Metasurface integrated non-terrestrial network transcoder with adaptive slicing for dynamic allocation of mesh resources
The technology described herein is directed towards a transcoder with bypass capabilities that can be used to couple non-terrestrial network satellites to user equipment (UEs), including by decoding and reencoding data packets at the packet level for existing Satcom satellites. A metasurface redirects signals from the satellite to a satellite radio frequency (RF) interface of the transcoder, with the transcoder also coupled by a UE RF interface to a UE, such as a computing device or cellphone. Dynamically slicing (subdividing) a metasurface associated with a transcoder node facilitates using part of the metasurface for UE or satellite communications, and another part to facilitate a mesh network among transcoder nodes, which helps with data privacy. The slicing can be adaptively performed based on learning and artificial intelligence models implemented into the transcoder nodes as edge computing devices. The transcoder can convert Satcom signals to and from new radio signals.
1 . A system, comprising:
at least one processor; and
at least one memory that stores executable instructions that, when executed by the at least one processor, facilitate performance of operations, the operations comprising:
obtaining first information corresponding to a first communication link between a user equipment and a first metasurface associated with a first node of a mesh network;
obtaining second information corresponding to a second communication link between a satellite and a second metasurface associated with a second node of the mesh network; and
based on the first information and the second information, subdividing the first metasurface into a first portion for communication between the first node and the user equipment via the first portion, subdividing the first metasurface into a second portion for communication between the first node and the second node via the second portion, subdividing the second metasurface into a third portion for communication between the second node and the satellite, and subdividing the second metasurface into a fourth portion for communication between the second node and the first node.
2 . The system of claim 1 , wherein the first information corresponds to the communication between the user equipment and the first node via the first portion, and comprises at least one of: second signal strength data representative of a signal strength corresponding to the communication, frequency band data representative of a frequency band corresponding to the communication, or beamforming and directionality data representative of at least one of a beamforming characteristic or a directionality corresponding to the communication.
3 . The system of claim 1 , wherein the second information corresponds to the communication between the satellite and the second node via the third portion, and comprises at least one of: second signal strength data representative of a signal strength corresponding to the communication, frequency band data representative of a frequency band corresponding to the communication, or beamforming and directionality data representative of at least one of a beamforming characteristic or a directionality corresponding to the communication.
4 . The system of claim 1 , wherein the operations further comprise detecting a change to the first information from a first information state to a second information state, and, based on the change to the second information state, redividing the first metasurface to change first dimensions of the first portion relative to second dimensions of the second portion.
5 . The system of claim 1 , wherein the obtaining of the first information and the subdividing of the first metasurface into the first portion and the second portion are performed using a first trained model of the first node, and wherein the obtaining of the second information and the subdividing of the second metasurface into the third portion and the fourth portion are performed using a second trained model of the second node.
6 . The system of claim 5 , wherein the operations further comprise deploying the first node as a first edge compute device.
7 . The system of claim 5 , wherein the training of the first trained model comprises using supervised learning to frequency band prediction.
8 . The system of claim 5 , wherein the training of the first trained model comprises using reinforcement learning for control of adaptive beamforming and directionality.
9 . The system of claim 5 , wherein the training of the first trained model comprises using graph neural networks for management of the mesh network.
10 . The system of claim 1 , wherein the operations further comprise:
detecting addition of a third node to the mesh network, wherein the third node comprises a third trained model; and
in response to the detecting of the addition of the third node:
obtaining third information associated with the third node, the third information comprising at least one of: third signal strength data representative of a signal strength corresponding to communication with the third node, third frequency band data representative of a frequency band corresponding to the communication with the third node, third beamforming and directionality data representative of at least one of a beamforming characteristic or a directionality corresponding to the communication with the third node, or network topology data representative of a network topology comprising the third node, and
training the first trained model, the second trained model, and the third trained model based on the first information, the second information, and the third information, respectively.
11 . The system of claim 10 , wherein the operations further comprise updating the first trained model, the second trained model, and the third trained model based on aggregating, via federated learning, first data associated with the first trained model, second data associated with the second trained model, and third data associated with the third trained model.
12 . The system of claim 10 , wherein the operations further comprise collecting performance data of the mesh network, and retraining the first trained model, the second trained model, and the third trained model based on the performance data.
13 . A method comprising,
obtaining, by a system comprising at least one processor and at least one trained model of a mesh network, current state data representative of a current state of the mesh network; and
based on the current state data,
configuring, by the system, first dimensions of a first portion of a first reconfigurable intelligent surface, associated with a first node of the mesh network, relative to second dimensions of a second portion of the first reconfigurable intelligent surface, wherein the first portion is used for first communication between the first node and a user equipment, and wherein the second portion is used for second communication between the first node and a second node of the mesh network;
configuring, by the system, third dimensions of a third portion of a second reconfigurable intelligent surface, associated with the second node of the mesh network, relative to fourth dimensions of a fourth portion of the second reconfigurable intelligent surface, wherein the third portion is used for third communication between the second node and a satellite, and wherein the fourth portion is used for fourth communication between the second node and the first node.
14 . The method of claim 13 , further comprising detecting, by the system, a change to the current state data corresponding to network conditions, and, in response to the detecting of the change, reconfiguring, by the system, the third dimensions of the third portion of the second reconfigurable intelligent surface, relative to the fourth dimensions of the fourth portion of the second reconfigurable intelligent surface.
15 . The method of claim 13 , further comprising collecting, by the system, network performance data associated with the mesh network, and retraining the at least one trained model of the mesh network based on the network performance data.
16 . The method of claim 13 , further comprising:
detecting, by the system, a change to the current state data corresponding to addition of a third node to the mesh network; and
in response to the detecting of the change, retraining the at least one trained model of the mesh network.
17 . The method of claim 13 , wherein the at least one trained model comprises a first trained model associated with the first node, a second trained model associated with the second node, and further comprising:
updating the first trained model and the second trained model via federated learning based on aggregating first data associated with the first trained model, and second data associated with the second trained model.
18 . A system, comprising:
a mesh network comprising a first node, a second node and a third node;
at least one trained model; and
a first metasurface associated with the first node, a second metasurface associated with the second node, and a third metasurface associated with the third node,
wherein the at least one trained model:
configures the first metasurface, comprising configuring a first portion of the first metasurface for first communication between the first node and a satellite, and configuring a second portion of the first metasurface for second communication between the first node and at least one other node of mesh network other than the first node, the second node and the third node;
configures the second metasurface, comprising configuring a third portion of the second metasurface for third communication between the second node and a user equipment, and configuring a fourth portion of the second metasurface for fourth communication between the second node and the at least one other node of mesh network; and
configures the third metasurface for fifth communication between the third node and the at least one other node of mesh network.
19 . The system of claim 18 , wherein the at least one trained model configures the first portion of the first metasurface for a first part of the first communication between the first node and the satellite at a first time based on first current network conditions associated with the mesh network, and wherein the at least one trained model reconfigures the first portion of the first metasurface for a second part of the first communication between the first node and the satellite at a second time based on second current network conditions associated with the mesh network.
20 . The system of claim 18 , wherein the first node comprises a primary node that controls the configuring of the first metasurface, the configuring of the second metasurface, and the configuring of the third metasurface.