A Critical Review of Position Fixing Techniques for Mobile Communications

Transcription

1 A Critical Review of Position Fixing Techniques for Mobile Communications B. Prabhakar 1, S. Balai 1, K.V.Murali Mohan, D.Krishna Reddy 3 1 Dept. of ECM, KL University, Viayawada Dept. of ECE, HITS, Hyderabad 3 Dept. of ECE, Chaithanya Bharathi Institute of Technology, Hyderabad, A.P., India. 1 bprabhakar008@gmail.com 3 dkreddi@rediffmail.com Abstract Mobile Station () positioning plays vital role in Location Based Services (LBS) of 3G communications. Accurate positioning leads to exact relevance to the services provided by LBS. For position fixing of, no single technique is superior, different techniques are used in different environments and for different services. This paper reviews various mobile positioning techniques. Keywords: LBS, Mobile positioning, and RSS I. INTRODUCTION The wireless mobile market is witnessing unprecedented growth fueled by technology revolution. The mobile network field is moving towards to integrate both voice and position data services. LBS is one such integrate service which brings great convenience and new allied services to subscribers and gained popularity because of its commercial potential. The mandatory requirement of E-911 of FCC, U.S. has prompted the growth of LBS. Due to complexity of mobile environment, positioning accurately is a challenge. In present highly competitive scenario, invariably the success, accuracy and economics of a LBS is dependent on the type of position finding techniques used. Positioning techniques are classified based on the role played by and BS in determining the location and also parameters used to find the position. These techniques should be highly reliable, accurate and fast. The various positioning methods used in maor cellular systems are critically discussed in this paper. II. NEED OF POSITION Location of the mobile user is fundamental to LBS. In the age of significant mobile communication competition, operators seek new ways to create special features and increase profits. One of the best way to deliver highly personalized services based on location. LBS delivers geographical information surrounding the for the variety of user needs such as delivery of spatial data for routing guidance (Kolodzie and Helm, 006). position is not only useful for LBS but also for other services as shown (Fig. 1). Services like E-911 in U.S, E-108/E-104 in India, E-110 in Japan and E-11 in Europe are enabled for both landline and mobile users. Enhanced 911, a location technology advanced by the FCC that enables cellular phones to process 911 calls and enable emergency services to locate geographic location of the caller. The accuracy required to E-911 services for terminal/network based mobile positioning techniques is shown in Table 1. Mobility management is an important sub layer in signaling layer of GSM network. It looks after the location updating and handoff mechanism. Handoff comes into the picture when user crossing the cell boundary during a call. The location information can also be used for distribution of radio resources dynamically to improve coverage and capacity of the cellular systems (Yamamoto et al., 001). It also helps in reducing co-channel interference, enhances frequency reuse and avoids unnecessary handoffs (Rappaport et al., 1996). In the context of security, location reporting and its mapping is essential in providing a defense against emerging national threats as well as handling conventional emergencies. Positioning systems will offer greater crime-fighting capabilities for law enforcement personnel. Remote positioning is useful in vehicle tracking, which is being exploited in fleet management. LBS become an integral component in Automatic Fleet Management Systems (AF), Intelligent Transposition System (ITS) and Automatic Vehicle Locator (AVL) by providing remote tracking facilities and offers turn-by-turn driving directions. An interesting ITS application is electronic road pricing (ERP) which exist in Germany. III. POSITIONING METHODS IN MOBILE SYSTE Based on where position calculation is conducted, positioning techniques are categorized (Sarikaya, 00) as, network based technique, handset based technique or hybrid technique. In network based technique the parameters used for computing location are measured at Base Station (BS) and transferred to a central facility for location determination (James et.al, 1998). Handset based technique requires handset modifications such as installation of GPS chipset in the. Hybrid techniques are the combination of network based and handset based techniques. Positioning methods implemented in maor cellular systems are tabled (Table ). All these methods are based on triangulation or trilateration techniques.

2 The data required to find location are angle of arrival of the signal, transit time of the signal from BS to and received signal strength or combination of these. (I) CELL OF ORIGIN (COO) METHOD It is simple and cheapest method to find position (Fig. ). When is in soft handover, radio network controller (RNC) decides appropriate cell id based on handoff history and signal strength (Jay, 004). Cell id is stored in network and. The error may be as large as network cell size radius R (Jagoe, 003). To improve accuracy, measuring signal Round Trip Time (RTT) is utilized (Jay, 004). Even though it is not that much accurate, this technique is widely used because of its simplicity and supports legacy terminals. (II) CELL OF ORIGIN WITH TIMING ADVANCE In TDMA systems (GSM), the known transit time of the signal is exploited for location estimation. Timing Advance (TA) is an estimate of the distance from the to the serving BTS (Fig. 3). TA is a 6-bit information (total 0-63 time slots correspond to 0-33 s) each with a radius of 550m (resolution is one GSM bit, which has the duration of 3.69 sec round trip from BS to and back, c m 550, c velocity of light) (35km/63 = 550m) (Helm, 003). TDMA cells know the time it takes for its signals to reach a mobile station. When no TA is available means 0 slot and last slot 63 indicates maximum TA. For example, if TA is 0, it means that is at a distance of 0 to 550 m from BS (Mattsson, 001). TA value is obtained on Random Access CHannel (RACH). In practice, the resolution of the TA may be further degraded by multipath fading. III) TIME OF ARRIVAL (TOA) METHOD The simplest way for positioning in GSM is ToA method based on handover measurements at BSs (Al-Jazzar et al., 00). For estimating RTT, a known signal (bursts) is sent to and received back by BS. Location Measurement Unit (LMU) measures time for signal to travel to and back to BS. When the BS receives the signals carrying time stamps from the it records time when it was received and sends it to a central server. The server gathers information from multiple BSs and by comparing the time of arrivals and the BSs positions, calculates the position of the by triangulation (Fig. 4). The intersection of two spheres is the circle, upon which the may be located. The accuracy of this method varies according to the accurate positioning of surrounding BSs locations, propagation of signals and synchronization of the clocks. Accuracy of is less and works with legacy terminals (Tayal, 005). For synchronization, each BS has to be equipped with GPS timing receiver. IV) ANGLE OF ARRIVAL (AOA) METHOD The direction finding systems estimate position by measuring angle at which the signal arrives at the BS. A line drawn from BS to is called as line of bearing (LoB) and is also known as radial line from BS to the. Intersection of two LoBs drawn from two different BSs defines a unique position of (Fig. 5). This technique requires no time synchronization and needs only two BSs. The difficulties in this method are estimation of angles which require complex antenna system such as installation of directional antennas or phased array antennas at BSs. This also suffers from Non Line of Sight (NLoS) propagation. V) TIME DIFFERENCE OF ARRIVAL (TDOA) ToA method generally fails to give required accuracy due to inaccurate estimation of time of arrival in mobile environment. TDoA technique avoids this difficulty, by measuring relative time difference instead of estimating the absolute time of the signal to reach from BS. TDoA is most commonly used network-based solution. It computes the location by measuring the difference between the arrival times of transmissions at individual BSs. LMU at BS notes the time when signal from arrived at BS 1 and BS (Mattsson, 001). This difference of time is called TDoA value. This TDoA is converted into distance difference between two BSs as foci to define a hyperboloid. Each TDoA measurement gives hyperboloid along which mobile may lie. The accuracy is affected by co-channel interference and multipath even when there is an LOS path (Ibnkahla, 005). For this method minimum of three BSs are needed to get two TDoA-measurements. The intersection of two hyperbolas pinpoint the position (Fig. 6). TDoA is a very precise method, but requires BSs are to be in LOS with. For ensuring time synchronization, each BS have to be equipped with GPS timing receiver. Moreover, the exact morphology of the area has to be known. The equation of such hyperboloid is written as (Krishna, 00), R i, ( X i x) ( Yi y) ( Z i z) where ( X x) ( Y y) ( Z z) (1) (X i, Y i, Z i ) and (X, Y, Z ) represent the coordinates of the i th and th BSs. (x,y,z) is location and R i, is distance equilent of TDoA. The modified handset TDoA method proposed for the CDMA system utilizes the pilot tones transmitted by different BSs (Reza, 000). VI) EOTD METHOD It is based on BCCH timing measurements from different cells at target in GSM (Drane et al., 1998). calculates difference between arrival times of signals from two different BSs called Observed Time Difference (OTD). This OTD is constant to two BSs and gives a hyperbola. This means that two pairs of BSs (at least three BSs) are required for defining unique position. Time difference of arrivals is given as (James et al., 1998)

3 Td i d R R c i, i Where d i, is the two BSs separation distance, R i, R are the distances of to i th and th BSs and c is the velocity of light. Resolution at which OTD measurements are reported is only 554m and the required synchronisation of the BSs is not guaranteed. In EOTD, as the transmission frames from BSs are not synchronized in GSM network, real time difference (RTD), that is the difference between transmitted times of BSs is needed. Geometric time difference (GTD), difference of OTD and RTD, also gives hyperbola for positioning (Fig. 7) (Mattsson, 001). OTD measurements are sent via S to LMU which performs the location calculations. LMU is at known distance from all BSs and calculates RTD. This method is less expensive than Disadvantage of this method is that EOTD software is required both at LMU and. If the propagation distance between BS 1 and is d 1 and propagation distance between BS and the is d, GTD is calculated as GTD 1 = (d - d 1 )/c, (3) where c is the light velocity. Response time (approximately 5s) depending on the network latency. VII) ASSISTED GPS (A-GPS) METHOD A-GPS technique uses both GPS satellites and mobile network infrastructure to provide a reliable positional accuracy (50m for indoor and 15m for outdoor) (Darnell and Wilczoch, 00). GPS receiver technology is matured enough to integrate with GSM technology for A-GPS method. This method requires both software and hardware modifications at and network. Basic idea is to establish a GPS reference network, whose receivers have clear view of sky and operate continuously. This reference network (location server) is connected to cellular network and continuously monitors GPS constellation status. Location server and handset interact each other for positional information exchange. It saves lot of time and battery life. On request from, data from GPS reference network is transmitted to GPS receiver chipset embedded in through cellular network (Fig. 8). This embedded receiver gets signals from available satellites and pseudorange measurements are given to position determination entity (PDE) of. The maor limitation of this technique is that up gradation of network in the form of GPS reference network, additional location determination units and modification of is necessary (Johan, 00). IX) MOBILE POSITIONING WITH RSS The range based model basically used the channel characteristics to find the propagation distance. The range free localization technique does not use the channel model for location calculation. Range based scheme is further classified according to measurements used for location estimation (time of arrival, angle of arrival and received signal strength indication). Positioning based on RSS are feasible because () control channel signals are present at for every 0.48s (Garg and Wilkes, 1999). Also, these methods do not need any expensive upgrades. Hence, RSS based methods are simple and economical for self-positioning of. The location is found from the RSS which gives the distance information. The distances from three BSs are used in the circular triangulation for location. Distance calculation using received signal strength (w) is obtained as (Patil et al., 005) p r kd 4 K is a constant which takes into account of carrier frequency and transmitted power, d is the distance between the th and the BS, =1,, and 3. is the path loss exponent. if K and are known, propagation distance can be estimated from Eq. 4. These methods suffer from several effects such as multipath, diffraction, weak signals and BS unavailability. Methods like Pattern recognition technique and Network measurement report (NMR) also similar disadvantages. IV. COMPARISON OF POSITIONING METHODS Each of the method has its own merits and demerits like greater accuracy but limited in coverage (Table 3). Every method may suffer due to multipath, diffraction and weak signal condition. EOTD and assisted-gps are reported to offer better accuracies so these methods are of the probable methods to LBS application. Hearability is a maor limiting factor for both network-based and handset-based methods and adversely affects the deployment of a location schemes (Ibnkahla, 005). Accuracy and response time are foremost performance measures in positioning technologies (Issac et al., 00). Accuracy of positioning methods can be evaluated to identify the suitable method for varying different environments and for different services. To measure positional accuracy Circular Error Probability (CEP), Geometric Dilution of Precision (GDOP), Mean Square Error (E) and Cramer-Rao Lower Bound (CRLB) are used (Liberti and Rappaport, 1999). V. CONCLUSIONS Mobile phones with positioning capabilities are rapidly spreading around the globe because of its commercial and safety applications. Network based methods offer good accuracy in dense urban environment but fails to offer same in rural environment. On the other hand, hand set based methods provide privacy (user controlled) to the customer. Among the methods discussed, hyperbolic and A-GPS methods have real time applications because of better accuracy. In TDMA systems, hyperbolic techniques exploit the timing information while A-GPS takes the advantage of hand to hand cooperation of GSM and GPS. RSS based methods consider all induced effects of mobile environment, making them reliable. Accuracy in turn depends on the location technologies used and on the network topology.

4 REFERENCES [1] Kolodzie, K.W. and Helm, J., Location positioning systems LBS applications and services, CRC press, Taylor&Francis Group, Boca Raton, FL, 006 [] Rappaport T.S., Reed J.H. and Woerner B.D., Position Location Using Wireless Communications on Highways of the Future, IEEE Communications Magazine, Vol.34, No.10, pp , October 1996). [3] Jagoe, A., Mobile location services the definitive guide, Pearson Education, Inc. NJ, 003 [4] Sarikaya, B., Geographic Location in the Internet, Springer publications, 00 [5] James, M., Steen, A. P., Julian, J. B., and Karen, D.M., Providing Universal Location Services Using a Wireless E911 Location Network, IEEE Communication Magazine, pp , April 1998 [6] Helm J., Convergence Architecture between GPS (Global Positioning Systems) and Positioning in Mobile Networks: Creating Location Services for Wireless web, Wiley Dreamtech India (P) Ltd., New Delhi, 003 [7] Reza, I.R., Data Fusion For Improved TOA/TDOA Position Determination in Wireless Systems, Master s thesis in Electrical Engineering, Virginia Polytechnic Institute and State University, Blacksburg, July 000 [8] Mattson, O., Positioning of a cellular phone using the SIM, Master s Thesis in Computer Science Royal Institute of Technology, Kungliga, Stockholm, 001 [9] Al-Jazzar, S., Caffery, J. and You, H.R., A Scattering Model Based Approach to NLOS Mitigation in TOA Location Systems, IEEE, 00 [10] Tayal, M., Location services in the GSM and UMTS Networks, IEEE International Conference on Personal Wireless Communications (ICPWC-005), January 3-5, 005 [11] Ibnkahla, M., Signal Processing for Mobile Communications Hand book, CRC Press LLC, 005 [1] Drane, C., Macnaughtan, M. and Scott, C., Positioning GSM Telephones, IEEE Communications Magazine, Vol. 36, No. 4, pp , Apr [13] Johan, H., Convergence architecture between GPS and Positioning in Mobile networks: Creating Location Services for the Wireless Web, John Wiley & Sons Inc, USA, 00 [14] Garg, V. K. and Wilkes, J. E., Principles & Applications of GSM, Prentice Hall PTR, 1999 [15] Darnell, C. and Wilczoch, C., Real Time Positioning; Construction and implementation of a GPS-Communicator, Master s thesis in Control and Communication, Institution for Systemteknik, Linkoping, June 00 [16] Patil, M.M., Shasha, U., Desai, U.B. and Merchant, S.N., Localization in Wireless Sensor Networks using Three Masters, IEEE International Conference on Personal Wireless Communications (ICPWC-005), January 3-5, 005 [17] Issac, K., Adusei, K. and Klaus, J., Mobile positioning technologies in cellular Networks: An evaluation of their performances Metrics, Proc. of MILCOM, Vol., pp139-44, 00 [18] Liberti, J.C., Jr. and Rappaport, T.S., Smart Antennas for Wireless Communications: IS-95 and Third Generation CDMA Applications, Prentice Hall Communications Engineering and Emerging Technologies Series, NJ, 1999 Positioning E Mobility management Lawful Interception LBS Fig. 1 Use of position for various services R BS BS Fig. Cell of origin method Fig. 3 Cell of origin with timing advance method

5 3 Base Station 1 Base Station Base Station 1 Base Station 3 Base Station 3 Base Station 1 Figure 4 Concept of ToA method Figure 5 Concept of AoA method d LMU 3 LMU d LMU t1 (X 1, Y 1, Z 1 ) Base Station 1 t3 (X 3, Y 3, Z 3 ) Base Station 3 Fig. 6 Time difference of arrival method t (x,y,z) Mobile Station (X, Y, Z ) Base Station (x 3,y 3 ) Base Station 3 Hyperbola for GTD1 d LMU 1 d 3 (x,y) d 1 (x 1,y 1 ) Base Station 1 d (x,y ) Base Station Hyperbola for GTD GPS Satellites Fig. 7 Enhanced observed time difference method Reference network BS Mobile network Fig. 8 Concept of A-GPS method

6 Table 1 Accuracy requirement for E-911 service (FCC Docket, 1996) Type of positioning solutions 67% of calls with accuracy 95% of calls with accuracy Terminal-based 50 m 150 m Network-based 100 m 300 m Table Positioning methods implemented in maor cellular systems System GSM IS-136 IS-95 UMTS Location technology Cell identity + time advance and ToA, EOTD A-GPS Cell identity and AoA, A-GPS Cell identity and Time Difference Of Arrival (TDOA) Advanced Forward Link Trilateration Cell identity + round trip time and AoA EOTD, time difference of arrival (TDOA) anda-gps Method CoO Execution Point Handset/ Network Table 3 Comparison of positioning methods (Mattsson, 001) Advantages High availability Only need contact with BSs No modifications needed Limitations Low accuracy CoO with Handset/ Not all handsets support TA Better accuracy than CoO TA Network Low accuracy ToA Network No changes needed in handsets Need of LMU at each BS TDoA Network No changes needed in handsets Expensive modifications, one LMU at each BS EOTD Hybrid Good accuracy Needs changes both at network and handset AoA Network Need only two BSs Complex antennas required at BS NMR GPS with Handset Handset High availability No changes needed Simple High accuracy No changes in network Not all handsets support Provides calculated guesses Less accurate Require LOS Expensive handset modification A-GPS Hybrid High accuracy Expensive up grades needed Pattern recognition Handset Simple and cost effective Poor accuracy