Time-series of attenuation on EHF and SHF fixed satellite links derived from meteorological and forecast data

2002 ◽  
Author(s):  
R.J. Watson
Keyword(s):  
Time Series ◽  
Satellite Links ◽  
Forecast Data

scholarly journals Short-term prediction of rain attenuation level and volatility in Earth-to-Satellite links at EHF band

2008 ◽  
Vol 15 (4) ◽  
pp. 631-643 ◽  
Author(s):  
L. de Montera ◽  
C. Mallet ◽  
L. Barthès ◽  
P. Golé
Keyword(s):  
Time Series ◽  
Prediction Error ◽  
Nonlinear Models ◽  
Garch Model ◽  
Rain Attenuation ◽  
Frequency Scaling ◽  
Attenuation Level ◽  
Satellite Links ◽  
Short Term Prediction ◽  
The Cost

Abstract. This paper shows how nonlinear models originally developed in the finance field can be used to predict rain attenuation level and volatility in Earth-to-Satellite links operating at the Extremely High Frequencies band (EHF, 20–50 GHz). A common approach to solving this problem is to consider that the prediction error corresponds only to scintillations, whose variance is assumed to be constant. Nevertheless, this assumption does not seem to be realistic because of the heteroscedasticity of error time series: the variance of the prediction error is found to be time-varying and has to be modeled. Since rain attenuation time series behave similarly to certain stocks or foreign exchange rates, a switching ARIMA/GARCH model was implemented. The originality of this model is that not only the attenuation level, but also the error conditional distribution are predicted. It allows an accurate upper-bound of the future attenuation to be estimated in real time that minimizes the cost of Fade Mitigation Techniques (FMT) and therefore enables the communication system to reach a high percentage of availability. The performance of the switching ARIMA/GARCH model was estimated using a measurement database of the Olympus satellite 20/30 GHz beacons and this model is shown to outperform significantly other existing models. The model also includes frequency scaling from the downlink frequency to the uplink frequency. The attenuation effects (gases, clouds and rain) are first separated with a neural network and then scaled using specific scaling factors. As to the resulting uplink prediction error, the error contribution of the frequency scaling step is shown to be larger than that of the downlink prediction, indicating that further study should focus on improving the accuracy of the scaling factor.

A Time Series Analysis and Forecasting Model in Dam Safety Monitoring

2013 ◽  
Vol 756-759 ◽  
pp. 589-593
Author(s):  
Rui Lian Hou ◽  
Hu Gao ◽  
Hui Li
Keyword(s):  
Time Series ◽  
Time Series Analysis ◽  
Safety Monitoring ◽  
Measured Data ◽  
Forecasting Model ◽  
Dam Safety ◽  
Model Structure ◽  
Series Analysis ◽  
Dam Safety Monitoring ◽  
Forecast Data

Based on the new time series analysis and the theory of dam safety, this paper proposed a new forecasting model in dam safety monitoring. First the paper introduced basic method of flat forecast and described its algorithm routine, and explained what is time series. Second the model structure of monitoring system based on time series analysis was given through the analysis of the method. The system can provide the comparison chart of the measured data, forecast data and trends. At last the model was realized based on VB and tested by the actual data. Experimental results show that this forecasting model has better prediction results in dam safety monitoring.

scholarly journals Model Space Time Autoregressive (STAR) dan Aplikasinya Terhadap Penyakit Demam Berdarah Dengue di Provinsi Sulawesi Barat

2019 ◽  
Vol 1 (2) ◽  
pp. 115
Author(s):  
Wahidah Sanusi ◽  
Maya Sari Wahyuni ◽  
Rahmat Setiawan
Keyword(s):  
Time Series ◽  
Dengue Fever ◽  
Time Series Data ◽  
Model Space ◽  
Space Time ◽  
Series Data ◽  
Star Model ◽  
Autoregressive Time Series ◽  
Forecast Data ◽  
Model Star

Abstrak. Model Space Time Autoregressive (STAR) merupakan data deret waktu yang mempunyai keterkaitan antar lokasi (space time). Tujuan dari penelitian ini adalah untuk mendapatkan model STAR yang sesuai dengan data jumlah penderita penyakit DBD di Provinsi Sulawesi Barat serta memperoleh data hasil ramalan untuk beberapa bulan kedepan. Data yang digunakan merupakan data bulanan penderita DBD di lima lokasi yaitu Kota Mamuju, Kabupaten Majene, Kabupaten Polmas, Kabupaten Mamuju Tengah, dan Kabupaten Mamuju Utara pada Januari 2014 sampai Juli 2016. Pendugaan parameter model STAR menggunakan metode kuadrat terkecil (MKT). Model STAR yang sesuai dengan data jumlah penderita penyakit DBD di Provinsi Sulawesi Barat adalah model STAR5(11). Pembobot yang digunakan merupakan bobot lokasi seragam. Pada hasil pengecekan parameter penduga dengan menggunakan bobot lokasi seragam didapatkan tiga model. Hal ini dilihat dari adanya pengaruh yang nyata terhadap lokasi yang berdekatan. Hasil ramalan dengan model STAR5(11) tentang jumlah penderita penyakit DBD di Provinsi Sulawesi Barat untuk dua bulan kedepan yaitu bulan Agustus sampai September 2016.Kata Kunci: Model STAR, ARIMA, Autoregressive, Deret WaktuAbstract. The Space Time Autoregressive (STAR) model is a time series data that has a link between locations (space time). The purpose of this study was to obtain a STAR model that was in accordance with the data on the number of dengue fever patients in West Sulawesi Province and also the forecast data for the next few months. Data in the form of DHF data in five locations, namely Mamuju City, Majene Regency, Polmas District, Central Mamuju Regency, and North Mamuju Regency from January 2014 to July 2016. STAR Estimation parameter model uses vertical squares (MKT) method. The STAR model that matches the data on the number of DHF patients in West Sulawesi Province is the STAR5 model (11). The weighting is a uniform location. In the estimator checking results using uniform location weight of three models. Things that happen between others. Forecast results with the STAR5 (11) model on the number of dengue fever patients in West Sulawesi Province for the next two months, namely August to September 2016, namely 9 people for Mamuju City and 12 people for Polman Regency.Keywords: STAR Model, ARIMA, Autoregressive, Time Series

scholarly journals Implementation of α-Sutte Indicator to Forecasting Consumer Price Index in Turkey

2017 ◽  
Author(s):  
Ansari Saleh Ahmar ◽  
Abdul Rahman ◽  
Usman Mulbar
Keyword(s):  
Time Series ◽  
Test Data ◽  
Price Index ◽  
Time Series Data ◽  
Reference Data ◽  
Consumer Price Index ◽  
Training Data ◽  
Series Data ◽  
Forecasting Method ◽  
Forecast Data

α-Sutte Indicator (α-Sutte) was originally from developed of Sutte Indicator. Sutte Indicator can use to predict the movement of stocks. As the development of science, then Sutte Indicator developed to predict not only the movement of stocks but also can forecast data on financial, insurance, and time series data. This method called α-Sutte Indicator (α-Sutte). α-Sutte was developed using the principle of the forecasting method of using the previous data. The data used in this research is Consumer Price Index in Turkey data from January 2003 - June 2017. This data is divided into 2 parts, namely training data and test data. Training data starts from January 2003 - October 2016 and test data from November 2016 - June 2017. To see the accuracy of α-Sutte, it will be done benchmarking the results of forecasting with other forecasting method is Automatic Time Series Forecasting: The forecast Package for R (AutoARIMA) developed by Hyndman-Khandakar (2008). Comparison of this accuracy is to compare the value of MSE forecasting result on test data by using training data as reference data. Results obtained from this study that the MSE value of α-Sutte is smaller (5.697723) than MSE from AutoARIMA (292.5125). This indicates that α-Sutte is more suitable for predicting Consumer Price Index in Turkey data.

scholarly journals PERBAIKAN MODEL SEASONAL ARIMA DENGAN METODE ENSEMBLE KALMAN FILTER PADA HASIL PREDIKSI CURAH HUJAN

2019 ◽  
Vol 19 (1) ◽  
pp. 9
Author(s):  
Dwi Anugrah Wibisono ◽  
Dian Anggraeni ◽  
Alfian Futuhul Hadi
Keyword(s):  
Time Series ◽  
Kalman Filter ◽  
Ensemble Kalman Filter ◽  
Estimation Method ◽  
Filter Method ◽  
Sarima Model ◽  
Forecast Data ◽  
Kalman Filter Method ◽  
Algorithm Implementation

Forecasting is a time series analytic that used to find out upcoming improvement in the next event using past events as a reference. One of the forecasting models that can be used to predict a time series is Kalman Filter method. The modification of the estimation method of Kalman Filter is Ensemble Kalman Filter (EnKF). This research aims to find the result of EnKF algorithm implementation on SARIMA model. To start with, preticipation forecast data is changed in the form of SARIMA model to obtain some SARIMA model candidates. Next, this best model of SARIMA applied to Kalman Filter models. After Kalman Filter models created, forecasting could be done by applying pass rainfall data to the models. It can be used to predict rainfall intensity for next year. The quality of this forecasting can be assessed by looking at MAPE’s value and RMSE’s value. This research shows that enkf method relative can fix sarima method’s model, proved by mape and rmse values which are smaller and indicate a more accurate prediction. Keywords: Ensemble Kalman Filter, Forecast, SARIMA

Sign in / Sign up

Export Citation Format

Share Document