Indoor and outdoor geolocation and time of arrival estimation
A time of arrival of a signal received by a receiving device over a wireless channel is estimated by identifying by the receiving device a location in the time domain of a first path component in a waveform of the received signal, which includes applying a filter to the waveform; identifying in the waveform an observation window associated with an energy rise in the waveform; correlating the identified observation window in the waveform with correlator sequences, and determining a time of arrival offset of the first path component from results of correlating the observation window with the correlator sequences. The properties of the correlator sequences depend on properties of the filter such that the correlator sequences account for the filtering of the waveform during the correlating, the correlator sequences reduce the effect of colored noise after hypothetical first path components; and the correlator sequences correspond to different time shifts of a hypothetical first path component.
1. A method for estimating a time of arrival of a signal received by a receiving device over a wireless channel, the method comprising the step of identifying by the receiving device a location in the time domain of a first path component in a waveform of the received signal; wherein identifying the location of the first path component comprises:
applying a filter to the waveform;
identifying in the waveform an observation window associated with an energy rise in the waveform;
correlating the identified observation window in the waveform with correlator sequences,
wherein the correlator sequences account for the filtering of the waveform during the correlating, and
wherein the correlator sequences reduce the effect of colored noise after hypothetical first path component
wherein the correlator sequences correspond to different time shifts of a hypothetical first path component;
determining a time of arrival offset of the first path component from results of correlating the observation window with the correlator sequences.
2. The method of claim 1 wherein the observation window has at most 20 samples.
3. The method of claim 1 wherein each of the correlator sequences has at most as many samples as the observation window.
4. The method of claim 1 wherein each of the correlator sequences has at most 20 samples.
5. The method of claim 1 further comprising selecting the correlator sequences from a set of pre-stored correlator sequences.
6. The method of claim 1 further comprising selecting the correlator sequences from a set of correlator sequences computed in real time.
7. The method of claim 1 wherein determining a time of arrival offset of the first path component from results of correlating the observation window with the correlator sequences comprises minimizing a cost function.
8. The method of claim 1 wherein the filter is a near-causal filter.
9. The method of claim 8 wherein the near-causal filter has a flat response in a region before a main lobe.