Abstract
The real‐time positioning accuracy of the TI‐4100 was tested by comparison with an accurate off‐shore positioning system. The reasons for trends in the differences were investigated in detail. A correlation was noticed between position errors and extremely low satellite elevations and also between height errors and clock bias errors. These are analysed and the stability of the Rubidium frequency standard is discussed.
During good satellite coverage—as will be normal after 1988—the TI‐4100 performance was good, though automatic satellite selection would be preferable over the present operator selection. Most often agreement with conventional positioning was better than 10 meters.
Under bad geometry conditions, errors became sometimes several hundred meters, even though the system still gave the uncertainty in the computed positions as better than 10 meters. With the present limited availability, software improvements are desirable, especially in the altitude‐hold and TBR‐hold (Time Bias Rate) modes.