Forecasting of ionospheric delay over Parit Raja Station, Johor, using statistical method

Abstract

GPS is of great importance in precise positioning, however, the accuracy is marked by error sources, i.e. ionospheric effects. The signal information is delayed and the carrier phase experiences an advance due to the dispersive character of the ionosphere. The delay can be more than 100 meters in the worst case scenario and tends to increase with increasing solar activity. This error can be corrected by processing the GPS data if it is known. This paper describes the possibility of using statistical methods to forecast the ionospheric delay which shows repeatable patterns in time series. The statistical method used is the Holt-Winter method due to its ability to forecast time series with repeated trends and seasonal patterns. Based on the TEC data collected during a period of a month, ionospheric delay forecast is generated for the following month, which is then compared with the real data. Results show that there is a 6% error between the forecast and the real ionospheric delay and the error correction for the delay can be more than 90%.