Antenna Pointing Algorithms for Non-Geostationary Satellite based UMTS Systems
A. Papathanassiou and P. T. Mathiopoulos’
Institute for Space Applications and Remote Sensing National Observatory of Athens, Athens, Greece Abstract
-
LEO/MEO satellite systems are playingan increasing important role in future worldwide mobile telecommunications. One of the difficult technical issues encountered in LEO/MEOs is the high speed of these satellites relative to the mobile terminals. Thus, accurate and efficient methods of tracking the satellites are required. In this paper, we propose, analyse and evaluate the performance of efficient antenna pointing algorithms allowing the mobile terminals to track their own movements as well as those of the LEO/MEO satellites.
I. INTRODUCTION
Terrestrial cellular networks, such as the GSM, provide mobile communications services within certain limited geographical regions. In contrast, satellite communications play their role where the terrestrial cellular networks are either not competitive (e.g. due to low traffic density), not applicable (e.g. for maritime and aeronautical services), or are not developed at all. In order to supplement such terrestrial systems, low/medium earth orbit (LEO/MEO) satellite systems have been designed and are being implemented. An integrated satellite/terrestrial mobile system will greatly extend the geographical range of mobile telecommunications. In the future, the mobile telecommunications systems (e.g. UMTS) with a fully integrated satellite component will provide seamless global personal communications [ 11.
As compared to geostationary earth orbit (GEO) satellite systems, the advantages of the LEOME 0 include low transmit power and relatively short transmission delay which permits direct communications between hand-held mobile terminals and the satellites [2]. However, there are
several technical issues which need to be addressed in order to make full use of the above features and guarantee certain quality of service. One of the technical difficulties with L E O M E 0 is the relative movement between the satellites and mobile terminals due to both high speed of the satellites and the unknown orientation change of the mobile terminals. Both movements could result in loss of the tracking of the satellites unless an accurate and efficient tracking subsystem is deployed. In the past, the issue of satellite tracking has been addressed in [4] for a GEO satellite system. Some very limited and rather add-hoc performance evaluation results have been also published in the same reference.
In this paper, we take a closer and systematic look at the problem of satellite tracking. More specifically, we propose and evaluate the performance of efficient antenna pointing algorithms allowing mobile terminals to effectively track the LEO/MEO satellites. The organization of the paper is as follows.
In Section 11, the antenna pointing algorithms for vehicle-based mobile terminals are described.
Various performance evaluation results of the proposed algorithm are presented in Section 111.
Finally, in Section IV the conclusions of the paper are given.
11.
ANTENNA POINTING ALGORITHMS Instead of a sophisticated gyroscopic system, the proposed antenna pointing system employs a simple control mechanism which maintains the orientation of a mobile terminal’s antenna in alignment with the moving satellites.We assume that the satellite transmits an undistorted pibt signal which is detected by the mobile antenna at regular time intervals. The measured power of the pilot signal corresponds to a certain deviation angle of the mobile
’
Also with the Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, Canada.0-7803-6253-5/00/$10.00 02000 IEEE. 849
antenna relative to the beam of the satellite. The value of the deviation angle can be calculated in accordance with the power-angle relationship of a given antenna radiation pattern. The maximum power is measured when the antenna is in alignment with the beam of the satellite. The antenna’s pointing error is thus proportional to the power difference between the measured power and the maximum power. The control system uses the information of the pointing errors to maneuver the antenna and resume the antenna’s orientation of the maximum power.
1. Mobility Model for Trucking
As illustrated in Fig. 1, we consider a vehicle- based mobile terminal model. Both the satellite movement and the movement of the mobile terminal are taken into account. On the one hand, the antenna orientation of the vehicle varies at a certain speed in both azimuth (cp) and elevation
(e)
directions due to unknown turninglpitching of the vehicle. We have adopted a worst case mobility scenario for the vehicle, as follows. The orientation change of the vehicle antenna in azimuth and elevation direction is assumed to be AcplA~ = 15 degreelsecond and AOIAo= 10 degreelsec, respectively.
On the other hand, the movement of the LEO/MEO satellites is deterministic. For instance, the satellites in the Iridium system travel around the earth in circular orbits. The orbital period is approximately 100 minutes [3]. The angular speed of the satellite is thus 0.06 degreelsec which is much smaller than that of the mobile vehicle antenna.
I I
Fig. 1 Vehicle-based mobile mobility model.
Y e s I
I I
4
Calculate Ag, AQ
1
I 1
1
Fig. 2 APA Flowchart.
2. The Antenna Trucking Algorithm
One of the technical issues in the antenna tracking is to decide the tracking direction. We employ two tracking methods which we refer to as
“straight trucking ” and “dithering trucking ”, to determine the tracking direction in our algorithm.
Straight Tracking: In this method, the system needs to measure two consecutive pilot signals, Piv , P q , and P i e , P2e in azimuth and elevation direction respectively, in order to decide on which side of the satellite beam the antenna stays. After the measurement of P i v and P i e , the antenna is moved by small angles, 6cpand 68, and then P2(pand P2e are measured. If P2+, > P l y , i.e. the movement of the antenna by 6 9 increases the received power, the antenna should turn, by the amount of Acp oc Pmax - PQ, in the same direction as 6 ~ . Otherwise, if Pq , < P 1
v,
the antenna should turnin the opposite direction as 6 q . The same process applies to the antenna motion in the elevation direction. The straight tracking process is represented on the left part of the flowchart in Fig. 2.
The pilot signals transmitted by the satellite may be continuous or in a TDMA signaling format. W e assume a TD MA signaling format as shown in Fig. 3. The mobile terminal receives data signals during
the
period ofT.
During the period of AT The APA detects twice the pilot signals in each direction. These two measurements are necessary for the system to determine the tracking direction of the mobile antenna. The dimensionless parameterK = AdT, i.e. the ratio of the pilot signal period to the data signal period, reflects the efficiency of the system. Either increasing T or reducing AT will result in a smaller K, and thus a larger efficiency. However, the value of T-cannot be too big due to the increase of the antenna pointing error with T. In the straight tracking algorithm, the mobile antenna needs to move a small angle (typically 3 degrees) angle within Ar. If AT is too small, the mechanically moving parts of the system may break due to a very fast movement. This problem is avoided by the application of the following dithering mechanism.
-4
+
AT T Tim
AT: pilot signal period.
T : dataperiod.
K =ATIT.
Pi(A, E) = P w P,o and P, (A, E) =P,,,, P , ~ . Fig. 3 The TDMA signaling format.
Dithering Tracking: Dithering is the mechanical tuming or rocking of the antenna by a small angle and at a very low speed [4]. Fig. 4 shows that a mobile terminal antenna spins at the rate of AydAt = 2 Hz. The spanning angle
w
is about two degrees.The distance between the antenna and the satellite beam changes continuously during the dithering. As a result, the average power levels of the pilot signals vary in different dithering phases, By comparing the average power levers in different phases, the antenna’s position relative to the satellite beam can
thus be determined. The dithering tracking process is illustrated on the right part of the flowchart in Fig. 2.
In contrast to the straight tracking, the dithering tracking does not require a fast movement
of
the antenna. The dithering increases the searching area of the antenna. Therefore, it increases the probability of tracking the satellite beam. However, it may also increqse the background pointing error due to the same reason./
satelliteFig. 4 The mechanism of dithering.
111. PERFORMANCE RESULTS AND
We have evaluated the performance of proposed APA system for various mobility conditions of the vehicle and with respect to the antenna radiation pattems, the gain of the control system and the TDMA signaling period for both the straight and dithering tracking. The results of the simulation are represented in terms of the pointing error of the mobile antenna. The pointing error is defined as the difference of the relative power between the measured and maximal power in decibel. The parameters used in the simulation are:
Azimuth tuming speed = 15 degreekecond;
Elevation tuming speed = 10 degreehecond;
Satellite angular speed = 0.06 degreekecond;
Power threshold (min. detectable power) = 0.5 dB.
DISCUSSION
1. Performance Evaluation of Straight Tracking Antenna Radiation Patterns: The antenna radiation pattems is found to be one of the most significant parameters which affects the performance
851
of the APA. The half-power-half-width (HPHW) of the antenna radiation pattern
RighWp RightRlp LeWDow LeWup
I 0 5 0 i - 0 5 -I
B g
0 1 2 3 4 5 6 7 8
Time, seconds
Right/Up RighUUp LeWDown LeWup I
0 5
0 0
i - 0 5 m
a -1.5
-'I
2 1
0 1 2 3 4 5 6 7
Time. seconds 8
Fig. 5 Tracking results of Antenna #1 & #2.
Antenna # I 1.5
I t *
0 5
'?
04 s I
Y
7 0 1 2 3 4 5 6 7 8
Time, s a n d s
Fig. 6 Tracking results of Antenna #1 with different gains of the control system.
of h e n n a #1 is 50 whereas for Antenna #2 is 150.
*' Figs. 5(a) and 5(b) present the tracking performance results of Antenna #1 and Antenna #2, respectively. The large antenna pointing errors shown in Fig. 5(a) correspond to the variation of the vehicle angular speed when the mobile terminal changes its turning direction. The numbers in the unit of degrees are the values of angles by which the antenna deviates from the satellite beam. The results demonstrate that for the same amount
of
the angle deviation, the pointing error of the
narrow Antenna #1 is an order of magnitude larger than that of the wide Antenna #2.
Gain of Control System: We try to find the optimum value of the gain of the control system. By observing the pointing error, the optimum value of the gain is found to be around 0.9. The simulation results in Fig. 6 , for instance, show that a 50%
decrease of the gain from 0.9 to 0.5 can result in about 100% increase of the antenna pointing error.
The APA
Antenna#l I
4 ' 0 I 2 4 5 6 7 8I
Time. scconds
Fig. 7 Tracking results of Antenna #1 with different data period T.
RighWp Right/Up LeRiDow Leluup
I S
+
I T
I 1
* O I 2 3 4 5 6 7 8
Time. seconds
Fig. 8 Tracking results of Antenna #1 with .and without dithering.
system will not be able to resume the position of the mobile antenna close enough to the satellite beam if the gain is too small. On the other hand, if the gain is too big, the antenna pointing error will also be big due to large overshoot.
TDMA Signaling Period The criteria of choosing the value of the parameter K (= AdT), depends on the maximum allowable antenna pointing error. The antenna may lose the tracking of the satellite if the data period is too big. The effect of the data period on the APA performance is presented in Fig. 7. The tracking is carried out for two different data periods, T = 0.1s and 0.15s, while the pilot signal period AT is
kept to be 1 ms in the both cases. It can be seen from the results that the antenna pointing error is increased when the data period increases. The effect is more significant at the time when the mobile terminal alters its turning direction. ’
2. Performance Evaluation of Dithering Tracking In contrast to the straight tracking in which the pilot signal period AT cannot be too small, the dithering tracking can receive the pilot signals within very short pilot signal period. Fig. 8 compares the tracking results between the straight tracking and the dithering tracking. Since the dithering introduces a small deviation angle into the system, the pointing error in the dithering tracking is usually slightly larger than that in the straight tracking. The dithering, on the other hand, increases the probability of searching the satellite beam, the antenna pointing error in the dithering tracking should be smaller than that in the straight tracking when the antenna has a big angle deviation from the satellite beam. These features can be seen in Fig. 8.
Iv.
CONCLUSIONSThe computer simulation results indicate that the APA system is feasible for a given worst-case I mobility scenario and can provide an accurate satellite tracking. Using the A P S system the satellite can be tracked by a vehicle-based mobile terminal whose turning direction change is unknown.
v.
ACKNOWLEGMENETSThis work has been supported by GSRT and by the European Commission under the PENED 99 Programme (99ED443).
REFERENCES
1. G. Losquadro and R. Sheriff, “Requirements of Multiregional Mobile Broadband Satellite Networks”, IEEE Personal Communications, April 1998.
2. G. Maral, J. Restrepo, E. Del Re, R. Fantacci and G. Giambene, “Performance Analysis for a Guaranteed Handover Service in an LEO Constellation with a ‘Satellite-Fixed Cell’
System”, IEEE Trans. Vehicular Tech. Vol.4 7, No.4, Nov. 1998.
3. M. Werner, A. Jahn, E. Lutz and A. Bottcher, “ Analysis of System Parameters for LEO/ICO- Satellite Communication Networks”, IEEE J.
Select. Areas Commun., vo1.13, no.2 pp.371-381, Feb. 1995
4. A. C. Densmore and V. Jamnejad, “A Satellite- Tracking K- and Ka- Band Mobile Vehicle Antenna System”, IEEE Trans. Veh. Technol., vo1.42, no.4, pp 502-513, Nov. 1993.
5. E. Del Re, R. Fantacci and G. Giambene,
“Handover Queuing Strategies with Dynamic and Fixed Channel Allocation Techniques in Low Earth Orbit Mobile Satellite Systems”, IEEE Trans. Commun. Vol.47, No. I , Jan. 1999.
853
