Back to results
US 12,744,316 B2Granted
AI control and allocation of network resources
OwnerKhue Duonginventor
Life of this patent
Estimating the term…
Abstract
A method for managing spectrum allocation in a heterogeneous wireless communication network includes employing artificial intelligence (AI) to predict interference between users and services; and adjusting spectrum allocations for users and services in the network based on the interference predictions to minimize interference.
Claims17
A method for managing spectrum allocation in a heterogeneous wireless communication network, comprising:
- collecting network usage data and signal quality indicators from multiple network types including satellite, 5G, WiFi, and unlicensed band transceivers for a neural network trained on historical data of interference patterns and user communication patterns;
- employing artificial intelligence (AI) to predict interference between users and services;
- and
- adjusting spectrum allocations for users and services in the network based on the interference predictions to minimize interference by classifying interference scenarios and select an allocation policy;
- evolving spectrum allocation strategies under the allocation policy;
- and resolving conflicts between competing spectrum demands from different users and services.
3 more elements
A system for managing handoffs in a heterogeneous wireless communication network, comprising:
- a data collection module configured to gather network condition data and user equipment (UE) parameters;
- an artificial intelligence (AI) module equipped with a machine learning model trained to predict optimal handoff targets based on the collected data;
- a handoff execution module configured to perform handoffs of UEs to predicted optimal networks comprising satellite, 5G, and WiFi, wherein the AI module is further configured to continuously update the machine learning model based on feedback received post-handoff to improve future handoff predictions, and to adjust spectrum allocations with a decision to classify interference scenarios, select an allocation policy;
- evolve spectrum allocation strategies under the allocation policy;
- and resolve conflicts between competing spectrum demands from different users and services and wherein the AI module is further configured to predict handoff targets for a plurality of UEs simultaneously.
2 more elements
A method for allocating communication resources in a wireless network, comprising:
- collecting data related to network conditions and terminal device requirements;
- applying a machine learning model to predict resource block allocations for terminal devices;
- encoding the predicted resource block allocations into resource indication information;
- sending, by a network device, resource indication information to a terminal device, wherein the resource indication information indicates frequency domain resources of a data channel;
- sending, by the network device, data on the data channel to the terminal device, or receiving, by the network device, data on the data channel from the terminal device;
- evolving spectrum allocation strategies to avoid interference;
- resolving conflicts between competing spectrum demands from different users and services;
- and
- transmitting allocation information to the user equipment using a compressed signaling format.
6 more elements
Dependencies are inferred from the claim text.