National University of Defense Technology

About: National University of Defense Technology is a education organization based out in Changsha, China. It is known for research contribution in the topics: Radar & Synthetic aperture radar. The organization has 39430 authors who have published 40181 publications receiving 358979 citations. The organization is also known as: Guófáng Kēxuéjìshù Dàxué & NUDT.

A reconfigurable fault-tolerant deflection routing algorithm based on reinforcement learning for network-on-chip

Abstract: We propose a reconfigurable fault-tolerant deflection routing algorithm (FTDR) based on reinforcement learning for NoC. The algorithm reconfigures the routing table through a kind of reinforcement learning---Q-learning using 2-hop fault information. It is topology-agnostic and insensitive to the shape of the fault region. In order to reduce the routing table size, we also propose a hierarchical Q-learning based deflection routing algorithm (FTDR-H) with area reduction up to 27% for a switch in an 8 x 8 mesh compared to the original FTDR. Experimental results show that in the presence of faults, FTDR and FTDR-H are better than other fault-tolerant deflection routing algorithms and a turn model based fault-tolerant routing algorithm.

CloudEyes: Cloud-based malware detection with reversible sketch for resource-constrained internet of things (IoT) devices

Abstract: Summary Because of the rapid increasing of malware attacks on the Internet of Things in recent years, it is critical for resource-constrained devices to guard against potential risks. The traditional host-based security solution becomes puffy and inapplicable with the development of malware attacks. Moreover, it is hard for the cloud-based security solution to achieve both the high performance detection and the data privacy protection simultaneously. This paper proposes a cloud-based anti-malware system, called CloudEyes, which provides efficient and trusted security services for resource-constrained devices. For the cloud server, CloudEyes presents suspicious bucket cross-filtering, a novel signature detection mechanism based on the reversible sketch structure, which provides retrospective and accurate orientations of malicious signature fragments. For the client, CloudEyes implements a lightweight scanning agent which utilizes the digest of signature fragments to dramatically reduce the range of accurate matching. Furthermore, by transmitting sketch coordinates and the modular hashing, CloudEyes guarantees both the data privacy and low-cost communications. Finally, we evaluate the performance of CloudEyes by utilizing both the campus suspicious traffic and normal files. The results demonstrate that the mechanisms in CloudEyes are effective and practical, and our system can outperform other existing systems with less time and communication consumption. Copyright © 2016 John Wiley & Sons, Ltd.

Robust Beamforming Design for Sum Secrecy Rate Maximization in Multibeam Satellite Systems

Abstract: This letter investigates the secure transmission optimization in a multibeam satellite downlink network with multiple eavesdroppers, where each legitimate user is wiretapped by a corresponding eavesdropper. Our design objective is to maximize the sum secrecy rate of the multibeam satellite network in the presence of imperfect channel state information (CSI) of eavesdroppers, while guaranteeing the total transmit power constraint on-board the satellite. Due to the nonconvexity and intractability resulting from the eavesdroppers’ CSI uncertainty, a robust beamforming scheme is proposed to transform the initial optimization problem into a convex framework by jointly applying the Taylor expansion, S-Procedure, and Cauchy–Schwarz inequality. Meanwhile, an iterative algorithm is introduced to obtain the optimal solution. Finally, the validity and superiority of the proposed scheme are confirmed through comparisons with the existing nonrobust and perfect CSI approaches.

SHREC'14 Track: Shape retrieval of non-rigid 3D human models

Abstract: We have created a new benchmarking dataset for testing non-rigid 3D shape retrieval algorithms, one that is much more challenging than existing datasets Our dataset features exclusively human models, in a variety of body shapes and poses 3D models of humans are commonly used within computer graphics and vision, and so the ability to distinguish between body shapes is an important shape retrieval problem In this track nine groups have submitted the results of a total of 22 different methods which have been tested on our new dataset