In this paper, we present a Machine Learning (ML) Mixture of Experts (MoE) technique to predict the number of iterations needed for a Pathfinder-based FPGA router to complete a routing problem. Given a placed circuit,...
详细信息
ISBN:
(纸本)9798350312058
In this paper, we present a Machine Learning (ML) Mixture of Experts (MoE) technique to predict the number of iterations needed for a Pathfinder-based FPGA router to complete a routing problem. Given a placed circuit, our technique uses features gathered on each routing iteration to predict if the circuit is routable and how many more iterations will be required to successfully route the circuit. This enables early exit for routing problems which are unlikely to be completed in a target number of iterations. Such early exit may help to achieve a successful route within tractable time by allowing the user to quickly retry the circuit compilation with a different random seed, a modified circuit design, or a different FPGA. We demonstrate our predictor in the VTR 8 framework;compared to VTR's predictor, our ML predictor incurs lower prediction errors on the Koios Deep Learning and Titan23 benchmark suites. Based on our tests, equipping VTR with our ML predictor would reduce time wasted on unroutable designs by 31% while also allowing 28% more routable designs to be completed.
In this paper, we propose a cost-effective laser frequency locking scheme based on frequency modulation spectroscopy (FMS) for precision measurements and experiments in various fields. We demonstrate that by digitally...
详细信息
In this paper, we propose a cost-effective laser frequency locking scheme based on frequency modulation spectroscopy (FMS) for precision measurements and experiments in various fields. We demonstrate that by digitally modulating the detected signal frequency using a field-programmable-gate-array (FPGA) driven by a home-built lock-in and a proportional-integral-derivative (PID) control system, our system achieves higher precision, user-friendliness, and versatility. Our system generates a 20 V peak-to-peak amplified voltage to a piezo transducer (PZT), which enables a mode-hop free laser scan of approximately 2 +/- 0.2 GHz. We directly modulate the detected saturation absorption signal with a sinusoidal waveform, then demodulate it to obtain multiple zero crossing locking points on the D2 transition of Cs at 852.35 nm. We optimized the system for three commonly used atomic transitions and found that researchers can select any of the zero crossing peaks for frequency stabilization depending on their experimental requirements, we locked the laser at the crossover transition. We measured the long-time frequency fluctuation and power spectral density and found that the frequency fluctuation of the laser is much less than the natural line width of the D2 transition of Cs. Our results demonstrate that our laser frequency locking scheme is effective and practical for precision measurements and experiments in various fields.
The development of the processing capability has enabled the implementation of advanced control systems for power electronics converters. In this framework, the modular multilevel converter has attracted the attention...
详细信息
ISBN:
(数字)9781665466189
ISBN:
(纸本)9781665466189
The development of the processing capability has enabled the implementation of advanced control systems for power electronics converters. In this framework, the modular multilevel converter has attracted the attention of industry and academia, thanks to the good performance in terms of power quality, wide voltage capability, and fault tolerance, which are key requirement for high power applications. This paper proposes a quasi distribute control architecture based on a FemtoCore platform, an optimized soft-core designed for modular power electronics applications. Latency estimates and simulation results shows the potential of this solution for the control of modular multilevel converters.
Using a frequency comb as a transfer oscillator, 'virtual' or 'transfer' beats between widely separated optical frequencies can be created. These real-time beats between a frequency and a rational frac...
详细信息
ISBN:
(数字)9781665497183
ISBN:
(纸本)9781665497183
Using a frequency comb as a transfer oscillator, 'virtual' or 'transfer' beats between widely separated optical frequencies can be created. These real-time beats between a frequency and a rational fraction of a second optical frequency can be employed for comparing or phase-locking lasers, independent of fluctuations of the frequency comb. We demonstrate an all-digital FPGA-based implementation for the creation of a transfer beat which is simple to scale and can be reconfigured with high flexibility. The noise performance of this all-digital transfer beat implementation is at least on par with its analogue counterpart.
Quantum algorithms can be efficiently emulated on classical hardware such as fieldprogrammablegatearrays (FPGAs), achieving significant speedup over software simulations. However, the increase in the required hardw...
详细信息
ISBN:
(纸本)9781538691700
Quantum algorithms can be efficiently emulated on classical hardware such as fieldprogrammablegatearrays (FPGAs), achieving significant speedup over software simulations. However, the increase in the required hardware resources for emulating quantum systems becomes a critical limitation as the number of qubits is increased. In this paper, we propose a scalable emulation framework for modeling quantum algorithms on FPGAs that employs efficient resource scheduling such as space and space-time scheduling. In addition, full floating point precision arithmetic and dataflow non-linear (dynamic) pipelining are also used to achieve higher accuracy and higher throughput. We propose scalable and optimized hardware architectures for Quantum Fourier Transform (QFT) and Grover's search algorithm and demonstrate the scalability of our framework by scaling the system up to 5 fully-entangled qubits. A multi-node (multi-FPGA), state-of-the-art high-performance reconfigurable computer (HPRC) was used for implementation of the proposed architectures. Our experimental results show that by employing efficient resource scheduling techniques, the hardware resource constraints could be mitigated and the proposed emulation framework could be made feasible for emulation of more complex, larger-scale quantum algorithms while maintaining higher accuracy and throughput than existing work.
Photothermal techniques allow the detection of characteristics of material without invading it. Researchers have developed hardware for some specific Phase and Amplitude detection (Lock-In Function) applications, elim...
详细信息
Photothermal techniques allow the detection of characteristics of material without invading it. Researchers have developed hardware for some specific Phase and Amplitude detection (Lock-In Function) applications, eliminating space and unnecessary electronic functions, among others. This work shows the development of a Digital Lock-In Amplifier based on a fieldprogrammablegatearray (FPGA) for low-frequency applications. This system allows selecting and generating the appropriated frequency depending on the kind of experiment or material studied. The results show good frequency stability in the order of 1.0 x 10(-9) Hz, which is considered good linearity and repeatability response for the most common Laboratory Amplitude and Phase Shift detection devices, with a low error and standard deviation.
New computational solutions are required to understand how atomic-scale properties affect magnetic behaviour at micrometer dimensions. We describe a field-programmable-gate-array (FPGA) based simulation of a dilute an...
详细信息
New computational solutions are required to understand how atomic-scale properties affect magnetic behaviour at micrometer dimensions. We describe a field-programmable-gate-array (FPGA) based simulation of a dilute antiferromagnet with a large number of Ising spins using Glauber dynamics. The simple atomic model qualitatively reproduces experimental findings when scaled up to sufficiently large spatial dimensions, and provides insight into the finite size thresholds separating nanoscale from microscale behaviour. A real-time visualisation module was used to study the dynamics of the fractal domain structure and non-exponential relaxation mechanism. A performance comparison with contemporary GPU and CPU implementations suggests that a FPGA route is a competitive alternative. Crown Copyright (C) 2011 Published by Elsevier B.V. All rights reserved.
暂无评论