Automatic modulation classification (AMC) of received unknown signals is critical in modern communication systems, enabling intelligent signal interception and spectrum management. In this paper, we propose a wavelet-based spectrum convolutional neural network (WS-CNN) model that integrates signal processing techniques with deep learning to achieve robust classification under challenging conditions, including noise, fading, and Doppler effects. The WS-CNN model is based on wavelet analysis and a convolutional neural network (CNN). Specifically, the proposed wavelet analysis, including wavelet threshold denoising, median filtering, and continuous wavelet transformation, is used for signal preprocessing to extract features and generate a compact 2D diagram. The 2D diagram is subsequently fed into the CNN for classification. The simulation results show that the proposed WS-CNN model achieves higher classification rates across a wide range of signal-to-noise ratios (SNRs) compared with existing methods.

High-precision Doppler measurement is essential for deep-space Unified X-band (UXB) tracking systems, yet digital implementations suffer from finite word-length quantization errors that degrade performance. This study analyzes frequency offset errors in UXB transponder systems, focusing on the phase-locked loop (PLL) and system-level digital processing. A digital system model is presented, featuring an FFT-based coarse acquisition and a digital Costas loop for carrier synchronization. The simulation results reveal that 32-bit quantization yields unacceptable frequency offset errors. By extending critical paths to 48 bits, the system reduces frequency offset error by approximately 2¹⁶ and achieves sub-0.01 mm/s velocity accuracy, significantly improving coherence and meeting deep-space measurement requirements.

With the intensifying competition among countries in lunar and deep-space exploration, the forward communication bands between Earth and the Moon are becoming increasingly congested. Spread-spectrum communication is poised to replace conventional unified-carrier wave systems as the dominant telemetry and tracking technology. Compared with Phase-Shift Keying (PSK) modulation, Gaussian-Filtered Minimum-Shift Keying (GMSK) offers lower transmit sidelobes and higher spectral efficiency, making it a promising alternative to Quadrature Phase-Shift Keying (QPSK) as a communication scheme integrated with spread-spectrum systems. This paper compares the direct-sequence spread-spectrum Gaussian-Filtered Minimum-Shift Keying (DSSS–GMSK) modulation scheme with the widely adopted DSSS–BPSK scheme and demonstrates the advantages of DSSS–GMSK, including higher spectral efficiency and improved signal acquisition performance over a wide Doppler dynamic range. This paper also proposes a deep-space DSSS–GMSK signal acquisition algorithm capable of achieving signal acquisition at a carrier-to-noise density ratio $(C/N_0)$ of 34 dB-Hz over a Doppler range from −350 kHz to +350 kHz, demonstrating the feasibility and potential of DSSS–GMSK. The algorithm requires minimal computational hardware resources, and its feasibility has been verified through simulations. Furthermore, its resistance to false locking is analyzed and discussed.