Context aware handovers
AssigneeIntel Corporation
InventorsAshwin Umapathy, Akshaya Ravishankar, Sanket Vinod Shingte
Life of this patent
Estimating the term…
Abstract
Disclosed embodiments provide a handover prediction scheme that is based on contextual awareness of a compute node, such as a mobile device. The contextual information is used to predict network availability in a predicted target location, which is an area that a compute node is likely to travel. The use of sensors embedded in or accessible by the compute node may be used to carry out aspects of the embodiments. A reinforcement learning recommendation model is used to determine an optimal network, radio access technology, and/or network access node to connect with ahead of arriving at the predicted target location at a predicted arrival time. Other embodiments are described and/or claimed.
Claims19
An apparatus for context-aware handovers (HOs), the apparatus comprising:
- processor circuitry communicatively coupled with one or more Radio Access Technology (RAT) interfaces, the processor circuitry arranged to:
- operate a context inference model (CIM) to determine context information of the apparatus, and predict a target location of the apparatus at a predicted time instance based on the context information,
- operate a Reinforcement Learning Recommendation Model (RLRM) to determine a recommended target network at the predicted target location based on the context information, and
- operate a session layer entity to, in response to detection of an HO trigger event, initiate an HO to the target network using an available RAT interface of the one or more RAT interfaces associated with the target network.
A system for facilitating context aware handovers (HOs), the system comprising
- a plurality of network access nodes (NANs) arranged to provide network connectivity in respective network coverage areas, each NAN of the plurality of NANs provides network connectivity according to at least one radio access technology (RAT);
- one or more network repositories communicatively coupled with the plurality of NANs, the one or more network repositories arranged to store network information of individual NANs of the plurality of NANs, the network information indicating network parameters of the individual NANs;
- and
- a client computing device to obtain network connectivity from at least one NAN of the plurality of NANs, the client computing device comprising:
- processor circuitry communicatively coupled with one or more RAT interfaces, the processor circuitry arranged to:
- operate a context inference model (CIM) to determine context information of the client computing device, and predict a target location of the client computing device at a predicted time instance based on the context information,
- operate a network discovery service (NDS) to query for network information associated with the predicted target location,
- operate a Reinforcement Learning Recommendation Model (RLRM) to determine a recommended target NAN of the plurality of NANs at the predicted target location based on the context information and the network information, and
- operate a session layer entity to, in response to detection of an HO trigger event, initiate an HO from a source NAN of the plurality of NANs to the target NAN using a RAT interface of the one or more RAT interfaces associated with a RAT of the target NAN.
Dependencies are inferred from the claim text.