Search

Metallic nanostructures array has promising potential applications in nano-photonic devices based on surface plasmon resonance. In this letter, we report the fabrication of silver nanostructured grating pattern on periodically poled lithium niobate (PPLN) single crystal by the site-selective photo-deposition with UV-light irradiation. The microstructure of the silver nanostructured grating pattern on the positive domains was characterized by the DIC microscope and AFM. The SNOM characterization shows the enhancement of optical field due to the surface plasmon resonance (SPR) on silver NPs. Optical transmission spectra of silver nanostructured grating pattern show the silver NPs enhance the absorption.

Transient chaotic neural networks (TCNNs) have shown promise in solving optimization problems but still suffer from slow convergence and being difficult to implement in hardware. In this paper, the HP memristor is introduced to a TCNN to develop a memristor-based transient chaotic neural network (MTCNN) model that is highly efficient, converges quickly, and has significant prospects for physical implementation. The proposed MTCNN makes full use of the nonlinearity and memory-related characteristics of memristors, and their conductance values are used as self-feedback connection weights that can be adjusted dynamically according to the annealing algorithm. The MTCNN model was applied to solve combinatorial optimization problems, including the channel assignment problem (CAP) of four cells and the traveling salesman problem (TSP) of 10 cities. In 500 runs, the MTCNN algorithm delivered a 5% higher optimal solution rate than the TCNN algorithm while using only 70% of its number of iterations in the CAP, and achieved a shorter average distance and a 40% higher convergence speed than the TCNN algorithm in the TSP.

Brain-inspired neuromorphic computing has attracted much attention for its advanced computing concept. However, the massive hardware cost in fully-connected architectures makes it challenging to build a large-scale neuromorphic system. In this work, we report a compact, programmable, versatile, and scalable neuromorphic architecture. To demonstrate the concept of the neuromorphic architecture, a neuromorphic system consisting of four cores is implemented on an FPGA platform. On the one hand, the neuromorphic system is extremely compact and hardware-saving. The computing block based on a simple digital leaky Integrate-and-Fire (LIF) model only costs 69 logic elements (LEs); only one physical neuron is implemented in each core, and it can be reused as hundreds of virtual neurons by time-division-multiplexing; only four 9-bit synaptic weights are assigned to each neuron, which effectively alleviates the hardware explosion in fully-connected architecture. On the other hand, the neuromorphic system is programmable and versatile, and can perform different neural network computing. The neuromorphic system mapped with a three-layer feedforward network successfully recognizes the MNIST handwritten digits with an accuracy of 96.26%, and it also effectively realizes different convolution operations which are basic computing operations in convolutional neural networks. Last but not least, each neuromorphic core has its own router module, making it convenient to scale up.

Although synaptic behaviours of memristors have been widely demonstrated, implementation of an even simple artificial neural network is still a great challenge. In this work, we demonstrate the associative memory on the basis of a memristive Hopfield network. Different patterns can be stored into the memristive Hopfield network by tuning the resistance of the memristors, and the pre-stored patterns can be successfully retrieved directly or through some associative intermediate states, being analogous to the associative memory behaviour. Both single-associative memory and multi-associative memories can be realized with the memristive Hopfield network.