
FASTRAK MANUAL
Rev. G
E-1
JUNE 2012
APPENDIX E. ‘Latency’ White Paper
TECHNICAL NOTE
Latency - 3SPACE™ FASTRAK
®
H. R. Jones
INTRODUCTION
ANSI/IEEE Std 100-1977 defines latent period as “The time elapsing between the
application of a stimulus and the first indication of a response”. The definition excludes the time
required to transmit the response. It is in this context that we define the latent periods (1)
between the application of a synchronization pulse and a response, and (2) between the
application of receiver motion and a response. The “response” for both cases occurs when the
receiver coordinate solution is made ready for output, and, as noted above, does not include the
time required to transmit the coordinates over the interface in use (e.g. RS-232, MIL-STD-1553,
IEEE-488, etc).
Polhemus 3SPACE FASTRAK magnetic 6DOF measurement systems emanate low
frequency magnetic fields from a stationary transmitting antenna and sense them with a movable
receiving antenna. The received magnetic field samples are subjected to analog and digital
processes, and are ultimately solved for the receiver’s position and orientation coordinates. The
solutions are formatted in varied ways according to user selections, then output over various
types of interfaces depending on the product.
The FASTRAK system’s latent period is due to the time required to sample the magnetic
fields, solve for the receiver coordinates, and make the solutions available for output. However,
from the user’s point of view, the latent period may appear longer than this due to delays in the
interface or in the user’s computer, or due to (incorrectly) configured FASTRAK filters which
can make the response appear to occur later. These topics are discussed in the following
paragraphs.
SYNC-TO-OUTPUT LATENT PERIOD
Application of an external synchronization pulse
initiates magnetic field sampling, a
period that lasts about 3.5 ms. It goes beyond the scope of this note to explain the sampling
process in more detail, so let it suffice that nine magnetic field samples are taken per cycle time.
The samples are then solved for receiver coordinates, a period that requires another 2 ms. The
solution is then placed in an output buffer and is made ready for transmission over the interface
in use. The total “sync-to-output” latent period is the sum of field sampling and coordinate
solution periods, or 5.5 ms, and is independent of update rate.
EFFECTIVE LATENT PERIOD
“Sync-to-output” latency is important for reasons of interface timing; however, it does
not quantify the effective latent period between receiver motion and output coordinate values.
1
If external synchronization and continuous print are not implemented, the environment is being run asynchronously
and the latent period cannot be defined precisely.