Location determination with a cloud radio access network
A communication system that includes a plurality of radio points is disclosed. Each radio point is configured to exchange radio frequency (RF) signals with a wireless device that transmits a Sounding Reference Signal (SRS) from a first physical location in a site; and extract at least one SRS metric from the SRS. The communication system also includes a baseband controller communicatively coupled to the plurality of radio points. The baseband controller is configured to determine a signature vector based on the at least one SRS metric from each of the plurality of radio points. The communication system also includes a machine learning computing system communicatively coupled to the baseband controller. The machine learning computing system is configured to use a machine learning model to determine location data for the first physical location of the wireless device based on the signature vector.
1. A communication system, comprising:
a plurality of radio points serving at least a cell, each radio point configured to:
exchange radio frequency (RF) signals with a wireless device that transmits a Sounding Reference Signal (SRS) from a first physical point location in a site; and
extract at least one SRS metric from the SRS;
a baseband controller communicatively coupled to the plurality of radio points, wherein the baseband controller is configured to determine a signature vector based on the at least one SRS metric from each of the plurality of radio points; and
a machine learning computing system communicatively coupled to the baseband controller, wherein the machine learning computing system is configured to:
use a machine learning model to determine location data for the first physical point location of the wireless device based on the signature vector for the wireless device and training signature vectors;
wherein the location data for the first physical point location of the wireless device is further based on user-indicated training physical point locations received from at least one training wireless device, wherein each user-indicated training physical point location is a single point; and
during a training phase, correlate one of the training signature vectors with training location data based on a first timestamp of when the training location data is associated with the training signature vector and a second timestamp of when the training signature vector is determined at the baseband controller.
2. The communication system of claim 1 , wherein the at least one SRS metric comprises an SRS power measurement and a channel impulse response, both of which are measured from the SRS.
3. The communication system of claim 1 , wherein the at least one SRS metric comprises at least one of an SRS power measurement, a channel impulse response measured from the SRS, an angle of arrival for the SRS, or at least one previous signature vector for the wireless device.
4. The communication system of claim 1 , wherein determining the location data for the first physical point location comprises matching the received signature vector to a closest-matching signature vector measured during the training phase, where each signature vector comprises an element for each of the plurality of radio points.
5. The communication system of claim 1 ,
wherein, during the training phase, one of the at least one training wireless device is configured to:
move to one of the user-indicated training physical point locations in the site;
tag the user-indicated training physical point location to associate the training location data for the user-indicated training physical point location with the first timestamp;
transmit the training location data for the user-indicated training physical point location to the machine learning computing system; and
transmit a training signal that is received by at least one of the plurality of radio points.
6. The communication system of claim 5 ,
wherein, during the training phase, each of the at least one radio point is configured to determine at least one training signal metric based on the training signal;
wherein, during the training phase, the baseband controller is further configured to determine one of the training signature vectors based on the at least one training signal metric from the at least one radio point;
wherein, during the training phase, the machine learning computing system is further configured to correlate the training signature vector with the location data further based on a radio network temporary identifier (RNTI) of the wireless device; and
wherein, during the training phase, the machine learning computing system is further configured to train the machine learning model based on the correlated signature vector and the training location data that are correlated together.
7. The communication system of claim 1 ,
wherein the wireless device is further configured to transmit an additional SRS at a second physical point location in the site;
wherein each of the plurality of radio points is further configured to:
receive the additional SRS from the wireless device; and
extract at least one additional SRS metric from the additional SRS.
8. The communication system of claim 7 ,
wherein the baseband controller is further configured to determine an additional signature vector based on the at least one additional SRS metric from each of the plurality of radio points; and
wherein the machine learning computing system is further configured to determine additional location data for the second physical point location based on the additional signature vector.
9. The communication system of claim 1 , wherein the plurality of radio points and the baseband controller implement a Long Term Evolution (LTE) Evolved Node B (eNB).
10. The communication system of claim 1 ,
wherein the baseband controller is further configured to track real-time location data for a plurality of wireless devices at the site over a period of time; and
determine, following the training phase, a heat map that aggregates one or more or more of channel conditions, traffic density, user density, and user movement patterns at the site as a function of location and time based on the real-time location data.
11. The communication system of claim 1 , wherein, during the training phase, at least one user tags a respective physical point location associated with each training signature vector by pointing to the respective physical point location on a map of the site displayed on the training wireless device.
12. The communication system of claim 1 , wherein the at least one SRS metric comprises at least one previous signature vector for the wireless device.
13. A method for determining a first physical point location of a wireless device, comprising:
transmitting, from the wireless device, a Sounding Reference Signal (SRS) from the first physical point location, wherein the SRS is received by a plurality of radio points in a communication system;
extracting, by each of the plurality of radio points serving at least a cell, at least one SRS metric from the SRS;
determining a signature vector based on the at least one SRS metric from each of the plurality of radio points; and
using a machine learning model to determine location data for the first physical point location of the wireless device based on the signature vector for the wireless device and training signature vectors;
wherein the location data for the first physical point location of the wireless device is further based on user-indicated training physical point locations received from at least one training wireless device,
wherein each user-indicated training physical point location is a single point
during a training phase, correlating one of the training signature vectors with training location data based on a first timestamp of when the training location data is associated with the training signature vector and a second timestamp of when the training signature vector is determined.
14. The method of claim 13 , wherein the at least one SRS metric comprises an SRS power measurement and a channel impulse response, both of which are measured from the SRS.
15. The method of claim 13 , wherein the at least one SRS metric comprises at least one of an SRS power measurement, a channel impulse response measured from the SRS, an angle of arrival for the SRS, or at least one previous signature vector for the wireless device.
16. The method of claim 13 , wherein determining the location data for the first physical point location comprises matching the received signature vector to a closest-matching signature vector measured during the training phase, where each signature vector comprises an element for each of the plurality of radio points.
17. The method of claim 13 , wherein the training phase comprises:
moving, by one of the at least one training wireless device, to one of the user-indicated training physical point locations in a site with the radio points;
tagging, by the training wireless device, the user-indicated training physical point location to associate the training location data for the user-indicated training physical point location with the first timestamp;
transmitting, by the training wireless device, the training location data for the user-indicated training physical point location to a machine learning computing system; and
transmitting, by the training wireless device, a training signal that is received by at least one of the plurality of radio points.
18. The method of claim 17 , wherein the training phase further comprises:
determining, by each of the at least one radio point, at least one training signal metric based on the training signal;
determining, by a baseband controller in the communication system, one of the training signature vectors based on the at least one training signal metric from the at least one radio point;
correlating the training signature vector with the location data further based on a radio network temporary identifier (RNTI) of the wireless device; and
training, by the machine learning computing system, the machine learning model based on the correlated signature vector and the training location data for the physical point location.
19. The method of claim 13 , further comprising:
transmitting, from the wireless device, an additional SRS at a second physical point location in a site with the radio points;
receiving, by each of the plurality of radio points, the additional SRS from the wireless device; and
extracting, by each of the plurality of radio points, at least one additional SRS metric from the additional SRS.
20. The method of claim 19 , further comprising:
determining, by a baseband controller in the communication system, an additional signature vector based on the at least one additional SRS metric from each of the plurality of radio points; and
determining additional location data for the second physical point location based on the additional signature vector.
21. The method of claim 13 , wherein the plurality of radio points and a baseband controller in the communication system implement a Long Term Evolution (LTE) Evolved Node B (eNB).