Independently Convolutional Gated Recurrent Neural Unit for Space based ADS-B Signal Separation with Single Antenna

Automatic Dependent Surveillance-Broadcast (ADS-B) is a critical technology to transform aircraft navigate through improving safety, efficiency, and overall effectiveness in the aviation industry. However, ADS-B signal overlapping is a large challenge, especially for space-based ADS-B system. The existing traditional methods cannot effectively separate overlapping signals with small difference such as power difference, carrier frequency difference. To solve this problem, we propose Independently Convolutional Gated Recurrent Neural Unit (Ind-CGRU) to build the encoder-decoder network. And Ind-CGRU explores the more temporal relationship of ADS-B signal to generate effective separation mask and improves the ADS-B signal separation performance. Experimental results on SRADSB dataset, demonstrate that the proposed Ind-CGRU achieves the good performance.