• 沒有找到結果。

Satellite TDMA Controller Design for Timing Acquisition and Timing Synchronization

N/A
N/A
Protected

Academic year: 2022

Share "Satellite TDMA Controller Design for Timing Acquisition and Timing Synchronization "

Copied!
5
0
0

加載中.... (立即查看全文)

全文

(1)

Satellite TDMA Controller Design for Timing Acquisition and Timing Synchronization

Inone Joo, Jae-Hoon Kim

Satellite Ground Control System Research Team Electronics and Telecommunications Research Institute

inone@,etri.re.kr, ihkim@etri.re.kr

Abstract- Satellite TDMA Controller(STC) Design for timing acquisition and timing synchronization in satellite TDMA system is presented, which utilizes the modified open-loop control method in the acquisition process and the closed-loop control method in the synchronization process. Since the modified open- loop control method affords a wide timing margin, the timing acquisition can be securely established. Besides, the closed-loop control method can maintain stable timing synchronization. The feasibility of STC Design is confirmed through experiments.

Keywords- satellite; TDMA; controller; synchronization;

acquisition

1. INTRODUCTION

Earth stations access a given satellite transponder sequentially in time so that each station uses the full power and bandwidth of the transponder during assigned time intervals.

However, a geostationary satellite is subjected to small perturbations caused by the moon and the sun and moves about a region centered at its nominal position. Therefore, these undoubtedly cause errors in the burst positions of the satellite transponder. Accordingly, the transmission timing of the bursts is carefully synchronized so that all the bursts arriving at the satellite transponder from a community of earth stations in the network are closely spaced in time hut do not overlap. Timing acquisition and timing synchronization have been an essential topic of satellite TDMA system and deeply investigated in the past [ 1-51, The closed-loop control method can maintain stable timing synchronization because this method can reduce the guard time between bursts[4]. However, a transmit burst timing(TBT) in the closed-loop control method cannot he obtained, so the burst overlap can occur in the acquisition process.

This paper presents Satellite TDMA Controller(STC) Design for timing acquisition and timing synchronization. STC Design basically utilizes the closed control method for timing synchronization. Moreover, in the acquisition process, the modified open-loop control method is utilized in order to securely establish timing acquisition by transmitting the short burst. STC is adopted in HiSTARS(High-speed Satellite TDMA gRound System) project which provides 45/155Mbps transmission rate over the Ka-Band KOREASAT-3. The feasibility of STC is confirmed through the experiment of HiSTARS.

11. PRELIMINARIES

In order to establish timing acquisition in the modified open-loop control method, the distance, dN from the N,h traffic station to the satellite is needed, which is calculated as follows.

d , =lOS-OMI (1)

where

E

is the vector from the center of the Earth to the is the vector from the center of the Earth

- -

satellite(S) and

to N,h traffic station(hi) in ( X ,

Y,

Z ) coordinate system.

A. Satellite Position Vector

( E )

Satellite(9, reference station(R) and N,h traffic station(i\i) in ( X, K Z ) , ( X , Y : Z’ ) and ( x, y , z ) coordinate system is shown in Fig. 1.

(X,

E

2 ) coordinate system is obtained by rotation of

(e,)

in ( X , Y’, Z’ ) coordinate system and the transformation matrix, 8 I

(e,

) can he expressed by

cos@, sing, 0

S 2 (8, )=

[

- s t 8, CO:

:]

(2)

where 6, = reference station longitude ( 6, > 0 for west longitude and 6, < 0 for east longitude ).

Fig. I . S,

Y’

I R, N in ( X Y, Z ) , ( X, Y : 2’ ) and ( x, y , I ).

0-7803-8255-2/04/$20.@J Qu)(ll IEEE 1934

(2)

Using Eqs. (2), (31, (S), and (6), we can rewrite Eq. (4) as follows.

cos$, sine, sine, cos0, cos@, cosEcosA os=d, -sin#,sin8, cosb', -sinBLcos8, cosEsinA

- [

-cos$, 0

R,cosB, cos0,

+

-R,sinb',cosQ,

[

R,sinB,

]

sin 8,

][

sinE

:]

Y (7)

Pig. 2. dislancc( &), elevation angle( E ), azimuth angle ( A ) in thc ( x, ,y, r )

coordinate system. B. Trafic Station Position Vector

(ohi)

The N,h traffic station position vector in the geocentric ( X ' , Y : Z' ) coordinate system is obtained by rotation of

(90" -8)) in (x, y , z ) coordinate system and the transformation matrix,

s,

(900 - 8,

)

can be expressed by

coordinate, ?%is obtained such that.

z

= %;(oL,,").

2

(8)

sins, 0 cose, -cos& 0 sin 6,

where SI = reference station latitude

( e ,

> 0 for north latitude and

el

< 0 for north latitude ).

From a relationship between three coordinates described above, the satellite position vector in the geocentric coordinate,

?%

is obtained from Eqs. (2) and (3) such that

(3)

?%

= %:(B,).['R, (90' -8,). RS+ OR'

- -1

(4)

where

k?

is the vector from the position of the reference station to that of the satellite in the ( x, y , z ) coordinate system and

OR

is the vector from the center of the Earth to the position of the reference station in the ( X: Y : Z ' ) coordinate system.

The satellite position vector in the ( x, y , z ) coordinates, is obtained from Fig. 2 such that

cos E cos A R S = ym = d , cosE sinA

- [ 1111 [

sin E

:]

where dR , E, and A denote the distance between the reference station and the satellite, the elevation angle, and the azimuth angle obtained from antenna encoder.

OR

means the position vector of the reference station in the ( X : Y'.

Z'

) coordinate system and is obtained from Fig. 1 such that

where Re =earth radius = 6378.155[Km]

where is the vector m the ( X : Y : 2' ) coordinate system from the center of the Earth lo the N,h traffic station.

The transformation matrix, s2(oL

,,)

is obtained similarly to Eq. (2) such that

-sineLr COSS, ,,,, (9)

where

e,,,,,

= N,* traffic station longitude ( .9L,N > 0 for west longitude and

e,,,,,

< 0 for east longitude ).

The means the geographical location vector of the N,h traffic station in the ( X : Y : Z' ) coordinate system and is obtained similarly to Eq. (6) such that

where = N,h traffic station latitude ( > 0 for north latitude and

e,,,

< 0 for south latitude ).

Using Eqs. (9) and (IO), we can rewrite Eq. (8) as follows

-

ON = R e -sin@,,, cos6,,N

I

c0soL.N c0soi.N

1

(11)

111. SATELLITE TDMA CONTROLLER(STC) DESIGN A . Modijkd Open-Loop Control Method for Timing

Acquisition

As the reference station controls and monitors all traffic station and satellite, it is apprised of their positions.

Accordingly, the distance, dN from the N,h traffic station to the

(3)

satellite is calculated using Eqs. (I), (7) and (1 I). The transmit frame delay, DN is needed to calculate the transmit burst timing(TBT) in the traftic station, which is calculated in the reference station. D,v means the time delay between the receive frame timing(RFT) and the transmit frame timing(TF0, which is calculated as follows.

D, = M T, - 2 x d,v I C (12) where &the smallest integer chosen such that D, 2 0 for all d,,, T,= TDMA frame period, and c=velocity of light

Then, the reference station supplies, via the transmit timing channel of its reference burst a transmit frame delay, DnJ to the N,h traffic station as shown in Fig. 3. Carrier and clock recovery enable all ground station demodulators to recover the carrier phase and regenerate the bit and symbol timing clock. And, unique word provides the generation of the receive frame timing(RFT).

Fig 3. RB s1mctllre including a transmit frame delay. D,v

The Nlh traffic station receives the reference hurst(RB) including D, and decodes it. Since DA, means the transmit frame timing in the N,h traffic station(TFTn,), a traffic burst transmitted at the moment of the transmit frame delay(D,) after the receive frame timing(RFT) will he coincident with the reference burst at the satellite. Accordingly, the transmit burst timing of the N,h traffic burst, TBTN is obtained such that

TBT, = RFT

+

D,

+

B, (13) where EN is the fixed offset for transmitting the N,h traffic burst according to the burst time plan(BTP).

Eq. (13) shows that the accuracy of DN is proportional to the accuracy of TFT,. The DN can be affected by error sources such as distance measurement, calibration, and computation error, since the DN is obtained by the ranging stations[4]. An excess of the error in DN over the guard time occurs the burst overlap.

In order to be more stable timing acquisition, the modified open-loop control method is utilized. The N,h traffic station transmits short burst(SB) including carrier&clock recovery and unique word, instead of a traffic burst(TB) as shown in Fig. 4.

mB(Traffl-)

Fig. 4. 78 and shod burst(SB) StNCNre

Transmitting the SB instead of the TB increases the timing margin and reduces the timing error, which can establish more

stable timing acquisition. The timing margin, (TB, - SB, ) is divided by 2 considering neighbor bursts of the TB,, Accordingly, the transmit short burst timing of the TBT(SB), is obtained such that

B. Closed-Loop Control Method for Timing Synchronization Once the traffic station has successfully positioned the short burst within the assigned traffic burst time slot, it can start timing synchronization to maintain the traffic burst position within the guard time. The TBT in the N,h traffic station is established directly from the receive frame timing(RF7) by observation of the traffic burst position error, E relative to the fixed offset, B, for the N,h traffic burst according to the burst time plan(BTP). After the roundtrip propagation delay &om transmitting short burst(SB,), the N,h traffic burst receives its own SBN and measures the receive short burst timing, RBT(SB),. Then, the traffic burst position error,€ is obtained such that

E = RBT(SB), - B, ( 1 9 The N,h traffic station compensates this error and determines the next transmit burst timing, and it transmits the TB instead of the SB. The burst timing for transmitting TB, TBT, is obtained such that

TBT, = TBT(SB), - E (16) The closed-loop method is continuously repeated for timing synchronization. The closed-loop control method can reduce the guard time between bursts and maintain stable timing synchronization since it compensate a timing error by measuring a receive burst timing(RB7).

Fig. 5 and Fig. 6 show the process and the timing graph of the control method used in Satellite TDMA Controller(STC).

As shown in Fig. 6, the modified open-loop control method is utilized for more stable timing acquisition, and the closed-loop control method is utilized for timing synchronization.

C. STC H/W Design

STC WW is designed to he used commonly in both the reference station and the traffic station, which is shown in Fig.

7. STC HM, was implemented as a daughter board, a PMC(PC1 Mezzanine Card) type, which communicated with a canier board, a processor board by PCI bus. STC function for timing acquisition and timing synchronization was implemented in FPGA(Field Programmable Gate Array), and the synchronization signal of STC was generated in CPLD(Comp1ex Programmable Logic Device). Clock in the reference station was generated by timing equipment which provided a high accurate clock by receiving GPS signal. Clock in the traffic station was selected in SoMHz with 1 ppm clock accuracy, which occurred a symbol error per 2011s. 50MHz clock in the traffic station is enough to he used in HiSTARS since RFT resets the counter logic of TBT calculation per TDMA frame period(2.7ms). ROM, PLL, and PCI Bridge were used as suh-components for STC operation.

(4)

Fig. 5 Pmces! ; of the control meUiod used in STC

Fig. 7. Hardware srmchrre of STC

IV. EXPERIMENTAL RESULTS

STC was adopted in HiSTARS(High-speed Satellite TDMA gRound System) project which provides 451155Mbps transmission rate over the Ka-Band KOREASAT-3.

451155Mbps TDMA frame structure used in HiSTARS project is shown in TABLE 1.

TABLE. I . 451155Mbps TDMA frame srmchm in HiSTARS

Fig. 8 shows the status of timing synchronization in 45Mbps transmission test of HiSTARS, while STC provides timing synchronization after timing acquisition. Fig. 8(a) shows the receive frame timing(RFZ) signal, an interface signal between STC and satellite MODEM. Fig. 8(b) is analog signal of 10 traffic bursts over the Ka-Band KOREASAT-3, which is measured at a input of AID converter in satellite MODEM.

Fig. 6 . Timing graph of the control method uscd in STC

1937

(5)

.. . . . . ... .. . .. .. . . . .. .. .. 4 . . .. . .

-

. .. ~ . . .. .... . . .. .

Fig. 8. Timing synchronization in 45Mbps HiSTARS

Fig. 9. Timing synchronization in 155Mbps HiSTARS

Fig. 9 shows the status of timing synchronization in 155Mbps transmission test of HiSTARS. Fig. 9(a) shows the receive frame timing(RFQ signal, and Fig. 9(b) shows a

unique word synchronization signal of 16 traffic bursts received from satellite MODEM, which is generated at the moment of unique word synchronization.

As shown in Fig3 and Fig.9, in 4Si155Mbps transmission rate over the Ka-Band KOREASAT-3, stable timing synchronization without the horst overlap demonstrates the feasibility of the closed-loop control method. But, the measurement of timing acquisition couldn’t be obtained because receive short burst timing, RBT(SB),” was disappeared in an instant and the timing synchronization process was converted. However, without the failure of timing synchronization process, establishing timing acquisition using the modified open-loop control method was comprehended.

V. CONCLUDNG REMARKS

Satellite TDMA Controller(STC) Design for timing acquisition and timing synchronization was presented in this paper, which utilizes the modified open-loop control method and the closed-loop control method. The modified open-loop control method establishes stable timing acquisition since it affords a wide timing margin by transmitting the short burst, and the closed control method maintains stable timing synchronization by compensating a timing error measured.

STC was adopted in HiSTARS(High-speed Satellite TDMA gRound System) project which provides 451155Mt~ps transmission rate over the Ka-Band KOREASAT-3. The feasibility of STC has been confirmed through experiments.

REFERENCES

[I]

[Z]

Tri T. Ha, “Digital Satellitc Communications”, McGraw -Hill, 1990.

Tri T. Ha, “Real Time Satcllitc Position Determination far TDMA Systems”, IEEE Trans on Aerospace and Elecuonic System Vol. AES- ZI,No.6,Novembcr751-156, 1985.

Campanclla S. J and Hodson K, “Open-loop TDMA Frame Acquisition and Synchronization”, Comsat Technical Rcview, Fall 341-384, 1979.

Lei T. R, “Synchronization Accuracy and Error Analysis of Open-loop TDMA Systems”, Comsat Tcchnical Review, Spring 59-76, 1981.

A m m i , M. and Garofalo, G. “Access and synchronization schemes in thcESAOBPsysfem”,GLOBECOM,206-211 vol.1, 1991

[3]

[4]

[5]

1938

參考文獻

相關文件

*In general, if a node is involved in the computation required by one or more periodic timing constraints, the process assigned to the node has a period equal to GCD of periods

(10), the output of IFFT appeared as a form of channel impulse response with the reference of integral frequency offset and the maximum IFFT output of the correlation between the

Conventional DFT-based channel estimations suffer from the timing offset and incorrect channel impulse response, and this results in means square error (MSE) floor of

The engine operation conditions of injection timing, injection pressure, fuel temperature, and amount of injection will be varied to evaluate the engine performances. The

The empirical results suggest that, whenever in the rising period, the dumping period or the whole period, there are very few funds that are equipped with the volatility

n Media Gateway Control Protocol Architecture and Requirements.

Biases in Pricing Continuously Monitored Options with Monte Carlo (continued).. • If all of the sampled prices are below the barrier, this sample path pays max(S(t n ) −

This paper aims to study three questions (1) whether there is interaction between stock selection and timing, (2) to explore the performance of &#34;timing and stock