Rank deficiency in coprime sampling

This paper mainly analyzes the rank deficiency in coprime sampling (CPS) which causes the failure of signal recovery. Due to the fact that the rank of convariance matrix has close relation with the way of arranging the signal matrix, the conditions of rank deficiency are different when using differe...

Full description

Saved in:
Bibliographic Details
Published in2019 14th IEEE International Conference on Electronic Measurement & Instruments (ICEMI) pp. 1740 - 1746
Main Authors Yu, Chen, Yijiu, Zhao, Haoyue, Yan, Jianguo, Huang, Yanze, Zheng, Sitao, Mei
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.11.2019
Subjects
compressive sensing
Conferences
coprime sampling
Covariance matrices
Delays
Estimation
Frequency measurement
Instruments
multicoset sampler
Multiple signal classification
MUSIC algorithm
rank deficiency
Online AccessGet full text

Cover

More Information
Summary:This paper mainly analyzes the rank deficiency in coprime sampling (CPS) which causes the failure of signal recovery. Due to the fact that the rank of convariance matrix has close relation with the way of arranging the signal matrix, the conditions of rank deficiency are different when using different processing methods. There are two popular methods used to recover signal from samples collected by CPS, i.e., direct estimation (DE) and multicoset sampler (MC) equivalence. These two methods are both based on Multiple Signal Classification (MUSIC) algorithm to search for the support of input signal. After analysis and comparison, the MC equivalence is much less likely to suffer from rank deficiency than DE. Furthermore, the MC equivalence has higher anti-noise capability and less computation. So it is concluded in this paper that MC equivalence is a preferred method employed in CPS whether the input is a multitone signal or a multiband signal.
DOI:10.1109/ICEMI46757.2019.9101823