Communications: directive radio wave systems and devices (e.g. – Directive – Position indicating
Reexamination Certificate
2001-04-23
2002-11-26
Phan, Dao (Department: 3662)
Communications: directive radio wave systems and devices (e.g.,
Directive
Position indicating
Reexamination Certificate
active
06486831
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to methods and apparatus for estimating the accuracy of measurement signals and, more particularly, to techniques for preventing use of spurious or low-quality range measurement signals in determining an object's position and for estimating the accuracy of acceptable range measurement signals based on measurement history information and operational parameters.
2. Description of the Related Art
Tracking filters are commonly used in a variety of contexts to estimate the present state of an entity by processing raw measurements relating to the entity. For example, the relative or absolute position of an object can be determined by taking measurements, such as series of range measurements, that indicate position. Typically, such measurements have some degree of inaccuracy due to the presence of noise or interference which introduces errors in the measurement values. By tracking the position of the object over time, a tracking filter essentially reduces the uncertainty caused by measurement noise and develops a more accurate estimate of the object's position than would be possible from simply assuming that each measurement accurately reflects the object's true position.
Minimal means-square-error (MSE) filters, such as the well-known Kalman filter, attempt to minimize errors in the tracked position of an object by appropriately weighting the impact of each measurement as a function of the reported accuracy of the measurement. When a new measurement is received, the filter predicts the position of the object at the present time by extrapolating from the previously estimated state of the object. The filter also estimates the accuracy of the predicted current position. The accuracy of the measured position is conventionally determined as a function of the received signal-to-noise ratio (or the signal-to-interference ratio where significant interference is present in addition to noise), with a higher signal-to-noise ratio translating into a higher measurement accuracy. To update the state of the object (e.g., the estimated position and velocity in three dimensions) with the new measurement, the filter must decide the relative extent to which it trusts the predicted current position and the new measurement. If the accuracy of the measurement is high relative to the accuracy of the predicted position, the filter will incorporate the measurement into the position solution using a high filter gain, meaning that the updated position estimate will rely more heavily on the measurement than on the predicted position. Conversely, if the accuracy of the measurement is low relative to the accuracy of the predicted position, the filter will incorporate the measurement into the position solution using a low filter gain, meaning that the updated position estimate will rely more heavily on the predicted position than on the measurement, such that the measurement will have less impact on the position estimate generated by the tracking filter.
Although the signal-to-noise ratio is conventionally relied upon to gauge the accuracy of position measurements in the filter updating process, there are circumstances in which the signal-to-noise ratio alone may not fully reflect the accuracy of the measurement or the extent to which the tracking filter should rely on the measurement. In the case of measuring the range to an object or another device, a precise determination of the signal propagation time between the devices must be made. The signal propagation time can be derived by knowing the transmission and reception times of one or more ranging signals traveling along a direct path between the devices.
For example, the well-known global positioning system (GPS) relies on measurement of the one-way propagation time of signals sent from each of a set of satellites to a receiving device in order to determine the range to each satellite and the position of the receiving device. Position location systems that relies on a two-way, round-trip ranging signal scheme are described in U.S. patent application Ser. No. 09/365,702, filed Aug. 2, 1999, entitled “Method and Apparatus for Determining the Position of a Mobile Communication Device Using Low Accuracy Clocks” and U.S. patent application Ser. No. 09/777,625 filed Feb. 6, 2001, entitled “Methods and Apparatus for Determining the Position of a Mobile Communication Device”, the disclosures of which are incorporated herein by reference in their entireties. In the ranging schemes described in these applications, a master mobile communication device transmits outbound ranging signals to plural reference communication devices which respond by transmitting reply ranging signals that indicate the location of the reference radio and the signal turn around time (i.e., the time between reception of the outbound ranging signal and transmission of the reply ranging signal). Upon reception of the reply ranging pulse, the master radio determines the signal propagation time, and hence range, by subtracting the turn around time and internal processing delays from the elapsed time between transmission of the outbound ranging pulse and the time of arrival of the reply ranging pulse. The accuracy of the position determined by these systems depends largely on the accuracy with which the receiving devices can determine the time of arrival of the ranging signals traveling along a direct path between the devices.
In an environment where multipath interference is significant, it is possible to mistakenly identify a strong multipath signal as the direct path signal. Since a multipath signal travels along an indirect path between the transmitter and receiver, the signal propagation time and, hence, the observed range differ from that of the direct path. In a position determining system relying on precise measurements of direct-path signal propagation time to determine range, erroneously interpreting a multipath signal as the direct path signal can drastically degrade performance. In particular, a multipath signal may result in a severely erroneous range measurement; nevertheless, if the multipath signal has a relatively high signal-to-noise ratio, the erroneous range measurement will be reported to the tracking filter as being highly accurate. Consequently, the filter will be misled into placing a high degree of reliance on a severely erroneous range measurement, thereby degrading the accuracy of the position estimate without the degraded accuracy being immediately known or reported.
As described in the aforementioned patent applications, one approach to avoiding the problem of accuracy degradation caused by multipath signals is to use frequency diversity to find a transmission frequency and phase that minimize multipath interference. A rake filter or equalizer can also be employed to separately identify the direct path signal and prominent multipath signals in order to separate or constructively combine these signals. Nevertheless, even with technique such as these, it is possible to measure range with a significant error that is not correctly represented by the signal-to-noise ratio of the ranging signal from which the range measurement is derived.
Even where the signal-to-noise ratio can be trusted as a indicator of measurement accuracy, there may be other measurement information available to supplement the signal-to-noise ratio in estimating the measurement accuracy. For example, the receiving device may have knowledge of the severity of multipath interference and the precision with which the signal arrival time is determined, and the history of recent measurements may suggest the extent to which the latest measurement should be relied upon. Failure to account for such factors in reporting the accuracy of the measurement to a tracking filter may result in a less accurate estimate of position. Accordingly, there remains a need to identify and prevent the use of spurious or unacceptably low accuracy measurements in systems that perform position estimation from measurement signals as well a
Galgano Steven
Martorana Marc J.
Martorana Scott C.
Edell, Shapiro, Finnan & Lytle LLC
ITT Manufacturing Enterprises Inc.
Phan Dao
LandOfFree
Methods and apparatus for estimating accuracy of measurement... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for estimating accuracy of measurement..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for estimating accuracy of measurement... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2951169