With the launch of an increasing number of satellites to establish complex satellite communication networks, automatic modulation recognition (AMR) plays a crucial role in satellite signal recognition and spectrum management. However, most existing AMR models struggle to handle signals in such complex satellite communication environments. Therefore, this paper proposes an adaptive multi-rate atrous convolution network with residual attentional feature fusion (RAFF-AMACNet) that employs the adaptive multi-rate atrous convolution (AMAC) module to adaptively extract and dynamically join more prominent multi-scale features, enhancing the model’s time-series context awareness and generating robust feature maps. On this basis, the pyramid backbone consists of multiple stacked residual attentional feature fusion (RAFF) modules, featuring a dual-attention collaborative mechanism designed to mitigate feature map shifts and increase the separation between feature clusters of different classes under significant Doppler effects and nonlinear influences. On our independently constructed RML24 dataset, a general-purpose dataset tailored for satellite cognitive radio systems, simulation results indicate that at a signal-to-noise ratio of 0 dB, the modulation recognition accuracy reaches 92.99%.

The rapid evolution of satellite communication systems necessitates the use of cognitive radio technologies for enhanced adaptability. However, current public datasets predominantly focus on terrestrial signal characteristics and fail to incorporate critical real-world channel impairments, significantly limiting their utility for space applications. This paper presents RML24, the first open-source benchmark dataset specifically designed for deep learning-based satellite signal processing. Unlike conventional datasets, RML24 integrates authentic Telemetry, Tracking, and Command (TTC) system waveforms with practical channel effects through a hybrid data collection approach combining software-defined radio (SDR) platforms and RF transceivers, providing a robust benchmark for satellite signal identification and demodulation. RML24 contains over 1.3 million samples across 22 modulation types. Comprehensive experimental results demonstrate that RML24 aligns more closely with the distribution of real-world signals. Compared to traditional training models, it achieves a 40% improvement in recognition performance under high signal-to-noise ratio (SNR) conditions.RML24 aims to advance intelligent, adaptive deep learning-driven satellite systems by offering a high-quality dataset tailored for satellite communications.

Aiming at the lack of publicly available datasets of satellite Specific Emitter Identification (SEI) and the problem of insufficient data under multi-receiver conditions, this paper constructs a multi-receiver one-dimensional I/Q dataset for Iridium downlink signals using six ADALM-PLUTO software-defined radio receivers. The dataset contains a total of 367,658 receiver-level valid signal observations covering 66 Iridium satellites. Each sample contains metadata such as SNR, acquisition timestamp, location information, and environmental information. The dataset preserves the unmodulated tone, BPSK field, and QPSK data segment of each detected Iridium burst as one-dimensional I/Q sequences, while the multi-receiver acquisition configuration enables the investigation of receiver-dependent effects under common observation conditions. Representative classification experiments with multiple deep-learning models, different input lengths, different signal-to-noise ratios, and cross-receiver evaluations are provided as technical validation to demonstrate the usability of the released data for downstream satellite specific emitter identification tasks. The dataset provides a publicly available data resource for satellite radio frequency fingerprinting, cross-receiver identification, and low-SNR signal studies.

Aiming at the lack of public dataset of satellite Specific Emitter Identification (SEI) and the problem of insufficient data under the condition of multiple receivers, this paper constructs a multi-receiver one-dimensional I/Q dataset for Iridium downlink signal through six software radio receivers with the same parameter configuration. The dataset contains 367,658 valid samples from 66 Iridium satellites. Each sample contains metadata such as SNR, acquisition timestamp, location information, and environmental information. At the same time, the validity of the dataset was evaluated by three experiments under seven deep learning models, different input lengths and different signal-to-noise ratios. The experimental results show that the dataset can effectively support the training and testing of different deep learning models, and can reflect the influence of signal length and signal quality on the performance of satellite individual identification. It provides data basis for satellite radio-frequency fingerprinting identification, low signal-to-noise ratio identification and model performance evaluation under the condition of multiple receivers.