Loop tracking method and system for spread spectrum telemetry signals based on GPU acceleration

In the satellite spread-spectrum telemetry system, traditional CPU software tracking devices struggle to meet the urgent requirements of real-time tracking, multi-task processing, and flexible parameter adjustment. To address this issue, this paper proposes a spread-spectrum telemetry signal loop tracking method based on a CPU+GPU heterogeneous architecture. The carrier tracking loop of the frequency-locked loop (FLL) assisted phase-locked loop (PLL) and the pseudo-code tracking loop of the delay-locked loop (DLL) are parallelized. Relevant computational kernels are designed to accelerate the operations. Meanwhile, a three-level state transition model is proposed, and the state switching of the carrier loop is achieved through the dynamic discrimination of the I/Q path energy ratio. Experimental results show that for spread-spectrum telemetry data with a sampling rate of 56 MHz, the processing time for 1 second of data using this architecture is reduced from about 13 seconds of traditional CPU devices to 0.4 - 0.5 seconds, with an acceleration ratio of approximately 30 times. It can meet the requirements of real-time loop updates and can track stably under different signal-to-noise ratios and signal parameter conditions. Compared with hardware such as FPGAs, it has lower development difficulty, a shorter development cycle, and more flexible parameter adjustment, which is in line with the development trend of software-based measurement, operation, and control.