Search

Considering the high-performance and low-power requirements of edge AI, this study designs a specialized instruction set processor for edge AI based on the RISC-V instruction set architecture, addressing practical issues in digital signal processing for edge devices. This design enhances the execution efficiency of edge AI and reduces its energy consumption with limited hardware overhead, meeting the demands for efficient large language model (LLM) inference computation in edge AI applications. The main contributions of this paper are as follows: For the characteristics of large language models, custom instructions were extended based on the RISC-V instruction set to perform vector dot product calculations, accelerating the computation of large language models on dedicated vector dot product acceleration hardware. Based on the open-source high-performance RISC-V processor core XiangShan Nanhu architecture, the vector dot product specialized instruction set processor Nanhu-vdot was implemented, which adds vector dot product calculation units and pipeline processing logic on top of the XiangShan Nanhu.The Nanhu-vdot underwent FPGA hardware testing, achieving over four times the speed of scalar methods in vector dot product computation. Using a hardware-software co-design approach for second-generation Generative Pre-Trained Transformer (GPT-2) model inference, the speed improved by approximately 30% compared to pure software implementation with almost no additional consumption of hardware resources and power consumption., Comment: 10 pages, in Chinese language, 6 figures

We propose a method of complex-amplitude Fourier single-pixel imaging (CFSI) with coherent structured illumination to acquire both the amplitude and phase of an object. In the proposed method, an object is illustrated by a series of coherent structured light fields which are generated by a phase-only spatial light modulator, the complex Fourier spectrum of the object can be acquired sequentially by a single-pixel photodetector. Then the desired complex-amplitude image can be retrieved directly by applying an inverse Fourier transform. We experimentally implemented this CFSI with several different types of objects. The experimental results show that the proposed method provides a promising complex-amplitude imaging approach with high quality and a stable configuration. Thus, it might find broad applications in optical metrology and biomedical science., Comment: 6 pages, 5 figures

We propose and experimentally demonstrate a high-efficiency single-pixel imaging (SPI) scheme by integrating time-correlated single-photon counting (TCSPC) with time-division multiplexing to acquire full-color images at extremely low light level. This SPI scheme uses a digital micromirror device to modulate a sequence of laser pulses with preset delays to achieve three-color structured illumination, then employs a photomultiplier tube into the TCSPC module to achieve photon-counting detection. By exploiting the time-resolved capabilities of TCSPC, we demodulate the spectrum-image-encoded signals, and then reconstruct high-quality full-color images in a single-round of measurement. Based on this scheme, the strategies such as single-step measurement, high-speed projection, and undersampling can further improve the imaging efficiency., Comment: 5 pages, 4 figures

We investigate the hydrostatic pressure dependence of interfacial superconductivity occurring at the atomically sharp interface between two non-superconducting materials: the topological insulator (TI) Bi2Te3 and the parent compound Fe1+yTe of the chalcogenide iron based superconductors. Under pressure, a significant increase in the superconducting transition temperature Tc is observed. We trace the pressure dependence of a superconducting twin gap structure by Andreev reflection point contact spectroscopy (PCARS), which shows that a large superconducting gap associated with the interfacial superconductivity increases along with Tc. A second smaller gap, which is attributed to proximity-induced superconductivity in the TI layer, increases first, but then reaches a maximum and appears to be gradually suppressed at higher pressure. We interpret our data in the context of a pressure-induced doping effect of the interface, in which charge is transferred from the TI layer to the interface and the interfacial superconductivity is enhanced. This demonstrates the important role of the TI in the interfacial superconductivity mechanism.

Superconducting proximity effect (SPE) in topological insulator (TI) and superconductor (SC) hybrid structure has attracted intense attention in recent years in an effort to search for mysterious Majorana fermions (MFs) in condensed matter systems. Here we report on the SPE in a Bi2Se3/NbSe2 junction fabricated with an all-dry transfer method. Resulting from the highly transparent interface, two sharp resistance drops are observed at 7 K and 2 K, respectively, corresponding to the superconducting transition of NbSe2 flake and the SPE induced superconductivity in Bi2Se3 flake. Experimentally measured differential conductance spectra exhibit a bias-independent conductance plateau (BICP) in the vicinity of zero bias below 7 K. As temperatures further decrease a zero bias conductance peak (ZBCP) emerges from the plateau and becomes more enhanced and sharpened at lower temperatures. Our numerically simulated differential conductance spectra reproduce the observed BICP and ZBCP and show that the SPE in topological surface states (TSS) is much stronger than that in the bulk states of Bi2Se3. The SPE induced superconducting gap for the TSS of Bi2Se3 is comparable to that of NbSe2 and gives rise to the observed BICP below 7 K. In contrast, the SPE induced superconducting gap for the bulk states of Bi2Se3 is an order of magnitude smaller than that of NbSe2 and superconducting TSS. These weakly paired bulk states in Bi2Se3 give rise to the ZBCP below 2 K. Our study has clearly unveiled the different roles of TSS and bulk stats in SPE, clarified the physical origin of the SPE induced features, and shined light on further investigation of SPE and MF in TI/SC hybrid structures.

The discovery of two-dimensional superconductivity in Bi2Te3/FeTe heterostructure provides a new platform for the search of Majorana fermions in condensed matter systems. Since Majorana fermions are expected to reside at the core of the vortices, a close examination of the vortex dynamics in superconducting interface is of paramount importance. Here, we report the robustness of the interfacial superconductivity and 2D vortex dynamics in four as-grown and aged Bi2Te3/FeTe heterostructure with different Bi2Te3 epilayer thickness (3, 5, 7, 14 nm). After two years' air exposure, superconductivity remains robust even when the thickness of Bi2Te3 epilayer is down to 3 nm. Meanwhile, a new feature at ~13 K is induced in the aged samples, and the high field studies reveal its relevance to superconductivity. The resistance of all as-grown and aged heterostructures, just below the superconducting transition temperature follows the Arrhenius relation, indicating the thermally activated flux flow behavior at the interface of Bi2Te3 and FeTe. Moreover, the activation energy exhibits a logarithmic dependence on the magnetic field, providing a compelling evidence for the 2D vortex dynamics in this novel system. The weak disorder associated with aging-induced Te vacancies is possibly responsible for these observed phenomena., Comment: 21 pages, article+supplemental information, published in Scientific Reports

Recently, a logarithmic decrease of conductivity has been observed in topological insulators at low temperatures, implying a tendency of localization of surface electrons. Here, we report quantum transport experiments on the topological insulator Bi2Te3 thin films with arrayed antidot nanostructures. With increasing density of the antidots, a systematic decrease is observed in the slope of the logarithmic temperature-dependent conductivity curves, indicating the electron-electron interaction can be tuned by the antidots. Meanwhile, the weak anti-localization effect revealed in magnetoconductivity exhibits an enhanced dominance of electron-electron interaction among decoherence mechanisms. The observation can be understood from an antidot-induced reduction of the effective dielectric constant, which controls the interactions between the surface electrons. Our results clarify the indispensable role of the electron-electron interaction in the localization of surface electrons and indicate the localization of surface electrons in an interacting topological insulator., Comment: 27 pages, article+supplemental materials, published in ACS Nano