Wireless Interference Classification Using FFT and KS Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless local area networks (WLANs) face significant interference from other devices and non-communication sources, which degrades performance and requires resource-intensive methods to detect and classify, often focusing on individual interference sources without an efficient trade-off between resource usage and performance improvement.
Innovation Solution
A method that uses fast Fourier transform (FFT) values to detect and classify interference as narrowband, wideband, or a combination of both, employing a two-sample Kolmogorov-Smirnov test to determine interference type and apply appropriate reduction techniques, optimizing scan budgets for desired throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional interference detection methods are used, then interference detection capability is provided, but network resources are consumed and performance is adversely affected
Solution Approach 1:
The interference detection process is segmented into two distinct phases: a light scanning phase that performs minimal resource-intensive operations, and a heavy detection phase that performs comprehensive analysis only when interference is suspected. This segmentation allows the system to maintain continuous monitoring capability while consuming minimal resources during normal operation, thus resolving the contradiction between detection capability and network performance.
Solution Approach 2:
The system implements periodic scanning at predetermined intervals rather than continuous heavy monitoring. During these periodic scans, the system performs lightweight operations to check for interference presence. This periodic approach ensures interference detection capability is maintained while significantly reducing overall resource consumption and preserving network performance during non-scan periods.
2Measurement precision
If comprehensive interference detection and classification is performed, then interference classification accuracy is improved, but resource consumption increases
Solution Approach 1:
The system performs preliminary lightweight scanning operations at predetermined intervals to detect the presence of interference before initiating comprehensive classification. This preliminary action identifies potential interference cases that then trigger the heavier classification process, ensuring that resource-intensive classification operations are performed only when necessary, thus improving classification accuracy while controlling resource consumption.
Solution Approach 2:
The system uses its own periodic scanning mechanism to identify when interference is present, then automatically triggers the appropriate classification depth based on what the scanning detects. This self-service approach allows the system to adapt its resource consumption dynamically based on actual interference conditions, performing comprehensive classification only when interference is detected rather than continuously.
Data Source
AI summary
Interference in a wireless network is detected by first applying during a detection stage a two-sample Kolmogorov-Smirnov (KS) test to samples obtained from a channel in the wireless network to obtain a KS statistic. When the KS statistic is larger than a predetermine threshold, it is determined that the samples include narrowband interference, and wideband otherwise.