Ari Kangas

System performance evaluation of mobile positioning methods

This paper presents a simulation methodology for evaluating the performance of location techniques for cellular systems. The simulation tool developed enables quantification of the final system performance for various choices of components such as measurement techniques, channel models and position computation methods. Positioning performance of a GSM and IS-136 cellular system is shown for a specific choice of measurements and algorithms in various radio environments. The variation in performance with the choices of system parameters such as number of measurements made per position computation is shown.

Propagation channel impact on LTE positioning accuracy: A study based on real measurements of observed time difference of arrival

The goal with this work is to evaluate the performance of the observed time difference of arrival (OTDOA) positioning method using real measured channel data in a macro-cellular urban LTE scenario. This has been made possible using a highly accurate channel sounder developed within Ericsson Research. A wideband (20 MHz bandwidth) measurement campaign at 2.66 GHz was performed using three separate base station sites in a realistic deployment. One main conclusion is that the measured channel allows a positioning accuracy of 20 m and 63 m at the median and 95% level respectively. This is a very encouraging result since the FCC requirements are fulfilled with a safe margin in this urban scenario.

Time of arrival estimation of narrowband TDMA signals for mobile positioning

Positioning services in cellular systems have received increased attention recently. Solutions of the positioning problem requiring time and angle of arrival measurements have been proposed. The problem of estimating the time of arrival (TOA) of the line of sight (LOS) component of narrowband digital signals in a multipath environment is addressed. A TOA estimation algorithm designed to operate under low SNR is proposed. The algorithm combats multipath propagation by exploiting the fading statistics of the mobile channel. Performance of the algorithm obtained through simulations using COST propagation models for GSM and IS-136 signals is shown and is promising.

Angle of Arrival Localization in {LTE} Using {MIMO} Pre-Coder Index Feedback

The paper presents a new angle of arrival (AoA) positioning technique for long term evolution (LTE) cellular systems. The method quantifies the directional correlation between TX antenna bore-sight and different multiple-input-multiple-output (MIMO) pre-coder indices. This correlation is used to develop an AoA positioning method that only builds on functionality and hardware available in standard LTE base stations. Measurements from a Swedish test network indicate that the method performs as expected.

Time-of-arrival estimation for E-OTD location in GERAN

In this paper, the problem of time-of-arrival (TOA) estimation for E-OTD location in a GSM/EDGE Radio Access Network (GERAN) is described. To increase the number of available bursts for TOA estimation (and therefore increase the accuracy), the mobile can use bursts from all time slots. However, this requires that the TOA estimation algorithm is able to handle GMSK and 8PSK bursts without a-priori knowledge about the modulation scheme. In this paper, the mutual effect of different modulation formats on the TOA estimation performance for E-OTD location is investigated. A TOA estimation algorithm is described, which does not require a blind detection of the modulation scheme.

Emergency Call Delivery Standards Impair Cellular Positioning Accuracy

Cellular location based services need to use a variety of positioning methods in order to fulfill regulatory requirements in terms of availability, response time and accuracy. Each positioning method generates results in its own format. In the wideband code division multiple access (WCDMA) standard, 7 geographical formats are therefore available for reporting from the radio network (RAN) to the core network (CN). Unfortunately, the standards used in the US for emergency call delivery support only the ellipsoid point with uncertainty circle format when delivering the calculated position to the public safety answer point (PSAP). This limitation therefore requires geographical shape conversion algorithms to be applied in the cellular network. Three new such conversions are derived in the paper. The paper also shows by several examples that the shape conversions that are needed due to the reporting restriction can increase the inaccuracy by more than a factor of 30 for A-GPS positioning and by more than a factor of 200 for RTT positioning, as measured by the area of the reported format. This makes the current location reporting standards a major bottleneck for fulfilling E911 Phase 2 requirements. The obvious solution is to enhance the number of reporting formats for emergency positioning reporting to the PSAP, and to provision supporting software systems for distribution of graphical position information to field personnel.

Enhanced WCDMA Fingerprinting Localization Using OTDOA Positioning Measurements from LTE

The focus of this paper is on the convergence of positioning technology between the WCDMA and the LTE cellular systems. The emerging possibility to use multi-RAT positioning nodes can be exploited to mitigate the lack of a high-precision indoor OTDOA positioning method in WCDMA, which today limits the coverage of self-learning databases for fingerprinting positioning in this cellular system. The problem at hand is solved by use of measured high precision OTDOA positions of opportunity from LTE.