Design Space Exploration of 1-D FFT Processor
- 211 Downloads
A design space exploration methodology of 1-D FFT processor is proposed to find the best hardware architecture in a quantitative way during early design. The methodology includes architecture candidate collection, coarse-grained architecture selection, and circuit level design optimizations. We show how to select a better architecture from candidates including different architectures (SDF, SDC, MDF, MDC and memory-based) with different degree of parallelism at different radices. The sub-level designs, including designs of rotator and data scaling module, are introduced for further optimizations. As a proof of concept, an FFT processor for 4G, WLAN and future 5G is designed supporting 16-4096 and 12-2400 point FFTs. Memory-based architecture with 16-datapath mixed-radix butterfly unit is selected to satisfy the demands for 1GS/s (4096) throughput. The synthesis result based on 65nm technology shows that the silicon cost and power consumption are 1.46mm2 and 68.64mW respectively. The proposed processor has better normalized throughput per area unit and normalized FFTs per energy unit than the state of the art available designs.
KeywordsFFT (Fast Fourier Transform) Design space exploration Twiddle factor BFP (Block Floating Point) Non-power-of-two-point FFT
The authors would like to thank Synopsys for their support in the use of ASIP Designer, which is used as a high-level synthesizer.
The finance supporting from National High Technical Research and Development Program of China (863 program) 2014AA01A705 is sincerely acknowledged by authors.
- 1.Guideline for 3.5GHz 5G System Prototype and Trial(Version 1.0). Tech. rep. CMCC (2017).Google Scholar
- 2.3GPP TS 36.211: Evolved Universal Terrestrial Radio Access (E-UTRA); LTE Physical Channels and Modulation (2012).Google Scholar
- 17.Garrido, M., & Grajal, J. (2007). Efficient memoryless CORDIC for FFT computation. In Proc. IEEE Int. Conf. acoustics, speech, and signal proceess. (ICASSP) (Vol. 2, pp. II–113–II–116).Google Scholar
- 25.Huang, S.J., & Chen, S.G. (2014). A new memoryless and low-latency FFT rotator architecture. In Int. Symp. on integrated circuits (ISIC) (pp. 180–183).Google Scholar
- 26.Humphries, B., Zhang, H., Sheng, J., Landaverde, R., Herbordt, M.C. (2014). 3D FFTs on a single FPGA. In IEEE 22nd Annual Int. symp. on field-programmable custom computing machines (pp. 68–71).Google Scholar
- 28.IEEE 802.11ac-2013: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications–Amendment 4: Enhancements for Very High Throughput for Operation in Bands below 6 GHz.Google Scholar
- 32.Jui, P.C., Wey, C.L., Shiue, M.T. (2013). Low-cost parallel FFT processors with conflict-free ROM-based twiddle factor generator for DVB-T2 applications. In IEEE Int. Midwest symp. circuits syst. (MWSCAS) (pp. 1003–1006).Google Scholar
- 33.Choi, J.-R., Park, S.-B., Han, D.-S., Park, S.-H. (2000). A 2048 complex point FFT architecture for digital audio broadcasting system. In Proc. IEEE Int. symp. circuits syst. emerging technol. for the 21st Century (Vol. 5, pp. 693–696).Google Scholar
- 36.Koutsoyannis, R., Milder, P.A., Berger, C.R., Glick, M., Hoe, J.C., Puschel, M. (2012). Improving fixed-point accuracy of FFT cores in O-OFDM systems. In Proc. IEEE Int. conf. acoustics, speech, and signal proceess. (ICASSP) (pp. 1585–1588).Google Scholar
- 38.Lakshmi, B., & Dhar, A.S. (2008). High speed architectural implementation of CORDIC algorithm. In TENCON 2008 - 2008 IEEE region 10 conf. (pp. 1–5).Google Scholar
- 39.Lee, Y.H., Yu, T.H., Huang, K.K., Wu, A.Y. (2006). Rapid IP design of variable-length cached-FFT processor for OFDM-based communication systems. In Proc. IEEE Workshop signal process. syst. design and implement. (pp. 62–65).Google Scholar
- 40.Lenart, T., & Owall, V. (2003). A 2048 complex point FFT processor using a novel data scaling approach. In Proc. IEEE Int. Symp. circuits syst. (ISCAS) (Vol. 4, pp. IV–45–IV–48).Google Scholar
- 47.Oh, J.Y., & Lim, M.S. (2005). Area and power efficient pipeline FFT algorithm. In Proc. IEEE Workshop signal process. syst. design and implement (pp. 520–525).Google Scholar
- 55.Shousheng, H., & Torkelson, M. (1998). Designing pipeline FFT processor for OFDM (de)modulation. In Proc. URSI int. symp. signals, syst. and elect. (pp. 257–262).Google Scholar
- 59.Thomas, L.H. (1963). Using a computer to solve problems in physics. Applications of digital computers. Boston: Ginn.Google Scholar