GB2334862A

GB2334862A - Mapping symbol points in a quadrature amplitude modulation

Info

Publication number: GB2334862A
Application number: GB9825213A
Authority: GB
Inventors: Heon Jekal
Original assignee: Daewoo Electronics Co Ltd
Current assignee: WiniaDaewoo Co Ltd
Priority date: 1998-02-27
Filing date: 1998-11-17
Publication date: 1999-09-01
Also published as: KR19990071095A; GB9825213D0

Abstract

In order to map a symbol point in a 16-Quadrature Amplitude Modulation ("QAM") constellation pattern in response to a mode selection signal, which represents either a Gray-coded symbol mapping or a differential-coded symbol mapping, a 2-bit codeword is either directly passed or differentially encoded (110) to a 2-bit differential codeword therefor to be selected (120) as the 2 most significant bits. After the selected 2 most significant bits and the 2 remaining bits are mapped (130) to a symbol point on the constellation pattern to generate an (I, Q) value and the 2 most significant bits and the 2 remaining bits are compared (150) to generate an I-Q control signal, the (I, Q) value is adjusted (140) based on the mode selection signal and the I-Q control signal.

Description

METHOD AND APPARATUS FOR MAPPING SYMBOL POINTS IN A QUADRATURE AMPLITUDE MODULATION The present invention relates to a quadrature amplitude modulation (QAM) ; and, more particularly, to a method and an apparatus for adaptively supporting both a Gray-coded symbol mapping and a differential-coded symbol mapping in the QAM.

Digital data, for example, digitized compressed television or high definition television signals, can be terrestrially transmitted via very high frequency (VHF), ultra-high frequency (UHF), or cable television analog channels to end users. However, an input waveform can easily be corrupted and transformed by the analog channel while being transmitted therethrough. Specifically, the waveform may be corrupted in the form of a linear, frequency-selective amplitude and phase distortion, nonlinear or harmonic distortion, and multiplicative fading. An additive corruption of the waveform, due to, e. g., statistical, thermal or impulse noise, may be countered by using forward error correction codes.

In general, in order to communicate digital data via an analog channel, the data is modulated by using, for example, a form of Pulse Amplitude Modulation (PAM). Typically, Quadrature Amplitude Modulation (QAM) is used to increase the amount of data that can be transmitted within an available channel bandwidth. The QAM is a quadrature or an orthogonal combination of two PAM signals. When viewed as coordinates of a plane, the combined PAM signals form a"constellation" of possible transmission levels. Each transmitted constellation point is called a symbol. For example, two independent, quadrature four-level AM signals form a 16-QAM constellation which encodes four bits. A 32-point constellation can be formed with dependent six-level AM quadrature signals, encoding five bits per symbol.

In PAM, each signal is a pulse whose amplitude level is selected from a fixed set of levels. In a 16-QAM, each of the quadrature PAM signals is selected from uniformly spaced, bipolar amplitudes scaled from amplitude levels-3,-1, 1 and 3. Bandwidth efficiency in a digital communication system is normally defined in terms of the number of transmitted bits per second per unit of bandwidth, i. e., the ratio of the data rate to the bandwidth. In short, QAM provides a bandwidth efficient modulation, which can provide very low bit error rates when used with high efficiency forward error correction codes such as trellis coded modulation (TCM).

Trellis coded modulation (TCM) has evolved as a combined coding and modulation technique for digital transmission over band-limited channels. While the traditional application of convolutional codes to two-level PAM increases the bandwidth used in transmission, TCM increases the constellation size instead. In TCM schemes, a sequence of"coded"bits are convolutionally encoded into a sequence of groups which partition the symbol constellation. For each encoded group, a number of"uncoded"bits are transmitted by selecting a unique constellation element of the group. At a receiver, the sequence of transmitted groups is decoded by a soft-decision maximum likelihood (ML) convolutional code decoder. Such TCM schemes can improve the robustness of digital transmission against additive noises by 3-6 dB or more, compared with the conventional uncoded modulation at a same information rate.

Most TCM schemes map one step of the convolutional code trellis to one transmission symbol which consists of two QAM components (I, Q). Such a two dimensional (2-D) code achieves a throughput of an integer number of information bits per 2-D symbol.

Referring to Fig. 1, there is shown a block diagram of a conventional 2"-QAM trellis encoder. In the conventional encoding system for QAM transmission, an (N-1) bit input symbol is first parsed into a first bit on a line 30 and the (N-2) remaining bits on a line 20 at a parsing block 10 shown in Fig. 1. The first bit is encoded by employing a rate 1/2 binary convolutional encoding algorithm at a convolutional encoder 40, to thereby provide on a line 45 a two-bit codeword that defines one of four subsets of a 2"-QAM constellation pattern, wherein each subset includes 2"/4 symbol points of the constellation pattern. The remaining bits correlate the input symbol with one of the 2"/4 symbol points included in the subset defined by the 2-bit codeword. Specifically, at a 2"-QAM mapper 50, the codeword is mapped with the N-2 remaining bits to provide a modulation function which includes I and Q components to indicate a specific point on the QAM constellation pattern. The modulation function is further processed, e. g., modulated with a carrier for transmission on a communication channel (see, e. g., U. S. Pat. No. 5, 233, 629 issued to W. H. Paik et al.).

At the 2N-QAM mapper, the 2-bit codeword is either directly inputted (a Gray-coded symbol mapping) or converted into a differential coded symbol (a differential-coded symbol mapping) to perform an I-Q mapping. However, the conventional 2-QAM mapper may be operated in either the Gray-coded symbol mapping mode or the differential-coded symbol mapping mode ; and it is difficult to change the mapping mode at the 2"-QAM mapper.

It is, therefore, a primary object of the present invention to provide a novel method and apparatus for supporting both a Gray-coded symbol mapping and a differential-coded symbol mapping in a 2"-QAM constellation pattern.

In accordance with the present invention, there is provided an apparatus for mapping a symbol point in a 2"-Quadrature Amplitude Modulation (QAM) constellation pattern in response to a mode selection signal, N being an integer greater than 2, wherein said 28-QAM constellation pattern is divided into 4 subsets designated by a 2-bit codeword, each subset including 29-2 symbol points of said constellation pattern designated by N-2 remaining bits, and the mode selection signal represents either a Gray-coded symbol mapping or a differential-coded symbol mapping, said apparatus comprising : a differential encoder for converting the 2-bit codeword to a 2-bit differential codeword therefor ; means for selecting either the 2-bit codeword or the 2-bit differential codeword based on the mode selection signal to generate a 2-bit selected codeword ; means for mapping the 2-bit selected codeword and the N-2 remaining bits to a symbol point on the 2-QAM constellation pattern to generate an (I, Q) value ; a comparator for comparing the 2-bit selected codeword and the N-2 remaining bits to generate an I-Q control signal ; and means for adjusting the (I, Q) value based on the mode selection signal and the I-Q control signal to generate a modulation function.

According to the present invention there is also provided a method for mapping a symbol point in a 2N-Quadrature Amplitude Modulation ("QAM") constellation pattern in response to a mode selection signal, N being an integer greater than 2, wherein said 2N-QAM constellation pattern is divided into 4 subsets designated by a 2-bit codeword, each subset including 2"-Z symbol points of said constellation pattern designated by N-2 remaining bits, and the mode selection signal represents either a Gray-coded symbol mapping or a differential-coded symbol mapping, said method comprising the steps of : (a) if the mode selection signal indicates the Gray-coded symbol mapping, selecting the 2-bit codeword as 2 most significant bits and, if otherwise, differential-encoding the 2-bit codeword to a 2-bit differential codeword therefor to be selected as 2 most significant bits ; (b) mapping said 2 most significant bits and the N-2 remaining bits to a symbol point on the 2"-QAM constellation pattern to generate an (I, Q) value ; (c) comparing said 2 most significant bits with the N-2 remaining bits to generate an I-Q control signal ; and (d) adjusting the (I, Q) value based on the mode selection signal and the I-Q control signal to generate a modulation function.

The above and other objects and features of the present invention will become apparent from the following description of preferred embodiments given in conjunction with the accompanying drawings, in which : Fig. 1 shows a block diagram of a conventional 2"-QAM trellis encoder ; Fig. 2 offers a 16-QAM mapper for supporting both the Gray-coded symbol mapping and the differential-coded symbol mapping in accordance with the present invention ; Fig. 3A presents a 16-QAM constellation pattern in a Gray-coded symbol mapping mode ; Fig. 3B provides another 16-QAM constellation pattern in a differential-coded symbol mapping mode ; and Fig. 4 illustrates a general coordinate in the real (I) and imaginary (Q) plane of a 16-QAM constellation pattern.

Referring to Fig. 2, there is shown a quadrature amplitude modulation (QAM) mapping apparatus 100 for adaptively supporting both a Gray-coded symbol mapping and a differential-coded symbol mapping based on a mode selection signal for a QAM modulation function corresponding to a symbol point on a 2", e. g., 16, QAM constellation pattern in accordance with the present invention, N being an integer larger than 2, wherein the mode selection signal represents either the Gray-coded symbol mapping or the differential-coded symbol mapping.

The modulation function represents N, e. g., 4, bit data, wherein a 2-bit codeword (I1, Q1) of the most significant bits (MSB's) thereof, encoded from a first bit of an input symbol with a rate 1/2 binary convolutional encoding algorithm, determines a subset of the 2N-QAM constellation pattern and (N-2), e. g., 2, remaining bits (I2, Q2) of the least significant bits (LSB's) thereof determine a symbol point included in the subset defined by the 2-bit codeword (1.,Q.).

The 2-bit codeword (I1, Q1) is coupled with a differential encoder 110 and a multiplexer (MUX) 120 ; and the 2 remaining bits (I2, Q2) are directly provided to an I-Q mapper 130 and an I-Q control signal generator 150.

The differential encoder 110, including a 2-bit register 111 and a differential codeword calculation block 112, performs a differential coding on the 2-bit codeword (I1, Q1).

The differential codeword calculation block 112 calculates a current differential codeword based on the 2-bit codeword and a previous differential codeword retrieved from the 2-bit register 111. The current differential codeword is provided to the multiplexer 120 as a 2-bit differential codeword (I1d, Qad) and, also, to the 2-bit register 111 to be stored therein as a previous differential codeword for a next 2-bit codeword.

Referring to Table 1, there is illustrated the relationship between the current differential codeword and the previous differential codeword.

Table 1

previous current 2-bit quadrant 2-bit quadrant differential differential codeword phase change codeword condeword 00 0 11 11 00 0 01 01 00 0 00 00 00 0 10 10 01 90 11 01 01 90 01 00 01 90 00 10 01 90 10 11 11 180 11 00 11 180 01 10 11 180 00 11 11 180 10 01 10 270 11 10 10 270 01 11 10 270 00 01 10 270 10 00 The multiplexer 120 selects either the 2-bit codeword (I1, Q1) or the 2-bit differential codeword (I1d, Q1d) based on the mode selection signal. The 2-bit codeword (Il. Q1) and the 2-bit differential codeword (I1d, Q1d) correspond to the Gray-coded symbol mapping and the differential-coded symbol mapping, respectively. The 2-bit selected codeword is provided to the I-Q mapper 130 and the I-Q control signal generator 150.

The I-Q mapper 130 performs the differential-coded symbol mapping on the 2-bit selected codeword and the 2 remaining bits as shown in Fig. 3B, regardless of the mode selection signal, to generate an (I, Q) value. For example, if a set of the 2-bit selected codeword and the 2 remaining bits is '0110'in sequence, an (I, Q) value of (-1, 3) is generated as shown in Fig. 3B. The (I, Q) value of the differential-coded symbol mapping is provided to an I-Q regulator 140.

Referring to Figs. 3A and 3B, there are illustrated symbol point mapping diagrams for a 16-QAM communication system employing the Gray-coded symbol mapping and the differential-coded symbol mapping, respectively. As is apparent from Fig. 3A, the respective symbol points are symmetrically positioned with respect to the respective I and Q coordinate axes in the Gray-coded symbol mapping.

To the contrary, the symbol points positioned in the respective quadrants are arranged in the differential-coded symbol mapping in such a manner that these symbol points are rotated with respect to those of the adjoining quadrants.

That is to say, at the differential-coded symbol mapping, each of the 4 binary values defined by the 2 most significant bits (phase bits) of each of the 16 point-designating 4-bit segments defines a different one of the 4 quadratures of the I-Q plane. Therefore, rotation of the I-Q plane by 90 , 180 or 270 will change the position of the 4 quadrature accordingly. However, each of the 4 binary values defined by the 2 least significant bits (invariant bits) of each of the 16 point-designating 4-bit segments defines a different one of a set of 4 constellation points that are the same for each of the 4 quadrants. Further, the binary values of the invariant bits of each of the 4 sets are symmetrically arranged so that they do not change their relative positions within a quadrant in response to a rotation in the I-Q plane by 90 , 180 or 270 . Thus it is the bit assignment of the phase bits, used for differentiating each of the 4 quadrants, that is employed by the prior art to render the received symbol constellation insensitive to phase rotation by 90 , 180 or 270 .

Referring back to Figs. 3A and 3B, constellation points surrounded by a dashed line in the Gray-coded symbol mapping are different from those in the differential-coded symbol mapping. In the sections surrounded by dashed lines, the 2-bit selected codeword, i. e., 2 most significant bits, have different logic values, i. e.,'0'and'1', from each other and the 2 remaining bits, i. e., 2 least significant bits, have different logic values, i. e., 0'and'1', from each other.

Therefore, in the sections surrounded by dashed lines, the (I, Q) value must be adjusted as will be described below.

First of all, the I-Q control signal generator 150 generates an I-Q control signal based on the 2-bit selected codeword and the 2 remaining bits, wherein the I-Q control signal represents whether or not the (I, Q) value is required to be regulated. For example, the I-Q control signal generator 150 includes a first and a second exclusive OR gates 151 and 152 and an AND gate 153. Specifically, the 2-bit selected codeword is exclusive-ORed at the first exclusive-OR gate 151 so that an output signal thereof is provided to the AND gate 153 ; and the 2 remaining bits are exclusive-ORed at the second exclusive-OR gate 152 so that an output signal thereof is provided to the AND gate 153. At the AND gate 153, two output signals fed from the first and the second exclusive-OR gates 151 and 152, respectively, are logically multiplied to generate the I-Q control signal. The I-Q control signal is provided to the I-Q regulator 140.

The I-Q regulator 140 adjusts the (I, Q) value based on the mode selection signal and the I-Q control signal. If the mode selection signal corresponds to the differential-coded symbol mapping, there is no adjustment to the (I, Q) value fed from the I-Q mapper 130 since the I-Q mapper 130 performs the differential-coded symbol mapping. If the mode selection signal corresponds to the Gray-coded symbol mapping, the (I, Q) value must be adjusted in accordance with the Gray-coded symbol mapping based on the I-Q control signal. Specifically, if the I-Q control signal represents'1', i. e., if the 2-bit selected codeword, i. e., the 2 most significant bits have different logic values, e. g.,'0'and'1', from each other and the 2 remaining bits have different logic values, e. g.,'0' and'1', from each other, the 2 remaining bits must be reversed in sequence as shown in Figs. 3A and 3B. In other words, if a set of the 2-bit selected codeword and the 2 remaining bits corresponds to one of'0101','0110','1010' and'1001', the 2 remaining bits are reversed in sequence so that a new set of the 2-bit selected codeword and the 2 remaining bits should be generated as one of'0110','0101, '1001'and'1010'and, therefore, at the I-Q regulator 140, the (I, Q) value is adjusted so that its adjusted value (Iadj, Qadi) iS given as follows : (Iadjr =(-Q,-I).

In general, if the constellation points are provided at I levels of I0, I1, I2 and I3 and at Q levels of Qo, Q, Q2 and Q3 as shown in Fig. 4, at the Gray-coded symbol mapping, Io and Il. I2 and I3, Qo and Q1 and Q2 and Q3 must be converted with each other, respectively.

While the present invention has been described with respect to the particular embodiments, it will be apparent to those skilled in the art that various changes and modifications may be made without departing from the scope of the invention as defined in the following claims.

Claims

Claims 1. An apparatus for mapping a symbol point in a 2N-Quadrature Amplitude Modulation ("QAM") constellation pattern in response to a mode selection signal, N being an integer greater than 2, wherein said 2N-QAM constellation pattern is divided into 4 subsets designated by a 2-bit codeword, each subset including 2N-
2 symbol points of said constellation pattern designated by N-2 remaining bits, and the mode selection signal represents either a Gray-coded symbol mapping or a differential-coded symbol mapping, said apparatus comprising : a differential encoder for converting the 2-bit codeword to a 2-bit differential codeword therefor ; means for selecting either the 2-bit codeword or the 2-bit differential codeword based on the mode selection signal to generate a 2-bit selected codeword ; means for mapping the 2-bit selected codeword and the N-2 remaining bits to a symbol point on the 2"-QAM constellation pattern to generate an (I, Q) value ; a comparator for comparing the 2-bit selected codeword with the N-2 remaining bits to generate an I-Q control signal ; and means for adjusting the (I, Q) value based on the mode selection signal and the I-Q control signal to generate a modulation function. 2. The apparatus of claim 1, wherein said differential encoder includes : means for calculating the 2-bit differential codeword based on the 2-bit codeword and a 2-bit previous differential codeword, wherein the 2-bit previous differential codeword represents a 2-bit differential codeword transmitted previously ; and means for storing the 2-bit differential codeword as a previous differential codeword for a next 2-bit codeword.
3. The apparatus of claim 2, wherein N is equal to 4.
4. The apparatus of claim 3, wherein said comparator includes : two exclusive-OR gates for exclusive-ORing the 2-bit selected codeword and the 2 remaining bits, respectively ; and an AND gate for logically multiplying two output signals fed from said two exclusive-OR gates with each other to generate the I-Q control signal.
5. The apparatus of claim 4, wherein said mapping means performs the differential-coded symbol mapping on the 2-bit selected codeword and the 2 remaining bits.
6. The apparatus of claim 5, wherein, if the mode selection signal indicates the Gray-coded symbol mapping and the I-Q control signal represents'1', the (I, Q) value is adjusted so that its adjusted value (Iij, Qij) is given as follows : (-Q.-I).
7. A method for mapping a symbol point in a 2N-Quadrature Amplitude Modulation ("QAM") constellation pattern in response to a mode selection signal, N being an integer greater than 2, wherein said 2"-QAM constellation pattern is divided into 4 subsets designated by a 2-bit codeword, each subset including 2N-2 symbol points of said constellation pattern designated by N-2 remaining bits, and the mode selection signal represents either a Gray-coded symbol mapping or a differential-coded symbol mapping, said method comprising the steps of : (a) if the mode selection signal indicates the Gray-coded symbol mapping, selecting the 2-bit codeword as 2 most significant bits and, if otherwise, differential-encoding the 2-bit codeword to a 2-bit differential codeword therefor to be selected as 2 most significant bits ; (b) mapping said 2 most significant bits and the N-2 remaining bits to a symbol point on the 2"-QAM constellation pattern to generate an (I, Q) value ; (c) comparing said 2 most significant bits with the N-2 remaining bits to generate an I-Q control signal ; and (d) adjusting the (I, Q) value based on the mode selection signal and the I-Q control signal to generate a modulation function.
8. The method of claim 7, wherein said differential-encoding step includes the steps of : (al) calculating a 2-bit differential codeword based on the 2-bit codeword and a 2-bit previous differential codeword, wherein the 2-bit previous differential codeword represents a 2-bit differential codeword transmitted previously ; and (a2) storing the 2-bit differential codeword as a previous differential codeword for a next 2-bit codeword.
9. The method of claim 8, wherein N is equal to 4.
10. The method of claim 9, wherein the 2-bit most significant bits and the 2 remaining bits are mapped to the (I, Q) value on a 16-QAM constellation pattern in accordance with the differential-coded symbol mapping.
11. The method of claim 10, wherein said comparing step (c) includes the steps of : (cl) exclusive-ORing the 2 most significant bits to generate a first exclusive-ORed signal ; (c2) exclusive-ORing the 2 remaining bits to generate a second exclusive-ORed signal ; and (c3) multiplying the first exclusive-ORed signal by the second exclusive-ORed signal to generate the I-Q control signal.
12. The method of claim 11, wherein said adjusting step (d) includes the steps of : (dl) if the mode selection signal indicates the differential-coded symbol mapping, passing the (I, Q) value as the modulation function ; (d2) if the mode selection signal indicates the Gray-coded symbol mapping and the I-Q control signal indicates '0', passing the (I, Q) value as the modulation function ; and (d3) if the mode selection signal indicates the Gray-coded symbol mapping and the I-Q control signal indicates '1', passing an (Iad'adj) value as the modulation function, wherein the (Iadj, Qadj) value is obtained from the (I, Q) value by using a formula given by (Iadj, Qadj) = (-Q, -I).
13. An apparatus constructed and arranged substantially as herein described with reference to or as shown in Figures 2 to 4 of the accompanying drawings.
14. A method substantially as herein described with reference to or as shown in Figures 2 to 4 of the accompanying drawings.