The indoor ranging signal recovery requires not only the high detection probability but also the excellent detection accuracy, which is related to the matched filter output SINR, especially for radio signal based location and positioning systems, where little deviation of time delay will yield radical error of position estimation. Furthermore, the ranging signals in multi-measurment channels demands the so-called sparse condition [1] for uniquely determining the results. Because the corresponding recovery matrix in terms of time-code frame is a wide matrix, which can not be orthogonalized between all columns any more. So far, many related recovery algorithms haven been developed, like optimization based l1-norm minimization and greedy approaches based OMP, ROMP and CoSaMP [2]. However, these algorithms are either not real-time enough or short of uniform performance in different scenarios. In this paper we will first introduce the novel ranging signals for higher time delay estimation, then develop the corresponding detection algorithm namely Regularized Compressive Sampling Mathing Pursuit (RCoSaMP), which outperforms the most conventional detection approaches.
展开▼
