期刊文献+

多核DSP信号处理并行设计 被引量:4

Signal Processing Parallel Design Based on Multicore DSP
下载PDF
导出
摘要 并行计算是实现高性能计算的一个重要发展方向。随着信号处理、通信等领域对处理能力需求的不断提升,DSP的并行开发技术也得到了较快发展。多器件并行和片上多核的方法可以有效提高处理性能。多核并行处理相对于传统单核DSP要进行多任务并行设计,使系统设计更加复杂。文中在探讨了利用8核处理器进行信号处理开发的关键技术的基础上,采用Round-Robin方式设计了一种多核并行信号处理模式,并对多核的同步、Cache一致性、任务并行分配等进行了论述。 Parallel computation is an important way to implement high performance computing. With the improving demands of signal processing, communications, etc. , parallel design technology based on DSP is developing rapidly. Multichip system and multicore on a chip system are efficient to increase processing performance. Compared with traditional single core DSP, multicore parallel processing requires multitask parallel design, which makes the system design more complicated. After discussing the key technology of the signal processing based on an eight-core processor, a Round-Robin mode is implemented to design multicore parallel signal processing. Multicore synchronization, maintaining Cache coherence and mapping application to multicore are also introduced.
出处 《雷达科学与技术》 2013年第6期617-620,共4页 Radar Science and Technology
关键词 并行计算 信号处理 同步 CACHE一致性 parallel computation signal processing synchronization Cache coherence
  • 相关文献

参考文献6

二级参考文献40

  • 1鲁千红,杨军.基于CPCI总线的雷达信号处理系统[J].空军雷达学院学报,2007,21(2):118-120. 被引量:3
  • 2Iyengar J R, Amer P D, and Stewart R. Coneurrent multipath transfer using SCTP multihoming over independent end-to-end paths[J]. IEEE/AC'M Transactions on Networking, 2006, 14(5): 951-964.
  • 3Bing W, Wei W, Zheng W, et al.. Multipath live streaming via TCP: schenle, perfbrmanee and benefits[J]. ACM Transactions on Multimedia Computing, Communications, and Applications, 2009, 5(3): 1-23.
  • 4Evensen K, Kaspar D, Engelstand P, et al. A network-layer proxy tbr bandwidth aggregation and reduction of IP packet reordering[C]. 34th Annual IEEE Conference on Local Computer Networks, Zurich, Switzerland, 2009: 585-592.
  • 5Bai V, Weiping Z, Botta A, et al. A markovian approach to multipath data transfer in overlay networks[J]. IEEE Transactions on Par'allel and Distributed Systerms, 2010, 21(10): 1398-1411.
  • 6Jie F, Zhipeng O, Lisong X, et al. Packet reordering in high-speed networks and its impact on high-speed TCP variants[J]. Computer Communications, 2009, 32(1): 62-68.
  • 7Iliadis I and Lien L Y C. Resequencing delay for a queueing system with two heterogeneous servers under a threshold- type scheduling[J]. IEEE Transactions on Communicaltions, 1988, 36(6): 692-702.
  • 8Chowdhury S. An analysis of virtual circuits with parallel links[J]. IEEE Transaetions on Communications , 1991, 39(8) 1184-1188.
  • 9Jean-Marie A and Grin L. Parallel queues with resequencing[J]. Journal of the A CM, 1993, 40(5): 1188 1208.
  • 10Tsoukatos K and Makowski A. Asymptotic optimality of the Round-Robin policy in multipath routing with resequencing[J]. Queueing S?lstems, 2006, 52(3): 199-214.

共引文献50

同被引文献16

  • 1田耘,徐文波.Xilinx FPGA开发实用教程[M].清华大学出版社,2008.
  • 2于凤芹.TMS320 C6000 DSP结构原理与硬件设计[M].北京航空航天大学大学出版社,2008.
  • 3Texas Instruments.TMS320C645xDSPexternal memory interface(EMIF)USeY' s guide[EBlOI].httpll:www, ti. com, 2006.
  • 43GPP TS 36.212 v10.6.0: Multiplexing and channel coding. (Release 9)[S].2012, 6.
  • 5CUMMINGIG,WONGFH.合成孔径雷达成像:算法与实现[M].洪文,胡东辉,译.北京:电子工业出版社,2012:219—230.
  • 6TMS320C6678 Multicore Fixed and Floating-Point Digital Signal Processor [Z]. Dallas, Texas: Texas Instruments, 2015.
  • 7KWARK K S, KWON 0 K. Power - Reduction technique using a single edge -tracking clock for multiphase clock and data recovery circuits[J]. IEEE Transaction on Circuits and Systems II: Express Briefs ,2014,61 (4) : 239 - 243.
  • 8包东智.LTE产业现状及发展趋势[J].移动通信,2008,32(16):43-47. 被引量:14
  • 9方冬,章国宝.基于EMIF接口的数据采集系统的设计[J].电子设计工程,2011,19(2):174-178. 被引量:9
  • 10刘燕,孙光才,邢孟道.大场景高分辨率星载聚束SAR修正ω-k算法[J].电子与信息学报,2011,33(9):2108-2113. 被引量:6

引证文献4

二级引证文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部