![]() |
FOURTH
ANNUAL |
Cross-Correlation Based
|
Athina Petropulu and Haralambos Pozidis Drexel University Electrical and Computer Engineering Department Philadelphia, PA 19104 tel: (215) 895-2358 fax: (215) 895-1695 email: athina@artemis.ece.drexel.edu Kenneth
Abend Abstract We propose a novel method for the reconstruction of an unknown signal that propagates in the presence of multipath. By oversampling the received sequence, or through spatial diversity, we first form a single input/two output problem. We then estimate the multipath effect by applying cross-correlation based operations of the two outputs, and recover the input signal via inverse filtering. The method is computationally attractive as compared to higher-order spectra based methods. It is not sensitive to errors in channel length estimates and performs better in low SNR's as compared to existing second-order statistics based methods. We show how the method can lead to a novel solution to the problem of estimating the propagation channels in radar applications. |
![]()
Direct comments and questions to: webmaster@ll.mit.edu
© MIT Lincoln Laboratory. All rights reserved.