An Enhanced Error Correction Algorithm Combined with Directed Density-Based Clustering for Satellite-Based ADS-B Signals

The minimum Hamming distance of satellite-based Automatic Dependent Surveillance-Broadcast (ADS-B) signals at low signal-to-noise ratios (SNRs) is only 6, which is inadequate to meet airspace surveillance requirements in terms of packet decoding probability (PD). An enhanced error correction algorithm combined with directed density-based clustering for satellite-based ADS-B signals is proposed to address this phenomenon, and its performance is verified by simulation. Firstly, the density-based clustering model will cluster a given signal sequence according to its partial minimum Hamming distance from other sequences, reducing the chance of undetectable errors. Secondly, the proposed error syndrome matrix built offline streamlines the Brute Force correction, preserving on-star resources. Finally, the {a, b} algorithm compensates for low SNR-induced unreliability of confidence arrays through error correction depth a and error correction capability b. The simulation results show that the {20, 10} error correction algorithm can achieve a PD of 86.4% at the minimum SNR of satellite-based ADS-B signals.