GPS World - January 2008 - (Page 18)

SYSTEM DESIGN & TEST | Receiver Design adjustments that may be necessary in this case in the “Subtleties” section. Availability This quantity is deﬁned as the relative percentage of a given testing interval during which a receiver has a valid position ﬁx. A further distinction is sometimes made between 3D and 2D ﬁxes; this issue typically arises in the presence of frequent signal blockages, signiﬁcant multipath, signal attenuation, or otherwise degraded signal conditions. It is worth noting that the criteria for declaring a given ﬁx valid are entirely determined by each individual RUT, so equal availability does not necessarily imply equal accuracy (that is, some receivers may have more conservative position validity thresholds than others). Accuracy This quantity is deﬁned as the magnitude of the distance between the true position and the position ﬁx reported by the RUT at a speciﬁc instant. Depending on the mode of operation, a collection of successive ﬁxes can be used to calculate several additional statistical accuracy metrics, including mean and maximum error, 50th percentile radius (also known as Circular Error Probability (CEP) and Spherical Error Probability (SEP) for 2D and 3D ﬁxes, respectively), and 95th percentile radius. These and other accuracy metrics are discussed in detail in a pair of GPS World articles by Frank van Diggelen (January 1998 and January 2007). To evaluate accuracy in static data sets, it is suﬃcient to have a reference coordinate (either a surveyed antenna location or the position ﬁx programmed into a simulator). For dynamic sets, it is necessary to have an accurately timetagged truth data set to compute any of the metrics described above. Although it is mathematically possible to calculate CEP and SEP for dynamic data, the physical interpretation of these results is not entirely intuitive since the reference position is constantly moving. erefore, these two quantities are typically evaluated for static data sets only. 18 GPS World | January 2008 Time to Fix An important aspect of receiver performance is how long it takes for a position ﬁx to be reported back to the user after a “start” command is issued. The command can be initiated by the user, an automated process (for example, a task scheduler or a timer), or by the receiver itself (in the case of signal loss or other triggers). The most common time intervals of interest are: Time to First Fix (TTFF). As described above. is metric varies, sometimes signiﬁcantly, as a function of signal conditions and information available in receiver memory and/or through assistance messages, as described in the “Receiever Starts” section. Time to Reacquisition (TTR). Subsequent to a signal loss during normal receiver operation, this is the amount of time required to regain a valid ﬁx from the moment signals are restored. is metric varies as a function of the outage duration, receiver motion during the outage, and signal conditions. Receiver Starts The TTFF for a particular receiver strongly depends on initial conditions. The terms factory start, cold start, warm start, and hot start are widely used to describe starts with various pieces of information presumed to be available. These pieces of information may include: Quasi-Static Data: almanac (valid for seven days or longer); ephemeris (valid for 2–4 hours, or longer if extended algorithms are used); and ionospheric model parameters (valid for 6–12 hours) Variable Data: estimated (or last known) position; receiver real-time clock; receiver clock drift; and tracking parameters (code phase, Doppler) for all satellites in view. Conventionally, a factory start is assumed to mean that none of the data above is available, and the receiver performs a full search over all PRN codes and time and frequency oﬀsets (assuming worst-case clock drift) upon startup. (In some cases, a hard-coded default almanac may be available. is is a reasonable assumption because even an almanac several months old can still be useful for predicting satellite visibility from a given position, and some receivers ship with a pre-stored almanac for precisely this reason.) A cold start assumes the availability of a usable almanac and receiver clock drift from the most recent run, but nothing else. A warm start assumes the availability of a recent almanac, approximate position, receiver time, and clock drift. A hot start further assumes valid ephemeris is available, and may include a receiver’s last known estimates of code phase and Doppler for each satellite, as well as the receiver’s most recently calculated clock drift. Unfortunately, diﬀerent manufacturers sometimes deﬁne these terms in slightly diﬀerent ways, making it diﬃcult to perform meaningful comparisons between receivers. To avoid potential ambiguities and ensure accurate comparative results, it is preferable to avoid “type of start” labels whenever possible. Instead, any characterization of a receiver’s acquisition performance should explicitly specify which of the abovementioned pieces of information are available, and for data in the second group, the approximate accuracy of each. Experimental Setup The hardware testing setup is shown in FIGURE 1. Normally, the LNA is not needed and the receivers under test can be connected as shown. However, if a RUT is designed to work with an active antenna, an LNA with appropriate gain must be used to ensure that a reasonable amount of power is delivered to the GPS Generator RF Attenuator LNA Splitter RUT 1 RUT 2 RUT 3 RUT 4 p FIGURE 1 Hardware testing setup www.gpsworld.com http://www.gpsworld.com

Table of Contents Feed for the Digital Edition of GPS World - January 2008

GPS - January 2008 Contents Out in Front Expert Advice The Money-Go-Round u-Nav Latest Acquisition Apples to Apples Global SBAS 2008 GPS Receiver Survey Advertisers Index & Company Directory The Manufacturer's Road Year of the Who Working Indoor Up and Down Good, Better, Best Marketplace Classifieds Seen + Heard

GPS World - January 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.