KR100650507B1 - Apparatus for symbol timming synchronixation of atsc digital tv system and method thereof - Google Patents

Abstract

An apparatus and a method for synchronizing the symbol timing of an ATSC(Advanced Television Systems Committee) D(Digital)TV system are provided to estimate the sampling offset by means of the mean of a jitter. A jitter mean calculator(80) calculates the mean of output values of a loop filter(60) for the constant time. An offset estimator(90) estimates an offset from the output of the jitter mean calculator. A bandwidth attenuator(100) sets an attenuation coefficient of the loop filter on the basis of the output of the jitter mean calculator in order to reduce an amount of jitter and reduces the bandwidth of the loop filter. At the initial operation, the bandwidth of the loop filter is largely set.

Description

Apparatus and Method for Synchronizing the Symbol Timing of an ATSC DST System {APPARATUS FOR SYMBOL TIMMING SYNCHRONIXATION OF ATSC DIGITAL TV SYSTEM AND METHOD THEREOF}

1 is a block diagram of a Gardner type symbol timing synchronization loop in a typical ATSC DTV system.

2 is a block diagram of a conventional Gardner sampling clock offset estimator.

3 is a configuration diagram of a symbol timing synchronization device of an ATSC DTV system according to the present invention;

4 is an operational flowchart of the present invention.

5 is a table showing a mobile receiving channel model for the experiment of the present invention.

6A-9A and 6B-9B illustrate the conventional Gardner method and the jitter amount and convergence time according to the present invention in a Brazilian A-D channel.

10A and 10B illustrate the conventional Gardner method and the amount of jitter and convergence time according to the present invention in Rayleigh channels.

11 is a table showing the convergence time according to the present invention and conventional channel models.

<Explanation of symbols for main parts of the drawings>

10: ADC 20: matched filter

30: pilot extraction filter 40: down sampling unit

50: sampling clock offset estimator 60: loop filter

70: VCO 80: Jitter Average Calculator

90: offset estimator 100: bandwidth attenuator

The present invention relates to an ATSC DTV system, and more particularly, to a method and apparatus for synchronizing a symbol timing in an ATSC DTV system which can reduce the size of jitter while maintaining the convergence speed by adjusting the bandwidth of a loop filter in a Gardner type symbol timing synchronization. It is about.

In general, the digital communication system can accurately restore the signal of the transmitter only when the sampling moment at the transmitter and the sampling moment restored at the receiver are the same. Similarly, in the ATSC (Advanced Television Systems Commttee) method, a DTV broadcasting standard adopted in North America and Korea, when the transmitter samples the 8-VSB (Vestigial Sideband) signal and the receiver recovers it, the transmitter 8- The VSB signal can be recovered correctly.

ATSC DTV systems have high data rates for high quality and high sound quality. In addition, as a method for solving the indoor and outdoor fixed reception as well as mobile and mobile reception service is gradually being studied, the synchronization of the receiver has been actively conducted.

In the receiver, symbol timing synchronization refers to a process of compensating for a sampling clock frequency offset and a phase offset caused by an error and a phase error of a sampling clock frequency.

The symbol timing synchronization method introduced in the ATSC standard uses segment synchronization with +5, -5, -5, and +5 inserted repeatedly every 77.3 ms.

The timing error information can be obtained by passing the filter of 0, +, 0,-signal. To reduce the error of the segment sync, dozens of segment syncs are averaged. This method has the advantage that it does not affect the long echo, but it is affected by the short echo and the channel changed by one segment sync every 77.3 ms. There is a disadvantage in that it is difficult to keep up with the real motivation.

In addition, if the phase offset remains in the frequency phase locked loop (FPLL) for frequency synchronization, symbol timing synchronization cannot be achieved.

ATSC DTV generally uses the Gardner method in addition to the segment synchronization method to obtain symbol timing synchronization.

Segment synchronization has less performance degradation than the Gardner method in a noisy environment, but more severe than the Gardner method in a noisy environment.

This is because segment synchronization estimates an offset every 832 symbols, while the Gardner method estimates every symbol offset, so it is robust to noise and multiple paths, and thus is widely used for symbol timing synchronization.

FIG. 1 is a block diagram illustrating a symbol timing synchronization loop of a Gardner method in a typical ATSC DTV system. The received signal is doubled after passing through a matching filter 20 through an analog digital converter (ADC) 10. Sampled.

The signal sampled more than twice is removed by the pilot extraction filter 30, the offset is estimated by the sampling clock offset estimator 50, and passes through the loop filter 60 to take an average of the offsets. The symbol timing offset is corrected using an interpolation filter (not shown) after passing through the VCO 70.

FIG. 2 illustrates a Gardner-based sampling clock offset estimator. Referring to FIG. 2, a double oversampled input signal undergoes two time delays and uses a difference between signals as shown in Equation (1) below. Estimate the sampling clock offset.

(1)

(One)

는 추정 샘플링 오차이고,

는 두 배로 샘플링 된 수신 데이터이다.

Is the estimated sampling error,

Is the received data sampled twice.

The estimated value as in Equation (1) is obtained through the loop filter 60 and the VCO 70 to obtain the symbol timing synchronization.

In the case of BPSK / QPSK signals, these Gardner algorithms have intermediate values of two symbols close to zero, but for multilevel signals such as 8-VSB modulation used in ATSC DTV systems, the intermediate samples have various values. do.

와

의 중간값인

이 0이 아닌 값을 가질 수 있기 때문에 타이밍 지터가 증가하게 된다.Therefore, since the average value of the transmitted signal is zero, the average of the accumulated intermediate samples is zero, but

Wow

Is the median of

This can have non-zero values, resulting in increased timing jitter.

In other words, the Gardner method suitable for symbol timing synchronization has a small performance degradation due to multipath and a fast convergence time independent of carrier synchronization, but has a problem in that performance is degraded due to jitter, which is interference between magnetic signals.

The present invention has been made in view of the above, and an object of the present invention is to increase the bandwidth of an initial loop filter during operation in Gardner type symbol timing synchronization, and then converge by estimating the sampling offset using the average value of jitter and then reducing the bandwidth. The present invention provides a symbol timing synchronization apparatus and method of an ATSC DTV system that can improve performance by improving speed and reducing jitter after convergence.

A symbol timing synchronization device of an ATSC DTV system according to the present invention for achieving the object of the present invention, in the ATSC DTV system employing a Gardner type symbol timing synchronization loop including a sampling clock offset estimator, a loop filter and a VCO, Jitter average calculating means for calculating an average of the output values for a predetermined time of the loop filter; Offset estimation means for estimating an offset from the output of the jitter average calculating means; And bandwidth attenuation means for adaptively reducing the bandwidth of the loop filter by setting the attenuation coefficient of the loop filter to reduce jitter based on the output of the jitter averaging means.

A symbol timing synchronization method of an ATSC DTV system of the present invention for achieving the above object is a symbol timing synchronization method of an ATSC DTV system employing a Gardner type symbol timing synchronization loop including a sampling clock offset estimator, a loop filter, and a VCO. A first step of setting the bandwidth of the loop filter to a predetermined bandwidth, calculating the average value of jitter by averaging the output of the loop filter for a predetermined time, and then estimating an offset; Reducing the bandwidth of the loop filter based on the average value of the jitter; A third step of estimating the offset by averaging the output of the loop filter whose bandwidth is reduced by the second step and recalculating the average value of jitter; And a fourth step of repeatedly performing the second step or less until the symbol timing synchronization is performed.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the following examples are merely to illustrate the present invention is not limited to the contents of the present invention.

The influence of jitter in ATSC DTV systems is related to the loop filter bandwidth. Larger bandwidths result in larger jitter but faster convergence. On the other hand, if the bandwidth is small, the jitter is small but the convergence speed is slow.

Accordingly, the present invention seeks to reduce the amount of jitter while maintaining the convergence speed by appropriately adjusting the bandwidth of the loop filter.

3 is a block diagram of a symbol timing synchronization device of an ATSC DTV system according to the present invention. The jitter average calculator 80, an offset estimator 90, and a bandwidth attenuator 100 are shown in a general ATSC DTV system as illustrated in FIG. 1. Is further configured.

The jitter average calculator 80 calculates an average of the output values of the loop filter 60 for a predetermined time, and the offset estimator 90 estimates an offset from the output of the jitter average calculator 80.

The bandwidth attenuator 100 reduces the bandwidth of the loop filter 60 by setting the attenuation coefficient of the loop filter 60 based on the output of the jitter average calculator 80 to reduce the jitter amount.

The present invention configured as described above will be described with the flowchart of FIG.

When the bandwidth is large, the bandwidth of the operation initial loop filter 60 is set to be large by using the feature of fast convergence (S10), and the jitter average calculator 80 averages the output of the loop filter 60 for a predetermined time to jitter. Calculate the average value of (S20).

The initial bandwidth is set to approximately 1000 ppm. In general, an error of one millionth of a sampling clock offset is referred to as ppm. For example, 1 ppm means an error of 100 Hz in a 100 MHz bandwidth.

A bandwidth of 1000 ppm of the initial loop filter 60 means that an error of 1000 ppm or less of the sampling clock offset may be estimated in proportion to the bandwidth. In this case, a large error can be estimated and a fast convergence speed, but large jitter occurs, resulting in performance degradation. In order to solve this problem, the present invention has adaptively made the bandwidth of the fixed loop filter.

The predetermined time used in the jitter average calculator 80 calculates the average of jitter during 1000 symbol intervals, that is, 0.1 ms, in the simulation. The time interval is more effective at 1000 symbol intervals when simulated with both 1000 and 500 symbols.

If the channel is good, the offset can be estimated even at 500 symbol intervals and shows good performance, but in a harsh environment, longer time is required. Therefore, intervals of 1000 symbols are appropriate to include both Brazilian channels C and D.

Then, the offset estimator 90 estimates an initial offset value using the average value of jitter calculated by the jitter average calculator 80, and then, in the bandwidth attenuator 100, the bandwidth attenuator 100 may reduce the jitter average calculator 80. The bandwidth of the loop filter 60 is reduced by setting the attenuation coefficient of the loop filter 60 based on the output (S30). Here, the attenuation coefficient is set to a value less than 1 to reduce the bandwidth.

As the bandwidth of the loop filter 60 is reduced, the size of the jitter becomes smaller, and accordingly, the size of the jitter is calculated again by the jitter average calculator 80 (S40).

Using the average value, the offset estimator 90 estimates the offset value, and then the bandwidth attenuator 100 resets the attenuation coefficient of the loop filter 60 to reduce the bandwidth of the loop filter 60 and to reduce the bandwidth. The filter coefficient of the loop filter 60 is changed, and this process is repeatedly performed until symbol timing synchronization is performed (S50).

The coefficient of the loop filter 60 is determined by the bandwidth as shown in the following equations (2)-(6).

은 루프 필터 대역폭,

는 루프 고유 주파수,

은 댐핑 상수를 나타내며,

을 결정하면

을 구할 수 있으며, 식(3), (4)와 같이 루프 필터(60)의 계수 K₁, K₂가 결정된다. In equation (2)

Loop filter bandwidth,

Is the loop natural frequency,

Represents the damping constant,

If you decide

The coefficients K ₁ and K ₂ of the loop filter 60 are determined as shown in equations (3) and (4).

은 루프 필터(60)의 동작 시간차, K_NCO와 K_pd는 각각 도시하지 않은 FPLL의 NCO(Numerically Controlled Oscillator)의 이득과 위상검출기의 이득을 나타낸다. 특정 관찰 시간 동안 대역폭

이 결정되면 다음 대역폭은 감쇄 계수

배 만큼 감소하게 되며,

시간 이후 대역폭은 식(6)과 같이 된다. 감쇄 계수

는 채널의 특징에 따라 적절히 정할 수 있다.K is the loop gain,

Denotes the operating time difference of the loop filter 60, K _NCO and K _pd , respectively, the gain of the NCO (Numerically Controlled Oscillator) of the FPLL and the gain of the phase detector. Bandwidth for Specific Observation Time

Once this is determined then the bandwidth is attenuation factor

Decreases by a factor

After time, the bandwidth becomes as shown in equation (6). Attenuation factor

Can be appropriately determined according to the characteristics of the channel.

The following looks at the experimental example of the present invention.

In general, the DTV uses a Brazilian channel model, where A and B are outdoor reception antenna environments, C is an outdoor and indoor intermediate point, D is an indoor antenna environment, and E is a single frequency network (SFN) environment.

For channel information on mobile reception, the channel profile shown in ITU-R M.1225 is used as shown in the table of FIG. 5. Here, channel B represents the worst condition in the same moving environment. The Doppler effect on the speed of the mobile station is simulated by using Jakes' model and the carrier frequency using 476MHz, UHF band channel 15.

6A to 9A and 6B to 9B are roads comparing the amount of jitter and convergence time for the conventional scheme and the scheme proposed according to the present invention for each channel, the conventional Gardner scheme in the Brazilian AD channel and the jitter according to the present invention. The amount and convergence time are shown.

200ppm을 주었다. 감쇄 계수는 브라질 채널에서 0.96, 레일리(Rayleigh) 채널에서 0.97을 사용하였다. 전반적으로 본 발명에 따른 방식의 지터 크기가 적으며 수렴속도 또한 빠른 것을 알 수 있다. In the conventional method, the bandwidth is 500 ppm, and in the proposed method, the initial bandwidth is 1000 ppm, and the initial sampling clock offset is used.

Gave 200 ppm. The attenuation factor was 0.96 in the Brazilian channel and 0.97 in the Rayleigh channel. In general, it can be seen that the jitter size of the scheme according to the present invention is small and the convergence speed is also fast.

6A and 6B compare the jitter size of the conventional scheme and the scheme according to the present invention in the Brazilian channel A. The convergence time is similar, but the size of the jitter is significantly reduced as shown in FIG. 6B. have.

Brazilian channels B, C and D not only reduced jitter but also converged time faster. In particular, it shows good performance in indoor reception, and shows the same performance as FIG. 9 in Brazilian channel D. In indoor reception environments, the performance of the initial jitter is large, but after some time, it has a fast convergence rate and a small amount of jitter.

These results showed similar results in the mobile reception environment. As shown in FIGS. 10A and 10B, when the speed is 120 km in the Rayleigh channel, the convergence time is increased by about 0.9 msec and the size of jitter is also reduced.

FIG. 11 is a table showing convergence times for fixed reception and mobile reception. The unit is msec. Brazil channel A has a similar convergence time, but Brazil channels B, C, and D differ by more than 0.6msec. In addition, the convergence time is 0.9msec. This result shows the fast convergence because the offset value is estimated using the fast convergence feature at the large loop bandwidth.

As described above, although described with reference to a preferred embodiment of the present invention, those skilled in the art various modifications of the present invention without departing from the spirit and scope of the invention described in the claims below Or it may be modified.

As described above, the present invention improves the symbol timing synchronization performance by increasing the bandwidth of the initial loop filter during operation to quickly converge, and then gradually reducing the size of the jitter by adjusting the bandwidth of the loop filter, thereby increasing the conventional jitter amount. This can improve performance degradation.

In addition, the present invention has the advantage that the possibility of industrialization is very great because it is possible to improve the performance degradation due to the jitter generation in the ATSC DTV system by adapting the loop filter bandwidth by adding a very simple configuration.

Claims (5)

In an ATSC DTV system employing a sampling clock offset estimator and a Gardner type symbol timing locked loop including a loop filter and a VCO, Jitter average calculating means for calculating an average of the output values for a predetermined time of the loop filter; Offset estimation means for estimating an offset from the output of the jitter average calculating means; And Bandwidth attenuation means for adaptively reducing the bandwidth of the loop filter by setting the attenuation coefficient of the loop filter to reduce jitter based on the output of the jitter average calculating means; Symbol timing synchronization apparatus of the ATSC DTV system characterized in that it further comprises. The method of claim 1, wherein the jitter average calculating means The symbol timing synchronizer of the ATSC DTV system, characterized in that the output value of the loop filter for 0.1ms. In a symbol timing synchronization method of an ATSC DTV system employing a Gardner type symbol timing synchronization loop including a sampling clock offset estimator, a loop filter, and a VCO, A first step of setting an bandwidth of the loop filter to a predetermined bandwidth at the initial stage of operation, calculating an average value of jitter by averaging the output of the loop filter for a predetermined time, and then estimating an offset; Reducing the bandwidth of the loop filter based on the average value of the jitter; A third step of estimating the offset by averaging the output of the loop filter whose bandwidth is reduced by the second step, recalculating the average value of jitter; And A fourth step of repeatedly performing the second step or less until symbol timing synchronization is achieved; Symbol timing synchronization method of the ATSC DTV system, characterized in that comprises a. 4. The bandwidth of claim 3, wherein the bandwidth of the loop filter in the first step is Symbol timing synchronization method of ATSC DTV system, characterized in that it is set to 1000ppm (ppm: one millionth error in sampling clock offset). The method of claim 3, wherein the predetermined time Symbol timing synchronization method of the ATSC DTV system, characterized in that 0.1ms.

Images