Design-Space Exploration of Application-specific Instruction-set Processor Design

Application-Specific Instruction-Set Processors (ASIPs) have established their processing power in the embedded systems. Since energy efficiency is one of the most important challenges in this area, coarse-grained reconfigurable arrays (CGRAs) have been used in many different domains. The exclusive program execution model of the CGRAs is the key to their energy efficiency but it has some major costs. The context-switching network (CSN) is responsible for handling this unique program execution model and is also one of the most energy-hungry parts of the CGRAs. In this paper, we have proposed a new method to predict important architectural parameters of the CSN of a CGRA, such as the size of the processing elements (PEs), the topology of the CSN, and the number of configuration registers in each PE. The proposed method is based on the high-level code of the input application, and it is used to prune the design space and increase the energy efficiency of the CGRA. Based on our results, not only the size of the design space of the CSN of the CGRA is reduced to 10%, but also its performance and energy efficiency are increased by about 13% and 73%, respectively. The predicted architecture by the proposed method is over 97% closer to the best architecture of the exhaustive searching for the design space.

Leader-following Synchronization Control of Multiple Electrohydraulic Actuators Under Switching Network Topologies

A leader-following quasi-synchronization control is proposed in multiple electrohydraulic actuators (MEHAs) under different switching network topologies to guarantee the follower electrohydraulic actuators (EHAs) tracking the leader motion. Firstly, each electro-hydraulic actuator (EHA) has a 3-order nonlinear dynamics with unknown external load. Then by using Lie derivative technique, the MEHAs nonlinear models with $n+1$ nodes are feedback linearized for convenient control design. Furthermore, the leader node is constructed as a virtual simulation model to be stabilized by PI controller. Meanwhile, a quasi-synchronized controller together with a disturbance observer is designed by LMI and Lyapunov techniques to guarantee that the synchronization errors between the n follower nodes and the leader node 0 are uniformly ultimate boundaries. Finally, the effectiveness of the leader-following quasi-synchronized controller is verified by a MEHAs experimental bench with 3 EHAs under switching network topologies.

Method for assessing the operating quality of a switching node in a technological transmission network data in the context of DDoS attacks by an intruder

The purpose of this study is to develop a mechanism to account for the effects of a distributed denial of service attack on a switching node, as well as to assess the quality of its functioning under destructive effects. Relevance stems from the possibility of disruption of regional economic complex management due to the impact on the elements of the technological network of data transmission attacker type “distributed denial of service”. Based on the mathematical apparatus of GERT-networks the authors propose an approach to assess the quality of switching nodes. The essence lies in the representation of the data flow service node switching network data transmission under attack by an intruder in the form of a stochastic network, setting the type of partial distributions, the definition of the equivalent function, followed by determining the distribution function delay time packets. The model proposed by the authors allows to evaluate the quality of switching nodes operation under the conditions of “distributed denial of service” intruder attacks, both when transmitting stationary Poisson and self-similar traffic, represented by the Weibull and Pareto flow models. The results obtained are in good agreement with the data given in previously published works. The model allows to analyze and develop directions to improve the quality of functioning of switching nodes of technological network of data transmission under conditions of destructive information impact of the intruder.

Fault Diagnosis Knowledge Reasoning of Switching Network in Distributed Generation Based on Petri Net

Telephone network based on IMS technology has been widely applied in power production and dispatching communication, especially in distributed power stations. Analysis and positioning failure of IMS network is arduous, because it’s dependent on IP data communication network. In this paper, we first introduced IMS switching network architecture and distributed generation communication network architecture, analyzed and summarized all kinds of network malfunction. Combining typical IMS network fault connection relations, we introduced an improved Petri net fault handling model and reasoning method. The diagnosis and positioning results could reflect the defects of equipment logic functions. This method on fault diagnosis and location of substation network has been proved to be effective through practical application.

A biologically inspired architecture with switching units can learn to generalize across backgrounds

Humans and other animals navigate different landscapes and environments with ease, a feat that requires the brain's ability to rapidly and accurately adapt to different visual domains, generalizing across contexts/backgrounds. Despite recent progress in deep learning applied to classification and detection in the presence of multiple confounds including contextual ones, there remain important challenges to address regarding how networks can perform context-dependent computations and how contextually-invariant visual concepts are formed. For instance, recent studies have shown artificial networks that repeatedly misclassified familiar objects set on new backgrounds, e.g. incorrectly labeling known animals when they appeared in a different setting. Here, we show how a bio-inspired network motif can explicitly address this issue. We do this using a novel dataset which can be used as a benchmark for future studies probing invariance to backgrounds. The dataset consists of MNIST digits of varying transparency, set on one of two backgrounds with different statistics: a Gaussian noise or a more naturalistic background from the CIFAR-10 dataset. We use this dataset to learn digit classification when contexts are shown sequentially, and find that both shallow and deep networks have sharply decreased performance when returning to the first background after experience learning the second -- the catastrophic forgetting phenomenon in continual learning. To overcome this, we propose an architecture with additional ``switching'' units that are activated in the presence of a new background. We find that the switching network can learn the new context even with very few switching units, while maintaining the performance in the previous context -- but that they must be recurrently connected to network layers. When the task is difficult due to high transparency, the switching network trained on both contexts outperforms networks without switching trained on only one context. The switching mechanism leads to sparser activation patterns, and we provide intuition for why this helps to solve the task. We compare our architecture with other prominent learning methods, and find that elastic weight consolidation is not successful in our setting, while progressive nets are more complex but less effective. Our study therefore shows how a bio-inspired architectural motif can contribute to task generalization across context.

CNF-SAT modelling for banyan-type networks and its application for assessing the rearrangeability

A banyan-type network is a switching network, which is constructed by placing unit switches with two inputs and two outputs in s (s > 1) stages. In each stage, 2 n – 1 (n > 1) unit switches are aligned. Past studies conjecture that this network becomes rearrangeable when s ≥ 2n-1. Although a considerable number of theoretical analyses have been done, the rearrangeability of the banyan-type network with 2n – 1 or more stages is not completely proved. As a tool to assess the rearrangeability, this study presents a CNF-SAT (conjunctive normal form - satisfiability) modelling scheme for banyan-type networks. In the proposed scheme, the routing is formulated to a SAT problem represented in CNF. By feeding the problem to a SAT solver, it is found whether the problem is satisfiable or unsatisfiable. If the problem is unsatisfiable for a certain request, the network is not rearrangeable. By contrast, if the problem is satisfiable for any requests, the network is rearrangeable. This study applies the CNF-SAT modelling scheme to various configurations of 2n – 1 stage banyan-type networks. These networks are assessed for rearrangeability by solving the SAT problems. The proposed method will be helpful to conduct future theoretical studies on banyan-type networks.