Content area
Full Text
Abstract: The firm track range (FTR) is a key parameter for a radar system. Traditionally, the FTR has been defined in terms of an M-of-N confirmation scheme, but newer systems often use a truncated sequential probability ratio test (SPRT) for this purpose. A definition for an SPRT-based FTR is proposed.
(ProQuest: ... denotes formulae omitted.)
1 Introduction
A key parameter of interest for any practical radar system is the 'firm track range' (FTR), which is considered to be the range at which a target can be reliably tracked. When a radar system receives a detection that cannot be correlated with any existing tracks, there is usually an automated attempt to verify whether a tme target is present. In the past an 'M-out-of-.V ' confirmation scheme was generally used for this purpose, whereby N additional beams are transmitted in the direction of the initial detection, and a target is declared 'confirmed' if at least M detections are received. The 'FTR' has been historically defined as the range where a specified target will be confirmed with a probability of 0.9 or better [1], with mutual agreement between the customer and the contractor as to the appropriate values of 'M' and N '. Presently, many of the newer radar systems are using a sequential probability ratio test (SPRT) for this purpose, which somewhat confuses the FTR definition.
With an SPRT-based system detections that cannot be correlated to any existing tracks trigger the creation of a 'tentative track', which is typically not revealed to the system operator, and a judgment on whether a target is truly present is deferred until there are enough observations to render a statistically sound decision [2-4]. Unlike an M-of-N test, the number of observations needed to make a confirmation decision is not set to a single value. An SPRT approach can have several advantages over an M-of-N test, including the potential to render a faster decision, an increased FTR and the ability to correctly identify true targets in high clutter environments [3, 4]. This kind of an approach could also be beneficial in that it can prevent the operator from having to contend with false track reports, although it can also hide a plethora of tentative tracks that are slow to reach...