with memory devices that are individually accessible. By way of example, a method of organizing a column in a memory array of a memory device protected by an error correction code comprises the step of maximizing a distance of the error correction code by maximizing a physical distance between memory bits associated with a memory line within the column protected by the error correction code. Other soft error protection techniques may include use of a feed forward error correction code or use of a memory operation (e.g., read or write operation) suppress and retry approach.

Abstract: A reconfigurable neural network circuit is provided. The reconfigurable neural network circuit comprises an electronic synapse array including multiple synapses interconnecting a plurality of digital electronic neurons. Each neuron comprises an integrator that integrates input spikes and generates a signal when the integrated inputs exceed a threshold. The circuit further comprises a control module for reconfiguring the synapse array. The control module comprises a global final state machine that controls timing for operation of the circuit, and a priority encoder that allows spiking neurons to sequentially access the synapse array.