3-D Spatial Spectrum Fusion Indoor Localization Algorithm Based on CSI-UCA Smoothing Technique
Open Access
- 4 October 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 6, 59575-59588
Abstract
In the indoor wireless localization environment, the non-line-of-sight receiving signal and low signal-to-noise ratio are usually strongly dominant issues due to the heavy existence of multipath signals, which also restrict the final wireless localization performance seriously. In this paper, a novel indoor localization algorithm based on 3-D multi-array spatial spectrum fusion (3-D-MSSF), which also uses the channel state information (CSI) under uniform circular array (UCA) structure, is proposed. First, in the data assembly process, the number of antennas and the number of observations can be virtually extended by applying a beam space transformation and smoothing technique by using the observed CSI information. Then, the existing multiple signal classification approach is applied on the smoothed data to jointly estimate the 2-D direction-of-arrival angles and the time-of-flight information from the resulting spatial spectrums at each UCA array. And in the grid-fusion explorer process, the estimated parameters are subsequently transmitted to the aggregation center to calculate the location results of each point relative to each access point, which introduces a grid-refinement algorithm in the search grid to improve the localization precision. While the parameters of interest for the final target position can be estimated from a single fused spatial spectrum, which results from fusing all maximum noise subspaces corresponding to the minimum error between each estimated point in the search grid and every set in the 3-D space-time searching grid. Computer simulation results together with the real application experiments in the indoor environment in terms of source position estimation and corresponding RMSE values are given. The proposed 3-D-MSSF method proves a significant indoor positioning performance, which can achieve the final localization accuracy below 1 m even if the line-of-sight signal is blocked, and there exist only multipath path signals at the receiver.Keywords
Funding Information
- Important National Science and Technology Specific Project of China (2016ZX03001022-006, 2018ZX03001020-005)
- Shanghai Association for Science and Technology (16DZ1100402, 17511106300)
- National Natural Science Foundation of China (61771308)
- Shanghai Economic and Information Committee, Ultra High Sensitivity UAV Communication System Based on Radar Related Technology
This publication has 27 references indexed in Scilit:
- Data Fusion of Acoustics, Infrared, and Marine Radar for Avian StudyIEEE Sensors Journal, 2015
- Constellation of multisensors in bearing-only location systemIEE Proceedings - Radar, Sonar and Navigation, 2005
- Generalizing MUSIC and MVDR for Multiple Noncoherent ArraysIEEE Transactions on Signal Processing, 2004
- Wireless Location2003
- Multipath time-delay detection and estimationIEEE Transactions on Signal Processing, 1999
- Overview of radiolocation in CDMA cellular systemsIEEE Communications Magazine, 1998
- An introduction to multisensor data fusionProceedings of the IEEE, 1997
- Eigenstructure techniques for 2-D angle estimation with uniform circular arraysIEEE Transactions on Signal Processing, 1994
- Multiple emitter location and signal parameter estimationIEEE Transactions on Antennas and Propagation, 1986
- A transformation between the phasing techniques required for linear and circular aerial arraysProceedings of the Institution of Electrical Engineers, 1965