FPGA Implementations of the Round Two SHA-3 Candidates

The second round of the NIST-run public competition is underway to find a new hash algorithm(s) for inclusion in the NIST Secure Hash Standard (SHA-3). This paper presents the full implementations of all of the second round candidates in hardware with all of their variants. In order to determine their computational efficiency, an important aspect in NIST's round two evaluation criteria, this paper gives an area/speed comparison of each design both with and without a hardware interface, thereby giving an overall impression of their performance in resource constrained and resource abundant environments. The implementation results are provided for a Virtex-5 FPGA device. The efficiency of the architectures for the hash functions are compared in terms of throughput per unit area. To the best of the authors' knowledge, this is the first work to date to present hardware designs which test for all message digest sizes (224, 256, 384, 512), and also the only work to include the padding as part of the hardware for the SHA-3 hash functions.

FPGA based Soft-core SIMD Processing: A MIMO-OFDM Fixed-Complexity Sphere Decoder Case Study

To enable reliable data transfer in next generation Multiple-Input Multiple-Output (MIMO) communication systems, terminals must be able to react to fluctuating channel conditions by having flexible modulation schemes and antenna configurations. This creates a challenging real-time implementation problem: to provide the high performance required of cutting edge MIMO standards, such as 802.11n, with the flexibility for this behavioural variability. FPGA softcore processors offer a solution to this problem, and in this paper we show how heterogeneous SISD/SIMD/MIMD architectures can enable programmable multicore architectures on FPGA with similar performance and cost as traditional dedicated circuit-based architectures. When applied to a 4×4 16-QAM Fixed-Complexity Sphere Decoder (FSD) detector we present the first soft-processor based solution for real-time 802.11n MIMO.

Generic low-latency NoC router architecture for FPGA computing systems

A novel cost-effective and low-latency wormhole router for packet-switched NoC designs, tailored for FPGA, is presented. This has been designed to be scalable at system level to fully exploit the characteristics and constraints of FPGA based systems, rather than custom ASIC technology. A key feature is that it achieves a low packet propagation latency of only two cycles per hop including both router pipeline delay and link traversal delay - a significant enhancement over existing FPGA designs - whilst being very competitive in terms of performance and hardware complexity. It can also be configured in various network topologies including 1-D, 2-D, and 3-D. Detailed design-space exploration has been carried for a range of scaling parameters, with the results of various design trade-offs being presented and discussed. By taking advantage of abundant buildin reconfigurable logic and routing resources, we have been able to create a new scalable on-chip FPGA based router that exhibits high dimensionality and connectivity. The architecture proposed can be easily migrated across many FPGA families to provide flexible, robust and cost-effective NoC solutions suitable for the implementation of high-performance FPGA computing systems. © 2011 IEEE.

Power Modeling and Capping for Heterogeneous ARM/FPGA SoCs

Low-power processors and accelerators that were originally designed for the embedded systems market are emerging as building blocks for servers. Power capping has been actively explored as a technique to reduce the energy footprint of high-performance processors. The opportunities and limitations of power capping on the new low-power processor and accelerator ecosystem are less understood. This paper presents an efficient power capping and management infrastructure for heterogeneous SoCs based on hybrid ARM/FPGA designs. The infrastructure coordinates dynamic voltage and frequency scaling with task allocation on a customised Linux system for the Xilinx Zynq SoC. We present a compiler-assisted power model to guide voltage and frequency scaling, in conjunction with workload allocation between the ARM cores and the FPGA, under given power caps. The model achieves less than 5% estimation bias to mean power consumption. In an FFT case study, the proposed power capping schemes achieve on average 97.5% of the performance of the optimal execution and match the optimal execution in 87.5% of the cases, while always meeting power constraints.

Verification and FPGA circuits of a block-2 fast path-based predictor

This paper formally derives a new path-based neural branch prediction algorithm (FPP) into blocks of size two for a lower hardware solution while maintaining similar input-output characteristic to the algorithm. The blocked solution, here referred to as B2P algorithm, is obtained using graph theory and retiming methods. Verification approaches were exercised to show that prediction performances obtained from the FPP and B2P algorithms differ within one mis-prediction per thousand instructions using a known framework for branch prediction evaluation. For a chosen FPGA device, circuits generated from the B2P algorithm showed average area savings of over 25% against circuits for the FPP algorithm with similar time performances thus making the proposed blocked predictor superior from a practical viewpoint.

FPGA organization for the fast path-based neural branch predictor

This paper develops cycle-level FPGA circuits of an organization for a fast path-based neural branch predictor Our results suggest that practical sizes of prediction tables are limited to around 32 KB to 64 KB in current FPGA technology due mainly to FPGA area of logic resources to maintain the tables. However the predictor scales well in terms of prediction speed. Table sizes alone should not be used as the only metric for hardware budget when comparing neural-based predictor to predictors of totally different organizations. This paper also gives early evidence to shift the attention on to the recovery from mis-prediction latency rather than on prediction latency as the most critical factor impacting accuracy of predictions for this class of branch predictors.

On the automatic integration of hardware accelerators into FPGA-based embedded systems

This paper proposes an automatic framework for the seamless integration of hardware accelerators, starting from an OpenMP-based application and an XML file describing the HW/SW partitioning. It extends a fully software architecture by generating and integrating the cores, along with the proper interfaces, and the code for scheduling and synchronization. Experimental results show that it is possible to validate different solutions only by varying the input code.

Laser microstructuring of Si surfaces for low-threshold field-electron emission

Dense arrays of high aspect ratio Si micro-pyramids have been formed by cumulative high intensity laser irradiation of doped Si wafers in an SF6 environment. A comparative study using nanosecond (XeCl, 308 nm) and femtosecond (Ti: Sapphire, 800 nm and KrF, 248 nm) laser pulses has been performed in this work. The influence of pulse duration and ambient gas pressure (SF6) is also presented. Scanning electron microscopy has shown that upon laser irradiation conical features appear on the Si surface in a rather homogenous distribution and with a spontaneous self alignment into arrays. Their lowest tip diameter is 800 nm; while their height reaches up to 90 mum. Secondary tip decoration appears on the surface of the formed spikes. Areas of 2 X 2 mm(2) covered with Si cones have been tested as cold cathode field emitters. After several conditioning cycles, the field emission threshold for the studied Si tips is as low as 2 V/mum, with an emission current of 10(-3) A/cm(2) at 4 V/mum. Even though these structures have smaller aspect ratios than good quality carbon nanotubes, their field emission properties are similar. The simple and direct formation of field emission Si arrays over small pre-selected areas by laser irradiation could lead to a novel approach for the development of electron sources. (C) 2003 Elsevier B.V. All rights reserved.

Efficient Gate-tunable light-emitting device made of defective boron nitride nanotubes: from ultraviolet to the visible

Boron nitride is a promising material for nanotechnology applications due to its two-dimensional graphene-like, insulating, and highly-resistant structure. Recently it has received a lot of attention as a substrate to grow and isolate graphene as well as for its intrinsic UV lasing response. Similar to carbon, one-dimensional boron nitride nanotubes (BNNTs) have been theoretically predicted and later synthesised. Here we use first principles simulations to unambiguously demonstrate that i) BN nanotubes inherit the highly efficient UV luminescence of hexagonal BN; ii) the application of an external perpendicular field closes the electronic gap keeping the UV lasing with lower yield; iii) defects in BNNTS are responsible for tunable light emission from the UV to the visible controlled by a transverse electric field (TEF). Our present findings pave the road towards optoelectronic applications of BN-nanotube-based devices that are simple to implement because they do not require any special doping or complex growth

Multiplier-less realization of a poly-phase filter using LUT-based FPGAs

In DSP applications such as fixed transforms and filtering, the full flexibility of a general-purpose multiplier is not required and only a limited range of values is needed on one of the multiplier inputs. A new design technique has been developed for deriving multipliers that operate on a limited range of multiplicands. This can be used to produce FPGA implementations of DSP systems where area is dramatically improved. The paper describes the technique and its application to the design of a poly-phase filter on a Virtex FPGA. A 62% area reduction and 7% speed increase is gained when compared to an equivalent design using general purpose multipliers. It is also compared favourably to other known fixed coefficient approaches.

Differential power analysis resistance of Camellia and countermeasure strategy on FPGAs

Security devices are vulnerable to Differential Power Analysis (DPA) that reveals the key by monitoring the power consumption of the circuits. In this paper, we present the first DPA attack against an FPGA implementation of the Camellia encryption algorithm with all key sizes and evaluate the DPA resistance of the algorithm. The Camellia cryptographic algorithm involves several different key-dependent intermediate operations including S-Box operations. In previous research, it was believed that the Camellia is stronger than AES due to the additional Whitening phase protecting the S-Box operation. However, we propose an attack that bypasses the Whitening phase and targets the S-Box. In this paper, we also discuss a lowcost countermeasure strategy to protect the Pre-whitening / Post-whitening and FL function of Camellia using Dual-rail Precharged Logic and to protect against attacks of the S-Box using Random Delay Insertion. © 2009 IEEE.

Shortening design time through multiplatform simulations with a portable OpenCL golden-model: the LDPC decoder case

Hardware designers and engineers typically need to explore a multi-parametric design space in order to find the best configuration for their designs using simulations that can take weeks to months to complete. For example, designers of special purpose chips need to explore parameters such as the optimal bitwidth and data representation. This is the case for the development of complex algorithms such as Low-Density Parity-Check (LDPC) decoders used in modern communication systems. Currently, high-performance computing offers a wide set of acceleration options, that range from multicore CPUs to graphics processing units (GPUs) and FPGAs. Depending on the simulation requirements, the ideal architecture to use can vary. In this paper we propose a new design flow based on OpenCL, a unified multiplatform programming model, which accelerates LDPC decoding simulations, thereby significantly reducing architectural exploration and design time. OpenCL-based parallel kernels are used without modifications or code tuning on multicore CPUs, GPUs and FPGAs. We use SOpenCL (Silicon to OpenCL), a tool that automatically converts OpenCL kernels to RTL for mapping the simulations into FPGAs. To the best of our knowledge, this is the first time that a single, unmodified OpenCL code is used to target those three different platforms. We show that, depending on the design parameters to be explored in the simulation, on the dimension and phase of the design, the GPU or the FPGA may suit different purposes more conveniently, providing different acceleration factors. For example, although simulations can typically execute more than 3x faster on FPGAs than on GPUs, the overhead of circuit synthesis often outweighs the benefits of FPGA-accelerated execution.

High-radix systolic modular multiplication on reconfigurable hardware

The overall aim of the work presented in this paper has been to develop Montgomery modular multiplication architectures suitable for implementation on modern reconfigurable hardware. Accordingly, novel high-radix systolic array Montgomery multiplier designs are presented, as we believe that the inherent regular structure and absence of global interconnect associated with these, make them well-suited for implementation on modern FPGAs. Unlike previous approaches, each processing element (PE) comprises both an adder and a multiplier. The inclusion of a multiplier in the PE means that the need to pre-compute or store any multiples of the operands is avoided. This also allows very high-radix implementations to be realised, further reducing the amount of clock cycles per modular multiplication, while still maintaining a competitive critical delay. For demonstrative purposes, 512-bit and 1024-bit FPGA implementations using radices of 2(8) and 2(16) are presented. The subsequent throughput rates are the fastest reported to date.