Many practical applications including speech and audio processing, signal processing, system identification, econometrics and time series analysis involve the problem of reconstructing a dynamic system model from data...
详细信息
Many practical applications including speech and audio processing, signal processing, system identification, econometrics and time series analysis involve the problem of reconstructing a dynamic system model from data observed with noise in all variables. We consider an important class of dynamic single-input single-output nonlinear systems where the system model is polynomial in observations but linear in parameters, which captures a wide range of such systems. Assuming white Gaussian measurement noise that is characterized by a magnitude and a covariance structure, we propose a nonlinear extension to the generalized Koopmans-Levin method that can estimate parameters of dynamic nonlinear systems with polynomial nonlinearities given a priori knowledge on the noise covariance structure. In order to estimate noise structure, we apply a covariance matching objective function. Combining the extended Koopmans-Levin and the covariance matching approaches, an identification algorithm to estimate both model and noise parameters is proposed. The feasibility of the approach is demonstrated by Monte-Carlo simulations.
In this paper we address the overall message delay analysis of IEEE 802.16 wireless metropolitan area network with contention-based multiple access of bandwidth requests. The overall delay consists of the reservation ...
详细信息
ISBN:
(纸本)9783642022043
In this paper we address the overall message delay analysis of IEEE 802.16 wireless metropolitan area network with contention-based multiple access of bandwidth requests. The overall delay consists of the reservation and scheduling components. Broadcast polling is used for bandwidth reservation with binary exponential backoff (BEB) collision resolution protocol and a simple scheduling is applied at;the base station. An analytical model is developed with Poisson arrival flow for the Non Real-Time Polling Service (nrtPS) class. The model enables asymmetric traffic flows, different message sizes at the subscriber stations Old also allows for Best Effort (BE) service class. An approximation of the mean overall delay is established for the nrtPS service class. The analytical Model is verified by Means Of simulation.
In the last few years many Web sites were created to build social networks. These sites have common behavior, for example they need their users to search and mark their friends, so those are the users who build the gr...
详细信息
In the last few years many Web sites were created to build social networks. These sites have common behavior, for example they need their users to search and mark their friends, so those are the users who build the graph themselves. This graph is fixed from a certain point of view: the connections are binary connections and the graph consists of only the explicit data given by the users. In this paper we define a social network based on e-mails which is distributed and asymmetric. The key differences between this type of network and usual networks are that using this network we do not need to register to a specific Web site and we do not need to mark our friends. The social graph can be built automatically based on the mass of e-mails and the address book belonging to us. By analyzing this huge amount of data we can reveal the connections between our contacts and build a weighted graph. Besides this we can create a graph from the e-mails using the same data to improve the search results when looking for a specific e-mail or contact. These weighted graphs can be considered as neural networks. We can create an adaptive method to modify the connections in the graphs with implicit data gathered from the behavior of the user. Moreover, we can define a search algorithm on this type of graph based on the activation model of the neural networks. In this paper we show a way to build this kind of networks and we define an algorithm which finds similar elements to a specific item.
An agent-based decentralized control of dual channel, isolated resonant dc-dc converters is presented. Discrete-time control will be implemented in the control agents. Selection process of the most suitable converter ...
详细信息
An agent-based decentralized control of dual channel, isolated resonant dc-dc converters is presented. Discrete-time control will be implemented in the control agents. Selection process of the most suitable converter parameters is also presented.
The paper describes a new agent-based, decentralized architecture and control for parallel dc-dc converters, supplying low and moderate power consumers in the range of 0-250 W. The system provides increased flexibilit...
详细信息
ISBN:
(纸本)9781424444328
The paper describes a new agent-based, decentralized architecture and control for parallel dc-dc converters, supplying low and moderate power consumers in the range of 0-250 W. The system provides increased flexibility with autonomic system reconstruction capability in the case of misoperation or fault, significantly contributing to the increase of reliability. The distributed intelligence facilitates the optimization of the converters' operation according to predefined criteria.
This Model-Driven Development (MDD) facilitates the synthesis of application programs from models created using customized, domain-specific model processors. Model compilers can be realized by graph rewriting-based mo...
详细信息
This Model-Driven Development (MDD) facilitates the synthesis of application programs from models created using customized, domain-specific model processors. Model compilers can be realized by graph rewriting-based model transformation. In Visual Modeling and Transformation System (VMTS), metamodel-based rewriting rules facilitate to assign OCL constraints to model transformation rules. This approach supports validated model transformation. Unfortunately, the validation introduces a new concern that often crosscuts the functional concern of the transformation rules. To separate these concerns, an aspect-oriented solution is applied for constraint management. This paper introduces the identification method of the crosscutting constraints in metamodel-based model transformation rules. The presented algorithms facilitate the better understanding of the transformation, their easier constraint-based configuration, and make both the constraints and the rewriting rules reusable.
This paper considers a real-life railway timetable related problem, where a set of interconnected railway junction points form a railway network, which is essentially a directed graph with corresponding vertex and edg...
详细信息
This paper considers a real-life railway timetable related problem, where a set of interconnected railway junction points form a railway network, which is essentially a directed graph with corresponding vertex and edge capacities representing railway tracks and railway *** dynamic behavior of this model is driven by a timetable. Unforeseen weather and other external effects may contribute to delays in the timetable, which alters the behavior of the whole system. In this scenario a hierarchical fuzzy system is proposed that can suggest a possible outgoing delay for each train by evaluating a set of fuzzy rule bases using two input data. The first proposal does not take into account the possible propagation of delay in the whole railway network. In this article a negative feedback is applied on the hierarchical fuzzy *** fuzzy sets are optimized by an evolution based global search metaheuristics.
When building document categorization in distributed mobile environments, feature selection methods need to be employed to have a compact representation for each document topic and to reduce noise during classificatio...
详细信息
When building document categorization in distributed mobile environments, feature selection methods need to be employed to have a compact representation for each document topic and to reduce noise during classification. When interaction occurs between the nodes, locally retrieved features representing the document topic and their attributes have to be shared to have a more accurate estimation of the global classifier at every node. The network traffic should be kept at a minimum to reduce costs. We propose a probabilistic model for a keyword selection method, which makes a more thorough analysis possible and can be considered as a basis when sharing information. It can be used for building up the local document topic representations incrementally ensuring minimal network traffic. The description of the probabilistic model is complemented by experimental results.
This article presents a new approach to sound source separation. The introduced algorithm is based on spectral modeling of real instruments. The separation of independent sources is carried out by dividing the energy ...
详细信息
This article presents a new approach to sound source separation. The introduced algorithm is based on spectral modeling of real instruments. The separation of independent sources is carried out by dividing the energy of the mixture signal based on these instrument models. This way it is possible to regain some of the information that was lost when the independent sources were mixed together into a single signal. The paper presents the theory behind the proposed separation system, then focuses on the instrument model that is the basic element of the approach. Measurement results are given for polyphony levels from 2 to 10 demonstrating the separation quality, with special regard to the effect of prints on the result.
Optimization methods known from the literature include gradient techniques and evolutionary algorithms. The main idea of gradient methods is to calculate the gradient of the objective function at the actual point and ...
详细信息
Optimization methods known from the literature include gradient techniques and evolutionary algorithms. The main idea of gradient methods is to calculate the gradient of the objective function at the actual point and then to step towards better values according to this value. Evolutionary algorithms imitate a simplified abstract model of evolution observed in nature. Memetic algorithms traditionally combine evolutionary and gradient techniques to exploit the advantages of both methods. Our current research aims to discover the properties, especially the efficiency (i.e. the speed of convergence) of particular evolutionary and memetic algorithms. For this purpose the techniques are compared by applying them on several numerical optimization benchmark functions and on fuzzy rule base identification.
暂无评论