CN100356786C - Signal transmission method in digital TV ground broadcast transmission system - Google Patents

Abstract

The present invention discloses a signal transmitting method in a digital ground broadcasting transmission system, which is characterized in that a data flow to be transmitted is split into a plurality of continuous data frames to be transmitted; each data frame comprises preset information with the length of 1024 bits, system information with the length of 128 bits and data information with the length of 16384 symbols, and thus, simple first-order circulation is formed. The present invention adopts a simple data frame structure as a data structure in digital TV ground broadcasting transmission, and data circulation using the frame as a cycle is simply realized. Rapid and reliable system synchronization and rapid and accurate channel estimation can be obtained by using the preset information composed of a PN1023; the system information is protected by frequency extension so that a receiving end is reliable identified and processed. The length of the preset information, the length of the system information and the length of the data information of the present invention which are equal to the power of two can nicely support the use of a single carrier frequency domain processing technique.

Description

Signal Transmission Method in Digital TV Terrestrial Broadcasting Transmission System

Technical Field The present invention belongs to the field of signal transmission, in particular to a method for transmitting signals in digital terrestrial broadcasting transmission.

Background Art Currently, the data structure of the digital TV terrestrial broadcasting system is generally divided into two categories: one is the data frame/field/segment structure represented by the American ATSC in the time domain of a single carrier, and the other is the data structure based on the European DVB- T is the data frame structure representing the characteristics of multi-carrier frequency domain transmission.

Referring to Figure 1, the ATSC data structure in the United States is based on a data segment composed of a segment sync header and a certain amount of data information as the minimum cycle unit, and 312 data segments and a field sync signal segment of the same length form a data field. Two data fields form 1 data frame. The data structure performs respective cycles in three cycles of segment, field and frame.

The form of 1, 0, 0, 1 is selected for the segment synchronization signal, which is beneficial to the timing recovery of the receiving end. The field sync signal consists of 1 PN511 sequence, 3 sequential PN63 sequences, and appropriate reserved bits. Among them, the second PN63 sequence is reversed every other field. PN63 in the field sync of the first data field of a data frame is not inverted, and PN63 in the field sync of the second data field is inverted. Thus, through the inversion of PN63, the boundary of the data frame is divided.

The data information in the data structure of the ATSC system is concentrated in the data segment. Considering the channel coding method adopted by the ATSC system, the data information stored in each data segment is actually the data of an MPEG packet after encoding and mapping (not including sync byte 47H). Therefore, the data information contained in a data field or data frame has an integer multiple relationship with the MPEG packet, that is, 312 encoded and mapped MPEG packets, excluding the field synchronization segment. It means that the data encoding method has a certain relationship with the data structure. In fact, the byte interleaving depth adopted by ATSC is exactly 52 data segments, and the synchronization signal does not participate in encoding.

The ATSC system data structure is huge and complex. 1 data field is about 24ms, which is not conducive to tracking faster channel changes. The data structure is complex, and the correlation with the coding method inhibits the possibility of the system adopting a more efficient coding method, lacks flexibility, and restricts the development space.

Fig. 2 shows the data frame structure of the typical frequency domain transmission of the DVB-T system. The system uses OFDM symbols as the smallest unit. 68 OFDM symbols form a data frame, and 4 data frames form a data superframe.

In 1 OFDM symbol, it is composed of the corresponding number of data carriers and their guard intervals. The guard interval is actually taken from a part of the data carrier, and its length determines the ability of the system to resist multipath interference. Pilot information used to track channel characteristics and TPS carrying system information are dispersed in data carriers according to certain rules. Due to the characteristics of multi-carrier frequency domain block processing, and the pilot used to track channel changes exists on the data carrier in a discrete form, it affects the system's ability to track channel changes, and it is impossible to use continuous PN as in the above-mentioned single-carrier system. The sequence comes and the data itself tracks fast channel changes point by point.

SUMMARY OF THE INVENTION The purpose of the present invention is to provide a new digital signal transmission method. The signal transmission method has a simple data structure, but can quickly and reliably realize system synchronization and quickly and accurately perform channel estimation, and the transmission method can be used in a digital terrestrial broadcasting transmission system.

The signal transmission method designed by the present invention is as follows: in the digital television terrestrial transmission system, the transmitted data stream is divided into continuous data frames for transmission. Each data frame contains three parts, which are preset information, system information and data information, and form a simple first-order cycle.

The preset information is used for system synchronization, clock recovery, channel estimation and equalizer training. In the digital TV terrestrial transmission system, the preset information can choose a two-level time-domain PN sequence with good correlation characteristics and flat spectrum to assist system synchronization and channel estimation. In order to facilitate the fast Fourier transform (FFT) operation of frequency domain processing, such as for adaptive frequency domain equalization, the length of the preset information should be selected as the power of 2. For this reason, the preset information can be satisfied by adding a two-level definite signal with an appropriate number of digits through the PN sequence.

In order to obtain rapid and reliable system synchronization and rapid and accurate channel estimation, the present invention selects PN1023 as the two-level PN sequence constituting the preset information. Considering that the preset information length is a power of 2, PN1023 adds a 1-bit two-level definite signal and maps it into preset information with a length of 1024 two-level symbols.

System information is expressed through two-level sequences. The length of the system information can be selected according to different business requirements and automatically recognized at the receiving end, but it should meet the power of 2. For example, in the wired transmission mode, the system information can be 32 or 64 bits long, while in the wireless transmission mode, the system information can be selected to be 64 or 128 bits long.

The system information in the data structure is used to determine the transmission mode of the frame data information, including modulation mode, coding mode, etc., so that the receiving end can make corresponding processing. Especially in mixed mode transmission, the identification of system information is very important. Therefore, when the system information is transmitted, it can be processed to ensure that it can be reliably identified at the receiving end. The invention adopts the spread spectrum protection technology to enhance the robustness of the transmission system information, and adopts the direct sequence spread spectrum method to complete it, and is used to fight against the bad channel fading in the digital TV ground transmission.

Data information can be two-level data or multi-level data. The length of the data information is determined by the transmission mode contained in the system information, but it should also satisfy the power of 2.

The digital TV terrestrial broadcasting system of the present invention transmits three kinds of services, namely fixed reception of television signals (fixed service), mobile reception of television signals (mobile service) and mobile reception of data signals (data service). Among them, 16-OQAM can be selected for fixed services; 4-OQAM or 4-OQAM-NR can be selected for mobile services; 4-OQAM-NR can be selected for data services. In the mixed transmission mode, the modulation modes of various services are arbitrarily mixed in units of frames, and the modulation mode of each frame is clearly indicated by the system information in the frame.

The present invention adopts a simple data frame structure as the data structure in the digital TV terrestrial transmission, and the data only circulates in frames, and each data frame is composed of preset information, system information and data information, so the realization is simple. Using PN1023 to form preset information can get more rapid and reliable system synchronization and rapid and accurate channel estimation; the system information after spread spectrum protection makes the identification and processing of the receiving end more reliable. The length of the preset information, the length of the system information and the length of the data information in the present invention are all powers of 2, which can well support the use of single carrier frequency domain processing technology.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Fig. 1 is the signal transmission method of the data frame/field/segment structure of ATSC in the United States.

Figure 2 shows the data transmission method of European DVB-T.

Fig. 3 is a signal transmission method of the data structure designed in the present invention.

Figure 4 shows the generation method and preset value of PN1023.

Figure 5 shows the generation method and preset value of PN127.

DETAILED DESCRIPTION Referring to Figure 3, in the digital television terrestrial transmission system, the transmitted data stream is divided into successive data frames for transmission. Each data frame contains three parts, which are preset information, system information and data information, and form a simple first-order cycle.

The preset information is used for system synchronization, clock recovery, channel estimation and equalizer training. In order to obtain rapid and reliable system synchronization and rapid and accurate channel estimation, the present invention selects PN1023 as the two-level PN sequence that constitutes the preset information, and adds a two-level definite signal at the same time to meet the length of the preset information. Power of 2. In order to keep the spectrum characteristics of the preset information flat after adding the definite signal, the selected definite signal should ensure that the level sum (mean value) of the preset information after mapping is zero. For this reason, the two-level definite signal added in this embodiment is '0', and after mapping, preset information with a length of 1024 two-level symbols is formed.

The generating polynomial of PN1023 is: X ¹⁰ +X ³ +1; the initial value is 0000000001. See Figure 4 for how PN1023 is produced.

System information is expressed through two-level sequences. The length of the system information can be selected according to different business requirements and automatically recognized at the receiving end, but it should meet the power of 2.

Since the system information contains information that is very important to the receiver processing module, such as the modulation method and the coding method, in this embodiment, a 128-order WALSH sequence is used to map into 128 two-level symbol sequences, and carry corresponding system information. In this way, the system information can be recovered more reliably at the receiving end, ensuring the normal operation of each module of the receiver.

Each system information is expressed by a 128-bit two-level sequence. There are 128 of these two-level sequences, derived from 128-bit long Walsh blocks. The basic Walsh block is shown in formula 2-1, and the systematic generation method of Walsh block is shown in formula 2-2.

$\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\\ \mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right]$

Equation 2-1 Basic Walsh Block

$\mathrm{<span\; class="notranslate">\; W</span>}<span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\mathrm{<span\; class="notranslate">\; h</span>}& \mathrm{<span\; class="notranslate">\; h</span>}\\ \mathrm{<span\; class="notranslate">\; h</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; h</span>}\end{array}\right]$

Formula 2-2 The systematic generation method of Walsh block, where H is the Walsh block of the previous order

Inverting the above 128 128-bit long Walsh vectors increases the number of Walsh vectors to 256. In order to obtain spectrum flatness, these 256 vectors are multiplied by a random sequence with a length of 128 to obtain 256 system information vectors. The random sequence is generated by a 7-bit shift register to generate a 7-order maximum length sequence with a length of 127 and then add a '0' to generate it. The method for generating the 127-bit maximum length sequence is shown in FIG. 5 . Its generating polynomial is: X ⁷ +X ³ +1, and the initial value is 0000001.

Each transmission mode corresponds to one of the 256 system information vectors, and the other vectors are reserved modes. The following are examples of definitions for each mode:

1st vector reserved

2nd vector reserved

3rd vector reserved

4th vector reserved

5th vector 16-OQAM

6th vector reserved

7th vector 4-OQAM

8th vector reserved

9th vector 4-OQAM-NR

10th vector reserved

other vectors reserved

Among them, an even vector is the negation of its previous odd vector. That is to say, the 6th vector = the inversion of the 5th vector, the 8th vector = the inversion of the 7th vector, the 10th vector = the inversion of the 9th vector... .

When the receiver receives the reserved mode, it does not process the signal and skips the frame.

Data information can be two-level data or multi-level data. The length of the data information is determined by the transmission mode contained in the system information, but it should be a power of 2. The digital TV terrestrial broadcasting system of the present invention transmits three kinds of services, namely fixed service, mobile service and data service. Among them, 16-OQAM can be selected for fixed services; 4-OQAM or 4-OQAM-NR can be selected for mobile services; 4-OQAM-NR can be selected for data services. In this embodiment, in each transmission mode, the length of data information is determined to be 16384 symbols. However, in the 16-OQAM transmission mode, the data information is selected as four-level data symbols, and each data symbol is 2 bits; while in 4-OQAM and 4-OQAM-NR, the data information is selected as two-level data Symbols, each data symbol is 1 bit. so:

In the 16-OQAM transmission mode, each data frame contains the preset information (PN1023+'0') of a two-level data symbol with a length of 1024; the spreading protection of a two-level data symbol with a length of 128 System information (corresponding to the above-mentioned fifth vector); the data information of each frame has a total of 16384 four-level symbols, and each data symbol is a 2-bit, four-level symbol.

In the 4-OQAM transmission mode, each data frame contains the preset information (PN1023+'0') of a two-level data symbol with a length of 1024; the spreading protection of a two-level data symbol with a length of 128 System information (corresponding to the seventh vector above); the data information of each frame has a total of 16384 two-level symbols, and each data symbol is a 1-bit, two-level symbol.

In the 4-OQAM-NR transmission mode, each data frame contains the preset information (PN1023+'0') of a two-level data symbol with a length of 1024; the spread spectrum of a two-level data symbol with a length of 128 Protected system information (corresponding to the 9th vector above); the data information of each frame has a total of 16384 two-level symbols, and each data symbol is a 1-bit, two-level symbol.

In the mixed transmission mode, the modulation modes of various services are arbitrarily mixed in units of frames, and the modulation mode of each frame is clearly indicated by the system information in the frame.

Claims (9)

1. The signal transmission method in the digital TV terrestrial broadcasting transmission system, the data stream is divided into a continuous data frame for transmission, and it is characterized in that: each data frame contains information that can be used for system synchronization, clock recovery, channel estimation and equalization Preset information for device training, system information that can be used to characterize the data information transmission mode of this frame, and data information are three parts, and form a simple first-order cycle; The preset information is a two-level symbol sequence with a length of 1024 bits, which is composed of a two-level time-domain PN sequence PN1023 and a two-level definite signal added; The system information is represented by a randomized 128-bit two-level sequence, and the generation method of the sequence is: using a 128-order WALSH codeword to generate 128 128-bit long WALSH vectors, and after the 128 WALSH vectors are reversed, Combined with the original 128 vectors to form 256 vectors; the 256 vectors obtained above are multiplied by a random sequence with a length of 128 to obtain 256 vectors for representing system information; each transmission mode corresponds to one of the 256 vectors one, and the rest of the vectors as reserved modes; The data information is a two-level sequence or a multi-level sequence, and the length satisfies a power of 2. 2. The signal transmission method according to claim 1, characterized in that: the data information length is 16384 symbols, and under the 16-OQAM transmission mode, the data information is a four-level symbol, and each data symbol is 2 bits ;In the 4-OQAM and 4-OQAM-NR transmission modes, the data information is a two-level symbol, and each data symbol is 1 bit; In the mixed transmission mode, the modulation modes of various services are arbitrarily mixed in units of frames, and the modulation mode of each frame is clearly indicated by the system information in the frame. 3. The signal transmission method according to claim 1 or 2, characterized in that: the generator polynomial of the PN1023 is X ¹⁰ +X ³ +1, and the initial value is 0000000001. 4. The signal transmission method according to claim 1 or 2, characterized in that: the 1-bit two-level definite signal value added in the preset information is selected as "0", so that the level and to zero. 5. The signal transmission method according to claim 3, characterized in that: the 1-bit two-level definite signal value added in the preset information is selected as "0", so that the level sum after the preset information mapping is zero . 6. The signal transmission method according to claim 1 or 2, characterized in that: a 7-bit shift register is used to generate a length of 127 7-order maximum length sequence, and its generating polynomial is X ⁷ +X ³ +1 , the initial value is 0000001; The 7th-order maximum length sequence obtained above is added with a "0" to form a random sequence with a length of 128, which is used to multiply 256 vectors to obtain 256 system information vectors. 7. signal transmission method according to claim 3, it is characterized in that: adopting a 7-bit shift register to produce a length is 127 7 order maximum length sequences, and its generating polynomial is X ⁷ +X ³ +1, initial The value is 0000001; The 7th-order maximum length sequence obtained above is added with a "0" to form a random sequence with a length of 128, which is used to multiply 256 vectors to obtain 256 system information vectors. 8. signal transmission method according to claim 4, it is characterized in that: adopting a 7-bit shift register to produce a length is 127 7 order maximum length sequences, and its generator polynomial is X ⁷ +X ³ +1, initial The value is 0000001; The 7th-order maximum length sequence obtained above is added with a "0" to form a random sequence with a length of 128, which is used to multiply 256 vectors to obtain 256 system information vectors. 9. The signal transmission method according to claim 5, characterized in that: adopting a 7-bit shift register to produce a 7-order maximum length sequence of 127 in length, its generating polynomial is X ⁷ +X ³ +1, initial The value is 0000001; The 7th-order maximum length sequence obtained above is added with a "0" to form a random sequence with a length of 128, which is used to multiply 256 vectors to obtain 256 system information vectors.

Images