123,267 research outputs found
Recurrent Neural Network Based Narrowband Channel Prediction
No full textIn this contribution, the application of fully connected recurrent neural networks (FCRNNs) is investigated in the context of narrowband channel prediction. Three different algorithms, namely the real time recurrent learning (RTRL), the global extended Kalman filter (GEKF) and the decoupled extended Kalman filter (DEKF) are used for training the recurrent neural network (RNN) based channel predictor. Our simulation results show that the GEKF and DEKF training schemes have the potential of converging faster than the RTRL training scheme as well as attaining a better MSE performance
An unscented Kalman filter for freeway traffic estimation
No full textThis paper addresses the problem of freeway tra±c flow estimation. The freeway is considered as a network of components representing different freeway stretches called segments. The evolution of the traffic in a segment is modelled as a dynamic stochastic system, influenced by states of neighbour segments. Measurements are received only at boundaries between some segments and averaged within regular time intervals. An Unscented Kalman filter is developed and its performance is compared with a particle filter both for synthetic data and for real traffc data. The intended application is to supply traffc control systems with the estimated traffc state
Dynamics Between Malaysian Equity Market And Macroeconomic Variables : An Application Of Kalman Filter Model With Heteroskedastic Error [QA402.3. C514 2007 f rb].
No full textSejak diperkenalkan oleh Kalman dan Bucy (1960), model penapis Kalman telah mendapat penggunaan yang luas dalam dalam program ruang angkasa dan bidang kejuteraan kawalan. Namun begitu, pengaplikasiannya dalam bidang siri masa kewangan masih jarang digunakan dan jauh ketinggalan.
Ever since the pioneering work of Kalman and Bucy (1960), Kalman filter model has become widely used in the space programme and control engineering. However, its applications in financial time series have been very few and far in between
Stochastic Surface Models for Commodity Futures: A 2D Kalman Filter Approach
No full textWe propose a two-dimensional Kalman filter approach that, additional to the information contained in futures prices evolution over time, makes use of information contained in the term structure of commodity futures along a second dimension of maturities. This time-maturity surface reflects a complete realization of the stochastic process as an alternative to standard Kalman filtering of a limited vector of futures prices along the one-dimensional time line. Thus, the proposed methodology may use the full information from the entire surface dynamics, including links from all available maturities per period, which eventually should lead to more accurate model parameter estimates. The technique is illustrated using coal futures prices.commodity prices, spatial analysis, two-dimensional Kalman filter, energy markets, futures markets, stochastic dynamic model
On Kalman Filtering with Nonlinear Equality Constraints
No full textThe state space description of some physical systems possess nonlinear equality constraints between some state variables. In this paper, we consider the problem of applying a Kalman filter-type estimator in the presence of such constraints. We categorize previous approaches into pseudo-observation and projection methods and identify two types of constraints-those that act on the entire distribution and those that act on the mean of the distribution. We argue that the pseudo-observation approach enforces neither type of constraint and that the projection method enforces the first type of constraint only. We propose a new method that utilizes the projection method twice-once to constrain the entire distribution and once to constrain the statistics of the distribution. We illustrate these algorithms in a tracking system that uses unit quaternions to encode orientation
Estimação de tempos de chegada de ônibus urbano utilizando filtros de Kalman
No full textDissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Tecnológico. Programa de Pós-Graduação em Engenharia de Automação e Sistemas.A fim de diminuir congestionamentos, poluição do ar, consumo de combustível, entre outros, tem-se buscado constantemente o desenvolvimento e modernização de sistemas de transporte público, pois sistemas mais eficientes, confortáveis e convenientes atraem um maior número de pessoas. A aplicação de linhas de ônibus em ambientes urbanos, tem sido um dos modos de transporte público mais utilizados. Para operação eficiente destas linhas, é importante conhecer a posição do veículo em tempo real, possibilitando o controle dos instantes de partida e a implantação de sistemas de informação sobre chegadas futuras, melhorando a percepção de qualidade do serviço prestado. A predição dos tempos de chegada do ônibus depende de uma série de fatores (por exemplo, atrasos em interseções sinalizadas, número de passageiros em pontos de parada, etc.). Estes fatores aumentam significativamente o nível das incertezas associadas ao processo e à medição. Este trabalho apresenta um algoritmo para predição dos tempos de chegada de ônibus urbano em pontos de parada, utilizando a abordagem de filtro de Kalman com análise de dados históricos. Fatores que aumentam o nível das incertezas associadas ao processo e à medição são considerados como propriedades estocásticas das perturbações do processo. A geração de dados de medição é realizada através de dois cenários distintos desenvolvidos e simulados no software de simulações microscópicas Aimsun 6.1. O teste de ajustamento de Kolmogorov-Smirnov é aplicado para análise das distribuições estatísticas destes dados. Os parâmetros necessários para configuração do filtro de Kalman são obtidos a partir de dados históricos através de dois métodos de análise estatística propostos: o método de análise longitudinal e o método de análise transversal. O filtro de Kalman é utilizado para estimação de dois estados do veículo, sua posição e sua velocidade. Por fim, é proposto um algoritmo que utiliza as estimações oriundas do filtro de Kalman para realizar a predição dos tempos de chegada do ônibus em pontos de parada.In order to reduce traffic congestions, air pollution, fuel consumption, and others, it has been constantly sought the development and modernization of public transportation systems, because more efficient, comfortable, and convenient systems attract more people. The application of bus lines in urban environments has been one of the public transportation modes most used. However, for efficient operation of bus lines, its important to know the vehicle position in real-time, enabling the control of departure times and implantation of information systems about future arrivals. The prediction of bus arrival time depends on a number of factors (e.g., delays at signalized intersections, number of passengers at bus stops, etc.). These factors increase significantly the level of uncertainties associated to process and measurement. This work presents an algorithm for prediction of bus arrival times at bus stops using Kalman filter with historical data analysis approach. Factors that increase the level of uncertainties associated to process and measurement are considerate stochastic properties of process disturbance. The generating of measurement data is performed by two different scenarios developed and simulated on the microscopic simulation software Aimsun 6.1. The Kolmogorov-Smirnov goodness of fit test is applied for analysis of the statistical distributions of these data. The parameters required for configuration of the Kalman filter are obtained from historical data through two proposed methods of statistical analysis: the method of longitudinal analysis, and the method of transversal analysis. The Kalman filter is used for estimation of two vehicle states, its position and its velocity. Finally, an algorithm is proposed that uses the estimates given by the Kalman filter to perform the prediction of bus arrival times at bus stops
A study on the effects of Kalman Filter on performance of IPMC-Based Active Vibration Control Scheme
No full textThis paper evaluates the effectiveness and performance of Ionic Polymer Metal Composite (IPMC) based active vibration control scheme equipped with the Kalman estimation algorithm. To assess the vibration attenuation efficiency, a rotating flexible manipulator has been modelled integrating two IPMC actuators following the modal approach. The elastic displacements as generalized coordinates for estimating optimal performance is carried out next by discretizing the elastic motion through the assumed mode technique and applying the Kalman filter. Simulations are then performed to demonstrate effective vibration attenuation using both IPMC and the Kalman filter. Kalman filter is employed for the whole vibrating system taking into account of the bending moment generated by the IPMC actuator. Experiment is conducted for the proposed damping scheme and the results are compared and verified with the simulation results
Restauração de imagens via filtragem de Kalman e considerações sobre a avaliação da qualidade de imagens restauradas
No full textTese (doutorado) - Universidade Federal de Santa Catarina. Centro Tecnológico. Programa de Pós-Graduação em Engenharia Elétrica.A presente tese trata da utilização de programação evolucionária (PE) em sistemas de restauração de imagens via filtragem de Kalman e da proposta de uma medida para avaliação da qualidade de imagens restauradas baseada na percepção visual humana. A PE é usada na etapa de estimação paramétrica do filtro de Kalman de modelo de ordem reduzida (reduced order model Kalman filter - ROMKF). Em conseqüência da função de controle da estimação paramétrica apresentar ótimos locais e da utilização de algoritmos de otimização sensíveis às condições iniciais, as estratégias tradicionais reiniciam o processo de restauração diversas vezes, com diferentes condições iniciais, na tentativa de contornar os problemas de convergência indesejável. Contudo, as simulações apresentadas mostram que a estratégia de reinícios é ineficiente e, por outro lado, uma única restauração via ROMKF-PE é suficiente para se obter uma imagem que é representativa do melhor que o sistema de restauração pode oferecer. Esta tese também propõe uma medida para avaliação da qualidade de imagens restauradas, denominada medida de qualidade composta (MQC), que é baseada na percepção visual humana. No desenvolvimento da MQC, são realizados experimentos que avaliam a correlação entre a percepção humana da qualidade em imagens e medidas objetivas dos efeitos de distorção em freqüência e injeção de ruído, considerados isoladamente. A MQC é baseada na medida de qualidade de ruído (noise quality measure - NQM) e na medida de qualidade de distorção em freqüência (MQD), sendo validada experimentalmente
An Economist´s guide to the Kalman filter
No full textAlmost since its appearance, the Kalman Filter (KF) has been successfully used in control engineering. Unfortunately, most of its important results have been published in engineering journals with language, notation and style proper of engineers. In this paper, we want to present the KF in an attractive way to economists by using information theory and Bayesian inference.
Kalman interpolation filter for channel estimation of LTE downlink in high-mobility environments
No full textThe estimation of fast-fading LTE downlink channels in high-speed applications of LTE advanced is investigated in this article. In order to adequately track the fast time-varying channel response, an adaptive channel estimation and interpolation algorithm is essential. In this article, the multi-path fast-fading channel is modelled as a tapped-delay, discrete, finite impulse response filter, and the time-correlation of the channel taps is modelled as an autoregressive (AR) process. Using this AR time-correlation, we develop an extended Kalman filter to jointly estimate the complex-valued channel frequency response and the AR parameters from the transmission of known pilot symbols. Furthermore, the channel estimates at the known pilot symbols are interpolated to the unknown data symbols by using the estimated time-correlation. This article integrates both channel estimation at pilot symbols and interpolation at data symbol into the proposed Kalman interpolation filter. The bit error rate performance of our new channel estimation scheme is demonstrated via simulation examples for LTE and fast-fading channels in high-speed applications
- …
