Parameters selection of FBLMS algorithm for noise FM-CW radar

Yuping Zhang, Meiguo Gao, Jiabin Liu, Yunjie Li

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)

Abstract

The direct wave and clutter wave cancellation is a key procedure for noise FM-CW (frequency-modulated continuous-wave) radar. Frequency-domain Block LMS (FBLMS) adaptive filters have better performance of clutter (include the direct wave) cancellation in low computational complex. In practical system, however, the chosen of step-size and the tap length of FBLMS are extremely important for the subsequent target detection. Although, the step-size bounds of FBLMS have been derived and presented by different expressions, the drawbacks of them as follows: on the one hand, the step-size bound are usually complex expressions including eigenvalues which are related with the covariance matrix of the input signal, it has high computational complexity to calculate the step-size; on the other hand, the relationship between step-size bound, the tap length of FBLMS and clutter attenuation (CA) has been not discussed comprehensively. To address above-mentions, in this paper, the step-size bound is re-derived based on Feuer's work. Furthermore, the relationship between the step-size, the tap length and CA is elaborated, it shows that the CA is dominated by the tap length of FBLMS, once the tap length is not less than the number of clutter path, the CA will be decreased as the step-size increases. In addition, the relationship is intuitively shown by three-dimensional plot. Numerical results show that the parameters selection of FBLMS is beneficial to implement in the system of noise FM-CW radar.

Original languageEnglish
Pages215-220
Number of pages6
DOIs
Publication statusPublished - 2014
Event2014 12th IEEE International Conference on Signal Processing, ICSP 2014 - Hangzhou, China
Duration: 19 Oct 201423 Oct 2014

Conference

Conference2014 12th IEEE International Conference on Signal Processing, ICSP 2014
Country/TerritoryChina
CityHangzhou
Period19/10/1423/10/14

Keywords

  • Clutter cancellation
  • FBLMS algorithm
  • Noise FM-CW radar
  • Parameters selection

Fingerprint

Dive into the research topics of 'Parameters selection of FBLMS algorithm for noise FM-CW radar'. Together they form a unique fingerprint.

Cite this