This work presents a 4-channel, mm-scale, electrostatic and piezoelectric actuator driver that uses < 1μA total quiescent bias current and can drive actuator loads up to 120-330V at frequencies over 1kHz. The driver achieves over 99% current efficiency and can operate untethered with an integrated photovoltaic array driven by a collimated or diffuse optical power source. The circuit is tested with an off-chip boost circuit, generating over 1.5kV with 85% power efficiency at 45mW load. The system uses a simple 4-bit CMOS logic level interface with 100 kHz clock to actuate high voltage channels; on-chip photovoltaics also power the digital controller, and I/O bus.
more »
« less
A 120-330V, sub-µA, optically powered microrobotic drive IC for DARPA SHRIMP
This work presents a 4-channel, mm-scale, electro-static and piezoelectric actuator driver that uses< 1 µA total quiescent bias current and can drive actuator loads up to 120-330 V at frequencies over 1kHz. The driver achieves over 99% current efficiency and can operate untethered with an integrated photovoltaic array powered by a collimated or diffuse optical power source. The circuit is demonstrated also as a driver for an off-chip boost circuit, generating over 1.5 kV with 85% power efficiency at 45mW load. The system uses a simple 4-bit CMOS logic level interface with 100 kHz clock to actuate high voltage channels; on-chip photovoltaics also power the digital controller, and I/O bus.
more »
« less
- Award ID(s):
- 1711077
- NSF-PAR ID:
- 10181240
- Date Published:
- Journal Name:
- GOMACTech 2020
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Brain-inspired cognitive computing has so far followed two major approaches - one uses multi-layered artificial neural networks (ANNs) to perform pattern-recognition-related tasks, whereas the other uses spiking neural networks (SNNs) to emulate biological neurons in an attempt to be as efficient and fault-tolerant as the brain. While there has been considerable progress in the former area due to a combination of effective training algorithms and acceleration platforms, the latter is still in its infancy due to the lack of both. SNNs have a distinct advantage over their ANN counterparts in that they are capable of operating in an event-driven manner, thus consuming very low power. Several recent efforts have proposed various SNN hardware design alternatives, however, these designs still incur considerable energy overheads.In this context, this paper proposes a comprehensive design spanning across the device, circuit, architecture and algorithm levels to build an ultra low-power architecture for SNN and ANN inference. For this, we use spintronics-based magnetic tunnel junction (MTJ) devices that have been shown to function as both neuro-synaptic crossbars as well as thresholding neurons and can operate at ultra low voltage and current levels. Using this MTJ-based neuron model and synaptic connections, we design a low power chip that has the flexibility to be deployed for inference of SNNs, ANNs as well as a combination of SNN-ANN hybrid networks - a distinct advantage compared to prior works. We demonstrate the competitive performance and energy efficiency of the SNNs as well as hybrid models on a suite of workloads. Our evaluations show that the proposed design, NEBULA, is up to 7.9× more energy efficient than a state-of-the-art design, ISAAC, in the ANN mode. In the SNN mode, our design is about 45× more energy-efficient than a contemporary SNN architecture, INXS. Power comparison between NEBULA ANN and SNN modes indicates that the latter is at least 6.25× more power-efficient for the observed benchmarks.more » « less
-
Miniaturization of the neuromodulation system is important for non-invasive or sub-invasive optogenetic application. This work presents an optimized wireless power transfer (WPT) system integrated with an on-chip rectification circuitry and an off-chip stimulation circuitry for optogenetic stimulation of freely moving rodents. The proposed WPT system is built using parallel transmitter (TX) coils on printed circuit board (PCB) and wire-wound based receiver (RX) coil followed by a seven-stage voltage doubler and a low dropout regulator (LDO) circuit designed in 180 nm standard Complementary Metal Oxide Semiconductor (CMOS) process. A pulse stimulation is used to stimulate the neurons which is generated using a commercially available off-the-shelf (COTS) components based oscillator circuit. The intensity of the stimulation is controlled by using a COTS based LED driver circuit which controls the current through the μ LED. The total dimension of the RX coil is 8 mm × 3.4 mm. The maximum power transfer efficiency (PTE) of the proposed WPT system is ∼ 35% and the power conversion efficiency (PCE) of the rectifier is 52%. The proposed system with reconfigurable stimulation frequency is suitable for exciting different brain areas for long-term health monitoring.more » « less
-
Spiking Neural Networks (SNN) are fast emerging as an alternative option to Deep Neural Networks (DNN). They are computationally more powerful and provide higher energy-efficiency than DNNs. While exciting at first glance, SNNs contain security-sensitive assets (e.g., neuron threshold voltage) and vulnerabilities (e.g., sensitivity of classification accuracy to neuron threshold voltage change) that can be exploited by the adversaries. We explore global fault injection attacks using external power supply and laser-induced local power glitches on SNN designed using common analog neurons to corrupt critical training parameters such as spike amplitude and neuron’s membrane threshold potential. We also analyze the impact of power-based attacks on the SNN for digit classification task and observe a worst-case classification accuracy degradation of −85.65%. We explore the impact of various design parameters of SNN (e.g., learning rate, spike trace decay constant, and number of neurons) and identify design choices for robust implementation of SNN. We recover classification accuracy degradation by 30–47% for a subset of power-based attacks by modifying SNN training parameters such as learning rate, trace decay constant, and neurons per layer. We also propose hardware-level defenses, e.g., a robust current driver design that is immune to power-oriented attacks, improved circuit sizing of neuron components to reduce/recover the adversarial accuracy degradation at the cost of negligible area, and 25% power overhead. We also propose a dummy neuron-based detection of voltage fault injection at ∼1% power and area overhead each.more » « less
-
Dynamic voltage and frequency scaling (DVFS) is a well-known technique to reduce the power and/or energy consumption of various applications. While most processors provide chip-level DVFS, where the frequency and voltage of the cores in a chip can only be changed all together; core-level DVFS, where each core can be controlled independently, requires core-level voltage regulators in hardware and only is supported in production in Haswell generation among Intel processors. The finer grained control that per-core DVFS provides can lead to higher energy efficiency compared to chip-level DVFS especially for the unsynchronized, unstructured parallel applications when carefully applied. Ability to do per-core DVFS opens up new doors for different optimizations within runtime systems. We implement an intelligent energy efficient runtime module which uses a fine-grained function level per-core DVFS approach. Our module finds the energy-optimal frequency for each phase/function/kernel of the application over the first few iterations and applies the optimal frequency for each function. We test our implementation on Haswell processors and show that our algorithm enables 4% to 35% energy reduction over chip-level DVFS with as much as performance.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
