Search

This letter introduces a framework for the automatic generation of hardware cores for Artificial Neural Network (ANN)-based chaotic oscillators. The framework trains the model to approximate a chaotic system, then performs design space exploration yielding potential hardware architectures for its implementation. The framework then generates the corresponding synthesizable High-Level Synthesis code and a validation testbench from a selected solution. The hardware design primarily targets FPGAs. The proposed framework offers a rapid hardware design process of candidate architectures superior to manually designed works in terms of hardware cost and throughput. The source code is available on GitHub., Comment: 5 Pages

Multipliers are widely-used arithmetic operators in digital signal processing and machine learning circuits. Due to their relatively high complexity, they can have high latency and be a significant source of power consumption. One strategy to alleviate these limitations is to use approximate computing. This paper thus introduces an original FPGA-based approximate multiplier specifically optimized for machine learning computations. It utilizes dynamically reconfigurable lookup table (LUT) primitives in AMD-Xilinx technology to realize the core part of the computations. The paper provides an in-depth analysis of the hardware architecture, implementation outcomes, and accuracy evaluations of the multiplier proposed in INT8 precision. Implementation results on an AMD-Xilinx Kintex Ultrascale+ FPGA demonstrate remarkable savings of 64% and 67% in LUT utilization for signed multiplication and multiply-and-accumulation configurations, respectively, when compared to the standard Xilinx multiplier core. Accuracy measurements on four popular deep learning (DL) benchmarks indicate a minimal average accuracy decrease of less than 0.29% during post-training deployment, with the maximum reduction staying less than 0.33%. The source code of this work is available on GitHub., Comment: 5 figures, 3 tables

The P4 language has drastically changed the networking field as it allows to quickly describe and implement new networking applications. Although a large variety of applications can be described with the P4 language, current programmable switch architectures impose significant constraints on P4 programs. To address this shortcoming, FPGAs have been explored as potential targets for P4 applications. P4 applications are described using three abstractions: a packet parser, match-action tables, and a packet deparser, which reassembles the output packet with the result of the match-action tables. While implementations of packet parsers and match-action tables on FPGAs have been widely covered in the literature, no general design principles have been presented for the packet deparser. Indeed, implementing a high-speed and efficient deparser on FPGAs remains an open issue because it requires a large amount of interconnections and the architecture must be tailored to a P4 program. As a result, in several works where a P4 application is implemented on FPGAs, the deparser consumes a significant proportion of chip resources. Hence, in this paper, we address this issue by presenting design principles for efficient and high-speed deparsers on FPGAs. As an artifact, we introduce a tool that generates an efficient vendor-agnostic deparser architecture from a P4 program. Our design has been validated and simulated with a cocotb-based framework. The resulting architecture is implemented on Xilinx Ultrascale+ FPGAs and supports a throughput of more than 200 Gbps while reducing resource usage by almost 10$\times$ compared to other solutions., Comment: Presented at ISFPGA'21, 2021 Source code available at : https://github.com/luinaudt/deparser/tree/FPGA_paper

Convolutional Neural Networks (CNNs) have proven to be extremely accurate for image recognition, even outperforming human recognition capability. When deployed on battery-powered mobile devices, efficient computer architectures are required to enable fast and energy-efficient computation of costly convolution operations. Despite recent advances in hardware accelerator design for CNNs, two major problems have not yet been addressed effectively, particularly when the convolution layers have highly diverse structures: (1) minimizing energy-hungry off-chip DRAM data movements; (2) maximizing the utilization factor of processing resources to perform convolutions. This work thus proposes an energy-efficient architecture equipped with several optimized dataflows to support the structural diversity of modern CNNs. The proposed approach is evaluated by implementing convolutional layers of VGGNet-16 and ResNet-50. Results show that the architecture achieves a Processing Element (PE) utilization factor of 98% for the majority of 3x3 and 1x1 convolutional layers, while limiting latency to 396.9 ms and 92.7 ms when performing convolutional layers of VGGNet-16 and ResNet-50, respectively. In addition, the proposed architecture benefits from the structured sparsity in ResNet-50 to reduce the latency to 42.5 ms when half of the channels are pruned., Comment: 12 pages

The emergence of P4, a domain specific language, coupled to PISA, a domain specific architecture, is revolutionizing the networking field. P4 allows to describe how packets are processed by a programmable data plane, spanning ASICs and CPUs, implementing PISA. Because the processing flexibility can be limited on ASICs, while the CPUs performance for networking tasks lag behind, recent works have proposed to implement PISA on FPGAs. However, little effort has been dedicated to analyze whether FPGAs are good candidates to implement PISA. In this work, we take a step back and evaluate the micro-architecture efficiency of various PISA blocks. We demonstrate, supported by a theoretical and experimental analysis, that the performance of a few PISA blocks is severely limited by the current FPGA architectures. Specifically, we show that match tables and programmable packet schedulers represent the main performance bottlenecks for FPGA-based programmable switches. Thus, we explore two avenues to alleviate these shortcomings. First, we identify network applications well tailored to current FPGAs. Second, to support a wider range of networking applications, we propose modifications to the FPGA architectures which can also be of interest out of the networking field., Comment: To be published in : IEEE International Conference on High Performance Switching and Routing 2020

The success of neural networks in image classification has inspired various hardware implementations on embedded platforms such as Field Programmable Gate Arrays, embedded processors and Graphical Processing Units. These embedded platforms are constrained in terms of power, which is mainly consumed by the Multiply Accumulate operations and the memory accesses for weight fetching. Quantization and pruning have been proposed to address this issue. Though effective, these techniques do not take into account the underlying architecture of the embedded hardware. In this work, we propose PoET-BiN, a Look-Up Table based power efficient implementation on resource constrained embedded devices. A modified Decision Tree approach forms the backbone of the proposed implementation in the binary domain. A LUT access consumes far less power than the equivalent Multiply Accumulate operation it replaces, and the modified Decision Tree algorithm eliminates the need for memory accesses. We applied the PoET-BiN architecture to implement the classification layers of networks trained on MNIST, SVHN and CIFAR-10 datasets, with near state-of-the art results. The energy reduction for the classifier portion reaches up to six orders of magnitude compared to a floating point implementations and up to three orders of magnitude when compared to recent binary quantized neural networks., Comment: Accepted in MLSys 2020 conference

Convolutional Neural Networks (CNNs) have shown outstanding accuracy for many vision tasks during recent years. When deploying CNNs on portable devices and embedded systems, however, the large number of parameters and computations result in long processing time and low battery life. An important factor in designing CNN hardware accelerators is to efficiently map the convolution computation onto hardware resources. In addition, to save battery life and reduce energy consumption, it is essential to reduce the number of DRAM accesses since DRAM consumes orders of magnitude more energy compared to other operations in hardware. In this paper, we propose an energy-efficient architecture which maximally utilizes its computational units for convolution operations while requiring a low number of DRAM accesses. The implementation results show that the proposed architecture performs one image recognition task using the VGGNet model with a latency of 393 ms and only 251.5 MB of DRAM accesses., Comment: 4 pages

High-Level Synthesis (HLS) brings FPGAs to audiences previously unfamiliar to hardware design. However, achieving the highest Quality-of-Results (QoR) with HLS is still unattainable for most programmers. This requires detailed knowledge of FPGA architecture and hardware design in order to produce FPGA-friendly codes. Moreover, these codes are normally in conflict with best coding practices, which favor code reuse, modularity, and conciseness. To overcome these limitations, we propose Module-per-Object (MpO), a human-driven HLS design methodology intended for both hardware designers and software developers with limited FPGA expertise. MpO exploits modern C++ to raise the abstraction level while improving QoR, code readability and modularity. To guide HLS designers, we present the five characteristics of MpO classes. Each characteristic exploits the power of HLS-supported modern C++ features to build C++-based hardware modules. These characteristics lead to high-quality software descriptions and efficient hardware generation. We also present a use case of MpO, where we use C++ as the intermediate language for FPGA-targeted code generation from P4, a packet processing domain specific language. The MpO methodology is evaluated using three design experiments: a packet parser, a flow-based traffic manager, and a digital up-converter. Based on experiments, we show that MpO can be comparable to hand-written VHDL code while keeping a high abstraction level, human-readable coding style and modularity. Compared to traditional C-based HLS design, MpO leads to more efficient circuit generation, both in terms of performance and resource utilization. Also, the MpO approach notably improves software quality, augmenting parametrization while eliminating the incidence of code duplication., Comment: 9 pages. Paper accepted for publication at The 27th IEEE International Symposium on Field-Programmable Custom Computing Machines, San Diego CA, April 28 - May 1, 2019

Due to the emergence of new network applications, current IP lookup engines must support high-bandwidth, low lookup latency and the ongoing growth of IPv6 networks. However, existing solutions are not designed to address jointly those three requirements. This paper introduces SHIP, an IPv6 lookup algorithm that exploits prefix characteristics to build a two-level data structure designed to meet future application requirements. Using both prefix length distribution and prefix density, SHIP first clusters prefixes into groups sharing similar characteristics, then it builds a hybrid trie-tree for each prefix group. The compact and scalable data structure built can be stored in on-chip low-latency memories, and allows the traversal process to be parallelized and pipelined at each level in order to support high packet bandwidth. Evaluated on real and synthetic prefix tables holding up to 580 k IPv6 prefixes, SHIP has a logarithmic scaling factor in terms of the number of memory accesses, and a linear memory consumption scaling. Using the largest synthetic prefix table, simulations show that compared to other well-known approaches, SHIP uses at least 44% less memory per prefix, while reducing the memory latency by 61%., Comment: Submitted to EEE/ACM Transactions on Networking

Quantum Key Distribution is a practically implementable information-theoretic secure method for transmitting keys to remote partners performing quantum communication. After examining various protocols from the simplest such as QC and BB84 we move on to describe BBM92, DPSK, SARG04 and finally MDI from the largest possible communication distance and highest secret key bitrate. We discuss how any protocol can be optimized by reviewing the various steps and underlying assumptions proper to every protocol with the corresponding consequence in each case., Comment: 29 pages and 10 figures

Packet parsing is a key step in SDN-aware devices. Packet parsers in SDN networks need to be both reconfigurable and fast, to support the evolving network protocols and the increasing multi-gigabit data rates. The combination of packet processing languages with FPGAs seems to be the perfect match for these requirements. In this work, we develop an open-source FPGA-based configurable architecture for arbitrary packet parsing to be used in SDN networks. We generate low latency and high-speed streaming packet parsers directly from a packet processing program. Our architecture is pipelined and entirely modeled using templated C++ classes. The pipeline layout is derived from a parser graph that corresponds a P4 code after a series of graph transformation rounds. The RTL code is generated from the C++ description using Xilinx Vivado HLS and synthesized with Xilinx Vivado. Our architecture achieves 100 Gb/s data rate in a Xilinx Virtex-7 FPGA while reducing the latency by 45% and the LUT usage by 40% compared to the state-of-the-art., Comment: Accepted for publication at the 26th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays February 25 - 27, 2018 Monterey Marriott Hotel, Monterey, California, 7 pages, 7 figures, 1 table

The measurement device independent (MDI) Quantum Key Distribution (QKD) is a practically implementable method for transmitting secret keys between respective partners performing quantum communication. SARG04 (Scarani-Ac\`{i}n-Ribordy-Gisin 2004) is a protocol tailored to struggle against photon number splitting (PNS) attacks by eavesdroppers and its MDI-QKD version is reviewed and optimized from secret key bitrate versus communication distance point of view. We consider the effect of several important factors such as error correction function, dark counting parameter and quantum efficiency in order to achieve the largest key bitrate versus longest communication distance., Comment: 10 pages and 7 figures

This paper presents a memory efficient architecture that implements the Multi-Scale Line Detector (MSLD) algorithm for real-time retinal blood vessel detection in fundus images on a Zynq FPGA. This implementation benefits from the FPGA parallelism to drastically reduce the memory requirements of the MSLD from two images to a few values. The architecture is optimized in terms of resource utilization by reusing the computations and optimizing the bit-width. The throughput is increased by designing fully pipelined functional units. The architecture is capable of achieving a comparable accuracy to its software implementation but 70x faster for low resolution images. For high resolution images, it achieves an acceleration by a factor of 323x., Comment: This paper was accepted and presented at Conference on Design and Architectures for Signal and Image Processing - DASIP 2016

P4 is an emergent packet-processing language with which the user can describe how the packets are to be processed in a switching element. This paper presents a way to implement complex operations that are not natively supported in P4. In this work, we explored two different methods to add extensions to P4: i) using new native primitives and ii) using extern instances. As a case study, an ROHC entity was implemented and invoked in a P4 program. The tests showed similar relative performance in both methods in terms of normalized packet latency. However, extern instances appear to be more suitable for target-specific switching applications, where the manufacturer/vendor can specify its own specific operations without changes in the P4 syntax and semantics. Extern instances only require changes in the target-specific backend compiler while keeping the P4 frontend compiler unchanged. The use of externs also results in a more elegant code solution since they are implemented outside the switch-core, thus reducing side effects risks that can be caused by a modification in a switch pipeline implementation., Comment: 6 pages, 4 figures, 3 listings

Implementing an accurate and fast activation function with low cost is a crucial aspect to the implementation of Deep Neural Networks (DNNs) on FPGAs. We propose a high-accuracy approximation approach for the hyperbolic tangent activation function of artificial neurons in DNNs. It is based on the Discrete Cosine Transform Interpolation Filter (DCTIF). The proposed architecture combines simple arithmetic operations on stored samples of the hyperbolic tangent function and on input data. The proposed DCTIF implementation achieves two orders of magnitude greater precision than previous work while using the same or fewer computational resources. Various combinations of DCTIF parameters can be chosen to tradeoff the accuracy and complexity of the hyperbolic tangent function. In one case, the proposed architecture approximates the hyperbolic tangent activation function with 10E-5 maximum error while requiring only 1.52 Kbits memory and 57 LUTs of a Virtex-7 FPGA. We also discuss how the activation function accuracy affects the performance of DNNs in terms of their training and testing accuracies. We show that a high accuracy approximation can be necessary in order to maintain the same DNN training and testing performances realized by the exact function., Comment: 8 pages, 6 figures, 5 tables, submitted for the 25th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (ISFPGA), 22-24 February 2017, California, USA

Secure communication based on message encryption might be performed by combining the message with controlled noise (called pseudo-noise) as performed in Spread-Spectrum communication used presently in Wi-Fi and Smartphone Telecommunication systems. Quantum communication based on entanglement is another route for securing communications as demonstrated by several important experiments described in this work. The central role played by the photon in unifying the description of Classical and Quantum noise as major ingredients of secure communication systems is highlighted and described on the basis of the classical and quantum fluctuation dissipation theorems., Comment: 19 pages, 9 figures. Published as a Review paper in European Journal of Physics

The Canonical Function Method (CFM) is a powerful accurate and fast method that solves the Schr\"{o}dinger equation for the eigenvalues directly without having to evaluate the eigenfunctions. Its versatility allows to solve several types of problems and in this work it is applied to the solution of several 1D potential problems, the 3D Hydrogen atom and the Morse potential., Comment: 10 pages, 9 figures. To submit to American Journal of Physics

The Canonical Function Method (CFM) is a powerful method that solves the radial Schr\"{o}dinger equation for the eigenvalues directly without having to evaluate the eigenfunctions. It is applied to various quantum mechanical problems in Atomic and Molecular physics with presence of regular or singular potentials. It has also been developed to handle single and multiple channel scattering problems where the phaseshift is required for the evaluation of the scattering cross-section. Its controllable accuracy makes it a valuable tool for the evaluation of vibrational levels of cold molecules, a sensitive test of Bohr correspondance principle and a powerful method to tackle local and non-local spin dependent problems., Comment: 30 pages, 12 figures- To submit to Reviews of Modern Physics

Quantum Defect theory is a well established theoretical concept in modern spectroscopy. We show that this approach is useful in electron impact ionization problems where state of the art theoretical methods are presently restricted mostly to simple atomic targets. For the well documented Argon ionization case in equal energy sharing geometry the approach suggested leads to significant improvements compared to previous calculations., Comment: 13 pages, 3 figures, published in: Many Particle Spectroscopy of Atoms, Molecules, Clusters and Surfaces, Kluwer (New-York 2001)

It is shown how the Canonical Function approach can be used to obtain accurate solutions for the distorted wave problem taking account of direct static and polarisation potentials and exact non-local exchange. Calculations are made for electrons in the field of atomic hydrogen and the phaseshifts are compared with those obtained using a modified form of the DWPO code of McDowell and collaborators: for small wavenumbers our approach avoids numerical instabilities otherwise present. Comparison is also made with phaseshifts calculated using local equivalent-exchange potentials and it is found that these are inaccurate at small wavenumbers. Extension of our method to the case of atoms having other than s-type outer shells is dicussed., Comment: Submitted to J. Phys. B (June 2002)

Vibrational spectra of long-range molecules are determined accurately and to arbitrary accuracy with the Canonical Function Method. The energy levels of the $0^-_g$ and $1_u$ electronic states of the $^{23}{\rm Na}_2$ molecule are determined from the Ground state up to the continuum limit. The method is validated by comparison with previous results obtained by Stwalley et al. using the same potential and Trost et al. whose work is based on the Lennard-Jones potential adapted to long-range molecules., Comment: 19 pages, 5 figures and 6 tables. To be published in the G. Herzberg memorial issue, Can. J. Physics Vol. 79 (2001)

The canonical function method (CFM) is a powerful means for solving the Radial Schrodinger Equation. The mathematical difficulty of the RSE lies in the fact it is a singular boundary value problem. The CFM turns it into a regular initial value problem and allows the full determination of the spectrum of the Schrodinger operator without calculating the eigenfunctions. Following the parametrisation suggested by Klapisch and Green, Sellin and Zachor we develop a CFM to optimise the potential parameters in order to reproduce the experimental Quantum Defect results for various Rydberg series of He, Ne and Ar as evaluated from Moore's data., Comment: Published in J. Phys. IV France (1999), 9 pages, 1 figure and 3 tables

Gao applied LeRoy and Bernstein semi-classical analysis for the energy levels in a potential of the form -C/r^n to sequences of scaled energy differences progressing towards low lying states and found a better agreement with the semi-classical prediction. We checked that for the energy levels obtained by Stwalley et al. with the same potential, the agreement with the semi-classical approximation is better for higher vibrational quantum numbers in agreement with Bohr's correspondence principle., Comment: Submitted to Phys. Rev. Lett. as a Comment on February 2000, 4 pages, no figures, 1 table

The Landau-Lifshitz (L-L) equation describing the time dependence of the magnetisation vector is numerically integrated fully without any simplifying assumptions in the time domain and the magnetisation time series obtained is Fourier transformed (FFT) to yield the permeability spectrum up to 10 GHz. The non linear results are compared to the experimental results obtained on magnetic amorphous thin films of Co-Zr, Co-Zr-Re. We analyse our results with the frequency response obtained directly from the Landau-Lifshitz equation as well as with the second order Gilbert frequency response., Comment: 4 pages, 5 figures, published in J. Phys.IV France Vol. 8, Pr2-331 (1998)