Cellular system
View Patent ↗A system includes a cellular transceiver to communicate with a predetermined target; one or more antennas coupled to the 5G transceiver each electrically or mechanically steerable to the predetermined target; a processor to control a directionality of the one or more antennas in communication with the predetermined target; and an edge processing module coupled to the processor and the one or more antennas to provide low-latency computation for the predetermined target.
1. A system to service mobile client devices with processing loads including machine learning processing load, extended reality processing load, or autonomous driving processing load, comprising:
a cellular transceiver to offload execution of a compute-intensive code object upon request from a client device with a processing load request;
one or more antennas coupled to the cellular transceiver each electrically or mechanically steerable to the client device;
a processor coupled to the one or more antennas in communication with the client device based on signals received from the client device, wherein the client device requests the processor to process at least a portion of a client processing load; and
an edge processing module coupled to the processor and mechanically coupled to the one or more antennas to process the compute-intensive code object on behalf of the client device and to wirelessly provide low-latency access for the client device,
wherein the edge processing module shares processing loads with a core processing module located at a 5G head-end and a cloud module located at a cloud data center such that processing loads requiring intermediate low latency access are transferred to the core processing module at the 5G head-end, while non-time sensitive large processing loads are offloaded to the cloud module at the cloud data center, and further wherein the edge processing module having a neural network with down sampled neural network parameters associated with processing loads from the client device.
2. The system of claim 1 , wherein the processor calibrates a radio link between a transceiver in a housing and the client device.
3. The system of claim 1 , wherein the processor is coupled to fiber optics cable to communicate with a cloud-based radio access network (RAN) or a remote RAN.
4. The system of claim 1 , wherein the processor calibrates a connection by analyzing RSSI (Received Signal Strength Indicator) and TSSI (Transmit Signal Strength Indicator) and moves the antennas until predetermined cellular parameters are reached.
5. The system of claim 1 , wherein the edge processing module comprises at least a processor, a graphical processing unit (GPU), a neural network, a statistical engine, or a programmable logic device (PLD).
6. The system of claim 1 , wherein the edge processing module and the antenna comprise one unit.
7. The system of claim 6 , comprising a pole, a building, or a light.
8. The system of claim 1 , wherein the cellular transceiver comprises a 5G transceiver.
9. The system of claim 1 , wherein the processor coordinates beam sweeping by the one or more antennas with radio nodes or user equipment (UE) devices based upon service level agreement, performance requirement, traffic distribution data, networking requirements or prior beam sweeping history.
10. The system of claim 9 , wherein the beam sweeping is directed at a group of autonomous vehicles or a group of virtual reality devices.
11. The system of claim 1 , comprising a neural network coupled to a control plane, a management plane, and a data plane to optimize 5G parameters.
12. The system of claim 1 , comprising one or more cameras and sensors in a housing to capture security information.
13. The system of claim 1 , comprising edge sensors including LIDAR (Light Detection And Ranging) and RADAR (Radio Detection And Ranging).
14. The system of claim 1 , comprising a camera for individual identity identification.
15. The system of claim 1 , wherein the edge processing module streams data to the client device to minimize loading the client device.
16. The system of claim 1 , comprising an edge learning machine in a housing to provide local edge processing for Internet-of-Things (IOT) sensors.
17. The system of claim 16 , wherein the edge learning machine uses pre-trained models and modifies the pre-trained models for a selected task.
18. The system of claim 1 , a cellular device for a person crossing a street near a city light or street light, the cellular device emitting a person to vehicle (P2V) or a vehicle to person (V2P) safety message.
19. The system of claim 1 , comprising a cloud trained neural network whose network parameters are filter count reduced before transferring to the edge neural network.