Error Resilient Deep Compressive Sensing
This addresses a specific issue in signal processing for applications requiring reliable transmission, but it is incremental as it builds on existing DCS methods.
The paper tackles the problem of deep compressive sensing (DCS) failing to recover signals when measurements are lost, by proposing a robust deep reconstruction network that preserves error-resilient properties under random measurement loss, using a measurement lost layer in an end-to-end framework.
Compressive sensing (CS) is an emerging sampling technology that enables reconstructing signals from a subset of measurements and even corrupted measurements. Deep learning-based compressive sensing (DCS) has improved CS performance while maintaining a fast reconstruction but requires a training network for each measurement rate. Also, concerning the transmission scheme of measurement lost, DCS cannot recover the original signal. Thereby, it fails to maintain the error-resilient property. In this work, we proposed a robust deep reconstruction network to preserve the error-resilient property under the assumption of random measurement lost. Measurement lost layer is proposed to simulate the measurement lost in an end-to-end framework.