Cellular Interference Localization Using PSD Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile networks face challenges in accurately and efficiently identifying and locating sources of interference within complex radio communication systems, leading to disrupted data communication and increased time and cost in resolving interference issues.
Innovation Solution
A method and apparatus utilizing power spectral density data to cluster cells based on abnormal signal features, determine correlation coefficients, and group cells with high received power to identify the source of interference, facilitated by unsupervised clustering and supervised learning to automate interference detection and localization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If field engineers use spectrum analyzers to manually search for interference sources, then measurement precision can be achieved, but loss of time increases significantly
Solution Approach 1:
The patent creates a virtual model of the radio frequency environment by collecting and processing PSD data from multiple base stations. This digital replica allows automated analysis of interference patterns without requiring physical field measurements, thereby maintaining detection accuracy while dramatically reducing the time needed to locate interference sources.
Solution Approach 2:
The patent replaces the mechanical field measurement process with an automated computational system. Instead of engineers physically traveling to locations with spectrum analyzers, the system uses algorithms to process PSD data from distributed base stations, substituting manual mechanical measurement with automated digital analysis.
2Productivity
If the number of cells operating in multiple frequency bands is increased to support more devices, then productivity increases, but device complexity increases due to interference management
Solution Approach 1:
The patent implements a universal interference detection system that operates across multiple frequency bands and cell types simultaneously. The PSD-based analysis method is band-agnostic and can process data from heterogeneous base stations using different radio access technologies, providing a unified approach to interference management that scales with network complexity.
Solution Approach 2:
The system enables the network to automatically detect, analyze, and locate interference sources without external intervention. By utilizing existing base station measurements and applying automated correlation algorithms, the network self-diagnoses interference issues, reducing the need for manual engineering efforts as the network expands.
3Measurement precision
If manual field deployment is used to identify interference sources, then measurement precision can be maintained, but ease of operation deteriorates due to the time-consuming nature of field searches
Solution Approach 1:
The patent creates a virtual model of the radio frequency environment by collecting and processing PSD data from multiple base stations. This digital replica allows automated analysis of interference patterns without requiring physical field measurements, thereby maintaining detection accuracy while dramatically reducing the time needed to locate interference sources.
Solution Approach 2:
The patent replaces the mechanical field measurement process with an automated computational system. Instead of engineers physically traveling to locations with spectrum analyzers, the system uses algorithms to process PSD data from distributed base stations, substituting manual mechanical measurement with automated digital analysis.
Data Source
AI summary
Methods, apparatuses and computer program products for locating sources of potential interference in frequency bands of carrier signals used by cells of a plurality of base stations of a cellular radio communication system are disclosed. The method comprises receiving data representative of power spectral density, PSD, in carrier signals across different frequency bands received at the base stations for cells of the communication system over time intervals. The cells are clustered based on features of the received PSD data that distinguish between a normal received signal and an abnormal received signal subject to potential interference. For each cell assigned to a cluster having features indicative of an abnormal received signal, a correlation coefficient is determined between a matrix of the signal strength values in the received PSD data for that cell in the different frequency bands over the time intervals with the signal strength values in the received PSD data for all other cells in the respective frequency bands over the same time intervals. Cells having correlation coefficients above a threshold value are grouped as being members of interference groups. For each interference group, the cell having the highest received power in a correlated frequency band is identified as the cell causing or nearest that cause of the interference affecting the cells in the interference group.