Positioning Measurement Reports Using Vendor-Agnostic Signal Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile and wireless telecommunication systems, such as 5G NR, face challenges in standardizing ML-based positioning frameworks due to vendor-specific implementation issues, leading to inconsistent measurement reporting and reduced inference performance.
Innovation Solution
A vendor-agnostic reporting framework is introduced, allowing UE and NW vendors to develop proprietary ML-based positioning solutions without exposing proprietary details, using an autoencoder for compressing and decompressing PRS/SRS signals, and configuring code generation for standardized reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If vendor-specific measurement reporting formats are used, then vendors can implement proprietary ML-based positioning solutions with optimized performance, but inconsistency in reporting formats across different vendors reduces standardization and interoperability
Solution Approach 1:
The measurement report is segmented into two distinct parts: a standardized header section that ensures consistency and interoperability across all vendors, and a flexible payload section that allows vendor-specific ML model details and proprietary formatting. This segmentation resolves the contradiction by maintaining standardization requirements while preserving vendor implementation flexibility.
Solution Approach 2:
A standardized intermediate format or translation layer is introduced between the vendor-specific ML models and the network positioning system. This intermediary converts various vendor-specific measurement formats into a common standardized structure, enabling consistent reporting while allowing vendors to maintain their proprietary solutions. The intermediary acts as a mediator that preserves both standardization and vendor autonomy.
2Measurement precision
If detailed proprietary ML model information is included in measurement reports, then inference performance can be optimized, but exposing proprietary details reduces vendor intellectual property protection
Solution Approach 1:
Instead of transmitting actual proprietary ML model parameters and detailed architectural information, the system uses standardized descriptors, metadata, and abstract representations that capture the essential characteristics needed for positioning inference. These copies or representations maintain positioning accuracy while protecting the underlying proprietary intellectual property from direct exposure.
Solution Approach 2:
Different levels of information detail are provided in different parts of the measurement report: standardized sections contain only necessary operational parameters for positioning, while vendor-specific sections use controlled descriptions that maintain precision requirements without revealing sensitive proprietary algorithms or detailed model architectures. This local differentiation of information quality protects IP while maintaining performance.
3Reliability
If standardized measurement reporting formats are enforced, then interoperability and consistency are improved, but vendor flexibility to implement proprietary ML solutions is reduced
Solution Approach 1:
The measurement reporting format is designed with dynamic characteristics where the structure adapts based on vendor capabilities and specific ML model requirements. The standardized framework provides consistent baseline elements, while allowing optional vendor-specific extensions and configurable parameters that enable proprietary solutions to be integrated within the standardized structure, thus maintaining both consistency and flexibility.
Solution Approach 2:
The standardized measurement report format is designed as a universal container that can accommodate multiple vendor-specific ML model types and proprietary formats through a common interface. The standardized structure serves multiple functions: ensuring interoperability, maintaining consistency, and providing a framework that can universally support various proprietary implementations through defined extension mechanisms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems, methods, apparatuses, and computer program products for a framework for handling the appropriately configured positioning measurement reports for use by a network element without requiring knowledge of vendor-specific or proprietary information are provided. For example, a method may include receiving from a network node, a message including a code configuration to configure collected samples measurements for reporting. The method may also include generating a code from the collected samples measurements in a predefined format according to the code configuration. The method may further include reporting the generated code to the network node.