C-RAN Load Balancing via Global Host Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing load balancing methods in mobile networks primarily focus on local optimization, neglecting global network knowledge, which limits their efficiency in managing non-uniform traffic distribution and leading to issues like high call blocking rates and packet dropping.
Innovation Solution
A load balancing method and device that utilizes global knowledge by determining and ranking base stations based on their shared and fixed mobile hosts, applying a greedy algorithm and constructing a bipartite graph to efficiently distribute mobile hosts across base stations, leveraging the centralized processing capabilities of Cloud-Radio Access Network (C-RAN) architecture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If local optimization methods (cell-breathing, channel borrowing, dynamic scheduling) are used for load balancing, then implementation complexity is reduced and ease of operation is improved, but load balancing efficiency deteriorates because only neighbor base stations are considered and global network knowledge is not utilized
Solution Approach 1:
The patent segments the mobile host population into shared mobile hosts and fixed mobile hosts for each base station. This segmentation allows the system to selectively migrate only shared mobile hosts between base stations while keeping fixed mobile hosts stationary, thereby simplifying the load balancing operation while improving efficiency by focusing resources on movable loads.
Solution Approach 2:
The patent transitions from local two-dimensional optimization (considering only neighbor base stations) to global multi-dimensional optimization by incorporating network-wide base station information. The centralized controller collects load information from all base stations and performs comprehensive load balancing decisions across the entire network, adding the dimension of global network knowledge to the optimization process.
2Productivity
If centralized global load balancing is implemented using C-RAN architecture, then load balancing efficiency is improved by exploiting global network knowledge, but device complexity and implementation difficulty increase
Solution Approach 1:
The patent introduces a centralized controller as an intermediary between base stations and mobile hosts. This controller collects load information from all base stations, performs global optimization calculations, and directs mobile host migrations. By centralizing the decision-making function, the system achieves global load balancing efficiency while managing complexity through a dedicated control entity rather than distributed complex interactions.
Solution Approach 2:
The patent implements self-service mechanisms where base stations automatically report their load information to the centralized controller, and mobile hosts autonomously migrate according to controller directives without requiring complex manual intervention. The system serves itself by leveraging existing reporting mechanisms and automated migration protocols, reducing implementation complexity despite the centralized architecture.
3Ease of manufacture
If all mobile hosts are treated equally in load balancing decisions, then simplicity of implementation is maintained, but manufacturing precision and optimization accuracy deteriorate because shared and fixed mobile hosts have different migration characteristics
Solution Approach 1:
The patent applies local quality by treating shared mobile hosts and fixed mobile hosts differently based on their specific characteristics. Shared mobile hosts are eligible for migration between base stations, while fixed mobile hosts remain associated with their serving base station. This differentiated treatment optimizes load balancing by focusing migration resources on mobile hosts that can actually be moved, improving optimization accuracy without significantly complicating implementation.
Data Source
AI summary
A load balancing method and device are provided by the present invention. The method includes that: for each base station, determining the number of shared mobile hosts shared by the base station with each neighbor base station of the base station and the number of fixed mobile hosts covered only by the base station; ranking all of base stations in an ascending order of the number of fixed mobile hosts; and balancing the number of fixed mobile hosts of each base station by applying a greedy algorithm on the ranked base stations. With the present invention, the load can be efficiently balanced by exploiting the global knowledge.