Abstract: Automatic modulation recognition is a critical prerequisite for achieving signal detection and demodulation in non-cooperative communication scenarios. In recent years, deep learning has demonstrated significant advantages in this field. However, existing studies have largely overlooked the challenges posed by random interference and jamming during communication. In fact, due to the open and broadcast nature of wireless communications, interference and jamming attacks have become major threats. To fully harness the potential of automatic modulation recognition in wireless communications, we have conducted an in-depth exploration of deep learning-based modulation recognition techniques under interference and jamming conditions. Specifically, we propose corresponding recognition methods based on interference recognition for scenarios where interference is either known or unknown, and we have validated the effectiveness of our proposed algorithms using the open dataset RML2016.10a.