UE Logged Data Reporting Across RRC States for AI/ML MDT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Legacy minimization of drive tests (MDT) in wireless communication systems are inadequate for supporting artificial intelligence (AI)/machine learning (ML) applications due to limitations in data logging and reporting, particularly in the RRC idle state, which restricts flexibility and continuity across RRC states, and does not support multiple measurement configurations.
Innovation Solution
Implementing multiple data collection configurations simultaneously at the user equipment (UE) for network performance data measurement, logging, and reporting across RRC idle, connected, and inactive states, with conditions for activating, deactivating, and filtering data to enhance flexibility and efficiency, and discarding logged data to reduce memory usage and network impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple data collection configurations are implemented simultaneously at the UE, then data collection flexibility and continuity across RRC states are improved, but device complexity and memory usage increase
Solution Approach 1:
The patent segments data collection configurations into multiple independent configurations, each with its own identification, measurement parameters, and logging settings. This allows the UE to manage complex data collection requirements by dividing them into manageable, independently controllable units that can be activated or deactivated based on specific conditions.
Solution Approach 2:
The patent implements dynamic configuration management where data collection configurations can be activated, deactivated, or modified based on RRC state transitions and network conditions. The UE can dynamically switch between different configurations depending on whether it is in idle, connected, or inactive state, allowing flexibility without permanent complexity.
2Quantity of substance
If logged measurement data is retained for AI/ML model training, then data availability for network optimization is improved, but memory usage and network impact increase
Solution Approach 1:
The patent implements a selective data retention and discarding mechanism where logged measurement data is kept in buffers for a configurable duration or until specific conditions are met. The network can request specific datasets for AI/ML training, and the UE can discard data that is no longer needed, thus managing memory usage while maintaining data availability when required.
Solution Approach 2:
The patent allows the UE to log and retain more measurement data than immediately needed, maintaining buffers of historical data that can be selectively reported to the network. This partial retention strategy ensures sufficient data is available for AI/ML model training without permanently storing all possible measurements, balancing memory usage with data availability.
3Loss of energy
If data filtering and selective reporting is implemented, then network impact is reduced, but reporting complexity increases
Solution Approach 1:
The patent implements preliminary filtering and organization of measurement data at the UE side before reporting. Data is pre-processed, aggregated, and organized according to configuration identifiers and reporting conditions, reducing the amount of raw data that needs to be transmitted to the network and simplifying subsequent network processing.
Solution Approach 2:
The patent applies different filtering and reporting strategies to different data configurations based on their specific characteristics and network requirements. Each data collection configuration can have customized reporting conditions, thresholds, and formats, allowing optimized network impact for each type of measurement data without a one-size-fits-all approach.
4Duration of action of stationary object
If continuous logging across RRC states is enabled, then measurement continuity is improved, but power consumption and processing load increase
Solution Approach 1:
The patent implements periodic or event-triggered logging instead of continuous logging across all RRC states. Measurement logging is activated based on periodic intervals or specific events such as RRC state transitions, cell changes, or quality thresholds, reducing power consumption while maintaining measurement continuity where needed.
Solution Approach 2:
The patent dynamically adjusts logging behavior based on the current RRC state and network conditions. In idle state, logging may be more aggressive with lower power consumption, while in connected state, logging can be more selective. The system adapts its measurement and logging intensity to balance continuity requirements with power consumption constraints.
Data Source
AI summary
Various aspects of the present disclosure relate to a user equipment (UE) for wireless communication with at least one memory and at least one processor coupled with the at least one memory and configured to cause the UE to transmit a first indication of a set of logging identifiers, wherein each logging identifier is associated with respective logged measurement data, receive a request message that identifies at least a subset of the set of logging identifiers and at least one criterion for reporting corresponding logged measurement data associated with the subset of the set of logging identifiers, and transmit a response message based at least in part on to the request message, wherein the response message comprises the corresponding logged measurement data associated with the subset of the set of logging identifiers and according to the at least one criterion.