WO2007108474A1

WO2007108474A1 - Wireless transmitting device and wireless transmitting method

Info

Publication number: WO2007108474A1
Application number: PCT/JP2007/055686
Authority: WO
Inventors: Zhanji Wu; Jifeng Li
Original assignee: Matsushita Electric Industrial Co., Ltd.
Priority date: 2006-03-20
Filing date: 2007-03-20
Publication date: 2007-09-27
Also published as: CN101043484A

Abstract

A wireless transmitting device and a wireless transmitting method are provided to simultaneously transmit a wireless signal to plural users with improved throughput. A transmitting device (100) is comprised of a multiple-user-bit stream forming unit (130) for allocating each structuring bit as a user usable bit in accordance with the degree of easiness for correcting errors per structuring bit in a unit bit train indicated by one symbol and situations in each user channel and for placing elements of parallel data series addressed to plural users in order following the allocation to form a bit stream addressed to the users; and a modulating unit (150) for forming a modulating signal in accordance with the unit bit train of the bit stream.

Description

Specification

Radio transmission apparatus and radio transmission method

Technical field

TECHNICAL FIELD [0001] The present invention relates to a wireless transmission device and a wireless transmission method, and more particularly to a wireless transmission device and a wireless transmission method to which a higher-order modulation scheme is applied.

Background art

[0002] Binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK) are modulation techniques that are commonly used in the current communications field. In order to further improve the efficiency of frequency spectrum utilization, High-order modulation techniques such as 8PSK, 16QAM, 64QAM, and 256QAM have also been introduced (see Non-Patent Document 1). On the constellation diagram of the higher-order modulation system, each modulation symbol is composed of several bits. For example, 8PSK, 16QAM, 64QAM, and 256QAM modulation symbols are composed of 3, 4, 6, and 8 bits, respectively. In order to reduce the bit error rate, Golay mapping is usually used as the mapping from bit strings to modulation symbols! Figure 1 shows the constellation map for 8PSK Golay coding. Each constellation point is represented by 3 bits (a a a), and the constellation is

one two Three

It is on the unit circle. Figure 2 shows the constellation map for 16QAM Golay coding, where each constellation point is represented by 4 bits (a a a a). Where X

1 2 3 4

The component of the axis is a a, the component of the Y axis is a a, and the energy of the constellation diagram

1 2 3 4

The normalization element is c = 1Z10. Figure 3 is 64QAM Golay (Golay) _n each cons Talay Chillon point is a diagram showing a Konsuta rate Chillon map encoding 6 bits (aaaaaa)

1 2 3 4 5 6 However, the X-axis component is a a a and the Y-axis component is a a a.

1 2 3 4 5 6

The energy normalization factor of the cision diagram is C = lZ42. In Golay mapping, the protection capability of each constituent bit in the bit string is uneven, and there are bits with strong protection capability and bits with weak protection capability.

[0003] In the conventional multi-user transmission technique, each modulation symbol is information of the same user. In actual operation, the quality of a certain user's channel is good, and the channel of each user is also unequal, as the channel of some users has a large signal-to-noise ratio (SNR). There is a possibility. For example, in the case of multi-user transmission in wireless communication, the user is close to the base station and the line attenuation is small, so the SNR is generally high, but at the cell edge, the user is generally low in SNR because the line attenuation is large. A perspective problem arises. Signals from users with inferior channels are often good channels! Users' signal strength is subject to severe interference. This type of problem is usually solved by power control. In other words, the base station increases the signal-to-interference noise ratio and improves the performance by increasing the transmission power to users with inferior channels.

FIG. 4 is a diagram illustrating a conventional multi-user high-order modulation multiple access transmission process. Each user independently performs coding, interleaving, parallel / serial conversion, high-order modulation, and power control, and then performs multiple access. Usually, access methods include time division (TDM), frequency division (FDM), orthogonal frequency division multiplexing (OFDM), code division (CDM), etc.

FIG. 5 is a diagram illustrating a conventional multi-user higher order modulation multiple access release reception process. Each user independently performs multiple access disconnection, higher-order demodulation, normal / serial conversion, Dinter leave, and decoding.

Non-Patent Document 1: Stephen G. Wilson, Digital Modulation and Coding, Prentice-Hall, 19 96

Disclosure of the invention

Problems to be solved by the invention

[0006] However, if the transmission power to a user with a poor channel is simply increased as in the conventional technique, the performance of the user is improved, but an extra multiple access is made even for a user with a good channel. Improve overall throughput efficiency, especially in code division multiple access (CDMA) systems that are subject to multiple access interference (MAI) constraints, because they cause interference, improve channel quality, and reduce user performance It is not possible. In addition, power control technology is not organically coupled with higher-order modulation technology, so there is a limit to improving performance.

[0007] An object of the present invention is to provide a wireless transmission device and a wireless transmission method for simultaneously transmitting wireless signals to a plurality of users, and to improve the throughput. Means for solving the problem

[0008] The wireless transmission device of the present invention assigns each constituent bit as a user's used bit based on the ease of error for each constituent bit of the unit bit string represented by one symbol and the channel status of each user, A bit stream forming means for forming a bit stream addressed to the plurality of users by arranging elements of parallel data sequences addressed to the plurality of users in an order in accordance with the unit bit string of the bit stream. And a modulation means for forming the structure.

[0009] The wireless transmission method of the present invention includes the steps of assigning each constituent bit as a user use bit based on the ease of error for each constituent bit of the unit bit string represented by one symbol and the channel status of each user; Arranging elements of parallel data sequences addressed to a plurality of users in an order according to the assignment, forming a bitstream addressed to the plurality of users; and a modulation signal corresponding to the unit bit string of the bitstream Forming a step.

The invention's effect

[0010] According to the present invention, it is possible to provide a wireless transmission device and a wireless transmission method for simultaneously transmitting a wireless signal to a plurality of users, which can improve the throughput.

Brief Description of Drawings

[0011] [FIG. 1] A diagram showing a constellation map of 8PSK Golay coding.

[Figure 2] Diagram showing the constellation map for 16QAM Golay coding

[Figure 3] Diagram showing constellation map of 64QAM Golay coding

[Fig.4] Diagram showing conventional multi-user high-order modulation multiple access transmission process

[FIG. 5] A diagram showing a conventional multi-user higher order modulation multiple access release reception process.

[Fig. 6] Diagram for explaining the method of distinguishing strong and weak constituent bits in high-order modulation

[Fig.7] Bit errors in 16-bit QAM Golay coded a-bit and a-bit AWGN channels

1 2

Diagram showing computer simulation curve of rate

[Figure 8] 64QAM Golay coded a-bit, a-bit, a-bit AWGN channel

one two Three

Figure showing a computer simulation curve of bit error rate [Figure 9] Diagram showing the main configuration of a transmitter that performs multi-access transmission in the multiuser higher-order modulation method

FIG. 10 is a diagram showing the main configuration of a transmission apparatus that performs multiple access transmission of 16-user 16QAM modulation method.

圆 11] Diagram for explaining multi-user higher-order modulation bitstream allocation multiplexing method [Fig. 12] Diagram for explaining 2-user 16QAM modulation bit-stream allocation multiplexing method 圆 13] Multiple access of multi-user higher-order modulation method Diagram showing the main configuration of a receiving device that performs cancellation reception

FIG. 14 is a diagram showing the main configuration of a receiving apparatus that performs multiple-user disconnection reception of 16QAM modulation of two users

圆 15] Diagram for explaining multi-user high-order modulation symbol transmission by time division method [Fig. 16] Diagram showing a general image of multi-user high-order modulation symbol transmission by time division method

圆 17] Used to explain multi-user higher-order modulation symbol transmission by frequency division

圆 18] Diagram for explaining reception of multi-user higher-order modulation symbol by wave number division method [Fig. 19] Diagram showing a general image of multi-user higher-order modulation symbol transmission by frequency division method

[Fig.20] Diagram showing a general image of multi-user higher-order modulation symbol reception by frequency division method

[Fig. 21] Diagram for explaining multi-user higher-order modulation symbol transmission by OFDM scheme [Fig. 22] Diagram for explanation of multi-user higher-order modulation symbol reception by OFDM scheme [Fig. 23] Multi-user higher-order modulation by OFDM scheme Diagram showing a general image of symbol transmission

[Fig.24] Diagram showing a general image of multi-user higher-order modulation symbol reception using OFDM.

[Fig.25] Diagram for explaining code division of multi-user higher-order modulation symbols by code division method FIG. 26 is a diagram for explaining despread reception of multi-user high-order modulation using a code division scheme.

[Fig.27] Diagram showing a general image of multi-user higher-order modulation symbol transmission by code division method

[Fig.28] Diagram showing a general image of multi-user higher-order modulation symbol reception using OFDM

[Fig.29] Code division of multi-user higher-order modulation symbol by OFDMA 'Diagram used to explain frequency division

[FIG. 30] A diagram showing reception of multi-user higher-order modulation symbols by the OFDMA system

[Fig.31] Diagram showing time slot format (data / pilot) arrangement of multiuser higher-order modulation symbols by time division multiple access

[Fig.32] Diagram showing time slot format (data / pilot) arrangement of multiuser higher-order modulation symbols by OFDM

[Fig.33] Diagram showing time slot format (data / pilot) arrangement of multi-user higher-order modulation symbols using CDMA

FIG. 34 is a simple signaling flowchart of a multiuser high-order modulation method. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that, in the embodiments, the same components are denoted by the same reference numerals, and the description thereof is omitted because it is redundant. It should be pointed out that the embodiments described here are for illustrative purposes only and do not limit the scope of the invention. The various numerical values described here are not intended to limit the present invention, and these numerical values can be appropriately changed as necessary by those skilled in the art. In this specification, it is obvious that the present invention can be applied to communication systems employing other M-QAM (M = 2m) modulation schemes such as 64QAM.

[0013] (Embodiment 1)

Hereinafter, in each of the embodiments, high-order modulation is performed for each of the constituent bits constituting the bit string represented by the non-uniform bit protection capability (that is, the signal point (symbol) on the constellation) This section describes the technology that effectively links to the ease of error) and the channel conditions of multiple users.

[0014] On the transmission side, first, it is necessary to clarify the protection capability of each component bit for each high-order modulation using different code mapping methods. A simple method is proposed to deal with this.

That is, all pairs of signal points on the constellation that minimize the Euclidean distance are found on the high-order modulation constellation map, and the sum of the hamming distances of each bit in the constellation is calculated. Since the sum of the protection capability of each constituent bit and its sum and ming distance are inversely proportional, it is defined as a weak bit if the sum is large and a strong bit if the sum is small. This is because the symbol error rate of high-order demodulation is mainly determined by the minimum Euclidean distance, and under the condition where the symbol error rate is the same, the bit error rate of each bit is a force proportional to its average and ming distance. is there. For this reason, the bit error rate of each constituent bit of the bit string represented by the signal point of the higher-order modulation is the sum of the ming distance and the sum of the constituent bits in the signal point pair on the constellation with the smallest Euclidean distance. Is directly proportional to On the transmitting side, the CQI (channel quality indicator) information of the different users fed back is used, and the channel conditions are good! They are divided into users (called strong users) and users with poor channel conditions (called weak users). For example, a user with a large signal to interference and noise ratio (SINR) is called a strong user, and a user with a small signal to interference and noise ratio is called a weak user. Alternatively, the user who fed back the ACK signal is classified as a strong user, and the user who fed back the NACK signal is classified as a weak user. If the user feeds back a NACK signal, this user is clearly defined as a weak user because his channel is bad and accurate decoding is not possible. For weak users, it is necessary to improve performance, so weak user data is mapped to strong bit positions in high-order modulation, and strong user data is mapped to weak bit positions in high-order modulation. . As described above, since information of a plurality of users is included in one high-order modulation symbol, multiple access interference (MAI) to strong users of weak users does not occur.

[0016] On the receiving side, for different users, corresponding multiple access cancellation processing, selective high-order modulation demodulation, parallel / serial conversion, demultiplexing, deinterleaving, and decoding are performed. Determination data can be obtained. The difference from the conventional method is that in order to perform high-order demodulation of a specific user, it is only necessary to demodulate the bit of that user. In other words, the corresponding LLR (Log Likelihood Ratio) is obtained using the Max-Log-Map decoding algorithm, and it is not necessary to demodulate all mapped bits on one modulation symbol! It is.

[0017] In this way, high-order modulation techniques can be effectively matched to the channel conditions of multi-user transmission, the performance of users with poor transmission channels can be improved, and the overall throughput efficiency in multi-user transmission can be improved.

[0018] FIG. 6 is a diagram for explaining a method of distinguishing strong and weak constituent bits in high-order modulation. First, a high-order modulation constellation map is determined based on a certain coding rule, and the minimum Euclidean distance between signal points on the constellation is found. Next, find a pair of signal points on all constellations corresponding to the minimum distance. Then, the sum of the Hamming distances is calculated for all the constituent bits for these pairs and arranged in order. When the obtained sum is small, it indicates that the protection capability is strong, so it is called a strong bit. When the obtained sum is large, it indicates that the protection capability is weak, so it is called a weak bit. Actually, the bit error rate of each component bit and the sum of its ming distances approximate a linear direct proportional relationship. The bit error rate increases as the sum of the Hamming distances increases.

(1) For example, in the 8PSK Golay coding constellation map shown in FIG. 1, the minimum Euclidean distance between the constellation signal points is 2 sin (w Z8). And all constellation signal point pairs with the smallest Euclidean distance are (11 1, 110), (110, 100), (100, 101), (101,001), (001,000), (000,010) , (010,011), (011,111).

[0020] For the a bit that is the first constituent bit in the bit string represented by the signal point, (101,001)

1

And (011, 111) changes between signal points. Since the Hamming distance is 1 for each of the two pairs, the sum of the Hamming distances is 2 for the a bit. a bit

For 1 2, a change occurs at (110,100) and (000,010), so that the sum of the distances is 2. The a bit is changed in (111, 110), (100, 101), (001,000), (010,011).

Three

Therefore, the sum of Hamming distances is 4. Therefore, a bit, a bit, a The ratio of bit sum and ming distance sum is 2: 2: 4 = 1: 1: 2. Ie, a bit, a bit

1 2 is a strong bit and a bit is a weak bit. And bit error of a bit, a bit, a bit

3 1 2 3

The rate should be directly proportional to 1: 1: 2.

(2) In the denormalized constellation map of 16QAM Golay coding shown in FIG. 2, the minimum Euclidean distance between constellation points is 2. The X-axis component is a a

1 2

Y-axis components are represented by a a and they are completely symmetric.

3 4

[0022] Therefore, when looking at all the constellation point pairs that minimize the Euclidean distance on the X axis, there are three pairs (11,10), (10,00), and (00,01). For the a bit, use (10,00)

1

A change occurs, so the sum of the Hamming distances is 1. For the a bit, (11, 10) and (

2

00,01), and the sum of the Hamming distances is 2. Similarly, a bit of Hami

The sum of 3 distances is 1, and the sum of a-bit Hamming distances is 2. Therefore, a bit, a

The ratio of the sum of the Hamming distances of 4 1 2 bits, a bits, and a bits is 1: 2: 1: 2. That is, a

3 4 1 bit, a bit is a strong bit, a bit, a bit is a weak bit. And a bit, a bit

The bit error rate of 3 2 4 1 2, a bit, a bit should be directly proportional to 1: 2: 1: 2. Also figure

3 4

7 includes 16 bits in the 16QAM Golay coded a bit and a bit in the AWGN channel.

1 2

The computer simulation curve of an error rate is shown. A bit, a bit depending on symmetry

3 The 4-bit error rate is equal to a and a bits, respectively. Bit error rate curve is a judgment method

1 2

Has proved its accuracy.

[0023] (3) In the denormalized constellation map of 64QAM Golay coding shown in Fig. 3, the minimum Euclidean distance between constellation points is 2. X axis component is a a a

one two Three

The Y-axis components are expressed as a a a, and they are completely symmetric.

4 5 6

[0024] Therefore, first, looking at all the constellation point pairs that minimize the Euclidean distance on the X axis, (111,110), (110,100), (100,101), (101,001), (001,000), (000,010) , (01 0,011) are 7 pairs. For the a bit, a change occurs at (101,001), so Hammin

1

The sum of the distances is 1. For the a bit, a change occurs between (110, 100) and (000,010).

2

Therefore, the sum of the Hamming distances is 2. For the a bit, (111,110), (100,101), (001,

Three

000) and (010,011), and the sum of the Hamming distances is 4. Therefore, the ratio of the sum of a bit, a bit, a bit, and ming distance is 1: 2: 4. Similarly, a bit The ratio of the sum of a bit, a bit, and ming distance is also 1: 2: 4. Therefore, a bit, a

5 6 1 2-bit, a-bit, a-bit, a-bit, a-bit Nominal distance of it, 1: 2: 4: 1: 2: 4

3 4 5 6

It becomes. Therefore, a bit, a bit are strong bits, a bit, a bits are weak bits, a

1 4 3 6 2 bits, a bit is in the middle. a bit, a bit, a bit, a bit, a bit, a bit

The bit error rate of 5 1 2 3 4 5 6 should be directly proportional to 1: 2: 4: 1: 2: 4. Figure 8 also shows the bit error rate for a 64-bit Q-A M Golay coded a-bit, a-bit, and a-bit AWGN channels.

one two Three

The computer simulation curve of is shown. A bit, a bit, a bit depending on symmetry

The bit error rate of 4 5 6 is equal to a bit, a bit, and a bit, respectively. The bit error rate curve is

one two Three

Similarly, prove the accuracy of the judgment method.

[0025] It should be pointed out that the above method can be similarly applied to a high-order modulation method based on arbitrary code mapping using a high-order modulation method using Golay codes. .

[0026] FIG. 9 is a diagram illustrating a main configuration of a transmission apparatus that performs multiple access transmission of a general multiuser higher-order modulation method according to the present embodiment. As shown in the figure, transmitting apparatus 100 includes N encoders 110-1 to 110-N that perform channel code processing on the source data sequences addressed to each user, and source data sequences after channel coding processing. N interleavers 120-1 to N that perform interleaving processing, multi-user bitstream formation unit 130, serial Z parallel (SZP) conversion unit 140, modulation unit 150, and multiple access processing unit 160 Have

[0027] The multi-user bitstream forming unit 130 inputs the source data sequence of each user, and also inputs the modulation scheme information and the channel state information of each user in the modulation unit 150 at the subsequent stage. The multi-user bitstream forming unit 130 assigns each constituent bit to the user based on the ease of error of the constituent bits of the bit string represented by one symbol in the modulation scheme represented by the modulation scheme information and the channel status of each user. Assigned as used bits. The multi-user bit stream forming unit 130 uses the source data sequences addressed to a plurality of users, and arranges the bits that are the elements of the source data sequence in the order according to the above allocation, thereby Form. In other words, in the multi-user bitstream, when attention is paid to each bit string (corresponding to a bit string representing one symbol) of the same number as the order corresponding to the modulation scheme information, each of the component bits is assigned the above-described allocation. It is configured by arranging the bits of the source data series addressed to the user according to the situation.

[0028] The serial Z parallel (SZP) conversion unit 140 converts the serial multi-user bit stream into the same number of parallel streams as the order corresponding to the modulation scheme information. Combining the bits of these multiple parallel streams results in a bit string represented by one symbol.

[0029] Modulating section 150 performs modulation corresponding to a symbol represented by a bit string obtained by combining the bits of the parallel stream to form a modulated signal. The multiple access processing unit 160 performs multiple access and transmits a modulation signal.

[0030] That is, the sequence U is obtained by first channel interleaving the source data sequence D (ie [1, N]) of each user with the channel code. Then, the multi-user bit data stream sequence V is obtained according to the bit stream allocation multiplexing method of multi-user higher-order modulation. Furthermore, if serial / parallel conversion is performed, corresponding high-order modulation mapping can be performed, and a modulation symbol sequence S can be obtained. Then, multiple access is performed and a multiple access transmission sequence Q is obtained. Normal access schemes include time division (TDM), frequency division (FDM), orthogonal frequency division multiplexing (OFDM), and code division (CDM), all of which are applicable to this embodiment. FIG. 10 shows one special example, and is a diagram showing a main configuration of a transmission apparatus that performs multi-access transmission of a 16-QAM modulation method for two users. As shown in the figure, in transmitting apparatus 100A, source data sequences D 2]) of users 1 and 2 are first encoded by channel encoder and interleaver in encoder 110, respectively.

Sequence U is obtained by performing channel interleaving at 120. Then, the multi-user bit data stream sequence V is obtained by the 16QAM modulation bit stream allocation multiplexing method of both users. Further, if serial / parallel conversion is performed and divided into four parallel branches (corresponding to the number of bits represented by one symbol in 16QAM), the corresponding 16QAM modulation mapping can be performed to obtain a modulation symbol sequence S. A multi-access transmission sequence Q is obtained by performing a main connection.

FIG. 11 is a diagram for explaining a general multiuser high-order modulation bitstream allocation multiplexing method according to the present embodiment.

[0032] First, on the transmission side, the signal-to-interference noise ratio, ACK / NACK feedback signal, etc. Which channel state information that different users have fed back is received. Then, the users are arranged in order based on the channel state as feedback information. Furthermore, user D (i≡ [1, N]) is grouped according to the channel state. For example, if the signal-to-interference noise ratio is high, the user channel is a strong user group, the signal-to-interference noise ratio is low, and the user channel is a weak user group. Alternatively, the user who fed back the ACK signal is set to the strong user group, and the user who fed back the NACK signal is set to the weak user group.

[0033] Next, the strong and weak users indicated by the channel status are assigned to the strong and weak bits of the higher-order modulation. In other words, strong user data is mapped to the positions of weak bits in high-order modulation, and weak user data is mapped to positions of strong bits in high-order modulation. After receiving the bit stream (source data sequence) of each user, the final multi-user bit stream sequence V is formed by multiplexing based on the matching result (allocation result) described above.

FIG. 12 is a diagram for explaining a bit stream allocation multiplexing method for 2-user 16QAM modulation in particular.

First, channel feedback information of users 1 and 2 is received. Assuming that the signal-to-interference noise ratio of user 1 is smaller than user 2, user 1 is a weak user and uses the strong bits (a bit, a bit) of the 16QAM modulation symbol. User 2 is a strong user, 16Q

13

Use the strong bit (a bit, a bit) of the AM modulation symbol. User 1 bitstream

twenty four

Receive sequence U and user 2 bitstream sequence U, and set both sequence lengths to N

1 2

Then, a bit stream sequence V multiplexed by the following equation is obtained.

[Equation 1] v (4k) = ul (k), v (4k + 2) = ul (k + N / 2),

v (4k + l) = u2 (k), v (4k + 3) = u2 (k + N / 2)

k e [0, N / 2-1]

[0036] Next, among multi-user higher-order modulation bitstream allocation multiplexing methods, in particular, 8PS

Examples for K, 16QAM, and 64QAM are described below.

[0037] (Example 1)

There are two users A and B, and user A is located at the cell edge and has a signal-to-interference / noise ratio. Suppose that user B, who is low, is close to the base station and has a high signal-to-interference and noise ratio. The base station defines user A as a weak user and user B as a strong user based on the signal-to-interference noise ratio of each user. User A's binary data block after encoding and interleaving in the channel is {a}, ie [0, N-1], and binary data block after encoding and interleaving in user B's channel is { b}, ie [0, N—l]. N is a block length, and N = 0 (mod3). The multi-user bit data stream after allocation multiplexing is {c}, ie [0, 2N-1]. When Golay mapping 8PSK is adopted, the multiplexing method represented by the following formula can be used.

[Equation 2]

2N

N

= Look at me. , ^

[0038] Then, {c} is mapped to an 8PSK constellation diagram with one set of 3 bits, modulated, and a modulated signal is transmitted.

[0039] (Example 2)

Assume that there are two users A and B, and user B is located at the cell edge and has a low signal-to-interference noise ratio. User B is close to the base station and has a high signal-to-interference noise ratio. The base station defines user A as a weak user and user B as a strong user based on the signal-to-interference noise ratio of each user. User A's binary data block that has been interleaved with channel code 匕 is {a}, i [0, N−l], and user B's binary data block that has undergone channel coding and interleaving is {b} }, i≡ [0, N—l]. N is a block length, and N = 0 (mod2). The multi-user bit data stream after the allocation multiplexing is {c}, i≡ [0, 2N−1]. When 16Q AM of Golay mapping is adopted, the multiplexing method represented by the following formula can be used.

[Equation 3] 4k = ^a k ^C 4k + 2 = ^a k + NI2

^C 4k + 1 = b _k , ^C 4k + 3 = only ifc + W / 2

[0, 1]

[0040] Then, {c} is mapped to a 16QAM constellation diagram with a set of 4 bits, modulated, and a modulated signal is transmitted.

[Example 3]

_Three user A, and B and C are present, the signal to interference noise ratio is assumed to be higher in that order. Based on the signal-to-interference / noise ratio of each user, the base station defines user A as a weak user, user C as a strong user, and user B as an intermediate user. Assume that the channel codes 匕 and interleaved binary data blocks of users A, B, and C are {a}, {b}, {c}, i £ [0, N−1], respectively. N is a block length, and N = 0 (mod2). The multiuser bit data stream after the multiplexing is {d}, ie [0, 3N-1]. When Goreimatsu 64QAM is adopted, the multiplexing method represented by the following formula can be used.

Dk = ° k «t ₊ 3 = ^a k ₊ N / 2 ^d M =

^ k + l ⁼ ° k + NI2, ^{= C} k, ^ k + 5 ^{= C} k + N I2

N

* e [o, † -i]

[0042] Then, {d} is mapped to a 64QAM constellation diagram with a set of 6 bits and modulated, and a modulated signal is transmitted.

Next, the receiving side will be described. FIG. 13 is a diagram illustrating a main configuration of a receiving apparatus that performs multiple access cancellation reception of a general multiuser higher-order modulation method according to the present embodiment. As shown in the figure, the receiving apparatus 200 includes a multiple access cancellation processing unit 210, a demodulation unit 220, a parallel Z-serial (PZS) conversion unit 230, and a decoding unit, which perform a cancellation process corresponding to the multiple connection made on the transmission side. Interrino 240 and decoder 250.

[0044] Demodulation section 220 outputs demodulated data related only to the configuration bits assigned to the own apparatus on the transmission side to PZS conversion section 230. In FIG. 13, the demodulator 220 performs other configuration bits, That is, a configuration is shown in which demodulation is performed even if the configuration bits assigned to another user on the transmission side are performed, and the demodulated data is discarded. In the demodulator 220, the transmission side is transmitted to another user. It is not always necessary to perform demodulation based on the configuration bits assigned in (1). That is, the demodulator 220 only needs to demodulate the configuration bits assigned to the own device on the transmission side.

[0045] When a plurality of configuration bits are assigned to the own device on the transmission side! /, The PZS conversion unit 230 converts a plurality of parallel demodulated data corresponding to the plurality of configuration bits into serial data. Data series. This data sequence corresponds to the source data sequence after interleaving on the transmitting side.

[0046] The Dinterleaver 240 performs Dinterleave corresponding to the interleave on the transmission side for the data series from the PZS conversion unit 230. Decoder 250 performs a decoding process corresponding to the encoding on the transmission side on the data sequence after the Dinterleave.

[0047] That is, each user terminal i receives the channel output sequence hat [Q], performs multiple access cancellation processing according to the transmission side, and obtains sequence hat [S]. Then, selective high-order demodulation and parallel / serial conversion are performed to obtain a sequence hat [U]. Then, Dinter leave and decryption are performed to finally obtain a judgment data sequence hat [D]. In the conventional method, it was necessary to output the LLR (Log Likelihood Ratio) value of all the bits on one higher order modulation symbol. In the power selective high order demodulation, only the log likelihood ratio of the bit of the user concerned is obtained. Other users' bits that just need to be output can be discarded. Where hat corresponds to the symbol Γ) in the figure

[0048] FIG. 14 is a diagram showing a main configuration of a receiving apparatus that specifically performs two-user 16QAM modulation multiple access cancellation reception. In the second embodiment, user A using 16QAM modulation is subjected to multiple access cancellation corresponding to the transmission side, selective higher-order demodulation, parallel / serial conversion, deinterleaving, and decoding. Determination data can be obtained.

[Equation 5]

^LLR (') = ^Ln -min i- 1 | 7-s (kf} — _m i _n {— || r-S () | ² } £ [l, 4], G [l, 16] [0049] Note that the complex vector Y = {y, y} is an additive white gauss of the higher-order modulation symbol at a certain time.

I Q

Represents the uss noise (AWGN) channel output value. The complex vector S (k) = (S (k), S (

I Q

k)} represents the kth constellation point on the 16QAM constellation diagram.

[0050] Assuming Y = S (t) + N, N represents complex Gaussian noise and S (t) is the transmission constellation point.

[0051] II Y-s (k) II represents the Euclidean distance between γ and s (k), and is expressed as follows.

[Equation 6]

|-S (k) f = (y ₇ -S ₇ (k)) ² + ( _e -S _Q (k) f

[0052] b (k, j) is the j-th bit value represented by the k-th constellation point S (k), b

(k, j) = represents {0, 1}.

[0053] Low complexity! As a selective demodulation technique, user A demodulation uses LLR (b) and LLR (b

13

), LLR (b) and LLR (b) need only be calculated for user B demodulation.

twenty four

(Embodiment 2)

Multiuser high-order modulation is widely applied as multiuser high-order modulation under various multiple access technologies. Conventional multiple access technology allocates multiple users orthogonally to resources in the time domain, frequency domain, code domain, spatial domain, etc., and multi-order higher-order modulation is based on this. This is a proof of the allocation of higher-order modulation bit resources.

[0055] Embodiment 2 is an embodiment in which multi-user higher-order modulation that assigns modulation bit resources based on the time domain is applied to a time division multiple access (TDMA) scheme.

[0056] On the transmission side, each modulation code in the time domain does not represent information of one user, but represents information of a plurality of users. The receiving terminal of each user must receive all modulation symbols including the user's information in the time domain as a multiple access cancellation process. Assuming that the coding length of each user is the same, only a part of the configuration bits are allocated to each user in the bit string represented by each modulation symbol. More modulation symbols and more A lot of time diversity is possible.

FIG. 15 is a diagram for explaining multiuser higher-order modulation symbol transmission by a time division scheme. The modulation symbol on time slot 1 represents user 1 and user 2 information. Similarly, the modulation symbols in timeslots 3 and 5 also represent user 1 and user 2 information. On the other hand, time slots 2, 4, and 6 represent user 3 and user 4 information. The receiving terminal of user 1 must receive the modulation symbols of the related time slots 1, 3, and 5 as the multiple connection cancellation process in the multiple connection cancellation processing section 210 of FIG. Similarly, the receiving terminal of user 2 receives the modulation symbols of the relevant time slots 1, 3, and 5, and the receiving terminals of users 3 and 4 must receive the modulation symbols of the relevant time slots 2, 4, and 6. Goodbye

FIG. 16 is a diagram showing a general image of multiuser higher-order modulation symbol transmission by the time division scheme. In FIG. 16, the high-order modulation symbol of the i-th time slot is Q, and there are N time slots in total. If the set of all users is U, u is a subset of U and u CU. If the number of users in u is m, the total number of users is L, and the order of higher-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 7]

\ ≤m _i ≤ miniM, ί

[Embodiment 3]

Embodiment 3 is a frequency division multiple access (FDMA) scheme! / ヽ is a multi-user higher-order allocation that allocates modulation bit resources on the basis of the frequency domain as compared to the orthogonal frequency division multiple (OFDM) scheme. It is an embodiment which applies modulation.

[0060] On the transmission side, each modulation symbol on each frequency band (subcarrier) does not represent information of a single user, but represents information of a plurality of users. The receiving terminal of each user performs coherent demodulation (in the case of FDM A) or FFT (in the case of OFDM) on the frequency domain as a process of multiple access cancellation, and all frequency domains including sub-user information (sub The modulation symbol on the carrier) is received. Assuming that each user has the same coding length, each modulation symbol allocates some bits to each user. Therefore, compared to conventional FDMA or OFDM, each user occupies more frequency bands (subcarriers) and more frequency diversity is possible.

FIG. 17 is a diagram for explaining multiuser higher-order modulation symbol transmission by the frequency division scheme. The modulation symbol on subcarrier 1 (input of the uppermost multiplier in the figure) represents user 1 and user 2 information. Similarly, the modulation symbols of subcarriers 3 and 5 also represent user and user ₂ information. On the other hand, the modulation symbols of subcarriers 2, 4, and 6 represent information of user 3 and user 4.

FIG. 18 is a diagram for explaining reception of multiuser higher-order modulation symbols by the frequency division scheme. The receiving terminal of user 1 receives the modulation symbols of the related subcarriers after performing coherent demodulation on subcarriers 1, 3, and 5 as a multiple access cancellation process before higher order demodulation. Similarly, the receiving terminal of user 2 receives modulation symbols of related subcarriers 1, 3, and 5, and the receiving terminals of users 3 and 4 receive modulation symbols of related subcarriers 2, 4, and 6. There must be.

FIG. 19 is a diagram showing a general image of multi-user higher-order modulation symbol transmission by the frequency division scheme. In FIG. 19, the higher-order modulation symbol transmitted on the i-th carrier f is Q N]), and there are N carriers in total. The entire set of users

Assuming U, u is a subset of U and u CU. If the number of users in u is m, the total number of users is L, and the order of higher-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 8]

\ ≤m _i ≤ mmiM, L)

FIG. 20 is a diagram showing a general image of multi-user higher-order modulation symbol reception by the frequency division scheme. For user j, first define the function F (u) as follows:

[Equation 9] [0065] When F (u) = 1, the circuit is on, otherwise it is off. The terminal of user j performs coherent demodulation on all subcarriers f including the user information, and obtains its higher-order modulation symbol sequence QN]).

[0066] FIG. 21 is a diagram for explaining transmission of multiuser higher-order modulation symbols by the OFDM scheme. The modulation symbol on subcarrier 1 (the uppermost input to the IFFT section in the figure) represents user 1 and user 2 information. Similarly, the modulation symbol of subcarrier 3 also represents user 1 and user 2 information. On the other hand, the modulation symbols of subcarriers 2 and 4 represent the information of users 3 and 4. The transmission sequence S (t) is obtained by IFFT calculation and parallel / serial conversion of the higher-order modulation symbol sequence.

FIG. 22 is a diagram for explaining reception of multiuser higher-order modulation symbols by the OFDM method. The receiving terminal of user 1 selects the output signal number on subcarriers 1 and 3 after serial / parallel conversion and FFT calculation as the multiple connection cancellation process before higher order demodulation.

[0068] FIG. 23 is a diagram showing a general image of multi-user higher-order modulation symbol transmission by the OFDM scheme. In FIG. 23, the higher-order modulation symbol transmitted on the i-th carrier f is Q N]), and there are N subcarriers in total. Then, the transmission sequence S (t) is obtained by converting Q into IFFT calculation and parallel / serial conversion. If the set of all users is U, u is a subset of U and u cu. If the number of users in u is m, the total number of users is L, and the order of higher-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 10]

\ ≤m _j ≤ mini, Z}

[0069] FIG. 24 is a diagram showing a general image of multi-user higher-order modulation symbol reception by the OFDM method. For user j, first define the function F (u) as follows:

[Equation 11]

"i [0070] When F (u) = 1, the circuit is connected, otherwise it is disconnected. User j's terminal selects the output signal numbers on all subcarriers including the user's information after serial / parallel conversion and FFT computation as the multiple connection release process before higher order demodulation.

[Embodiment 4]

The fourth embodiment is an embodiment in which multi-user higher-order modulation that assigns modulation bit resources based on the code domain is applied to the code division multiple access (CDMA) system.

[0072] On the transmission side, each modulation symbol on the code area represented by the spreading code does not represent information of a single user, but represents information of a plurality of users. Each user's receiving terminal performs a despreading process by matched filtering using a corresponding spreading code in the code area as a multiple access cancellation process, and all modulation symbols on the code area including the user information are received. Receive and take the average value.

FIG. 25 is a diagram for explaining the code division of multiuser higher-order modulation symbols by the code division scheme. Modulation symbol Q represents user 1 and user 2 information,

12

Spread by using spreading code C of user 1 and spreading code C of user 2 respectively.

1 2

Add on the area. Like normal CDMA systems, C and C are orthogonal, and C, C>

1 2 1 2

= 0. <X, y> represents obtaining the correlation value between the sequences X and y. In this example, the information of user 1 and user 2 is also orthogonal in the code area, and it can be understood that the following relationship holds.

[Equation 12]

<Q _{1 2} CQ _{1 2} C ₂ > = IQ i 2 I ² Q i 2 <C _1? C ₂ ) = 0 [0074] Similarly, the modulation symbol Q represents the information of user 3 and user 4, respectively. User 3

34

Spreading code PN and user 4 spreading code PN, and added in the time domain.

3 4

It is. It can be seen that the information of users 1, 2, 3, and 4 is orthogonal two by two on the code area. For example, the correlation value in the code area of user 1 and user 3 information satisfies the following relationship. [Equation 13]

(Q ^ C i, Q _{3 4} C ₂ ) = IQ _{1 2} I ² Q 4 <C _1; C ₂ > = 0 where Q * is the complex conjugate of Q.

34 34

FIG. 26 is a diagram for explaining despread reception of multi-user high-order modulation using a code division scheme. For user 1, the spreading codes C and C are used to

1 2

Perform tide filtering (despreading process) and take the average value. Since the transmitted modulation symbol is obtained from the orthogonality principle of the code area, the multiple disconnection process shown in FIG. 13 is completed.

FIG. 27 is a diagram illustrating a general image of multi-user higher-order modulation symbol transmission by the code division scheme. In FIG. 27, the high-order modulation symbol transmitted to the i-th user set u is Q (i≡ [1, N]), and there are N user sets in total. Entire user

U1

If the set of is U, u is a subset of U and u CU. Assuming that the user sets are orthogonal, the following relationship holds.

[Equation 14]

M- Π Μ. = Φ, V / ≠ j

[0077] If the number of users in u is m, the total number of users is L, and the order of high-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 15]

l≤m _i ≤ mm {M, L}

[0078] For Q, spreading is performed using all spreading codes [C,-, C] belonging to the user set u.

ui 1 ui (l) ui (mi)

And add in the time domain. In this way, spreading and adding are performed for all N user sets, and a transmission sequence S (t) is obtained.

FIG. 28 is a diagram showing a general image of multiuser higher-order modulation symbol reception by the OFDM scheme. For user j, assume jE _U. User j's terminal uses all the spreading codes [C, ---, C] belonging to user set u as the multiple access release process before performing higher-order decoding, respectively, and performs matched filtering processing (despreading process). ) Line ui (l) ui, mi) Take the average value.

[0080] (Embodiment 5)

In Embodiment 5, multi-user higher-order modulation is performed for orthogonal frequency division 'code division multiple access system (OFCDMA), in which modulation bit resources are allocated on the basis of a code domain and a frequency domain two-dimensional domain. This is an embodiment to be applied.

[0081] On the transmission side, each of the modulation symbols in the frequency domain 'code domain represented by spreading codes and subcarriers does not represent information of a single user, but represents information of a plurality of users. At the receiving terminal of each user, the FFT process is performed in the frequency domain as a process of multiple access cancellation, and then the code domain is performed in the code domain, and the despreading process by matched filtering is performed using the spreading code. In the code domain and the frequency domain, all modulation symbols including information on the user are received.

[0082] FIG. 29 is a diagram for explaining code division * frequency division of multiuser higher-order modulation symbols by the OFCDMA scheme. If the total number of OFDM subcarriers is Nc and the spreading factor of the spreading code is SF, Nc is an integer multiple of SF. Also, the higher-order modulation symbol transmitted to the i-th user set u is Q (i≡ [1, N]), and there are N user sets in total. If the set of all users is U, u is a subset of U and u CU. Further, assuming that the user sets are orthogonal, the following relationship is established.

[Equation 16] w, Π Μ = Φ, ≠ j

[0083] If the number of users in u is m, the total number of users is L, and the order of high-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 17]

\ ≤m _i ≤ min {,}

[0084] First, serial / parallel conversion is performed on Q and divided into Nc / SF parallel branches.

. For each parallel branch, all spreading codes [C,-, C] belonging to user set u

2) Spreading is performed, and the same chip phase is added in the time domain to obtain SF parallel output signals and used as IF inputs of IFFT. In this way, Nc / SF parallel Chic All Nc parallel output signals are obtained, and the Nc-point IFFT operation and parallel / serial conversion are performed to obtain the transmission sequence of the user set u. In this way, the transmission sequences of N user sets are obtained and added in the time domain to obtain the entire transmission sequence.

[0085] FIG. 30 is a diagram illustrating reception of multiuser higher-order modulation symbols by the OFCDMA system. For user j, assume jE _U. User j's terminal first performs serial / parallel conversion and FFT operation to obtain Nc parallel output signals. Then, SF signals can be grouped into Nc / SF groups in total. For each set of signals, despreading in the time domain is performed using all spreading codes [C,-, C] belonging to the user set u.

ui, l) ui (mi no

Addition is performed to obtain one output value. By performing parallel / serial conversion on Nc / SF output values, a higher-order modulation symbol sequence hat [Q] related to user j is obtained.

[0086] (Embodiment 6)

In the case of multi-user high-order modulation, the transmission side can synthesize and frame the different user time slot formats in the same modulation symbol to share pilot data. On the receiving side, channel estimation of the combined frame can be performed using pilot data shared by different users. Since the pilot data shared by multiple users is larger than the conventional pilot data dedicated to a single user, the accuracy of channel estimation increases.

FIG. 31 is a diagram showing a time slot format (data / pilot) arrangement of multiuser higher-order modulation symbols by time division multiple access.

As shown in the figure, in the conventional time division multiple access method, user 1 and user 2 occupy different time slots 1 and 2, respectively. In time slot 1, user 1 pilot p is placed in the middle of the time slot and data Q is at both ends.

1 1

It has been placed. In timeslot 2, pilot 2 of user 2

2

Placed in the middle, data Q is placed at both ends. P and P may not be the same,

2 1 2

It is specified by higher level signaling. On the receiving side, User 1 uses P to channel

1

User 2 performs channel estimation using P.

2

However, in the framework of multi-user high-order modulation, as shown in the figure, modulation symbol Q represents information of user 1 and user 2, and user 1 and user 2 are combined into a frame. And can share the same pilot P. On the receiving side, user 1 and user 2 share the same

12

Channel estimation can be performed using the same pilot P. Change frequency band utilization

12

Since more pilot data can be used on the assumption that there is no channel estimation, the accuracy of channel estimation is higher. Also, it can be seen that the number of modulation symbols occupied by the user in the time domain is larger than in the conventional method, so that more time diversity is possible and the error correction performance is improved. Similarly, the higher-order modulation symbol transmitted to the i-th user set U is Q (i≡ [1, N]), and there are N user sets in total. If the set of all users is U, u is a subset of U and u CU. If the number of users in u is m, the total number of users is L, and the order of high-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 18] l≤ _i ≤ mm {, Z}

[0090] Similarly, m time slots in the conventional method may be combined into a frame, and m same pilots p may be shared.

[0091] (Embodiment 7)

FIG. 32 is a diagram showing the time slot format (data / pilot) arrangement of multiuser higher-order modulation symbols by OFDM. In the conventional OFDM method, user 1 and user 2 occupy different frequency subbands 1 and 2 respectively. In subband 1, user 1 pilot P is placed in the middle of the frequency subband.

1

Data Q is placed at both ends. In subband 2, user 2's pilot

1

P is placed in the middle of the frequency subband, and data Q is placed at both ends. P and P

2 2 1 2 May not be the same. On the receiving side, User 1 uses P to perform channel estimation,

1

User 2 uses P to perform channel estimation.

2

[0092] However, in the framework of multi-user high-order modulation, as shown in the figure, modulation symbol Q represents information of user 1 and user 2, and user 1 and user 2 are combined and framed.

12

And can share the same pilot P. On the receiving side, user 1 and user 2 share the same

12

Channel estimation can be performed using the same pilot P. Similarly frequency band utilization Since more pilot data can be used on the assumption that no change is made, the accuracy of channel estimation becomes higher. In addition, it can be seen that the number of modulation symbols occupied by the user in the frequency domain is larger than that of the conventional method, so that more frequency diversity is possible and the error correction performance is improved. Similarly, the higher-order modulation symbol transmitted to the second user set U is Q (i≡ [1, N]), and there are N user sets in total. If the set of all users is U, u is a subset of U and u CU. If the number of users in u is _m , the total number of users is L, and the order of higher-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Equation 19]

\ ≤m _i ≤ va \ ^ L}

[0093] Similarly, m subbands in the conventional method may be combined and framed, and m identical pilots P may be shared.

[0094] (Embodiment 8)

FIG. 33 is a diagram showing a time slot format (data / pilot) arrangement of multi-user higher-order modulation symbols according to the CDMA scheme. In the conventional CDMA system, user 1 and user 2 occupy different orthogonal codes C and C, respectively, and are spread simultaneously.

1 2

To send. In time slot 1, user 1 pilot P is in time slot 1.

1

Located in the middle, data Q is placed at both ends. Time slot 2 transmitted at the same time

1

The pilot P of user 2 is also placed in the middle of timeslot 2 and data Q

2 2 is placed at both ends. P and P may not be the same. On the receiving side, user 1

1 2

First, despread using C, and then use P to estimate the time domain channel as

1 1

Do.

[Equation 20]

z γρ 'r>

IN where <x, y> represents the correlation function between sequences x and y, Ν is the spreading length, and Y is the received sequence. Similarly, User 2 first despreads using C and then uses Ρ Inter-channel estimation is performed.

[Number 21]

, < ₂ >

| P ₂ | ² N

However, in the framework of multi-user high-order modulation, as shown in the figure, modulation symbol Q represents information of user 1 and user 2, and user 1 and user 2 are combined and framed.

12

And can share the same pilot P. On the receiving side, both User 1 and User 2 are C and P

12 1

Ν

Nolot P that uses 2 to solve and share the composition

12 can be used for channel estimation in the time domain as shown below.

[Number 22]

_H <> I <yp _{2 2} >

2 | N 2 \ P ₂ \ ² N

Similarly, on the premise that the frequency band utilization rate is not changed, since more lot data can be used, the accuracy of channel estimation becomes higher. Also, it can be seen that the number of modulation symbols occupied by the user in the frequency domain is larger than in the conventional method, so that more code diversity is possible and the error correction performance is improved. Similarly, the higher-order modulation symbol transmitted to the i-th user set u is Q Ν]), and there are ユーザ user sets in all. If the set of all users is U, u is a subset of U and u CU. Assuming that the user sets are orthogonal, the following relationship holds.

[Equation 23]

If the number of users in u is _m , the total number of users is L, and the order of higher-order modulation (number of bits in one symbol) is M, m satisfies the following relationship.

[Number 24]

1≤ m _i ≤ min {, Z} [0098] Similarly, the number of spreading code spaces in the conventional method may be combined into a frame and the same number of pilots p may be shared.

[0099] FIG. 34 shows a simple signaling flowchart of the multi-user higher order modulation method. In the first step, the mobile station periodically transmits channel state information (CSI), eg, signal-to-interference and noise ratio (SINR), and ACKZNACK information to the base station. In the second step, the base station periodically analyzes and organizes information such as CSI and ACKZNACK of these mobile stations, and weak bits are given to users with good channels ('ACK signals with high SIRN). Users who are inferior (SINR is low and 'NACK signal) are assigned strong bits, and other multiple access resources such as time slot format, spreading code, frequency band, and nolot of each user are assigned. In the third step, the base station periodically sends multiple connection resources and bit correlation information to the mobile station via the control channel. In the fourth step, the base station transmits the multi-user high-order modulated data to the mobile station via the traffic channel. The overall flow is an adaptive process for channel changes, and when changes occur in the channel of the mobile station, the base station accordingly sets the strong and weak bits in the higher-order modulation and the associated multiple access source. New allocation improves transmission performance for weak users and maximizes system throughput.

[0100] The non-uniform multi-user high-order modulation described in the above embodiments has the following advantages and effects in comparison with the conventional method.

[0101] First, it is a process that can adapt to channel changes. For mobile stations with low signal-to-interference and noise ratios, the base station can assign strong bits in higher-order modulation, and weak This has the same effect as improving the signal-to-interference and noise ratio, and can improve the transmission performance of weak users, helping alleviate the perspective problem. For mobile stations that have fed back a NACK signal (the NACK signal reflects that the channel condition is usually poor), the base station can still assign strong bits in higher-order modulation, and ARQ (automatic repeat request) It helps improve the performance of weak users in the process and increases throughput.

[0102] Secondly, on the assumption that the code length and the frequency spectrum utilization rate do not change, the number of bits of each user possessed by each higher-order modulation symbol decreases, so that the number of modulation symbols occupied by each user decreases. When receiving diversity and receiving diversity, each user has more time Can be more frequency diversity (for frequency division systems) or more code diversity (for code division systems), and each user can Improve overall system throughput by sharing more pilot data with other users in the same group and providing more accurate channel estimation.

[0103] According to each of the above embodiments, an unequal multiuser higher-order modulation method including the following steps has been presented. That is, the unequal multi-user high-order modulation method performs encoding and interleaving on source data sequences from a plurality of users to generate a plurality of interleaved code data sequences, A step of arranging the users in order based on the state of the fed back channel, a step of arranging the symbol bits in the modulation symbol in order based on the order of the strength of the protection capability, and a symbol bit based on the number of symbol bits and the number of users. A step of dividing each user into symbol bit groups and user groups corresponding to each other, and interleaving code data sequences from user groups with good channel conditions are assigned to symbol bit groups with weak protection capability and multiplexed. And bad channel status! Interleaving code data sequences of リー by assigning them to symbol bit groups with strong protection capability, forming one modulation symbol from each parallel symbol bit, generating a modulation symbol sequence, and modulation symbols Performing multiple access to the sequence, and generating a multiple access transmission data sequence.

[0104] Preferably, the channel state of the user is determined by the signal-to-interference and noise ratio. The higher the signal-to-interference / noise ratio, the better the channel state of the user. Alternatively, the user's channel state is determined by the ACK / NACK feedback signal. If the user feeds back an ACK feedback signal, the user's channel condition is good, and if the user feeds back a NACK feedback signal, the user's channel condition is poor.

[0105] Preferably, the level of protection capability of each symbol bit is determined by the sum of the symbol and symbol distance of each symbol bit. However, the greater the sum of the Hamming distances, the weaker the protection capability of the bit. [0106] Preferably, the multiple access is one of the following multiple access schemes: time division multiple access, frequency division multiple access, orthogonal frequency division multiple access, code division multiple access, orthogonal frequency 'code division multiple access Adopt one.

[0107] Preferably, time slots, subbands, or spreading code spaces of a plurality of users within the same modulation symbol are combined into a frame. Multiple users in this modulation symbol share pilot data.

[0108] Corresponding to these, a multiple access cancellation high-order demodulation method comprising the following steps for receiving a multiple access transmission data sequence generated by the above-described multiuser high-order modulation method was presented. That is, the multiple access cancellation higher-order demodulation method includes receiving multiple access transmission data sequences, performing multiple access cancellation processing on the received multiple access transmission data sequences, and generating multiple access cancellation modulation symbol sequences; For each multiple access cancellation modulation symbol in the disconnection modulation symbol sequence, the symbol bit corresponding to the reception user in the multiple access cancellation modulation symbol based on the correspondence between the reception user and the symbol bit. A demodulating data sequence corresponding to the receiving user from the demodulated data by selectively demodulating the data, discarding symbol bits corresponding to other users in the multiple access modulation symbol, and And a step of performing deinterleaving and decoding on the demodulated data sequence to generate a decision data sequence.

[0109] The disclosures of the description, drawings and abstracts contained in the Chinese application of No. 200610071417.2 filed on March 20, 2006 are hereby incorporated by reference.

Industrial applicability

[0110] The wireless transmission device and wireless transmission method of the present invention are a wireless transmission device and a wireless transmission method for simultaneously transmitting wireless signals to a plurality of users, and are useful for improving throughput.

Claims

The scope of the claims

[1] Based on the ease of error of each constituent bit of the unit bit string represented by one symbol and the channel status of each user, each constituent bit is assigned as a user's use bit, and multiple bits are arranged in the order according to the assignment. By arranging elements of parallel data sequences addressed to a plurality of users, bit stream forming means for forming a bit stream addressed to the plurality of users, and modulation means for forming a modulation signal according to the unit bit string of the bit stream; A wireless transmission device comprising:

2. The radio transmission apparatus according to claim 1, wherein the bit stream forming means assigns the error bit V and the constituent bits as used bits to the user because the channel condition is bad!

[3] A step of assigning each constituent bit as a user use bit based on the ease of error of each constituent bit of the unit bit string represented by one symbol and the channel status of each user,

Forming a bitstream addressed to the plurality of users by arranging elements of parallel data sequences addressed to the plurality of users in an order according to the assignment; and

Forming a modulated signal corresponding to the unit bit string of the bit stream.