Distributed Cache Radio Resource Allocation for Large Wireless Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Centralized RRM systems are processor limited, leading to sub-optimal wireless network performance, while decentralized RRM systems optimize each AP individually, also resulting in sub-optimal overall network performance.
Innovation Solution
A distributed radio resource management system that utilizes a distributed cache formed from the memories of multiple APs to share processing load, assigning different RRM functions to various APs, including a controller, ranker, resource manager, and calculator APs to optimize operating parameters across the network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centralized RRM system is used, then the system can optimize operating parameters across the entire wireless network, but the system becomes processor limited and cannot handle large networks
Solution Approach 1:
The centralized RRM system is segmented into multiple distributed RRM entities, each responsible for a subset of access points. This segmentation distributes the processing load across multiple processors while maintaining coordinated optimization across the entire network, resolving the contradiction between comprehensive network optimization and processor resource requirements.
Solution Approach 2:
The system transitions from a single-dimensional centralized processing model to a multi-dimensional distributed architecture where RRM functions are spread across multiple spatial locations (different processors and access points). This dimensional change allows the system to maintain global optimization capabilities while distributing computational burden across multiple processing nodes.
2Device complexity
If a distributed RRM system is used, then the system can handle large networks, but each AP is optimized individually without considering the entire network
Solution Approach 1:
The distributed RRM system implements feedback mechanisms where each distributed RRM entity exchanges information with neighboring entities and receives feedback about network-wide optimization goals. This feedback loop enables individual AP optimizations to be coordinated with overall network performance, ensuring that local decisions contribute to global optimization rather than creating sub-optimal outcomes.
Solution Approach 2:
Each distributed RRM entity is designed with multi-functionality, capable of performing both local AP optimization and participating in network-wide coordination. This universality allows the system to simultaneously handle individual AP requirements while maintaining awareness of and contribution to overall network optimization objectives.
3Productivity
If RRM functions are distributed across multiple APs, then processing load is shared, but the system requires complex role assignment and coordination
Solution Approach 1:
The system dynamically changes operational parameters such as role assignments, data collection frequencies, and optimization intervals based on network conditions and processor loads. This parameter adaptability allows the distributed system to maintain high processing efficiency while managing coordination complexity through flexible, condition-based adjustments rather than rigid complex protocols.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A network system comprises a plurality of access points (APs) and a distributed cache. The distributed cache is formed using memory in the plurality of APs. The plurality of APs are configured to measure telemetry data and store the telemetry data in the distributed cache. One of the plurality of APs is assigned as a controller AP configured to assign, based on the telemetry data stored in the distributed cache, multiple APs of the plurality of APs to different roles to analyze the plurality of APs and update resource configurations of the plurality of APs.