17965294. COORDINATED LOAD BALANCING IN MOBILE EDGE COMPUTING NETWORK simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
COORDINATED LOAD BALANCING IN MOBILE EDGE COMPUTING NETWORK
Organization Name
Inventor(s)
Yi Tian Xu of Mount Royal (CA)
Gregory Lewis Dudek of West Mount (CA)
COORDINATED LOAD BALANCING IN MOBILE EDGE COMPUTING NETWORK - A simplified explanation of the abstract
This abstract first appeared for US patent application 17965294 titled 'COORDINATED LOAD BALANCING IN MOBILE EDGE COMPUTING NETWORK
Simplified Explanation
The abstract describes a method for load balancing in a wireless network using a neural network. Here are the key points:
- The method involves obtaining policy parameters of a neural network that correspond to a load balancing policy.
- Trajectories are received for each mobile device in the network, which represent the sequence of states of each device generated by its existing policy interacting with the network.
- Advantage functions are estimated for each mobile device based on their respective trajectories.
- The policy parameters are updated based on the estimated advantage functions, allowing the load balancing policy to be determined based on the states of each mobile device.
Potential Applications
- Load balancing in wireless networks: This method can be applied to improve the distribution of network resources among mobile devices, ensuring efficient utilization of the network capacity.
Problems Solved
- Inefficient resource allocation: The method addresses the problem of uneven distribution of network resources among mobile devices, which can lead to congestion and poor performance.
Benefits
- Improved network performance: By dynamically updating the load balancing policy based on the states of each mobile device, the method can optimize resource allocation and improve overall network performance.
- Enhanced user experience: By ensuring a more balanced distribution of resources, the method can provide a better user experience by reducing congestion and improving network responsiveness.
Original Abstract Submitted
A method includes obtaining at least one policy parameter of a neural network corresponding to a load balancing policy, receiving trajectories for each mobile device in a plurality of mobile devices of the wireless network, each trajectory corresponding to a sequence of states of a respective mobile device, wherein the sequence of states is generated based on a continuous interaction of an existing policy of the respective mobile device with the wireless network, estimating advantage functions for each mobile device in the plurality of mobile devices based on the trajectories for each respective mobile device, and updating the at least one policy parameter based on the estimated advantage functions such that the load balancing policy is determined based on states of each mobile device in the plurality of mobile devices.
