Artificial Intelligence and Robotics Commons^™

Application-Aware Cross-Layer Enrgy-Eficient Routing Scheme, Xu Fang, Hiyin Zhang, Wang Jing, Xu Ning, Zhijong Wang, Deng Min

Journal of System Simulation

Abstract: An aplication- aware cross-layer energy-effcient routing scheme (ACER) was presented to minimize the effect of supplt of terminals in ad hoc network composed of smart mobile devices. Features of energy consumption were sensed by application aware energy model with the application monitor and the remaining energy monitor. Stability of network path was monitored by link stability monitoring module in Data link layer. As the scheme used the idea of cross-layer design, network topology information, application-aware information in application layer and link-stability were utilized synthetically to make routing deisions in network layer;Simulations were crried out on NS2 platform. The …

Trnek Befor Detet Agoritmn For Weak Iret Based On Ergodic Hogh Tasform, Yunfei Guo, Xiaofeng Zheng, Dongliang Peng, Zebin Zeng

Journal of System Simulation

Abstract: The performance of track-before-detect (TBD) algorithm based on standard Hough Transform is not satisfied due to low detection probability and long elapsed time. For this problem, an Ergodic Hough Transform based TBD method was proposed. The raw radar echo whose amplitude exceeded the first threshold was stored separately and stacked into a data space; all diferent data were combined to calculate line parameter which then was mapped into a parameter space and the amplitude of each paired wae added to an accumulator. All cells whose amplitude exceeded the second threshold was used to estimate the last line parameter and …

Test Data Compression Based On Logical Operationand Dimidiate Partition, Wu Qiong, Huang Li

Journal of System Simulation

Abstract: A new test data compression based on logical operation and dimidiate partition was proposed. The whole test was partitioned to several length-fixed blocks. Then logical operation was applied into runs of Is or 0s or alternating bits. They were turned into all 0s. For the others, dimidiate partition technique was used until they got to the minimum length. The length of code words could be expressed by the times of dimidiate and it is easy to coding when they are turned into all 0s. Compared with traditional data-coding methods the depression structure is simple and reduced the costs of …

Study Of Improved Generalized Predictive Control In Ball Mill Application, Lingfang Sun, Jingmiao Sun

Journal of System Simulation

Abstract: Direct-fired pulverizing system with double inlets and outlets is regarded as important generating equipment in power plant, which has characteristics of nonlinear, multivariable, strong coupling and time-varying. Started from the mechanism of law, the mechanism mathematical model of the ball mill was established and a certain amount of disturbance transfer ftinction model was added in a mathematical model of the mechanism of input variables to give step disturbance tests to establish the ball mill system. Combining the most widely used basic control--PID control as industrial process control, and on the basis of general generalized predictive control algorithm, the generalized …

New Identify Based Proxy Signature Scheme, Xiaojing Hong, Bin Wang

Journal of System Simulation

Abstract: Proxy signature schemes allow a proxy signer to generate proxy signatures on behalf of an original signer. Mambo, et ah first introduced the notion of proxy signature and a lot of research work can be found on this topic nowadays. To simplify key management, many identity based proxy signature schemes were proposed. However, some existing schemes are vulnerable to proxy key exposure attack. It is necessary to propose a security model for identity based proxy signature schemes against proxy key exposure attack. Then an efficient scheme based on pairings was presented, which is provably secure in the random oracle …

Efficient Real-Time Traffic Signal Control Algorithm, Zhongcheng Yang, Ye Chen, Zhenyu Yang

Journal of System Simulation

Abstract: For real time taffic signal contol isus, a mathematical model was proposed to minimize the waiting ime, meanwhile a huristi search aigorihm was given to sove the optinal sluton Simulatin Tresuts show that the hcurisi scarceh algorim sufrfs fom being couataoalal complx and umstable, therefore, a multi sage decisin optimization algorihm is added, and the serehes adopt in all stages subject to a time limit, which ensures a stable and real-time algorithm, and also a soution in fixed time. Simulation resuts based on actal tafic data show that the waiting time can be reduced in comparison with that of …

Modeling Emergence Of Network Radar Countermeasure System, Huang Chen, Jianqing Qi, Fangzheng Liu

Journal of System Simulation

Abstract: Network radar countermeasure system (NRCS) is a new kind of integrated electronic warfare system with the integrated network developing trend of radar and EW equipment, and emergence is the typical complex feature of NRCS. The connotation and composition of the NRCS emergence was interpreted. Four sources of the NRCS emergence were analyzed: composition effect, structural effect, interaction effect and environmental effect. The conceptual models and mathematical models of the NRCS emergence were built from three aspects: system reconnaissance detection emergence, system target identification emergence and system jamming emergence. Simulation results coincide with the NRCS emergence source analysis, which verifies …

Simulation Of Energy Optimization For Cooling Coil In Central Air-Conditioning System, Qinglong Meng, Xiuying Yan

Journal of System Simulation

Abstract: The optimized control of minimizing energy consumption in one air-conditioning system was studied. The main models that needed in the simulation software HVACSIM+ for air handling unit were introduced. The AHU and rooms of the first floor of certain Building were taken as the simulation target. Then the function describing the relation between the consumption of cooling coil and the chilled water velocity was found and taken as the objective function. With improved cyclic variable method, the controller^ parameters were optimized. Results show that the simulation system runs steadily while the controller works with the optimal parameters, and the …

Gait Simulation Of Snake Robot Based On Cpg Method, Gao Qin, Zhelong Wang, Weijian Hu, Lanying Zhao

Journal of System Simulation

Abstract: Biological snakes in nature have a variety of periodic motion patterns such as serpentine motion, linear motion and lateral motion. Gaits diversity has greatly improved the adaptability of natural snakes to complex environment. Biologists has proved that such rhythmic movements of vertebrate animals are generated by CPG (the central neural pattern generator). Special mechanical structure of a snake robot with high degree offreedom and locomotion characteristics of different gaits was considered to bulid a suitale CPG network model. Hopf oscillators were chosen as neuron models of a central pattern generator owing to their stable features. A snake robot prototype …

Optimized Analysis Of Milling Thin-Wall Parts Based On Shell Element Models, Yusong Liao, Han Jiang

Journal of System Simulation

Abstract: As to milling the thin-wall part accurately, here with the finite element model composed of SHELL elements, the effects of lowering position, size of the part and cutting parameters to the deformations of the thin-wall part were analyzed and compared, and the corresponding theoretical analysis was provided. The conclusions can be got as; the model composed of SHELL elements can analyze the factors causing the deforroations of the thin-wall part and optimize the cutting method and cutting parameters effectively to improve the machining accuracy and efficiency.

Modeling And Simulation Of Traction System In Hybrid Shunting Locomotive, Kun Shen, Wang Ling, Wang Jian, Xiaoyang Yao

Journal of System Simulation

Abstract: The structure of traction system in hybrid shunting locomotive was analyzed, and the working principles of traction system under multiform power models were introduced. Based on which, the models of traction system and power accumulators of this hybrid shunting locomotive were built by MATLAB/Simulink, then the simulation experiments on hybrid shunting locomotive with the power models of hybrid power, and pure diesel generator power or pure accumulator power were done respectively. The simulation results show that hybrid shunting locomotive can achieve reliable operation in diflFerent conditions by the designed main circuit structure, parameters and system energy management strategy.

Impact Of Number Of Solar Cells In Parallel/Series And Temperature On Junction Capacitance, Zhigang Zhao, Chunjie Zhang, Gao Pu, Hutang Sang, Xiaoqian Li

Journal of System Simulation

Abstract: The study to the dynamic parameters of the photovoltaic cell is of crucial importance for the design of the corresponding afterward stage controller in the photovoltaic power generation system. Beginning with the analysis to physical mechanism of the photovoltaic cell, more accurate equivalent formulation of the junction voltage and bias voltage of the photovoltaic cells /module/array was deduced, and the explicit formulation between the bias voltage and output voltage was received by using the Lambert W function based on equivalent series resistance and the saturation current. The formulation was adopted which combined the engineering mathematics model and intrinsic carrier …

Evaluation And Analysis Of New Method Of Measurement Of Target Scattering Matrix, Zhenyu Huang, Zhao Bo, Huanyao Dai, Liandong Wang, Xuequan Zhou

Journal of System Simulation

Abstract: The basic theory of radar polarization signal processing is measurement of the target scattering matrix. The current algorithm of measurement of the target scattering matrix is gotten through orthogonal dual polarization channel in time-sharing or simultaneously, which demands orthogonal polarization measurement signals with complex coding. The measuring accuracy is higher. The complexity and the cost of the system are relatively higher. Scattering matrix can also be obtained by making use of the spatial polarization characteristics of the antenna to processing the radar echo without structure reformation of the radar. It only. needs to renewal measurement technology. The comparison and …

Molecule Optimization By Explainable Evolution, Binghong Chen, Tianzhe Wang, Chengtao Li, Hanjun Dai, Le Song

Machine Learning Faculty Publications

Optimizing molecules for desired properties is a fundamental yet challenging task in chemistry, material science, and drug discovery. This paper develops a novel algorithm for optimizing molecular properties via an Expectation-Maximization (EM) like explainable evolutionary process. The algorithm is designed to mimic human experts in the process of searching for desirable molecules and alternate between two stages: the first stage on explainable local search which identifies rationales, i.e., critical subgraph patterns accounting for desired molecular properties, and the second stage on molecule completion which explores the larger space of molecules containing good rationales. We test our approach against various baselines …

Multi-Modal Classification Using Images And Text, Stuart J. Miller, Justin Howard, Paul Adams, Mel Schwan, Robert Slater

SMU Data Science Review

This paper proposes a method for the integration of natural language understanding in image classification to improve classification accuracy by making use of associated metadata. Traditionally, only image features have been used in the classification process; however, metadata accompanies images from many sources. This study implemented a multi-modal image classification model that combines convolutional methods with natural language understanding of descriptions, titles, and tags to improve image classification. The novelty of this approach was to learn from additional external features associated with the images using natural language understanding with transfer learning. It was found that the combination of ResNet-50 image …

Low Light Image Enhancement Via Global And Local Context Modeling, Aditya Arora, Muhammad Haris, Syed Waqas Zamir, Munawar Hayat, Fahad Shahbaz Khan, Ling Shao, Ming-Hsuan Yang

Computer Vision Faculty Publications

Images captured under low-light conditions manifest poor visibility, lack contrast and color vividness. Compared to conventional approaches, deep convolutional neural networks (CNNs) perform well in enhancing images. However, being solely reliant on confined fixed primitives to model dependencies, existing data-driven deep models do not exploit the contexts at various spatial scales to address low-light image enhancement. These contexts can be crucial towards inferring several image enhancement tasks, e.g., local and global contrast, brightness and color corrections; which requires cues from both local and global spatial extent. To this end, we introduce a context-aware deep network for low-light image enhancement. First, …

Using Torchattacks To Improve The Robustness Of Models With Adversarial Training, William S. Matos Díaz

Cybersecurity: Deep Learning Driven Cybersecurity Research in a Multidisciplinary Environment

Adversarial training has proven to be one of the most successful ways to defend models against adversarial examples. This process consists of training a model with an adversarial example to improve the robustness of the model. In this experiment, Torchattacks, a Pytorch library made for importing adversarial examples more easily, was used to determine which attack was the strongest. Later on, the strongest attack was used to train the model and make it more robust against adversarial examples. The datasets used to perform the experiments were MNIST and CIFAR-10. Both datasets were put to the test using PGD, FGSM, and …

Fireeye: Cybersecurity In Action, Singapore Management University

Perspectives@SMU

FireEye built its success on its ‘Human + AI’ philosophy. But can a cybersecurity firm get ahead of the attackers and predict an attack…on itself?

Fairer Evaluation Of Zero Shot Action Recognition In Videos, Kaiqiang Huang, Sarah Jane Delany, Susan Mckeever

Conference Papers

Zero-shot learning (ZSL) for human action recognition (HAR) aims to recognise video action classes that have never been seen during model training. This is achieved by building mappings between visual and semantic embeddings. These visual embeddings are typically provided via a pre-trained deep neural network (DNN). The premise of ZSL is that the training and testing classes should be disjoint. In the parallel domain of ZSL for image input, the widespread poor evaluation protocol of pre-training on ZSL test classes has been highlighted. This is akin to providing a sneak preview of the evaluation classes. In this work, we investigate …

Deep Learning Methods For Fingerprint-Based Indoor And Outdoor Positioning, Fahad Alhomayani

Electronic Theses and Dissertations

Outdoor positioning systems based on the Global Navigation Satellite System have several shortcomings that have deemed their use for indoor positioning impractical. Location fingerprinting, which utilizes machine learning, has emerged as a viable method and solution for indoor positioning due to its simple concept and accurate performance. In the past, shallow learning algorithms were traditionally used in location fingerprinting. Recently, the research community started utilizing deep learning methods for fingerprinting after witnessing the great success and superiority these methods have over traditional/shallow machine learning algorithms. The contribution of this dissertation is fourfold:

First, a Convolutional Neural Network (CNN)-based method for …