RAN State Trajectory Control Within Network Coherence Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RAN controllers struggle to efficiently and timely manage the complex dynamics of wireless networks with heterogeneous devices and varying traffic, leading to suboptimal radio resource management due to the challenges of defining network configurations within the coherence time of the network.
Innovation Solution
A RAN controller utilizing Self-Organizing Radio Maps (SORMs) to learn and determine proximal network configurations within the coherence time, incorporating a network state manager, change state detector, and network configuration manager to anticipate and implement configuration changes based on best matching units and trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing RAN controllers use traditional control methods to manage network configurations, then the system structure remains simple, but the controller cannot efficiently and timely manage the complex dynamics of wireless networks with heterogeneous devices and varying traffic
Solution Approach 1:
The RAN controller is segmented into three functional modules: network state manager, change state detector, and network configuration manager. Each module handles specific tasks - the state manager processes network data, the detector identifies configuration changes, and the configuration manager executes updates. This segmentation enables efficient handling of complex network dynamics while maintaining manageable system architecture.
Solution Approach 2:
The system performs preliminary actions by anticipating network state changes through the change state detector, which identifies patterns and predicts future configurations before they occur. The network configuration manager prepares and schedules configuration updates in advance, ensuring changes are implemented within the coherence time constraint without disrupting ongoing network operations.
2Measurement precision
If the RAN controller implements comprehensive monitoring and analysis of network states, then the accuracy of configuration changes is improved, but the time required for processing increases and may exceed coherence time
Solution Approach 1:
The network state manager processes only the essential components of network data required for configuration decisions, rather than analyzing every possible parameter. The change state detector focuses on identifying significant state changes that trigger configuration updates, filtering out redundant information. This partial action approach maintains sufficient accuracy while reducing processing time to fit within coherence time constraints.
Solution Approach 2:
The system skips unnecessary processing steps by directly transitioning from data reception to configuration change detection when the network state manager identifies relevant patterns. The change state detector rapidly analyzes network states and immediately triggers configuration updates when changes are detected, minimizing the time spent on intermediate analysis and ensuring timely response within coherence time.
3Reliability
If the RAN controller frequently updates network configurations to adapt to changing conditions, then the network performance is optimized, but the frequency of updates may exceed the coherence time window
Solution Approach 1:
The change state detector continuously monitors network states and identifies patterns that predict future changes before they occur. The network configuration manager schedules updates in advance based on these predictions, ensuring configurations are updated at optimal moments that maintain reliability while staying within the coherence time window. This preliminary action prevents unnecessary updates and ensures timing accuracy.
Solution Approach 2:
The system implements feedback mechanisms where the network state manager continuously reports on network conditions, and the change state detector analyzes these reports to determine when configuration updates are necessary. The feedback loop ensures updates are triggered only when actual changes occur or are predicted to occur, maintaining network performance optimization while respecting coherence time constraints through intelligent timing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is provided a communication system (100) comprising a core network and a Radio Access Network (RAN) controller (6) comprising : - a network state manager (60) configured to receive data inputs at different instant times, in an observation time window, a data input received at a given instant time belonging to an input observation space and characterizing the state of the network at said given instant time, the network state manager (60) being configured to apply at least one self-organizing radio map (70A) to each data input received at a given instant time, said at least one self-organizing radio map (70A) comprising an array of a set of units arranged in a grid describing said input observation space, which provides a best matching state unit at said given instant time, said determined best matching state units forming a state trajectory in the grid of the self-Organizing radio map (70A); - a change state detector (61) configured to determine a state change vector from the best matching state units determined for the data inputs received in said observation time window, the state change vector comprising components, each component state of the change state vector comprising an instant time and an associated best matching state unit, among said determined best matching state units; and - A network configuration manager (62) configured to detect if a configuration change is to be triggered in the RAN network from the state change vector, within a coherence time of the RAN network.