This paper proposes a parallel programming scheme for the cross-point array with resistive random access memory (RRAM). Synaptic plasticity in unsupervised learning is realized by tuning the conductance of each RRAM c...
详细信息
This paper proposes a parallel programming scheme for the cross-point array with resistive random access memory (RRAM). Synaptic plasticity in unsupervised learning is realized by tuning the conductance of each RRAM cell. Inspired by the spike-timing-dependent-plasticity (STDP), the programming strength is encoded into the spike firing rate (i.e., pulse frequency) and the overlap time (i.e., duty cycle) of the pre-synaptic node and post-synaptic node, and simultaneously applied to all RRAM cells in the cross-point array. Such an approach achieves parallel programming of the entire RRAM array, only requiring local information from pre-synaptic and post-synaptic nodes to each RRAM cell. As demonstrated by digital peripheral circuits implemented in 65 nm CMOS, the programming time of a 40 kb RRAM array is 84 ns, indicating 900X speedup as compared to state-of-the-art software approach of sparse coding in image feature extraction.
In this work, we report experimental results regarding the recycling of optical pumps signal for energy conversion, namely optical pumps from Raman amplifiers and fiber optic parametric amplifiers. Therefore, we propo...
详细信息
A new architecture for the real-time detection of scale-invariant features in image sequences is presented. The system is based on a low-cost smart-camera custom board, developed to target robotic vision applications....
详细信息
A new architecture for the real-time detection of scale-invariant features in image sequences is presented. The system is based on a low-cost smart-camera custom board, developed to target robotic vision applications. Several optimizations of the SIFT detection procedure are proposed in order to achieve robust keypoint detection with high repeatability and recall values. As a result, a high accuracy and resource-efficient implementation of the SIFT detector is presented. The system is pipelined and streams pixel data using a 45 MHz clock, allowing keypoint detection at 150 frames per second, in video sequences with resolution 640×480. Integrating a commodity CMOS sensor, the prototype system displays keypoints at video rate, using only a fraction of the resources of a low-cost FPGA device.
This research deals with the deeper analysis of the novel concept of a multi-chaos-driven evolutionary algorithm Differential Evolution (DE). This paper is aimed at the embedding and alternating of set of two discrete...
详细信息
We propose a fabrication-tolerant SOI polarization splitter-rotator with cascaded MMI couplers and an assisted bi-level taper. The performance remains very stable even for a large deviation of ±50 nm in terms of ...
详细信息
In this study the performance of two popular evolutionary computational techniques (particle swarm optimization and differential evolution) is compared in the task of batch reactor geometry optimization. Both algorith...
详细信息
In this work suitable architectures and high-throughput FPGA implementations of Volterra Decision Feedback Equalizers (VDFEs) for optical communication links are presented. Two VDFE configurations were selected based ...
详细信息
In this work suitable architectures and high-throughput FPGA implementations of Volterra Decision Feedback Equalizers (VDFEs) for optical communication links are presented. Two VDFE configurations were selected based on the available resources of the employed FPGA devices, and two multiplexer-based architectures were developed for each of them in order to achieve the target throughput. The comparison of the experimental results with respect to different VDFE configurations, throughput, and FPGA devices points out the platform-specific design characteristics. The introduced architectures meet the desired 10Gb/s throughput, so it is demonstrated that the FPGA is a suitable platform for high-speed optical fiber communication systems.
Dynamic routing is considered an attractive solution for many network applications such as monitoring and tracking. In dynamic routing the path between the source and the destination can change dynamically following t...
详细信息
Dynamic routing is considered an attractive solution for many network applications such as monitoring and tracking. In dynamic routing the path between the source and the destination can change dynamically following the network conditions. Next node selection criteria can have a great impact on the selection of each path and as a consequence on the network performance. Moreover, each network application may have different Quality of Service (QoS) requirements. In this paper, we examine the impact of the next node selection criteria on the network performance. Three routing protocols with different approaches along with their routing algorithms are presented. To examine the performance of each approach a discrete event simulator is used. It is shown that the selection criteria should be carefully designed following the QoS requirements of the network application. Furthermore, the obtained performance results can be used as an indicator of the appropriateness of a dynamic protocol for a variety of applications.
Low-complexity Volterra Decision Feedback Equalizers (VDFEs) for optical fiber links are proposed. By properly discarding large blocks of coefficients in the feedforward and feedback sections of the equalizer, a signi...
详细信息
Low-complexity Volterra Decision Feedback Equalizers (VDFEs) for optical fiber links are proposed. By properly discarding large blocks of coefficients in the feedforward and feedback sections of the equalizer, a significant complexity reduction is achieved without affecting its efficiency. Moreover, suitable architectures and high-performance FPGA implementations are provided. It is demonstrated that the efficiency of the proposed VDFEs in terms of BER is similar to the counterpart full-size VDFEs for links of the same length, while, they demand 50% less arithmetic resources. Also, the proposed DFEs meet the desired 10 Gb/s rate and in certain cases achieve rates of 17 Gb/s and 25 Gb/s.
During last few decades wind energy has become one of the fastest growing renewable sources. Recent research interest in wind power forecasting community is focused on wind ramps, sudden changes in wind power output. ...
详细信息
During last few decades wind energy has become one of the fastest growing renewable sources. Recent research interest in wind power forecasting community is focused on wind ramps, sudden changes in wind power output. Conventional ramp prediction methods derive future ramp estimates from power forecast series. We suggest to analyze real wind power series or other weather parameters searching for specific patterns and dependencies signalizing forthcoming wind ramp event. In this paper, principal component analysis of wind speed time series is presented. Analyzed power production and wind speed data were collected from August 2011 to July 2012 on sample wind farm located close to Let bridge, AB Canada. Sensitivity of analysis is defined by length of selected time frame. Frequency tables for first three principal components are used to evaluate conditional probability of forthcoming wind ramp event. Advantage of proposed methodology compared to conventional methods is that numerical weather prediction (NWP) model producing wind forecasts is not required.
暂无评论