The continuous proliferation of more complex and various security threats leads to the conclusion that new solutions are required. Intrusion Detection Systems can be a pertinent solution because they can deal with the...
详细信息
The continuous proliferation of more complex and various security threats leads to the conclusion that new solutions are required. Intrusion Detection Systems can be a pertinent solution because they can deal with the large data volumes of logs gathered from the multitude of systems and can even identify new types of attacks if based on anomaly detection. In this paper we propose an IDS model which includes two stages: feature selection with information gain and detection with Support Vector Machines (SVM). A draw-back of SVM is that its performance results are influenced by its user input parameters. Therefore, in order to better the classifier we exploit the advantages of a recent Swarm Intelligence (SI) algorithm, the Bat Algorithm (BA), which we improve by enhancing its randomization with Lévy flights. We test our model for the NSL-KDD dataset and prove that it can outperform the original BA, ABC or the popular PSO.
The paper investigates the expressive power of the temporal query language LH. We show that First-Order Logic is unable to formulate queries such as temporal connectivity, which can be naturally expressed by LH. The p...
详细信息
The paper investigates the expressive power of the temporal query language LH. We show that First-Order Logic is unable to formulate queries such as temporal connectivity, which can be naturally expressed by LH. The paper describes in detail our application of the Ehrenfeucht-Fräýsse method, which is used to examine limitations in the expressive power of First-Order Logic.
The paper focuses on building the mathematical model of a filling/drainage process (namely ASTANK2), by analytical approach. Models are subsequently tested by simulation. The process is versatile: it consists of two t...
详细信息
The paper focuses on building the mathematical model of a filling/drainage process (namely ASTANK2), by analytical approach. Models are subsequently tested by simulation. The process is versatile: it consists of two tanks with different geometric shape, which could be considered in independent or coupled setup and are supplied by an adjustable water distribution pipelines network. The proposed model is nonlinear and can be defined in various configurations, by connecting the desired component models, from the available ones. A simulator that integrates these components is implemented in MATLAB-SIMULINK environment. Its goal is twofold: to simulate the plant dynamics according to several scenarios, on one hand, and to validate the proposed analytical model, on the other hand. Some simulation results are presented in the end.
The adaptive-robust control strategy, proposed in this paper, combines the advantages of robust and adaptive control, while minimizing the computational load. Usually, a lot of mathematical operations are necessary, w...
详细信息
The adaptive-robust control strategy, proposed in this paper, combines the advantages of robust and adaptive control, while minimizing the computational load. Usually, a lot of mathematical operations are necessary, which limit the real time applicability of such strategies. By developing a simple, yet useful rule, the number of operations needed to determine if a new controller is needed has been greatly reduced. That is the most repetitive part of an adaptive - robust algorithm, because the rule check has to be done at every sampling moment. To test the new rule, a simulation software was developed, which proves the efficiency of the new rule. There are no disadvantages for the new adaptive - robust strategy, when compared to the usual approach.
OSPF is one of the most commonly used interior gateway protocols due to the ability to run between pieces of equipment produced by different vendors and its capacity of splitting the network into areas and fast conver...
详细信息
OSPF is one of the most commonly used interior gateway protocols due to the ability to run between pieces of equipment produced by different vendors and its capacity of splitting the network into areas and fast convergence. In this paper, we propose an approach for setting OSPF weights, using Cisco onePK API, in order to minimize congestion in small networks. It is based on a genetic algorithm, combined with two evaluation functions. The first evaluation function is based on getting loads on links depending on costs, while the other is based on getting the maximum flow in the network. Information about link load is obtained through onePK SDK, a Software Defined Networking solution developed by Cisco.
This paper presents different approaches for developing a speaker recognition system to be used in a voice control interface for an assistive domotic system. Experimental research has been carried out in Matlab, using...
详细信息
This paper presents different approaches for developing a speaker recognition system to be used in a voice control interface for an assistive domotic system. Experimental research has been carried out in Matlab, using the Voicebox toolbox for data preparation and feature extraction and using libSVM for speaker recognition.
The main goal of network-connected machines is to improve the quality of lives in numerous ways: from speeding up traffic, reducing fuel consumption, saving lives, improving access to healthcare, down to making everyd...
详细信息
The main goal of network-connected machines is to improve the quality of lives in numerous ways: from speeding up traffic, reducing fuel consumption, saving lives, improving access to healthcare, down to making everyday life more convenient and interesting. The development will make possible to include a chip in everything, enabling the developers and innovators in diverse industries to collaborate and to drive the growth in connections. In this paper, we review the functionalities of the M2M ecosystem together with the standard models and explore the possibilities to use a specific connection management platform enabled by 3GPP mobile networks and utilizing OMA Lightweight M2M Enabler (LWM2M) for device management. Although this technology targets mainly constrained devices with low power and small amount of memory, it can be also utilized with powerful devices and sensors that can benefit from its efficient communication and strong security mechanisms. Being based on the very last standards in M2M communication there is no other documented approach or platform to set against.
Autonomous manipulation is one the the most complex fields of humanoid robotics research, because it involves several domains such as: object recognition, object tracking, manipulator positioning, grasping, etc. In th...
详细信息
Autonomous manipulation is one the the most complex fields of humanoid robotics research, because it involves several domains such as: object recognition, object tracking, manipulator positioning, grasping, etc. In this paper a tactile controller based on fuzzy logic is proposed, capable of complex inhand manipulation tasks. This fuzzy logic approach assures finger motion smoothness by dealing with sensor nonlinearities and uncertainties, thus reducing the need of knowing the object dimensions or exact position in relation to the robot hand.
This work treats the problem of blood pressure regulation in post cardiac surgery patients with an improved model and reinforcement learning. The latter is the focus of the present study, specifically showing that it ...
详细信息
This work treats the problem of blood pressure regulation in post cardiac surgery patients with an improved model and reinforcement learning. The latter is the focus of the present study, specifically showing that it is suitable for the automaticcontrol of blood pressure in post cardiac surgery patients. The context is the direct use of clinical data, with no need for a patient model. Classical control, with PID or Polynomial RST algorithms, has also been used for blood pressure regulation. An existing model of patient is used here and improved. We use a model here for the only purpose of generating pseudo-clinical data for the use of reinforcement learning. The obtained results confirm that this approach can be an alternative to the classical control.
This paper addresses the development of general purpose game agents able to learn a vast number of games using the same architecture. The article analyzes the main existing approaches to general game playing, reviews ...
详细信息
This paper addresses the development of general purpose game agents able to learn a vast number of games using the same architecture. The article analyzes the main existing approaches to general game playing, reviews their performance and proposes future research directions. Methods such as deep learning, reinforcement learning and evolutionary algorithms are considered for this problem. The testing platform is the popular video game console Atari 2600. Research into developing general purpose agents for games is closely related to achieving artificial general intelligence (AGI).
暂无评论