摘要
捕获灵敏度和实时性是衡量软件接收机基带信号处理性能的关键指标,然而这两个指标之间却存在着此消彼长的关系。文中提出了一种基于CPU和GPU异构计算平台的北斗B1C信号快速捕获方法,可以较好地平衡捕获灵敏度和实时性能之间的矛盾。通过分析B1C信号捕获模块的各个部分访存量和计算量,该方法可以针对性地对计算和带宽资源消耗较大的部分进行GPU并行优化。复杂的逻辑判断和状态转换依靠CPU实现,而大量的彼此之间独立的数据计算则依赖于GPU实现。与CPU实现的软件接收机相比,文中的实现不仅在捕获总耗时上大幅度减少,而且在捕获灵敏度上也比捕获传统的B1I信号提高了将近8个d B。
Acquisition sensitivity and real-time performance are key indicators for measuring the performance of baseband signals processing in software receivers. However,there is a trade-off between these two indictors. The paper proposes a fast acquisition method for Beidou B1 C signals based on CPU and GPU heterogenous computing platform,which can balance the contradiction between acquisition sensitivity and real-time. By analyzing the amount of memory access and computation of each part in B1 C signal acquisition module,this method can be used to GPU parallel optimize the parts with larger computation and bandwidth resource consumption. Complex logic decisions and state transitions rely on CPU implementation, while a large number of independent data calculations depend on the GPU implementation. Compared with the software receiver implemented by the CPU,the implementation of this paper not only greatly reduces the total acquisition time,but also improves the acquisition sensitivity by nearly 8 d B compared with the acquisition of the traditional B1 I signal.
引文
[1]谢钢.GPS原理与接收机设计[M].北京:电子工业出版社,2013:349-374.
[2]Borio D,O’Driscoll C,Lachapelle G.Coherent,Noncoherent,and Differentially Coherent Combining Techniques for Acquisition of New Composite GNSS Signals[J].Aerospace and Electronics System,IEEE Transactions on,2009(45):1227-1240.
[3]Borio D,Presti L.Data and Pilot Combining for Composite GNSS Signal Acquisition[J].International Journal of Navigation and Observation,2008:1-12.
[4]Huang Bin,Yao Zheng,Guo Fu,et al.STARx–A GPU Based Multi-System Full-Band Real-Time GNSS Software Receiver[J].ION GNSS+2013,Institute of Navigation,Nashville,Tennessee,2013:1549-1559.
[5]Xu Liang-chun,Ziedan N,Guo Wen-fei,et al.NAVSDR:A GPUbased Modular GPS Software Receiver[J].ION GNSS+2015,Institute of Navigation,Tampa,Florida,2015:3198-3205.
[6]Xu Liang-chun,Ziedan N,Niu Xiao-ji,et al.Correlation Acceleration Software Receivers Using a CUDA-enabled GPU[J].GPS Solutions,2017,21(1):225-236.
[7]Parkinson B,Spilker J,Axelrad P,et al.Global Positioning System:Theory and Applications[M].American Institute of Astronautics and Aeronautics,1996.
