Abstract: Methods and systems for monitoring a communication network using machine-learning techniques are disclosed. In some implementations, a forecasted amount of traffic for a communication network is determined using one or more network traffic forecasting models being configured to generate the forecasted amount of traffic based on data indicating one or more previous amounts of traffic for the communication network. A measure of network health is generated based on a measured amount of traffic and the forecasted amount of traffic. Data indicating one or more characteristics of the communication network is processed using one or more machine learning models to generate a predicted measure of network health for a future time period. An indication of the predicted measure of network health for the future time period is provided.

Abstract: Some implementations of the disclosure relate to dynamic switching of a satellite inroute data path between a Time Division Multiple Access (TDMA) method and a Time Division Multiplexing (TDM) method. In one implementation, a satellite terminal comprises one or more processors; and one or more non-transitory computer-readable storage media configured with instructions executable by the one or more processors to cause the satellite terminal to perform operations comprising: communicating, using the satellite terminal, over an inroute TDM channel; determining, based on an ingress traffic rate to the satellite terminal or a determination that the satellite terminal has not received any traffic flows classified for communication using TDM, to switch communications from the inroute TDM channel to an inroute TDMA channel; and after determining to switch communications, switching, at the satellite terminal, from communicating over the inroute TDM channel to communicating over the inroute TDMA channel.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for machine learning models for adjusting communication parameters. In some implementations, data for each device in a set of multiple communication devices is obtained. A machine learning model is trained based on the obtained data. The model can be trained to receive an indication of a geographic location and predict a communication setting capable of providing at least a minimum level of efficiency. After training the machine learning model, an indication of a predicted communication setting for a particular communication device is generated. A determination is then made whether to change a current communication setting for the particular communication device based on the predicted communication setting.

Abstract: A method for balancing inroute traffic load that contains both guaranteed QoS and best effort traffic. Hierarchical grouping levels are defined with the lowest level corresponding to inroutes within the system. Certain levels have common symbol rates, modulation rates, or both. When a new terminal requires admission, it is assigned to entries in the different hierarchical levels so that the inroute traffic load across all levels are balanced. Terminals are admitted to inroutes based, in part, on their channel quality indicator. Inroute traffic load can periodically rebalance based on elapsed time or terminal redistribution.

Abstract: Some implementations of the disclosure relate to dynamic switching of a satellite inroute data path between a Time Division Multiple Access (TDMA) method and a Time Division Multiplexing (TDM) method. In one implementation, a satellite terminal comprises one or more processors; and one or more non-transitory computer-readable storage media configured with instructions executable by the one or more processors to cause the satellite terminal to perform operations comprising: communicating, using the satellite terminal, over an inroute TDM channel; determining, based on an ingress traffic rate to the satellite terminal or a determination that the satellite terminal has not received any traffic flows classified for communication using TDM, to switch communications from the inroute TDM channel to an inroute TDMA channel; and after determining to switch communications, switching, at the satellite terminal, from communicating over the inroute TDM channel to communicating over the inroute TDMA channel.

Abstract: Some implementations of the disclosure relate to dynamic switching of a satellite inroute data path between a Time Division Multiple Access (TDMA) method and a Time Division Multiplexing (TDM) method. In one implementation, a method comprises: communicating, using a satellite terminal, over an inroute TDMA channel; determining, at the satellite terminal, based at least on an aggregate ingress traffic rate to the satellite terminal, to switch communications from the inroute TDMA channel to an inroute TDM channel; and after determining to switch communications, sending, from the satellite terminal to a Gateway Earth Station, a request to be allocated an inroute TDM channel.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training and using machine learning clustering models to determine conditions a satellite communication system. In some implementations, feature vectors for a time period are obtained. Each feature vector includes feature values that represent properties of a satellite communication system at a respective time during the time period. Each feature vector is provided as input to a machine learning model that assigns the feature vector to a based on the properties of the satellite communication system represented by the feature vector. Each cluster corresponds to a respective potential operating condition of the satellite communication system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training and using machine learning models to detect problems in a satellite communication system. In some implementations, one or more feature vectors that respectively correspond to different times are obtained. The feature vector(s) are provided as input to one or more machine learning models trained to receive at least one feature vector that includes feature values representing properties of the satellite communication system and output an indication of potential causes of a condition of the satellite communication system based on the properties of the satellite communication system. A particular cause that is indicated as being a most likely cause of the condition of the satellite communication system is determined based on one or more machine learning model outputs received from each of the one or more machine learning models.

Abstract: Some implementations of the disclosure relate to dynamic switching of a satellite inroute data path between a Time Division Multiple Access (TDMA) method and a Time Division Multiplexing (TDM) method. In one implementation, a method comprises: communicating, using a satellite terminal, over an inroute TDMA channel; determining, at the satellite terminal, based at least on an aggregate ingress traffic rate to the satellite terminal, to switch communications from the inroute TDMA channel to an inroute TDM channel; and after determining to switch communications, sending, from the satellite terminal to a Gateway Earth Station, a request to be allocated an inroute TDM channel.

Abstract: A method and system for installing a terrestrial antenna for a satellite communication network. In the system and method, a remote unit is provided to an installation location for the terrestrial antenna. The remote unit is configured to communicate with a satellite of the satellite communication network and includes a memory in which is stored antenna information pertaining to positioning of the terrestrial antenna with respect to a virtual beam generated by the satellite. The information is accessible by a code. Thus, the antenna information is access from the memory at the installation location using the code, and the terrestrial antenna in relation to a virtual beam generated by the satellite based on the antenna information accessed from the memory at the installation location.

Abstract: Methods and systems for monitoring a communication network using machine-learning techniques are disclosed. In some implementations, a forecasted amount of traffic for a communication network is determined using one or more network traffic forecasting models being configured to generate the forecasted amount of traffic based on data indicating one or more previous amounts of traffic for the communication network. A measure of network health is generated based on a measured amount of traffic and the forecasted amount of traffic. Data indicating one or more characteristics of the communication network is processed using one or more machine learning models to generate a predicted measure of network health for a future time period. An indication of the predicted measure of network health for the future time period is provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for machine learning models for adjusting communication parameters. In some implementations, data for each device in a set of multiple communication devices is obtained. A machine learning model is trained based on the obtained data. The model can be trained to receive an indication of a geographic location and predict a communication setting capable of providing at least a minimum level of efficiency. After training the machine learning model, an indication of a predicted communication setting for a particular communication device is generated. A determination is then made whether to change a current communication setting for the particular communication device based on the predicted communication setting.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for machine learning models for adjusting communication parameters. In some implementations, data for each terminal in a set of multiple satellite terminals is obtained. A machine learning model is trained based on the obtained data. The model can be trained to receive an indication of a geographic location and predict a satellite beam capable of providing at least a minimum level of efficiency for communication at the geographic location. After training the machine learning model, an indication of a predicted satellite beam for a particular location is generated for a particular geographic location. A determination is then made whether to change the current satellite beam for a terminal at the particular geographic location based on the predicted satellite beam.

Abstract: Methods and systems for monitoring a communication network using machine-learning techniques are disclosed. In some implementations, a forecasted amount of traffic for a communication network is determined using one or more network traffic forecasting models being configured to generate the forecasted amount of traffic based on data indicating one or more previous amounts of traffic for the communication network. A measure of network health is generated based on a measured amount of traffic and the forecasted amount of traffic. Data indicating one or more characteristics of the communication network is processed using one or more machine learning models to generate a predicted measure of network health for a future time period. An indication of the predicted measure of network health for the future time period is provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training and using machine learning models to detect problems in a satellite communication system. In some implementations, one or more feature vectors that respectively correspond to different times are obtained. The feature vector(s) are provided as input to one or more machine learning models trained to receive at least one feature vector that includes feature values representing properties of the satellite communication system and output an indication of potential causes of a condition of the satellite communication system based on the properties of the satellite communication system. A particular cause that is indicated as being a most likely cause of the condition of the satellite communication system is determined based on one or more machine learning model outputs received from each of the one or more machine learning models.

Abstract: Methods and systems for spoken language interface for network management are disclosed. In some implementations, data indicating a transcription of a spoken request from a user of a voice-response interface is received. Status information for a communication system is received. The request is interpreted based on the transcription and the status information for the communication system. A response to the request is generated based on the status information for the communication system, and data for a synthesized speech utterance of the response is provided in response to the spoken request.

Abstract: A method for balancing inroute traffic load that contains both guaranteed QoS and best effort traffic. Hierarchical grouping levels are defined with the lowest level corresponding to inroutes within the system. Certain levels have common symbol rates, modulation rates, or both. When a new terminal requires admission, it is assigned to entries in the different hierarchical levels so that the inroute traffic load across all levels are balanced. Terminals are admitted to inroutes based, in part, on their channel quality indicator. Inroute traffic load can periodically rebalance based on elapsed time or terminal redistribution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training and using machine learning models to detect problems in a satellite communication system. In some implementations, one or more feature vectors that respectively correspond to different times are obtained. The feature vector(s) are provided as input to one or more machine learning models trained to receive at least one feature vector that includes feature values representing properties of the satellite communication system and output an indication of potential causes of a condition of the satellite communication system based on the properties of the satellite communication system. A particular cause that is indicated as being a most likely cause of the condition of the satellite communication system is determined based on one or more machine learning model outputs received from each of the one or more machine learning models.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training and using machine learning clustering models to determine conditions a satellite communication system. In some implementations, feature vectors for a time period are obtained. Each feature vector includes feature values that represent properties of a satellite communication system at a respective time during the time period. Each feature vector is provided as input to a machine learning model that assigns the feature vector to a based on the properties of the satellite communication system represented by the feature vector. Each cluster corresponds to a respective potential operating condition of the satellite communication system.

Abstract: A method for balancing inroute traffic load that contains both guaranteed QoS and best effort traffic. Hierarchical grouping levels are defined with the lowest level corresponding to inroutes within the system. Certain levels have common symbol rates, modulation rates, or both. When a new terminal requires admission, it is assigned to entries in the different hierarchical levels so that the inroute traffic load across all levels are balanced. Terminals are admitted to inroutes based, in part, on their channel quality indicator. Inroute traffic load can periodically rebalance based on elapsed time or terminal redistribution.