Acceleration of signal processing in modern radar based on GPU platform
DOI:
https://doi.org/10.54939/1859-1043.j.mst.111.2026.60-70Keywords:
Radar; DSP; GPU Processing; GPU Acceleration; CUDA Programming.Abstract
Modern radar systems aiming for high resolution, wide bandwidth, and real-time processing of large data volumes pose significant computational challenges to traditional signal processing approaches. Implementations based on CPUs, FPGAs (Field-Programmable Gate Array), or dedicated DSPs (digital signal processors) often fail to provide sufficient throughput and computational resources for intensive tasks such as matched filtering, fast Fourier transforms (FFTs), Doppler processing, digital beamforming, and synthetic aperture radar (SAR) image formation. To address these limitations, this paper proposes the use of graphics processing units (GPUs) as an acceleration platform for radar signal processing algorithms by exploiting the massive parallelism inherent in GPU architectures. The paper further presents performance measurements and evaluations conducted on representative radar datasets using various signal processing algorithms. The results demonstrate that GPU-based implementations can achieve speedups ranging from tens to hundreds of times compared to MATLAB-based CPU implementations. These findings indicate that GPU-accelerated signal processing is a promising solution for meeting real-time processing requirements in modern radar systems. In addition, computational complexity analysis and numerical accuracy validation between CPU and GPU implementations are provided to ensure the correctness and scientific rigor of the reported performance improvements.
References
[1]. Kong, Fanxing, Yan Rockee Zhang, Jingxiao Cai, and Robert D. Palmer. “Real-time radar signal processing using GPGPU (general-purpose graphic processing unit)”. Radar Sensor Technology XX, Vol. 9829, pp. 311-317, (2016).
[2]. Venter, Christian Jacobus. “Software-defined pulse-doppler radar signal processing on graphics processors”. PhD diss., University of Pretoria, (2014).
[3]. Yu, Xining, Yan Zhang, Ankit Patel, Allen Zahrai, and Mark Weber. “An implementation of real-time phased array radar fundamental functions on a DSP-focused, high-performance, embedded computing platform”. Aerospace, vol. 3, no. 3, p. 28, (2016).
[4]. NVIDIA. “CUDA C Programming Guide”. (2023).
[5]. Jin, Xingxing, and Seok-Bum Ko. “GPU-based parallel implementation of SAR imaging”. 2012 International Symposium on Electronic System Design (ISED), pp. 125-129, (2012).
[6]. Rupniewski, Marek, Gustaw Mazurek, Jacek Gambrych, Marek Nałęcz, and Rafał Karolewski. “A real-time embedded heterogeneous GPU/FPGA parallel system for radar signal processing”. 2016 Intl IEEE Conferences on Ubiquitous Intelligence & Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People, and Smart World Congress (UIC/ATC/ScalCom/CBDCom/IoP/SmartWorld), pp. 1189-1197, (2016).
[7]. Benson, Thomas M., Ryan K. Hersey, and Edwin Culpepper. “GPU-based space-time adaptive processing (STAP) for radar”. 2013 IEEE High Performance Extreme Computing Conference (HPEC), pp. 1-6, (2013).
[8]. Zhao, Xinyu, Peng Liu, Bingnan Wang, and Yaqiu Jin. “Gpu-accelerated signal processing for passive bistatic radar”. Remote Sensing, vol. 15, no. 22, p. 5421, (2023).
[9]. Liu, Hang. “Signal parallel processing in high frequency surface wave radar based on CUDA”. International Conference on Electronic Information Engineering and Computer Science (EIECS 2022), vol. 12602, pp. 171-176, (2023).
[10]. Hoffmann, Marcel, Theresa Noegel, Christian Schüßler, Lars Schwenger, Peter Gulden, Dietmar Fey, and Martin Vossiek. “Implementation of real-time automotive SAR imaging”. 2023 20th European Radar Conference (EuRAD), pp. 327-330, (2023).
[11]. Yang, Tao, Xinyu Zhang, Qingbo Xu, Shuangxi Zhang, and Tong Wang. “An embedded-gpu-based scheme for real-time imaging processing of unmanned aerial vehicle borne video synthetic aperture radar”. Remote Sensing, vol. 16, no. 1, p. 191, (2024).
[12]. Li, Wenda, Chong Tang, Shelly Vishwakarma, Karl Woodbridge, and Kevin Chetty. “Design of high‐speed software defined radar with GPU accelerator”. IET Radar, Sonar & Navigation, vol. 16, no. 7, pp. 1083-1094, (2022).
[13]. Zhao, Min, Qianshun Zou, Bing Sun, Youbin Song, and Jing Ma. “CPU+ GPU architecture radar real-time signal processing method based on signal description technology”. IET Conference Proceedings CP874, vol. 2023, no. 47, pp. 2365-2369, (2023).
[14]. Bu, Zirong, Lijun Wang, and Huijie Zhu. “Research on GPU Parallel Acceleration of Efficient Coherent Integration Processor for Passive Radar”. International Conference on Artificial Intelligence for Communications and Networks, pp. 415-422, (2021).
