An optimized reconfigurable power spectral density converter for real-time shrew DDoS attacks detection

The shrew DDoS attacks are stealth low-rate TCP-targeted DDoS attacks, which conceal their malicious activities into normal traffic. Although the good pretense eludes them from being detected in time domain, the existent energy exposes them in frequency domain. Online Power Spectral Density (PSD) analysis necessitates real-time PSD data conversion is a must. In this paper, an optimized FPGA based real-time PSD converter is proposed, which is based on our innovative component-reusable Auto-Correlation (AC) algorithm and the adapted 2N-point real-valued Discrete Fourier Transform (DFT) algorithm. Further optimization is achieved through the exploration of algorithm characters and hardware parallelism for this case. The evaluation results from both simulation and synthesis are provided. The conversion of a 512-point data sequence can be finished in the time interval of sampling one data point; and the overall design could be easily fitted in a Xilinx Virtex2 Pro FGPA.

Figure optionsDownload as PowerPoint slideHighlights
► We proposed an optimized FPGA based real-time PSD converter.
► An innovative component-reusable AC algorithm is designed.
► An adapted 2N-point real-valued DFT algorithm is implemented.
► The optimized converter achieved a speedup of 61.8%, consuming only 11% of resources on Xilinx Virtex-2 Pro XC2VP50 FPGA.