Usage-based network connection management
A computing device is described that includes at least one processor, a network interface, and a storage device that stores instructions executable by the at least one processor to obtain a usage profile generated by at least applying a machine learning model to historical feature usage information of the computing device collected while the computing device was wirelessly connected to a companion computing device. The instructions may further cause the one or more processors to determine a timeout value based on the usage profile. The instructions may further cause the one or more processors to initiate a connection to a network using the network interface responsive to determining that the computing device is no longer wirelessly connected to the companion computing device and after an amount of time specified by the timeout value has elapsed.
1 . A method comprising:
obtaining, by a first computing device, historical feature usage information of a second computing device collected while the first computing device is wirelessly connected to the second computing device;
generating, by the first computing device and based on the historical feature usage information, a usage profile;
determining, by the first computing device and based on the usage profile, a timeout value; and
sending, by the first computing device and to the second computing device, instructions to cause the second computing device to initiate a modem connection to a network responsive to the second computing device determining an amount of time specified by the timeout value has elapsed.
2 . The method of claim 1 , wherein the historical feature usage information includes at least one of: a frequency at which the first computing device or the second computing device sends and receives signals to establish one or more audio calls, availability of a user of the second computing device based on a schedule of the user, or a signal strength indication of the second computing device at a particular time of day, wherein the signal strength indication comprises a signal strength of a wireless connection between the first computing device and the second computing device.
3 . The method of claim 1 , wherein the first computing device is wirelessly connected to the second computing device via a personal area network.
4 . The method of claim 1 , further comprising:
continuously updating the usage profile by providing a machine learning model additional historical feature usage information of the second computing device
determining, by the first computing device and based on an updated usage profile, an updated timeout value; and
sending, by the first computing device and to the second computing device, updated instructions to cause the second computing device to initiate the modem connection to the network responsive to the second computing device determining an amount of time specified by the updated timeout value has elapsed.
5 . The method of claim 1 , wherein generating the usage profile comprises:
sending, by the first computing device and to a remote computing system, the historical feature usage information;
obtaining, by the first computing device, the usage profile, wherein the usage profile is generated by the remote computing system.
6 . The method of claim 1 , further comprising:
sending, by the first computing device and to the second computing device, a notification responsive to the second computing device initiating the modem connection to the network based on the amount of time specified by the timeout value or responsive to the first computing device reconnecting to the second computing device after the first computing device is disconnected from the second computing device.
7 . The method of claim 6 , wherein the notification includes one or more indications of an incoming voice call or a message not received by the second computing device while the first computing device was disconnected from the second computing device.
8 . The method of claim 1 , wherein a service supporting a voice channel modem connection to the network allows calling over a wireless local area network, and wherein the first computing device updates the timeout value to an indefinite length of time responsive to the wireless local area network establishing a voice channel modem connection to the network.
9 . The method of claim 1 , wherein the first computing device provides a machine learning model one or more instances of a user of the second computing device activating the modem connection to the network prior to the second computing device activating the modem connection to the network based on the amount of time specified by the timeout value, and wherein the machine learning model updates the usage profile based on the one or more instances.
10 . A computing device comprising:
at least one processor; and
a storage device that stores instructions executable by the at least one processor to:
obtain historical feature usage information of a companion computing device collected while the computing device was wirelessly connected to the companion computing device;
generate, based on the historical feature usage information, a usage profile;
determine, based on the usage profile, a timeout value; and
send, to the companion computing device, connection instructions to cause the companion computing device to initiate a modem connection to a network responsive to the companion computing device determining an amount of time specified by the timeout value has elapsed.
11 . The computing device of claim 10 , wherein the historical feature usage information includes at least one of: a frequency at which the computing device or the companion computing device sends and receives signals to establish one or more audio calls, availability of a user of the companion computing device based on a schedule of the user, or a signal strength indication of the companion computing device at a particular time of day, and wherein the signal strength indication comprises a signal strength of a wireless connection between the computing device and the companion computing device.
12 . The computing device of claim 10 , wherein the computing device is wirelessly connected to the companion computing device via a personal area network.
13 . The computing device of claim 10 , wherein the instructions executable by the at least one processor further cause the at least one processor to:
continuously update the usage profile by providing a machine learning model additional historical feature usage information of the companion computing device;
determine, based on an updated usage profile, an updated timeout value; and
send, to the companion computing device, updated connection instructions to cause the companion computing device to initiate the modem connection to the network responsive to the companion computing device determining an amount of time specified by the updated timeout value has elapsed.
14 . The computing device of claim 10 , wherein the instructions further cause the at least one processor to send, to the companion computing device, a notification responsive to the companion computing device initiating the connection to the network based on the amount of time specified by the timeout value or responsive to the computing device reconnecting to the companion computing device after the computing device is disconnected from the companion computing device.
15 . The computing device of claim 14 , wherein the notification includes one or more indications of an incoming voice call or a message not received by the companion computing device while the computing device was disconnected from the companion computing device.
16 . A computer-readable storage medium storing instructions that, when executed, cause at least one processor of a first computing device to:
obtain historical feature usage information of a second computing device collected while the first computing device was wirelessly connected to the second computing device;
generate, based on the historical feature usage information, a usage profile;
determine, based on the usage profile, a timeout value; and
send, to the second computing device, connection instructions to cause the second computing device to initiate a modem connection to a network responsive to the second computing device determining an amount of time specified by the timeout value has elapsed.
17 . The computer-readable storage medium of claim 16 , wherein the historical feature usage information includes at least one of: a frequency at which the first computing device or the second computing device sends and receives signals to establish one or more audio calls, availability of a user of the second computing device based on a schedule of the user, or a signal strength indication of the second computing device at a particular time of day, wherein the signal strength indication comprises a signal strength of a wireless connection between the first computing device and the second computing device.
18 . The computer-readable storage medium of claim 16 , wherein the first computing device is wirelessly connected to the second computing device via a personal area network.
19 . The computer-readable storage medium of claim 16 , wherein the instructions further cause the at least one processor to:
continuously update the usage profile by providing a machine learning model additional historical feature usage information of the second computing device;
determine, based on an updated usage profile, an updated timeout value; and
send, to the second computing device, updated connection instructions to cause the second computing device to initiate the modem connection to the network responsive to the second computing device determining an amount of time specified by the updated timeout value has elapsed.
20 . The computer-readable storage medium of claim 16 , wherein the instructions further cause the at least one processor to: send, to the second computing device, a notification responsive to the second computing device initiating the modem connection to the network based on the amount of time specified by the timeout value or responsive to the first computing device reconnecting to the second computing device after the first computing device is disconnected from the second computing device.