Reservoir Subspace Injection for Online ICA under Top-n Whitening

Researchers developed Reservoir Subspace Injection (RSI) to improve online Independent Component Analysis under nonlinear mixing conditions. The study identifies performance bottlenecks in top-n whitening and proposes a guarded RSI controller that preserves system performance while achieving 1.7 dB improvement over vanilla online ICA methods.

• →Reservoir expansion improves online ICA performance but top-n whitening can discard valuable injected features.
• →RSI diagnostics identified that stronger injection increases performance metrics but reduces passthrough energy retention.
• →A guarded RSI controller successfully preserves passthrough retention and recovers baseline performance within 0.1 dB.
• →RE-OICA demonstrates +1.7 dB improvement over standard online ICA under nonlinear mixing conditions.
• →The method achieves positive SI-SDR performance on super-Gaussian benchmarks with +0.6 dB improvement.