Journal of Xidian University

Previous Articles     Next Articles

Detection method for the burst signal in a low SNR

CHEN Dan1;LIU Mingqian2;ZHANG Yang2,3;HU Ye2   

  1. (1. School of Electronics and Control Engineering, Chang'an Univ., Xi'an 710064, China,
    2. State Key Lab. of Integrated Service Networks, Xidian Univ., Xi'an 710071, China;
    3. National Mobile Communications Research Laboratory, Southeast Univ., Nanjing 210096, China)
  • Received:2016-04-18 Online:2016-12-20 Published:2017-01-19

Abstract:

A detection method for the burst signal based on the cepstrum of the power spectrum is proposed to solve the problem of the poor burst signal detection performance in the low SNR environment. First, the maximum value of the spectrum is used as the test statistic and is smoothed by the smooth window. Then, the K-mean clustering algorithm is used to classify the decision to distinguish the signal and the noise. Finally, the detection of the burst signal is completed by the three-state transition based on the length. Simulation results show that the proposed method has a better detection performance and lower computational complexity in low SNR environments.

Key words: burst signal, signal detection, cepstrum, K-mean clustering, tristate conversion