Techniques for building digital twins of cellular networks and for navigating therewith
View Patent ↗Systems and methods for building digital twins and navigating using digital twins. Reference signal and identifier data sent via antennas is obtained. Timing advance parameters for signals sent to the antennas are received. One or more machine learning models are applied to the reference signal data and timing advance parameters. Each machine learning model is trained to output antenna positions, antenna configurations, or both, and may be trained using training antenna configurations in the form of reference models representing different potential antenna configurations. Based on the outputs of the machine learning models, potential interference for devices occupying various locations within the network are determined. Navigation decisions, gaps in network coverage, or both, are determined based on the potential interference. As a result, interference by one or more devices moving within the network is reduced, thereby improving signal quality.
1 . A method for optimizing navigation using digital twins, comprising:
building a digital twin by applying at least one machine learning model to inputs of a network including reference signal strength values and timing and frequency synchronization parameters, wherein the at least one machine learning model is trained to output at least a position and a configuration of each antenna in the network, wherein the digital twin is a model of the network including a mapping of a plurality of antennas of the network and a plurality of respective antenna configurations for the plurality of antennas;
determining an interference value for each of a plurality of locations within the network based on the digital twin of the network; and
optimizing at least a portion of network communications within the network based on the determined interference value for each of the plurality of locations within the network.
2 . The method of claim 1 , wherein optimizing the network communications further comprises:
determining at least one navigation decision based on the determined interference value for each of the plurality of locations within the network; and
causing a device navigating within the network to act according to the at least one navigation decision.
3 . The method of claim 2 , further comprising:
determining an optimal navigation route for the device navigating within the network, wherein the at least one navigation decision is determined based on the optimal navigation route.
4 . The method of claim 1 , wherein optimizing the network communications further comprises:
identifying at least one gap in coverage for the network based on the determined interference value for each of the plurality of locations within the network, wherein each gap in coverage is a location within the network having an interference value above a threshold, wherein the network communications are optimized based on the at least one gap in coverage.
5 . The method of claim 4 , further comprising:
modifying a configuration of at least one of the plurality of antennas based on the at least one gap in coverage.
6 . The method of claim 4 , further comprising:
selecting at least one optimal base station for a device navigating within the network based on the determined interference value for each of the plurality of locations within the network, wherein the at least one optimal base station is selected to maximize stability of signal strength for communications with the device; and
utilizing the selected at least one optimal base station to send signals to the device.
7 . The method of claim 1 , wherein the at least one machine learning model is trained using a training set including a plurality of labels corresponding to respective training reference signal values and training timing and frequency synchronization parameters, the plurality of labels including labels indicating a plurality of antenna configurations and labels indicating a plurality of antenna positions.
8 . The method of claim 1 , wherein building the digital twin further comprises:
determining a plurality of shapes of radiation patterns within the network based on outputs of the at least one machine learning model; and
identifying a position of at least one antenna of the plurality of antennas based on at least one intersection between shapes of the plurality of shapes, wherein the mapping is based on the position of the at least one antenna.
9 . A non-transitory computer readable medium having stored thereon instructions for causing a processing circuitry to execute a process, the process comprising:
building a digital twin by applying at least one machine learning model to inputs of a network including reference signal strength values and timing and frequency synchronization parameters, wherein the at least one machine learning model is trained to output at least a position and a configuration of each antenna in the network, wherein the digital twin is a model of the network including a mapping of a plurality of antennas of the network and a plurality of respective antenna configurations for the plurality of antennas;
determining an interference value for each of a plurality of locations within the network based on the digital twin of the network; and
optimizing at least a portion of network communications within the network based on the determined interference value for each of the plurality of locations within the network.
10 . A system for optimizing navigation using digital twins, comprising:
a processing circuitry; and
a memory, the memory containing instructions that, when executed by the processing circuitry, configure the system to:
build a digital twin by applying at least one machine learning model to inputs of a network including reference signal strength values and timing and frequency synchronization parameters, wherein the at least one machine learning model is trained to output at least a position and a configuration of each antenna in the network, wherein the digital twin is a model of the network including a mapping of a plurality of antennas of the network and a plurality of respective antenna configurations for the plurality of antennas;
determine an interference value for each of a plurality of locations within the network based on the digital twin of the network; and
optimizing at least a portion of network communications within the network based on the determined interference value for each of the plurality of locations within the network.
11 . The system of claim 10 , wherein the system is further configured to:
determine at least one navigation decision based on the determined interference value for each of the plurality of locations within the network; and
cause a device navigating within the network to act according to the at least one navigation decision.
12 . The system of claim 11 , wherein the system is further configured to:
determine an optimal navigation route for the device navigating within the network, wherein the at least one navigation decision is determined based on the optimal navigation route.
13 . The system of claim 10 , wherein the system is further configured to:
identify at least one gap in coverage for the network based on the determined interference value for each of the plurality of locations within the network, wherein each gap in coverage is a location within the network having an interference value above a threshold, wherein the network communications are optimized based on the at least one gap in coverage.
14 . The system of claim 13 , wherein the system is further configured to:
modify a configuration of at least one of the plurality of antennas based on the at least one gap in coverage.
15 . The system of claim 13 , wherein the system is further configured to:
select at least one optimal base station for a device navigating within the network based on the determined interference value for each of the plurality of locations within the network, wherein the at least one optimal base station is selected to maximize stability of signal strength for communications with the device; and
utilize the selected at least one optimal base station to send signals to the device.
16 . The system of claim 10 , wherein the at least one machine learning model is trained using a training set including a plurality of labels corresponding to respective training reference signal values and training timing and frequency synchronization parameters, the plurality of labels including labels indicating a plurality of antenna configurations and labels indicating a plurality of antenna positions.
17 . The system of claim 10 , wherein the system is further configured to:
determine a plurality of shapes of radiation patterns within the network based on outputs of the at least one machine learning model; and
identify a position of at least one antenna of the plurality of antennas based on at least one intersection between shapes of the plurality of shapes, wherein the mapping is based on the position of the at least one antenna.