Xiangwei Li
Carol Jingyi Li
Philip H. W. Leong
Abstract
The spectral correlation density (SCD) function is the time-averaged correlation of two spectral components used for analyzing periodic signals with time-varying spectral content. Although the analysis is extremely powerful, it has not been widely adopted in real-time applications due to its high computational complexity. In this article, we present an efficient FPGA implementation of the FFT accumulation method (FAM) for estimating the SCD function and its alpha profile. The implementation uses a linear systolic array with a bi-directional datapath consisting of DSP-based processing elements (PEs) with a dedicated instruction schedule, achieving a PE utilization of 88.2%.
The 128-PE implementation achieves a clock frequency in excess of 530 MHz and consumes 151K LUTs, 151K FFs, 264 BRAMs, 4 URAMs, and 1,054 DSPs, which is less than 36% of the logic fabric on a Zynq UltraScale+ XCZU28DR-2FFVG1517E RFSoC device. It has a modest 12.5W power consumption and an energy efficiency of 4,832 MOPS/W, which is 20.6× better than the published state-of-the-art GPU implementation. In terms of throughput, it achieves 15,340 windows/s (15,340 windows/s × 2,048 samples/window = 31.4 MS/s), which is a 4.65× improvement compared to the above-mentioned GPU implementation and 807× compared to an existing hybrid FPGA-GPU implementation.
- [1] . 2007. Cyclic spectral analysis in practice. Mechan. Syst. Sig. Process. 21, 2 (2007), 597–630.Google Scholar
Cross Ref
- [2] . 2009. Cyclostationarity by examples. Mechan. Syst. Sig. Process. 23, 4 (2009), 987–1036.Google ScholarCross Ref
- [3] . 2012. Detection of surface ships from interception of cyclostationary signature with the cyclic modulation coherence. IEEE J. Ocean. Eng. 37, 3 (2012), 478–493.Google ScholarCross Ref
- [4] . 2017. Fast computation of the spectral correlation. Mechan. Syst. Sig. Process. 92 (2017), 248–277.Google ScholarCross Ref
- [5] . 2015. GPU and FPGA based architecture design for real-time signal classification. In Proceedings of the Wireless Innovation Forum Conference on Wireless Communications Technologies and Software Defined Radio (WInnComm’15). Springer, 70–79.Google Scholar
- [6] . 2016. Reducing memory requirements for high-performance and numerically stable Gaussian elimination. In Proceedings of the ACM/SIGDA International Symposium on Field-programmable Gate Arrays. 244–253.Google ScholarDigital Library
- [7] . 2018. A faster algorithm for the calculation of the fast spectral correlation. Mechan. Syst. Sig. Process. 111 (2018), 113–118.Google ScholarCross Ref
- [8] . 1993. Digital implementations of spectral correlation analyzers. IEEE Trans. Sig. Process. 41, 2 (1993), 703–720.Google ScholarDigital Library
- [9] . 1996. Detection and Identification of Cyclostationary Signals.
Technical Report . Naval Postgraduate School, Monterey, CA.Google Scholar - [10] . 1986. The spectral correlation theory of cyclostationary time-series. Sig. Process. 11, 1 (1986), 13–36.Google ScholarDigital Library
- [11] . 1989. Statistical Spectral Analysis: A Nonprobabilistic Theory. Prentice-Hall, Englewood Cliffs, NJ.Google Scholar
- [12] . 1994. Cyclostationarity in Communications and Signal Processing. IEEE Press, New York.Google Scholar
- [13] . 2006. Cyclostationarity: Half a century of research. Sig. Process. 86, 4 (2006), 639–697.Google ScholarDigital Library
- [14] . 2008. A parallel computing based spectrum sensing approach for signal detection under conditions of low SNR and Rayleigh multipath fading. In Proceedings of the 3rd IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks. IEEE, 1–10.Google ScholarCross Ref
- [15] 2018. RF Datasets for Machine Learning. Retrieved from https://www.deepsig.ai/datasets.Google Scholar
- [16] 2021. UltraScale Architecture DSP Slice User Guide. Retrieved from https://docs.xilinx.com/v/u/en-US/ug579-ultrascale-dsp.Google Scholar
- [17] . 1985. VLSI array processors. IEEE ASSP Magazine 2, 3 (1985), 4–22.Google Scholar
- [18] . 2012. Parallelization of spectrum sensing algorithms using graphic processing units. In Proceedings of the Cross Strait Quad-Regional Radio Wireless Conference (CSQRWC’12). IEEE, 35–39.Google Scholar
- [19] . 2003. Design of a parameterizable silicon intellectual property core for QR-based RLS filtering. IEEE Trans. Very Large Scale Integ. Syst. 11, 4 (2003), 659–678.Google ScholarDigital Library
- [20] . 2018. GPU based quarter spectral correlation density function. In Proceedings of the Conference on Design and Architectures for Signal and Image Processing (DASIP’18). IEEE, 88–93.Google ScholarCross Ref
- [21] . 2020. GPGPU-based parallel implementation of spectral correlation density function. J. Sig. Process. Syst. 92, 1 (2020), 71–93.Google ScholarDigital Library
- [22] . 2014. Spectrum sensing based on spectral correlation for cognitive radio systems. In Proceedings of the 4th International Conference on Wireless Communications, Vehicular Technology, Information Theory and Aerospace & Electronic Systems (VITAE’14). IEEE, 1–4.Google ScholarCross Ref
- [23] . 1996. VLSI systolic arrays for adaptive nulling [radar]. IEEE Sig. Process. Mag. 13, 4 (1996), 29–49.Google ScholarCross Ref
- [24] . 2009. Automatic modulation classification for cognitive radios using cyclic feature detection. IEEE Circ. Syst. Mag. 9, 2 (2009), 27–45.Google ScholarDigital Library
- [25] . 1991. Computationally efficient algorithms for cyclic spectral analysis. IEEE Sig. Process. Mag. 8, 2 (1991), 38–49.Google ScholarCross Ref
- [26] . 2009. Identification and Classification of OFDM Based Signals Using Preamble Correlation and Cyclostationary Feature Extraction.
Technical Report . Naval Postgraduate School, Monterey, CA.Google Scholar - [27] . 1999. The strip spectral correlation algorithm for spectral correlation estimation of digitally modulated signals. In 4th International Conference on Telecommunications in Modern Satellite, Cable and Broadcasting Services. TELSIKS’99 (Cat. No. 99EX365), Vol. 1. IEEE, 277–280.Google ScholarCross Ref
- [28] . 2020. Real-time automatic modulation classification using RFSoC. In Proceedings of the International Parallel and Distributed Processing Symposium Workshops (IPDPSW’20). IEEE, 82–89.Google ScholarCross Ref
- [29] . 2010. Parallelized cyclostationary feature detection on a software defined radio processor. In Proceedings of the International Symposium on Signals, Systems and Electronics. IEEE, 1–4.Google ScholarCross Ref
- [30] . 2017. Automated systolic array architecture synthesis for high throughput CNN inference on FPGAs. In Proceedings of the 54th Annual Design Automation Conference. 1–6.Google ScholarDigital Library
- [31] . 1967. The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Trans. Aud. Electroacoust. 15, 2 (1967), 70–73.Google ScholarCross Ref
- [32] . 2020. The RFML ecosystem: A look at the unique challenges of applying deep learning to radio frequency applications. arXiv preprint arXiv:2010.00432 (2020).Google Scholar
- [33] . 2012. Portable and scalable FPGA-based acceleration of a direct linear system solver. ACM Trans. Reconfig. Technol. Syst. 5, 1 (2012), 1–26.Google ScholarDigital Library
Index Terms
- A Scalable Systolic Accelerator for Estimation of the Spectral Correlation Density Function and Its FPGA Implementation
Recommendations
A reconfigurable HexCell-based systolic array architecture for evolvable hardware on FPGA
AbstractEvolvable hardware is a system that modifies its architecture and behavior to adapt with changes of the environment. It is formed by reconfigurable processing elements driven by an evolutionary algorithm. In this paper, we study a reconfigurable ...
FPGA-based configurable systolic architecture for window-based image processing
Image processing requires more computational power and data throughput than most conventional processors can provide. Designing specific hardware can improve execution time and achieve better performance per unit of silicon area. A field-programmable-...
An FPGA-based fault-tolerant 2D systolic array for matrix multiplications
Transactions on computational science XIIIThis paper proposes a method to implement fault-tolerant self-reconfigurable 2D systolic arrays to calculate matrix multiplications on FPGAs. The array uses a 1.5-track switching network for reconfiguration. The array implemented is compared to the ...
Comments