Constructive Synthesis of Memory-Intensive Accelerators for FPGA From Nested Loop Kernels

Field-programmable gate arrays are ideal hosts to custom accelerators for signal, image, and data processing but de- mand manual register transfer level design if high performance and low cost are desired. High-level synthesis reduces this design burden but requires manual design of complex on-chip and off-chip memory architectures, a major limitation in applications such as video processing. This paper presents an approach to resolve this shortcoming. A constructive process is described that can derive such accelerators, including on- and off-chip memory storage from a C description such that a user-defined throughput constraint is met. By employing a novel statement-oriented approach, dataflow intermediate models are derived and used to support simple ap- proaches for on-/off-chip buffer partitioning, derivation of custom on-chip memory hierarchies and architecture transformation to ensure user-defined throughput constraints are met with minimum cost. When applied to accelerators for full search motion estima- tion, matrix multiplication, Sobel edge detection, and fast Fourier transform, it is shown how real-time performance up to an order of magnitude in advance of existing commercial HLS tools is enabled whilst including all requisite memory infrastructure. Further, op- timizations are presented that reduce the on-chip buffer capacity and physical resource cost by up to 96% and 75%, respectively, whilst maintaining real-time performance.

FPGA-based soft-core processors for image processing applications

With security and surveillance, there is an increasing need to process image data efficiently and effectively either at source or in a large data network. Whilst a Field-Programmable Gate Array (FPGA) has been seen as a key technology for enabling this, the design process has been viewed as problematic in terms of the time and effort needed for implementation and verification. The work here proposes a different approach of using optimized FPGA-based soft-core processors which allows the user to exploit the task and data level parallelism to achieve the quality of dedicated FPGA implementations whilst reducing design time. The paper also reports some preliminary
progress on the design flow to program the structure. An implementation for a Histogram of Gradients algorithm is also reported which shows that a performance of 328 fps can be achieved with this design approach, whilst avoiding the long design time, verification and debugging steps associated with conventional FPGA implementations.

Time-Independent Discrete Gaussian Sampling for Post-Quantum Cryptography

As the development of a viable quantum computer nears, existing widely used public-key cryptosystems, such as RSA, will no longer be secure. Thus, significant effort is being invested into post-quantum cryptography (PQC). Lattice-based cryptography (LBC) is one such promising area of PQC, which offers versatile, efficient, and high performance security services. However, the vulnerabilities of these implementations against side-channel attacks (SCA) remain significantly understudied. Most, if not all, lattice-based cryptosystems require noise samples generated from a discrete Gaussian distribution, and a successful timing analysis attack can render the whole cryptosystem broken, making the discrete Gaussian sampler the most vulnerable module to SCA. This research proposes countermeasures against timing information leakage with FPGA-based designs of the CDT-based discrete Gaussian samplers with constant response time, targeting encryption and signature scheme parameters. The proposed designs are compared against the state-of-the-art and are shown to significantly outperform existing implementations. For encryption, the proposed sampler is 9x faster in comparison to the only other existing time-independent CDT sampler design. For signatures, the first time-independent CDT sampler in hardware is proposed. 

Feature Study on a Programmable Network Traffic Classifier

Monitoring and tracking of IP traffic flows are essential for network services (i.e. packet forwarding). Packet header lookup is the main part of flow identification by determining the predefined matching action for each incoming flow. In this paper, an improved header lookup and flow rule update solution is investigated. A detailed study of several well-known lookup algorithms reveals that searching individual packet header field and combining the results achieve high lookup speed and flexibility. The proposed hybrid lookup architecture is comprised of various lookup algorithms, which are selected based on the user applications and system requirements.

Ultracompact Retrodirective Antenna Arrays With Superdirective Radiation Patterns

It is shown that the direction-of-arrival (DoA) information carried by an incident electromagnetic (EM) wave can be encoded into the evanescent near field of an electrically small resonance antenna array with a spatial rate higher than that of the incident field oscillation rate in free space. Phase conjugation of the received signal leads to the retrodirection of the near field in the antenna array environment, which in turn generates a retrodirected far-field beam toward the original DoA. This EM phenomenon enables electrically small retrodirective antenna arrays with superdirective, angular super-resolution, auto-pointing properties for an arbitrary DoA. A theoretical explanation of the phenomenon based on first principal observations is given and full-wave simulations demonstrate a realizability route for the proposed retrodirective terminal that is comprised of resonance dipole antenna elements. Specifically, it is shown that a three-element disk-loaded retrodirective dipole array with 0.15\lambda spacings can achieve a 3.4-dBi maximal gain, 3-dBi front-to-back ratio, and 13% return loss fractional bandwidth (at the 10-dB level). Then, it is demonstrated that the radiation gain of a three-element array can be improved to approximately 6 dBi at the expense of the return loss fractional bandwidth reduction (2%).

Ordered arrays of multiferroic epitaxial nanostructures

Epitaxial heterostructures combining ferroelectric (FE) and ferromagnetic (FiM) oxides are a possible route to explore coupling mechanisms between the two independent order parameters, polarization and magnetization of the component phases. We report on the fabrication and properties of arrays of hybrid epitaxial nanostructures of FiM NiFe(2)O(4) (NFO) and FE PbZr(0.52)Ti(0.48)O(3) or PbZr(0.2)Ti(0.8)O(3), with large range order and lateral dimensions from 200 nm to 1 micron. METHODS: The structures were fabricated by pulsed-laser deposition. High resolution transmission electron microscopy and high angle annular dark-field scanning transmission electron microscopy were employed to investigate the microstructure and the epitaxial growth of the structures. Room temperature ferroelectric and ferrimagnetic domains of the heterostructures were imaged by piezoresponse force microscopy (PFM) and magnetic force microscopy (MFM), respectively. RESULTS: PFM and MFM investigations proved that the hybrid epitaxial nanostructures show ferroelectric and magnetic order at room temperature. Dielectric effects occurring after repeated switching of the polarization in large planar capacitors, comprising ferrimagnetic NiFe2O4 dots embedded in ferroelectric PbZr0.52Ti0.48O3 matrix, were studied. CONCLUSION: These hybrid multiferroic structures with clean and well defined epitaxial interfaces hold promise for reliable investigations of magnetoelectric coupling between the ferrimagnetic / magnetostrictive and ferroelectric / piezoelectric phases.

Do Changes in Current Flow as a Result of Arrays of Tidal Turbines Have an Effect on Benthic Communities?

Arrays of tidal energy converters have the potential to provide clean renewable energy for future generations. Benthic communities may, however, be affected by changes in current speeds resulting from arrays of tidal converters located in areas characterised by strong currents. Current speed, together with bottom type and depth, strongly influence benthic community distributions; however the interaction of these factors in controlling benthic dynamics in high energy environments is poorly understood. The Strangford Lough Narrows, the location of SeaGen, the world’s first single full-scale, grid-compliant tidal energy extractor, is characterised by spatially heterogenous high current flows. A hydrodynamic model was used to select a range of benthic community study sites that had median flow velocities between 1.5–2.4 m/s in a depth range of 25–30 m. 25 sites were sampled for macrobenthic community structure using drop down video survey to test the sensitivity of the distribution of benthic communities to changes in the flow field. A diverse range of species were recorded which were consistent with those for high current flow environments and corresponding to very tide-swept faunal communities in the EUNIS classification. However, over the velocity range investigated, no changes in benthic communities were observed. This suggested that the high physical disturbance associated with the high current flows in the Strangford Narrows reflected the opportunistic nature of the benthic species present with individuals being continuously and randomly affected by turbulent forces and physical damage. It is concluded that during operation, the removal of energy by marine tidal energy arrays in the far-field is unlikely to have a significant effect on benthic communities in high flow environments. The results are of major significance to developers and regulators in the tidal energy industry when considering the environmental impacts for site licences.

Multiscale Analysis for Field-Effect Penetration through Two-Dimensional Materials

Gate-tunable two-dimensional (2D) materials-based quantum capacitors (QCs) and van der Waals heterostructures involve tuning transport or optoelectronic characteristics by the field effect. Recent studies have attributed the observed gate-tunable characteristics to the change of the Fermi level in the first 2D layer adjacent to the dielectrics, whereas the penetration of the field effect through the one-molecule-thick material is often ignored or oversimplified. Here, we present a multiscale theoretical approach that combines first-principles electronic structure calculations and the Poisson–Boltzmann equation methods to model penetration of the field effect through graphene in a metal–oxide–graphene–semiconductor (MOGS) QC, including quantifying the degree of “transparency” for graphene two-dimensional electron gas (2DEG) to an electric displacement field. We find that the space charge density in the semiconductor layer can be modulated by gating in a nonlinear manner, forming an accumulation or inversion layer at the semiconductor/graphene interface. The degree of transparency is determined by the combined effect of graphene quantum capacitance and the semiconductor capacitance, which allows us to predict the ranking for a variety of monolayer 2D materials according to their transparency to an electric displacement field as follows: graphene > silicene > germanene > WS2 > WTe2 > WSe2 > MoS2 > phosphorene > MoSe2 > MoTe2, when the majority carrier is electron. Our findings reveal a general picture of operation modes and design rules for the 2D-materials-based QCs.