System and method for predicting wireless channel path loss
A system and method for applying supervised learning to model a second wireless channel environment based upon data collected for a first wireless channel environment. In various embodiments, regression techniques are used to overcome known channel modeling issues. Using the data of one particular communication environment, it is possible to predict a path loss model of a different communication environment. As such, the required number of measurements and the complexity of the model prediction is greatly reduced.
1. A computer implemented method for predicting a path loss model for a wireless channel, the method comprising:
creating a first wireless channel environment comprising a plurality of channel parameters;
collecting measurement data for the first wireless channel environment;
performing data processing to cleanse the measurement data of the first wireless channel environment;
performing AI-based techniques on the cleansed measurement data to predict a path loss for a second wireless channel environment, wherein the second wireless channel environment has one or more different channel parameters than the first wireless channel environment;
evaluating the predicted path loss for the second wireless channel environment; and
generating a path loss model for the second wireless channel environment.
2. The method of claim 1 , wherein the plurality of channel measurement parameters are selected from, distance, frequency, bandwidth, transmission power, scenario, polarization, transmission array type, receiver array type, antenna, antenna Azimuth and elevation.
3. The method of claim 1 , wherein collecting measurement data for the first wireless channel environment comprises collecting measurement data using a wireless channel simulator.
4. The method of claim 1 , wherein the measurement data is selected from T-R (transmitter-receiver) separation distance (m), time delay (ns), receiver power (dBm), RMS (root-mean-square) delay spread (ns), elevation AoD (degree), Azimuth AoD (degree), Azimuth AoA (degree) and elevation AoA (degree).
5. The method of claim 1 , wherein the AI-based techniques are selected from linear regression, multiple linear regression, support vector machine regression, decision tree regression and random forest regression.
6. The method of claim 1 , wherein evaluating the predicted path loss for the second wireless channel environment comprises using one or more of mean square error (MSE) and R-square comparison techniques.
7. The method of claim 1 , wherein the first wireless channel environment is an urban macro channel environment and the second wireless channel environment is an urban micro channel environment.
8. A non-transitory computer-readable medium, the computer-readable medium having computer-readable instructions stored thereon that, when executed by a computing device processor, cause the computing device to:
create a first wireless channel environment comprising a plurality of channel parameters;
collect measurement data for the first wireless channel environment;
perform data processing to cleanse the measurement data of the first wireless channel environment;
perform AI-based techniques on the cleansed measurement data to predict a path loss for a second wireless channel environment, wherein the second wireless channel environment has one or more different channel parameters than the first wireless channel environment;
evaluate the predicted path loss for the second wireless channel environment; and
generate a path loss model for the second wireless channel environment.
9. The non-transitory computer-readable medium of claim 8 , wherein the plurality of channel measurement parameters are selected from, distance, frequency, bandwidth, transmission power, scenario, polarization, transmission array type, receiver array type, antenna, antenna Azimuth and elevation.
10. The non-transitory computer-readable medium of claim 8 , wherein collecting measurement data for the first wireless channel environment comprises collecting measurement data using a wireless channel simulator.
11. The non-transitory computer-readable medium of claim 8 , wherein the measurement data is selected from T-R (transmitter-receiver) separation distance (m), time delay (ns), receiver power (dBm), RMS (root-mean-square) delay spread (ns), elevation AoD (degree), Azimuth AoD (degree), Azimuth AoA (degree) and elevation AoA (degree).
12. The non-transitory computer-readable medium of claim 8 , wherein the AI-based techniques are selected from linear regression, multiple linear regression, support vector machine regression, decision tree regression and random forest regression.
13. The non-transitory computer-readable medium of claim 8 , wherein evaluating the predicted path loss for the second wireless channel environment comprises using one or more of mean square error (MSE) and R-square comparison techniques.
14. A system for predicting a path loss model for a wireless channel, the system comprising:
a processor; and
a non-transitory computer-readable medium operably coupled to the processor, the computer-readable medium having computer-readable instructions stored thereon that, when executed by the processor, cause the system to;
create a first wireless channel environment comprising a plurality of channel parameters;
collect measurement data for the first wireless channel environment;
perform data processing to cleanse the measurement data of the first wireless channel environment;
perform AI-based techniques on the cleansed measurement data to predict a path loss for a second wireless channel environment, wherein the second wireless channel environment has one or more different channel parameters than the first wireless channel environment;
evaluate the predicted path loss for the second wireless channel environment; and
generate a path loss model for the second wireless channel environment.
15. The system of claim 14 , wherein the plurality of channel measurement parameters are selected from, distance, frequency, bandwidth, transmission power, scenario, polarization, transmission array type, receiver array type, antenna, antenna Azimuth and elevation.
16. The system of claim 14 , wherein collecting measurement data for the first wireless channel environment comprises collecting measurement data using a wireless channel simulator.
17. The system of claim 14 , wherein the measurement data is selected from T-R (transmitter-receiver) separation distance (m), time delay (ns), receiver power (dBm), RMS (root-mean-square) delay spread (ns), elevation AoD (degree), Azimuth AoD (degree), Azimuth AoA (degree) and elevation AoA (degree).
18. The system of claim 14 , wherein the AI-based techniques are selected from linear regression, multiple linear regression, support vector machine regression, decision tree regression and random forest regression.
19. The system of claim 14 , wherein evaluating the predicted path loss for the second wireless channel environment comprises using one or more of mean square error (MSE) and R-square comparison techniques.