VSB transmission system

Abstract

A vestigial sideband (VSB) modulation transmission system and a method for encoding an input signal in the system are disclosed. According to the present invention, the VSB transmission system includes a convolutional encoder for encoding an input signal, a trellis-coded modulation (TCM) encoder for encoding the convolutionally encoded signal, and a signal mapper mapping the trellis-coded signal to generate a corresponding output signal. Different types of the convolutional encoders are explored, and the experimental results showing the performances of the VSB systems incorporating each type -of encoders. reveals that a reliable data transmission can be achieved even at a lower input signal to noise -ratio when-.a convolutional encoder is used as an error-correcting encoder in a VSB system.

Description

BACKGROUND OF THE INVENTION.

1. Field of the Invention

The present invention relates to a digital communication system, and more particularly, to a vestigial sideband (VSB) modulation transmission system including a TCM (Trellis-Coded Modulation) encoder and an additional ½ rate convolutional encoder having a superior state transition property when connected to the TCM encoder in the system.

2. Background of the Related Art

The TCM coded 8-VSB modulation transmission system has been selected as a standard in 1995 for the U.S. digital terrestrial television broadcasting, and the actual broadcasting incorporating the system has started since the second half of the year 1998.

In general, a digital communication system performs error correcting processes to correct the errors occurred at the communication channels. The total amount of the transmitting data is increased by such error correcting coding processes since it creates additional redundancy bits added to the information bits. Therefore, the required bandwidth is usually increased when using an identical modulation technique. Trellis-coded modulation (TCM) combines multilevel modulation and coding to achieve coding gain without bandwidth expansion. Also an improved signal to noise ratio can be achieved by using the trellis-coded modulation (TCM) technique.

FIG. 1A and FIG. 1B illustrate a typical TCM encoder used in a typical ATSC 8-VSB system and corresponding set partitions used by the TCM encoder. According to the FIG. 1A, an input bit d₀ is output as c₁ and c₀ after trellis-coded modulation, and then a subset is selected among (−7,1), (−5,3), (−3,5), and (−1,7). Thereafter, an input bit d₁ selects a signal within the selected subset. In other words, when d₁ and d₀ are inputted, one of eight signals (−7,−5,−3,−1,1,3,5,7) is selected by c2, c1, and c0 generated by the TCM encoder. d1 and d0 are called an uncoded bit and a coded bit, respectively.

FIG. 1B illustrates the set partitions used by the TCM encoder used in the ATSC 8-VSB system. Eight signal levels are divided into four subsets, each of which including two signal levels. Two signals are assigned to each subset such that the signal levels of each subset are as far as possible from each other as shown in FIG. 1B.

SUMMARY OF THE INVENTION

Accordingly, the present invention is directed to a VSB transmission system and a method for encoding an input signal in the VSB transmission system that substantially obviates one or more problems due to limitations and disadvantages of the related art.

An object of the present invention is to provide a VSB transmission system that can transmit data reliably even at a lower signal to noise ratio and can have an optimal state transition property when connected to the TCM encoder by using a ½ rate convolutional encoder as an additional error correcting encoder in the system.

Another object of the present invention is to provide a method for encoding an input signal in a VSB modulation transmission system enabling a data sender to achieve more reliable data transmission at a lower signal to noise ratio and to have an optimal state transition property of a ½ convolutional encoder, which is concatenated to the TCM encoder for error correcting in the system.

Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

To achieve these objects and other advantages and in accordance with the purpose of the invention, as embodied and broadly described herein, a vestigial sideband (VSB) modulation transmission system includes a convolutional encoder encoding an input signal; a trellis-coded modulation (TCM) encoder encoding the convolutionally encoded input signal; and a signal mapper mapping the trellis-coded input signal to generate a corresponding output signal.

In another aspect of the present invention, a vestigial sideband (VSB) modulation transmission system includes a ½ rate convolutional encoder encoding an input signal to generate first and second output signals; a ⅔ rate trellis-coded modulation (TCM) encoder encoding the first and second output signals to generate third, forth and fifth output signals; and a signal mapper mapping the third, forth, and fifth output signals.

There are three different types of ½ rate convolutional encoders that can be used in this aspect of the present invention. The first type includes a plurality of multipliers, each ith multiplier multiplying the input signal by a constant k₁ to generate an ith multiplier value; a plurality of memories, a first memory storing the previous second output value as a first memory value and each i+1th memory storing an i+1th memory value obtained by adding an ith memory value stored in a ith memory and the ith multiplier value; and a plurality of adders, each ith adder adding the ith memory value and the ith multiplier value, where i=1,2,3, . . . , n, and a n+1th memory value stored in a n+1th memory is the second output signal.

The second type of the ½ rate convolutional encoder includes a first memory storing the input signal as a first memory value; a second memory storing the first memory value as a second memory value; a first adder adding the input signal and the second memory value to generate the first output signal; and a second adder adding the input signal and the first and second memory values to generate the second output signal.

Finally, the third type of the ½ rate convolutional encoder includes a first memory storing the previous second output value as a first memory value; an adder adding the input signal and the first memory value; and a second memory storing a result from the adder as a second memory value, the second memory value being the second output signal.

In another aspect of the present invention, a method for encoding an input signal in a vestigial sideband (VSB) modulation transmission system includes the steps of encoding the input signal by the convolutional encoder; encoding the convolutionally encoded input signal by the TCM encoder; and generating a final output signal my mapping the trellis-coded input signal.

In a further aspect of the present invention, a method for encoding an input signal in a vestigial sideband (VSB) modulation transmission system includes the steps of generating first and second output signals by encoding the input signal using the ½ convolutional encoder; generating a third, forth, and fifth output signals by encoding the first and second output signals using the ⅔ rate TCM encoder; and generating a final output signal by mapping the third, forth, and fifth output signals.

The second output signal can be generated using three different methods in the last aspect of the present invention described above. The first method for generating the second output signal includes the steps of multiplying the input signal by a constant k₁ to generate an ith multiplier value for i=1,2,3 . . . n; storing the previous second output value as a first memory value; and storing an i+1 th memory value obtained by adding an ith memory value and the ith multiplier value for i=1,2,3 . . . n, where the second output signal is an n+1 th memory value.

The second method for generating the second output signal includes the steps of storing the input signal as a first memory value; storing the first memory value as a second memory value; generating the first output signal by adding the input signal and the second memory value; and generating the second output signal by adding the input signal and the first and second memory values.

Finally, the third method for generating the second output signal includes the steps of storing the previous second output value as a first memory value; adding the input signal and the first memory value; storing the value resulted from the adding step as a second memory value; and outputting the second memory value as the second output signal.

It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principle of the invention. In the drawings;

FIG. 1A illustrates a typical trellis-coded modulation (TCM) encoder used in a ATSC 8VSB transmission system according to the related art;

FIG. 1B illustrates set partitions used by a typical TCM encoder of a ATSC 8VSB transmission system according to the related art;

FIG. 2 illustrates an error correcting encoder concatenated to a ⅔ rate TCM encoder in a ATSC 8-VSB transmission system according to the present invention;

FIG. 3A illustrates a ½ rate convolutional encoder concatenated to a ⅔ TCM encoder to be used as an error correcting encoder in a ATSC 8-VSB transmission system according to the present invention;

FIG. 3B illustrates ⅔ and ⅓ rate convolutional encoders used as an error correcting encoder in a ATSC 8-VSB transmission system according to the present invention;

FIG. 4 illustrates a first type of a ½ rate convolutional encoder concatenated to a ⅔ TCM encoder in a ATSC 8-VSB transmission system according to the present invention;

FIG. 5A illustrates a second type of a ½ rate convolutional encoder used in a ATSC 8-VSB transmission system according to the present invention and its corresponding state transition diagram;

FIG. 5B illustrates a third type of ½ rate convolutional encoder used in a ATSC 8-VSB system according to the present invention and its corresponding state transition diagram;

FIG. 6 illustrates a VSB receiving system corresponding to a ATSC 8-VSB transmission system according to the present invention;

FIG. 7A illustrates Euclidean distances of a set of output signals generated from the ½ rate convolutional encoder shown in FIG. 5A;

FIG. 7B illustrates Euclidean distances of a set of output signals generated from the ½ rate convolutional encoder, shown in FIG. 5B; and

FIG. 8 illustrates performances of ATSC 8-VSB transmission systems when each of the ½ rate convolutional encoders shown in FIG. 5A and FIG. 5B is used.

DETAILED DESCRIPTION OF THE INVENTION

Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings.

FIG.2 illustrates a VSB transmission system in which an error correcting encoder is concatenated to a ⅔ rate TCM encoder according to the present invention. By adding an additional error correcting encoder to the ⅔ rate TCM encoder in the VSB system, it is possible to achieve a reliable data transmission even at a lower signal to noise ratio than that of the conventional ATSC TCM coded 8VSB system. In the present invention, a ½ rate convolutional encoder is used for the additional error correcting encoder. In addition, a multiplexer located between the error correcting encoder and the ⅔ rate TCM encoder classifies the data received from each of the error correcting encoder and a ATSC encoder and inputs each data to the TCM encoder. The additional error-corrected data will be regarded as an error by the ATSC receiver and will be discarded.

FIGS. 3A and 3B illustrate a ½ rate encoder used as an additional error correcting encoder shown in FIG. 2. According to FIG. 3A, an input bit u is processed in the ½ rate encoder to generate two output bits d₁ and d₀, and these are inputted to a ⅔ rate TCM encoder. In FIG. 3B, each of ⅔ and ⅓ rate encoders is connected to a ⅔ mate TCM encoder. Since the bit error rate of uncoded bits u₁ is lower than that of a coded bit u₀, the encoder having a higher code rate is used for u₁, and the other encoder is used for u₀. This will compensate the difference between two input bits u₀ and u₁. In addition, the ⅔ and ⅓ rate encoders can be considered as being a ½ rate encoder since it has three input bits and six output bits. Thus, combining encoders having different code rates can reduce the bit error rate of the whole system. As a result, the additional encoder can be any one of the ½ rate encoder and the combination of the ⅔ rate encoder and the ⅓ rate encoder shown in FIG. 3A and FIG. 3B, respectively. By adding the additional encoder, the performance of the system can be enhanced, and this will be shown later in this section. Considering the signal mapping of the TCM encoder, the error correcting encoder must be designed so that it has the optimal state transition property when connected to the TCM encoder.

FIG. 4 illustrates a first type of a ½ rate convolutional encoder concatenated to a ⅔ TCM encoder in a VSB transmission system according to the present invention. The ½ rate convolutional encoder receives an input bit u and generates a first output bit d₁ by bypassing u. A second output bit d₀ is the value of the N+1 th memory m_i+1. The ½ rate convolutional encoder includes N multipliers, N adders, and N+1 memories. The first memory m₁ stores a previous second output value, the first multiplier g₁ multiplies the input bit u by a first constant k₁, and the first adder adds the outputs from g₁ and m₁. Similarly, each i+1 th memory m_i+1 stores the output from the ith adder, the ith multiplier g_i multiplies the input bit u by an ith constant k_i, and the ith adder adds the outputs from g_i and m_i, where i=2,3,4, . . . ,N . Finally, the N+1th memory m_i+1 stores the output from the N th adder. Then the value stored in m_i+1 is output as a second output bit (current). In addition, the second output bit (current) is feedback to the first memory m₁ for calculating a next second output value. N can be greater than or equal to two and can be determined as one wishes to design the system. As shown in the FIG. 4, the ½ rate convolutional encoder receives u and outputs d₀ and d₁·d₀ and d₁ then become the output bits c₁ and c₂ of the TCM encoder. Therefore, when d₁d₀=00, c₂c₁=00, and the corresponding 8VSB symbol becomes 7(c₂c₁c₀=000) or −5 (c₂c₁c₀=001) depending on the value of c₀·c₀ is equal to the value stored in a second memory s₁ and is obtained by adding s₀ and d₀, where s₀ is the value stored in a first memory. The 8VSB symbols for d₁d₀=01, 10, 11 are (−3,−1), (1,3), and (5,7), respectively.

FIG. 5A illustrates a non-systematic ½-rate convolutional encoder used in a VSB system according to the present invention and its corresponding state transition diagram. This type of encoder is often used because of its long free-distance property. In the state transition diagram shown in FIG. 5A, a transition from the state S_k at t=k to the state S_k+1 at t=k+1 is denoted as a branch, and the value indicated above each branch corresponds to the output of the branch. The probability of receiving a signal r when a signal z having zero mean and variance σ² is sent through a AWGN channel can be obtained by using the equation: $\begin{array}{cc}p\left(r|z\right)=\frac{1}{\sqrt{2\pi {\sigma }^2}}\exp \left(\frac{-|r-z{|}^2}{2{\sigma }^2}\right)& \left[\mathrm{Equation}1\right]\end{array}$ where z represents a branch output. A branch metric is a probability measure of receiving r when the branch output z is sent from the encoder. It is an Euclidean distance between r and z, and can be obtained by the following equation:Branch Metric ∝ Log (p(r/z))=|r−z|².  [Equation 2]

A metric corresponding to a path including S₀, S₁, S₂, . . . , S_k can be calculated by the equation: $\begin{array}{cc}\mathrm{Path}\mathrm{Metric}=\sum _{t=0}^{t=k}\mathrm{Branch}\mathrm{Metric}.& \left[\mathrm{Equation}3\right]\end{array}$ The path metric is an accumulated value of the branch metrics of the branches included in a path and represents a probability of the path.

As shown in the state transition diagram of FIG. 5A, two branches are divided from each S^._k , and two branches are merged into, each S_k+1. A viterbi decoder that decodes a convolutional code first calculates the path metrics of the two paths that are merging into each state and selects the path having a lower path metric. The path metric selected using this technique represents the lowest path metric of the paths starting from an initial state (t=0) to each S_k.

When selecting a path between two paths merging into one state, the probability of the path selection becomes higher as the difference between the metrics of the two paths is larger. Since a path metric represents the sum of metrics of the branches included in a path, it is desired to have the largest difference between the branch metrics in order to maximize the performance of the encoder.

The ½ rate convolutional encoder shown in FIG. 5A includes a first memory for storing an input bit u as a first memory value s₀; a second memory for storing s₀ as a second memory value s₁; a first adder for adding u and s₁; and a second adder for adding u, s₀, and s₁. The output from the first and second adders becomes a first output bit d₁ and a second output bit d₀.

FIG. 5B illustrates a systematic convolutional encoder used in a VSB transmission system and its corresponding state transition diagram. A first output bit d₁ is generated by bypassing an input bit u, and a second output bit d₀ is generated by adding and delaying u. The systematic ½ rate convolutional encoder includes a first memory for storing a previous second output bit value as a first memory value s₀, an adder for adding the input bit u and s₀, and a second memory for storing the output from the adder as a second memory value s₁ and outputting s₁ as the second output bit d₀.

According to FIG. 5A, the combination of the branch outputs dividing from a state at t=k or merging into a state at t=k+1 is (00,11) or (01,10). According to the trellis-coded modulation fundamental, the encoder has a better performance as the difference between branch metrics of the combination is larger. A larger difference between the branch metrics means that the corresponding Euclidean distance is larger. The Euclidean distance of (00,11) is larger than that of (01,10). When the output is either 01 or 10, the error often occurs during the path selection. Therefore, it is desired to have, the combination of the branch outputs of (00,10) and (01,11) so that the difference between the branch metrics is large. This is shown in FIG. 5B. Therefore, the convolutional encoder of FIG. 5B has a better encoding performance than that of FIG. 5A.

FIG. 6 illustrates a VSB receiving system corresponding to the VSB transmission system of the present invention.

FIG. 7A and FIG. 7B illustrate Euclidean distances corresponding to the output combinations generated from the encoders shown in FIG. 5A and FIG. 5B, respectively. As it can be shown from both figures, the Euclidean distances of (00,10) and (01,11) are much larger than the that of (01,10). Therefore, the convolutional encoder of FIG. 5B has a better performance when connected to the ⅔ rate TCM encoder in the VSB transmission system.

FIG. 8 illustrates performances of ATSC 8-VSB transmission systems when each of the convolutional encoders shown in FIG. 5A and FIG. 5B is used in the system. For a bit error rate of 1 e−3, the signal to noise ratio is reduced by 2 dB and 4 dB when the convolutional encoders shown in FIG. 5A and FIG. 5B are used as an additional error-correcting encoder in the VSB system. Therefore, a bit error rate can be reduced by using a ½ rate convolutional encoder as an outer encoder of the TCM encoder, and the encoder shown in FIG. 5B has a better bit error rate reduction property.

In conclusion, data can be transmitted at a lower signal to noise ratio by concatenating a ½ rate, convolutional encoder to the TCM encoder in a VSB transmission system according the present invention.

The forgoing embodiments are merely exemplary and are not to be construed as limiting the present invention. The present teachings can be readily applied to other types of apparatuses. The description of the present invention is intended to be illustrative, and not to limit the scope of the claims. Many alternatives, modifications, and variations will be apparent to those skilled in the art.

Claims

1-16. (canceled)

17. An enhanced VSB transmitter comprising: a convolutional encoder encoding an input bit to generate first and second output bits; and a trellis-coded modulation (TCM) encoder encoding the first and second output bits and outputting third, fourth, and fifth output bits, wherein the convolutional encoder comprises: N multipliers, each ith multiplier multiplying the input bit by a constant ki to generate an ith multiplier value for i=1, 2, 3, . . . , N; N+1 memories including a 1st memory pre-storing a 1st memory value being a previous second output bit outputted from the convolutional encoder, each (i+1)th memory storing an (i+1)th memory value obtained by adding an ith memory value stored in an ith memory with the ith multiplier value for i=1, 2, 3, . . . , N; and N adders, each ith adder adding the ith memory value with the ith multiplier value for i=1, 2, 3, . . . , N, wherein the first output bit is the input bit and the second output bit is an (N +1)th memory value stored in an (N+1)th memory.

18. The enhanced VSB transmitter of claim 17, further comprising a VSB mapper selecting one of a plurality of signal levels based on the third, fourth, and fifth output bits outputted from the TCM encoder.

19. The enhanced VSB transmitter of claim 17, wherein the TCM encoder comprises: a first TCM memory pre-storing a first TCM memory value being a previous fifth output bit outputted from the TCM encoder; a TCM adder adding the second output bit with the first TCM memory value to obtain a second TCM memory value; and a second TCM memory storing the second TCM memory value, wherein the third output bit is the first output bit outputted from the convolutional encoder, the fourth output bit is the second output bit outputted from the convolutional encoder, and the fifth output bit is the second TCM memory value stored in the second TCM memory.

20. The enhanced VSB transmitter of claim 17, wherein ki is for i=1, 2, 3, . . . , N.

21. The enhanced VSB transmitter of claim 17, wherein N=1.

22. A method of encoding an input signal in an enhanced VSB transmitter, the method comprising: encoding an input bit using a convolutional encoder to output first and second output bits; and encoding the first and second output bits outputted from the convolutional encoder using a trellis-coded modulation (TCM) encoder to output third, fourth, and fifth output bits, wherein the encoding an input bit using a convolutional encoder comprises: multiplying the input bit by a constant ki to generate an ith multiplier value for 1 for i=1, 2, 3, . . . , N; pre-storing a 1st memory value in a 1st memory, the 1st memory value being a previous second output bit outputted from the convolutional encoder; storing an (i+1 )th memory value in an (i+1)th memory for i=1, 2, 3, . . . , N, the (i+1)th memory value being obtained by adding an ith memory value stored in an ith memory with the ith multiplier value, wherein the first output bit is the input bit and the second output bit is an (N+1)th memory value stored in an (N+1)th memory.

23. The method of claim 22, further comprising selecting one of a plurality of signal levels based on the third, fourth, and fifth output bits outputted from the TCM encoder using a VSB mapper.

24. The method of claim 22, wherein encoding the first and second output bits outputted from the convolutional encoder using a trellis-coded modulation (TCM) encoder comprises: pre-storing a first TCM memory value in a first TCM memory, the first TCM memory value being a previous fifth output bit outputted from the TCM encoder; adding the second output bit with the first TCM memory value to obtain a second TCM memory value; storing the second TCM memory value in a second TCM memory, wherein the third output bit is the first output bit outputted from the convolutional encoder, the fourth output bit is the second output bit outputted from the convolutional encoder, and the fifth output bit is the second TCM memory value stored in the second TCM memory.

25. The method of claim 22, wherein ki is 1 for i=1, 2, 3, . . . , N.

26. The method of claim 22, wherein N=1.

27. An enhanced VSB transmitter comprising: a convolutional encoder encoding an input bit to generate first and second output bits; and a trellis-coded modulation (TCM) encoder encoding the first and second output bits to generate third, fourth, and fifth output bits, wherein the convolutional encoder comprises: N+1 memories including a 1st memory storing a 1st memory value which is a previous second output bit outputted from the convolutional encoder, each (i+1)th memory storing an (i+1) memory value obtained by adding an ith memory value stored in an ith memory with the input bit for i=1, 2, 3, . . . , N; and N adders, each ith adder adding the ith memory value with the input bit for i=1, 2, 3, . . . , N, wherein the first output bit is the input bit and the second output bit is an (N+1)th memory value stored in an (N+1)th memory.

28. The enhanced VSB transmitter of claim 27, further comprising a VSB mapper selecting one of a plurality of signal levels based on the third, fourth, and fifth output bits outputted from the TCM encoder.

29. The enhanced VSB transmitter of claim 27, wherein the TCM encoder comprises: a first TCM memory pre-storing a first TCM memory value being a previous fifth output bit outputted from the TCM encoder; a TCM adder adding the second output bit with the first TCM memory value to obtain a second TCM memory value; and a second TCM memory storing the second TCM memory value, wherein the third output bit is the first output bit outputted from the convolutional encoder, the fourth output bit is the second output bit outputted from the convolutional encoder, and the fifth output bit is the second TCM memory value stored in the second TCM memory.

30. The enhanced VSB transmitter of claim 27, wherein N=1.

31. A method of encoding an input signal in an enhanced VSB transmitter, the method comprising: encoding an input bit using a convolutional encoder to output first and second output bits; and encoding the first and second output bits outputted from the convolutional encoder using a trellis-coded modulation (TCM) encoder to output third, fourth, and fifth output bits, wherein the encoding an input bit using a convolutional encoder comprises: storing a 1st memory value in a 1st memory, the 1st memory value being a previous second output value outputted from the convolutional encoder; storing an (i+1)th memory value in an (i+1)th memory for i=1, 2, 3, . . . , N, the (i+1)th memory value being obtained by adding an ith memory value stored in an ith memory with the input bit, wherein the first output bit is the input bit and the second output bit is an (N+1)th memory value stored in an (N+1)th memory.

32. The method of claim 31, further comprising selecting one of a plurality of signal levels based on the third, fourth, and fifth output bits outputted from the TCM encoder using a VSB mapper.

33. The method of claim 31, wherein encoding the first and second output bits outputted from the convolutional encoder using a trellis-coded modulation (TCM) encoder comprises: pre-storing a first TCM memory value in a first TCM memory, the first TCM memory value being a previous fifth output bit outputted from the TCM encoder; adding the second output bit with the first TCM memory value to obtain a second TCM memory value; storing the second TCM memory value in a second TCM memory, wherein the third output bit is the first output bit outputted from the convolutional encoder, the fourth output bit is the second output bit outputted from the convolutional encoder, and the fifth output bit is the second TCM memory value stored in the second TCM memory.

34. The method of claim 31, wherein N=1.