Mobile location estimation with Non-Line-of-Sight mitigation using Kalman filtering | |
| Author | Le Bao Long |
| Call Number | AIT Thesis no.TC-02-22 |
| Subject(s) | Kalman filtering Mobile communication systems |
| Note | A thesis submitted in partial fulfillment of the requirement for the degree of Master of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. TC-02-22 |
| Abstract | Mobile location estimation has attracted much interest over the past few years due to its potential for new wireless applications and network performance improvement. While the most promising technique for mobile location using cellular mobile networks may be Time Different of Arrival (TDOA) method, the major challenged problems which inhibit accurate mobile location estimation are multipath and non line of sight (NLOS) problems. These problems make the estimated range biased from the true range and finally lead to big errors in mobile location estimates. Range data due to NLOS Base Stations (BS) can be simply ignored but the number of available BSs may not always be enough and range data from as many BSs as possible should be used to overcome large location estimation errors due to undesirable geometric BS layout. In this thesis, techniques for LOSINLOS detection on a sample per sample basis, which are called Real Time Mode and Delayed Mode are proposed and suitable range processing approaches using Kalman filtering are also devised to smooth range data depending on the LOSINLOS detection result. The proposed techniques for LOSINLOS recognition and range processing have the advantage that an acceptable delay is used while other previously proposed techniques using the time historic information of range data required much delay, which would be unfavorable for real time location applications. The famous Chan's algorithm is employed to estimate the mobile location using the processed range data. A further Kalman filter, which was previously proposed by another author is used to smooth the obtained mobile trajectory. The proposed range processing technique in a mobile location architecture is tested by running simulations for different mobile trajectory types under the influence of exponentially distributed and uniformly distributed NLOS noise. The obtained location estimation performance was shown to well satisfy the 67% and 95% error requirements set by US Federal Communications Commission (FCC) in 1999 for straight and zigzag trajectory kinds. For challenged long trajectory, the location mandates were only violated in completely NLOS or nearly NLOS wireless environments. The Delayed Mode was also shown to work little better than Real Time Mode in these tough environments. |
| Year | 2002 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. TC-02-22 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Telecommunications (TC) |
| Chairperson(s) | Ahmed, Kazi Mohiuddin ; |
| Examination Committee(s) | Erke, Tapio J. ;Fernando, W. A. C. ; |
| Scholarship Donor(s) | Keihin Electric Express Railway Company Limited (KEIKYU); |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2002 |