JP2018160730A - Wireless communication system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem that depending on the transmission system used, it is sometimes difficult to have the EXIT characteristic of a demodulator and the EXIT characteristic of a decoder efficiently matched.SOLUTION: A wireless communication system of the present invention is constituted by a base station and a mobile station. The base station comprises a variable encoder for performing encoding using repetition code whose repetition ratio is varied. The mobile station comprises a demodulator for performing demodulation on the result of reception by a receiving unit and calculating likelihood to code bits, a decoder for performing repeated decoding on the result of demodulation by the demodulator, repetition processing means for processing repetition while propagating information between the demodulator and the decoder, an EXIT analysis unit for calculating the amount of mutual information of the demodulation result for each repetition processing by the repetition processing means, and an optimum code configuration search unit for calculating an optimum repetition ratio on the basis of the EXIT analysis result of the EXIT analysis unit. The variable encoder performs repeated encoding on the basis of the optimum repetition ratio.SELECTED DRAWING: Figure 1

Description

  The present disclosure relates to wireless communication systems.

  In the field of wireless communication, a technique called BICM-ID (Bit Interleaved Coded Modulation with Iterative Decoding) has been proposed, and thereby high transmission efficiency can be realized. For example, as described in Non-Patent Document 1, BICM-ID is realized by repeatedly performing demodulation processing and decoding processing via an interleaver. For the characteristic analysis of the BICM-ID, a technique based on an EXT (Extrinsic Information Transfer) chart is used (for example, JP 2010-87707 A).

JP 2010-87707 A

X.Li and J.A.Ritcey, “Bit-Interleaved Coded Modulation with Iterative Decoding”, IEEE Communications Letters, vol.1, pp.169-717,1997 Takashi.Yano, Tad Matsumoto, “Arithmetic Extended-Mapping for BICM-ID with Repetition Codes”, International ITG Workshop on Smart Antennas WSA, 2009 Sarah J. Johnson, “Iterative Error Correction: Turbo, Low-Density Parity-Check and Repeat-Accumulate Codes”, Cambridge University Press, November 2009, pp. 201-236

Depending on the transmission method used, it may be difficult to efficiently match the EXIT characteristics of the demodulator and the EXIT characteristics of the decoder.
Other problems and novel features will become apparent from the description of the specification and the accompanying drawings.

The outline of a representative one of the present disclosure will be briefly described as follows.
That is, the radio communication system is composed of a base station and a mobile station. The base station includes: a variable encoder that performs encoding using a repetition code whose repetition ratio is variable; a mapping unit that sets an IQ plane for an output signal of the variable encoder; and a mapping signal from the mapping unit And a transmitter that modulates and transmits the data. The mobile station receives a transmission signal from the base station, demodulates the result received by the receiving unit, calculates a likelihood for a code bit, and demodulates the demodulator A decoder for iteratively decoding the result, an iterative processing means for performing iterative processing while propagating information between the demodulator and the decoder, and a mutual information amount of a demodulation result for each iteration of the iterative processing means An EXIT analysis unit that calculates the optimal code configuration search unit that calculates a repetition rate based on the EXIT analysis result of the EXIT analysis unit, and modulates and transmits the repetition rate searched by the optimal code configuration search unit as transmission information Transmitting means. The base station further demodulates a reception unit that receives a transmission signal from the mobile station and a reception signal from the reception unit of the base station, performs demodulation for modulation of the transmission unit of the mobile station, and A demodulation unit that restores the repetition rate transmitted from the mobile station. The variable encoder performs iterative encoding based on the repetition rate restored by the demodulator.

  According to the present invention, the EXIT characteristic of the demodulator and the EXIT characteristic of the decoder can be efficiently matched.

1 is a block diagram illustrating a configuration of a wireless communication system according to an embodiment. EXIT chart EXIT chart during MIMO transmission The figure for demonstrating the encoding using a repetition code | symbol EXIT chart The figure which shows the structure of the optimal code structure search part 17 of FIG. The figure which shows the structure of the decoder EXIT calculation part 171 of FIG. Diagram for explaining the J '() function Diagram for explaining trapezoidal approximation of EXIT area

  An example of the aforementioned EXIT chart will be described with reference to FIG. In the EXIT chart, the vertical axis represents the mutual information amount of the demodulator output, and the horizontal axis represents the mutual information amount of the decoder. The range of any axis is 0 to 1, and the closer to 1, the higher the reliability. A mutual information amount of 1 indicates that the ambiguity of the information is 0, and a mutual information amount of 1 at the decoder output indicates that there is no decoding error.

  The iterative processing of BICM-ID is performed between the demodulator and the decoder as indicated by the arrow between the demodulator EXIT characteristic (solid line in FIG. 2) and the decoder EXIT characteristic (dotted line in FIG. 2). While propagating information repeatedly, the characteristics are improved by gradually increasing the amount of information.

  By using this EXIT chart, the following two points can be analyzed.

  First, when the EXIT characteristic of the demodulator and the EXIT characteristic of the decoder intersect, it becomes impossible to increase the amount of information further, and an error occurs in the decoder. The second point is that the smaller the area between the EXIT characteristic of the demodulator and the EXIT of the decoder, the more efficient transmission is possible, and it is known whether the transmission capacity of the propagation path can be fully utilized. be able to.

  Thus, by using the EXIT chart, it is possible to analyze the matching between the demodulator and the decoder.

  In the EXIT analysis described above, the EXIT characteristics of the demodulator differ depending on the Gray mapping or non-Gray mapping in which the hamming distance between adjacent mapping points is 1, or the extended mapping of Non-Patent Document 2. Also, the EXIT characteristic varies depending on the received signal to noise power ratio (SNR) and the propagation path characteristic.

  On the other hand, the EXIT characteristic of the error correction code is uniquely determined by the configuration of the encoder used. For this reason, it is common to perform an EXIT chart analysis of the demodulator based on the assumed SNR and propagation path characteristics, design in advance an encoder configuration that matches those characteristics, and apply it to a wireless transmission system.

  As described above, when the configuration of the encoder is determined, when the propagation path characteristics fluctuate, the demodulator and the decoder are not matched, and decoding error and transmission efficiency decrease occur.

  In order to solve this problem, the bidirectional radio transmission system uses adaptive modulation coding (AMC) that adaptively varies the modulation scheme and coding rate based on the characteristics of the propagation path. . This is because the reception unit determines the reception quality from the modulation error ratio (MER), the received signal strength (RSSI), or the error rate of the error correction result, and no error occurs. Thus, an appropriate modulation scheme and coding rate are determined. The determined modulation method and coding rate are fed back to the transmission side, and the transmission unit sets the modulation method and coding rate based on the obtained result.

  Application of AMC makes it possible to match the EXIT characteristic of the demodulator and the EXIT characteristic of the decoder. However, depending on the transmission method used, it may be difficult to match efficiently.

  For example, when Gray mapping and a low density parity check (LDPC) error correction code are applied to MIMO (Multiple Input Multiple Output) using a plurality of transmission / reception antennas, the exit characteristics of the demodulator are shown in FIG. It becomes such a characteristic. As shown in FIG. 3, when the received SNR is relatively high, the demodulator EXIT characteristic exceeds the EXIT characteristic of the decoder and does not cross, so that no decoding error occurs. However, the area of the EXIT is wide and the transmission efficiency is reduced. Furthermore, in MIMO mobile transmission, etc., the slope of the EXIT characteristic of the demodulator varies from time to time according to the characteristics of the MIMO channel, so it is difficult to match the EXIT characteristics of decoders such as LDPC and turbo codes. is there.

  The radio communication system according to the embodiment analyzes the EXIT characteristic of the demodulator on the receiver side in a propagation path whose characteristics change every moment, such as mobile transmission, and searches for an encoder configuration suitable for the EXIT characteristic of the demodulator. The searched encoder configuration is fed back, and an encoder is configured based on the encoder configuration obtained from the feedback information. According to the embodiment, it is possible to realize highly efficient transmission that can fully utilize the channel capacity of the propagation path.

  Examples will be described below with reference to the drawings. However, in the following description, the same components may be denoted by the same reference numerals and repeated description may be omitted.

  A wireless communication system according to an embodiment will be described in detail with reference to FIG. The radio communication system according to the embodiment includes a base station BS and a mobile station MS. The base station BS includes a variable encoder 1, an interleaver 2, a mapping unit 3, a transmission unit 4, a circulator 5, a transmission / reception antenna 6, a reception unit 7, and a demodulation unit 8. The mobile station MS includes a transmission / reception antenna 9, a circulator 10, a reception unit 11, a demodulator 12, a deinterleaver 13, a decoder 14, an interleaver 15, an EXIT analysis unit 16, an optimum code configuration search unit 17, a modulation unit 18, and a transmission. The unit 19 is provided.

In the transmitter of the base station BS unit, an information bit sequence b = {b ₁ , b ₂ ,..., B _k } to be transmitted to the variable encoder 1 is input, and a demodulating unit that analyzes feedback information from the mobile station MS unit An error correction encoder is configured based on the code configuration signal from 8, error correction encoding is performed by the configured encoder, and a code sequence c is output. Here, the operation of the code component signal from the demodulator 8 will be described later.

The error correction code in the variable encoder 1 is encoded using a repetition code. An example of this repetition code will be described with reference to FIG. A repetition code is applied to the input information bit sequence b = {b ₁ , b ₂ ,..., B ₈ }, and the code configuration is determined by the number of repetitions of each bit. In the example of FIG. 4, the ratio of the number of repetitions of 5 (applied to b ₁ and b ₂ ) is 2/8, the ratio of the number of repetitions of 4 (applied to b ₃ ) is 1/8, and the number of repetitions is 3 (b ₄ , the ratio of _{b 5} to the application) is 2/8, the ratio of the number of repetitions applied in _{2 (b 6, b 7,} b 8) is a code structure to be 3/8. The code configuration information received from the demodulator 8 receives the ratio of each repetition number, and performs encoding based on the ratio. In the example of FIG. 4, the code length K = 8, and the ratio of the number of repetitions from 2 to 5 is 3/8, 2/8, 1/8, 2/8 in order. Here, if the vector indicating the repetition rate is a, a = (a ₂ , a ₃ , a ₄ , a ₅ ) = (3/8, 2/8, 1/8, 2/8).

  The decoding characteristic is determined by the repetition rate ratio a, and the EXIT characteristic of the decoder is also uniquely determined. In addition, the present embodiment focuses on optimizing the code configuration according to the characteristics of the propagation path, and the repetition rate a is optimized so as to match the characteristics of the propagation path. The optimization of the repetition rate ratio a will be described later.

  Furthermore, this code configuration is desirably updated in a short period such that the propagation path characteristics can be regarded as the same when the propagation path characteristics such as mobile transmission fluctuate.

  The code c from the variable encoder 1 is input to the interleaver 2, and the interleaver 2 changes the order of the code c in order to fully exhibit the composite capability of the error correction encoder. A random pattern is often used as the interleave pattern, but a regular pattern may be used.

  The code sequence rearranged by the interleaver 2 is input to the mapping unit 3, and the mapping unit performs mapping processing for associating code bits with points on the I and Q planes. The I and Q mapping includes QAM (Quadrature Amplitude Modulation) arranged in a lattice pattern, mapping method arranged on the circumference, and the like.

  In addition, the mapping point associated with the code bit is a Gray mapping in which the Hamming distance of the code bit corresponding to the adjacent mapping point is 1 or a non-Gray designed so that the Hamming distance is a large value of 2 or more. There is mapping. Furthermore, there is an extended mapping in which a plurality of code bits are associated with one mapping point proposed in Non-Patent Document 2. In the embodiment, the extended mapping will be described as an example, but it is also possible to adapt to Gray or non-Gray mapping. The feature of the extended mapping is that the EXIT characteristic of the demodulator is greatly inclined to the right, and has strong consistency with the EXIT characteristic of the repetition code used in the variable encoder 1.

  The mapping signal is converted into a modulated signal by the transmitter 4. As a modulation method, it is possible to apply OFDM (Orthogonal Frequency Division Multiplexing), single carrier, FBMC (Filter Bank Multiple Carrier), UFMC (Universal Filtered MultiCarrier), etc., which are being considered as next-generation waveform, and there are restrictions. Never happen.

  It is also possible to apply MIMO in which a plurality of streams are multiplexed and transmitted in the spatial domain using a plurality of transmission / reception antennas.

  The modulated signal is transmitted from the transmitting / receiving antenna 6 via the circulator 5. The radio transmission signal is received by the transmission / reception antenna 9 via a propagation path that varies from moment to moment.

  The received signal is input to the receiving unit 11 via the circulator 10. The reception unit 11 performs reception processing suitable for the modulation processing performed by the transmission unit 4. Specifically, transmission / reception synchronization processing, IFFT (Inverse Fast Fourier Transform) processing, etc. are used when OFDM is used. Since these reception processes are not essential, detailed description thereof is omitted.

The demodulator 12 performs an equalization process for reconfiguring the transmission-side mapping points on the received signal. In SISO (Single Input Single Output) using one transmission / reception antenna one by one, equalization processing that divides the received signal by the estimated propagation path characteristic, and in MIMO, the estimated propagation path characteristic matrix for the reception signal received by multiple antennas There is a spatial filtering process using. The likelihood of the coded bit is calculated based on the distance between the reconfigured reception mapping point and the ideal point mapped on the transmission side. The log likelihood ratio L _E ^DEM (b _i ) of the sign bit b _i is expressed by equation (1), which is defined as a demodulator external log likelihood ratio (LLR) L _E ^DEM (b _i ). To do.

In the equation (1), P (b _i = 0) indicates the probability that b _i is 0, and P (b _i = 1) indicates the probability that b _i is 1. Here, the index i of b _i indicates the bit sequence number.

  Further, as processing in the demodulator 12, a plurality of reception candidate points (replicas) are calculated from the transmission mapping points and the estimated propagation path (matrix) characteristics, and based on the obtained Euclidean distance between the reception points and the reception candidate points. There is an MLD (Maximum Likelihood Detection) method for calculating code bit likelihood.

The demodulator external LLR L _E ^DEM (b _i ) from the demodulator 12 is connected to the deinterleaver 13. The deinterleaver 13 performs processing for restoring the order rearranged by the interleaver 2. The demodulator external LLR L _E ^DEM (b _i ) is rearranged by the deinterleaver 13 to become a decoder prior LLR L _A ^DEC (b _n ) that is prior information to the decoder 14.

  The output of the deinterleaver 13 is connected to the EXIT analysis unit 16 and the decoder 14.

The EXIT analysis unit 16 calculates the demodulator mutual information I _A ^DEM by performing the calculation of Expression (2) based on the decoder prior LLR L _A ^DEC (b _n ).

The demodulator mutual information amount I _A ^DEM indicates the reliability of the output of the demodulator 12, and the EXIT chart shown in FIG. In FIG. 5, numbers are surrounded by circles, but in the present specification, numbers are surrounded by ().

The decoder 14 performs decoding for the repetition code based on a code structure of a decoder in advance LLR _L ^{A DEC} _{(b n)} and the optimum code structure search section 17, a decoder external LLR _L ^{E DEC} and _{(b n)} calculate. The details of the decoding method for iterative codes are described in Non-Patent Document 3.

  As shown in Expression (2), the mutual information amount of the output of the decoder 14 is shown in FIG. 5 (dotted line). The mutual information amount (1 point) of the demodulator 12 is obtained, and by performing decoding based on that, the mutual information amount of the decoder output increases from the point (1) to the point (2). Yes.

Decoder external LLR L _E ^DEC from the decoder 14 (b _n) are input to the interleaver 15 performs a sort in the same pattern as the interleaver 2 interleaving patterns on the transmitting side is fed back to the demodulator 12. As in the case of the decoder 14, the decoder external LLR L _E ^DEC (b _n ) is rearranged by the interleaver 15, and when input to the demodulator 12, the demodulator pre-LLR L _A ^DEM ( b _i ).

  The above processing shows one iteration processing between the demodulator and the decoder.

  Next, the second iteration process will be described.

The demodulator 12 calculates a demodulator external LLR L _E ^DEM (b _i ) from the log likelihood ratio shown in Equation (1) and the demodulator prior LLR L _A ^DEM (b _i ) input as feedback. The difference from the first iteration is that the demodulator prior LLR L _A ^DEM (b _i ) is not obtained in the first time, and therefore the demodulator external LLR L _E ^DEM (b _i ) is a logarithmic likelihood of the equation (1). The frequency ratio is output as it is. However, in the second and subsequent iterations, the logarithmic likelihood ratio of equation (1) and the demodulator prior LLR L _A ^DEM (b _i ) to the demodulator external LLR L _E ^DEM (b _i ) Is calculated. A detailed calculation method is described in Non-Patent Document 3.

The demodulator external LLR L _E ^DEM (b _i ) is input to the EXIT analysis unit 16 via the deinterleaver 13 and calculates the mutual information amount described in Expression (2) in the same manner as in the first iteration process. To do. Since the demodulator pre-LLR L _A ^DEM (b _i ) is obtained during the second iteration, higher mutual information can be obtained than during the first iteration. This is the characteristic of (3) in FIG. 5, and the amount of mutual information obtained is increased by obtaining the demodulator prior LLR L _A ^DEM (b _i ) ((2) → (3) in FIG. 5). )

In the decoder 14, the decoding process is performed based on the decoder prior LLR L _A ^DEC (b _n ) as in the first iteration process. Decoding results in a decoder external LLR _L ^{E DEC} _{(b n)} is increased mutual information than when one iteration ((3 in FIG. 5) → (4)).

The decoder external LLR L _E ^DEC (b _i ) is converted into a demodulator pre-LLR L _A ^DEM (b _n ) by the deinterleaver 13 and input again to the demodulator 12.

  The above process is the second iterative process.

  By performing the above iterative process a plurality of times, in the EXIT chart of FIG. 5, the iterative process is repeated in the order of (4) → (5), (5) → (6), (6) → (7),. As the amount of mutual information improves.

  Further, in the above-described multiple iterations, the trajectory of mutual information obtained by the EXIT analysis unit 16 ((1), (3), (5), (7),..., FIG. 5) It is the EXIT characteristic of the demodulator obtained discretely.

  As described above, the smaller the area of the EXIT characteristic of the demodulator and the EXIT characteristic of the decoder, the better the transmission efficiency (the transmission capacity of the propagation path can be fully utilized). Variable on the BS unit 20 side so that it matches the EXIT characteristic of the demodulator whose path characteristics have changed (the EXIT characteristic of the demodulator and the EXIT characteristic of the decoder do not cross each other and the area between them is small). The encoder 1 is configured adaptively.

  Therefore, the optimum code configuration search unit 17 searches for an encoder configuration having a minimum EXIT area from the exit characteristics of the demodulator obtained by the EXIT analysis unit 16. As shown in FIG. 6, the optimum code configuration search unit 17 includes a decoder EXIT calculation unit 171 and a minimum EXIT area search unit 172. Further, as shown in FIG. 7, the decoder EXIT calculation unit 171 includes a d2: J ′ () function generator 1711, a d3: J ′ () function generator 1712, a dM: J ′ () function generator 1713, a multiplication. Units 1714 to 1716, an adder 1717, and a repetition rate adjustment unit 1718.

  The decoder EXIT calculation unit 171 calculates the EXIT characteristic of the decoder corresponding to each repetition rate a while changing the repetition rate a. The EXIT characteristic of the encoder does not depend on the propagation path and is uniquely determined from the configuration of the decoder. The EXIT characteristic of this decoder is expressed by equation (3).

a _i : Repeating ratio d _i : Number of repetitions Further, the J () function of Expression (3) is expressed by Expression (4).

J (√ (d _i −1) · J ⁻¹ (I _A ^DEC )) in equation (3) is determined by the decoder mutual information amount I _A ^DEC from 0 to 1 and the iteration number d _i of iterative decoding. Function.

We define _{d i} · J a _{(√ (d i -1) ·} J -1 (I A DEC)) J '(I A DEC, di) and. Further, Σ _i a _i · d _i of the denominator in the equation (3) matches the code length K. Furthermore, when the definition of the combination vector a having the repetition rate a _i is re-displayed, a = (a ₂ , a ₃ , a ₄ ,..., A _{M + 1} ) and Σa _i = 1. Taking these into account, Equation (3) is simplified to Equation (5).

The d2: J ′ () function generator 1711, d3: J ′ () function generator 1712, and dM: J ′ () function generator 1713 of the decoder EXIT calculation unit 171 have J ′ (for each iteration number d _i. ) is to generate a function, J '() for the function is uniquely determined by d _i, as shown in FIG. 8, J per advance repetition number d _i' by calculating the () function, ROM (a Read The calculation load can be reduced by storing the data in the (Only Memory).

As shown in the equation (5), the EXIT characteristics of the decoder are generated by the d2: J ′ () function generator 1711, the d3: J ′ () function generator 1712, and the dM: J ′ () function generator 1713. The result is a weighted average of the J ′ (I _A ^DEC , di) function with the repetition ratio a _i . Therefore, the decoder EXIT calculation unit 171 inputs the repetition rate a _i calculated by the repetition rate adjustment unit 1718 to the multipliers 1714, 1715 and 1716, and d2: J ′ () function generator 1711, d3: J ′ ( ) Function generator 1712, dM: J ′ () Multiplies the output of the function generator 1713. Each multiplication result is added by an adder 1717, whereby the EXIT characteristic of the decoder described in Expression (5) can be obtained.

As described above, the EXIT characteristic of the decoder can be obtained by the combination of the repetition rate a _i set by the repetition rate adjustment unit 1718.

The repetition rate a _i set by the decoder EXIT calculation unit 171 is sequentially output by the repetition rate adjustment unit 1718 in various combinations of a _i .

To output all combinations, the upper limit of the number of repetitions d _i is M, if the information bit length is N, a combination of N ^M. For example, if the information bit length N is 8 and the upper limit M of the number of repetitions d _i is ^16, there are 816 combinations, but there are a plurality of combinations with the same EXIT characteristics of the decoder. This combination can be omitted.

  Also, combinations that are very similar in EXIT characteristics can be omitted.

In this way, a plurality of patterns of repetition rate a _i are output, and the EXIT characteristic I _E ^DEC (I _A ^DEC , a) of the decoder is calculated.

From the decoder EXIT calculation unit 171, the EXIT characteristics of the decoder of each repetition rate a are input to the minimum EXIT area search unit 172, and the other input of the minimum EXIT area search unit 172 is a demodulator from the EXIT analysis unit 16. EXIT characteristic I _A ^DEM is input.

MINIMUM EXIT area searching section 172, EXIT characteristics of the resulting demodulator _I ^A is below the EXIT characteristics of the decoder than ^DEM, besides EXIT characteristics of the decoder ^{_{I E DEC (I A DEC,}} a) and the Search for the repetition rate a _opt that minimizes the area. For example, as shown in FIG. 9, the EXIT area is calculated by using discretely obtained two-point demodulator EXIT characteristics I _A ^DEM (white circles in FIG. 9) and decoder EXIT characteristics I _E ^DEC (I _A The area between ^DEC and a) is approximated by a trapezoid, and the sum of the areas of the trapezoids for all points of the obtained discrete demodulator EXIT characteristic I _A ^DEM is used. The area is obtained at each repetition rate a, and the repetition rate a that minimizes the area is set as the optimum repetition rate a _opt .

By configuring the variable encoder 1 based on the optimum repetition rate a _opt , the T characteristic of the demodulator and the XIT characteristic of the decoder can be matched.

Next, a second method relating to the calculation method of the optimum repetition rate a _opt of the optimum code configuration search unit 17 will be described.

The EXIT characteristic I _A ^DEM of the obtained demodulator is a vector q, the J ′ () function J ′ (I _A ^DEC , d _i ) at each iteration number d _i is the j _i vector, and the J ′ () function matrix is j. Here, j = (j ₂ , j ₃ , j ₄ ,..., J _{M + 1} ).

In the second approach relates to a method of calculating the optimum repetition ratio _{a opt} in norms that minimizes the squared difference between EXIT characteristic _I ^{E DEC} of the decoder and EXIT characteristic _I ^{A DEM} demodulator _(I ^{A DEC,} a), The optimal repetition ratio a _opt is calculated. The square difference is defined by equation (6).

  In order to obtain the optimum repetition rate a that minimizes ν in equation (6), equation (6) is differentiated to zero.

  The calculation result of Expression (7) is shown in Expression (8).

Here, since (JJ ^T ) ⁻¹ J is a determined value, when calculation is performed in advance and the calculation result is a matrix L, Expression (8) is expressed by Expression (9).

In the second method relating to the calculation method of the optimal repetition rate a _opt of the optimal code configuration search unit 17, the product of the EXIT characteristic q of the demodulator obtained by the EXIT analysis unit 16 and the matrix L calculated in advance is calculated. It ’s fine. Although the optimum repetition rate a _opt obtained by the equation (9) is calculated as a real number, the minimum resolution of the repetition rate a _i is the reciprocal of the code length K, so that the quantization process shown in the equation (10) is performed. Necessary.

Furthermore, it is necessary to finely adjust the result of Expression (10) so that the sum of each element of a ′ _opt becomes 1.

The optimum repetition rate a _opt calculated by the optimum code configuration search unit 17 is input to the modulation unit 18 and the decoder 14. The modulator 18 modulates the optimum repetition rate a _opt for feedback to the base station BS, and the transmitter 19 performs processing such as frequency conversion to generate a radio frequency signal. Thereafter, feedback transmission is performed to the base station BS side via the circulator 10 and the transmission / reception antenna 9. On the base station BS side, the signal is received by the transmitting / receiving antenna 6 and input to the receiving unit 7 via the circulator 5. The receiver 7 performs signal power amplification and frequency conversion, and inputs the result to the demodulator 8. The demodulator 8 performs demodulation processing corresponding to the modulator 18 and reproduces the optimum repetition rate a _opt transmitted on the mobile station MS side. The result of the demodulator 8 is input to the variable encoder 1, and the encoder is configured with the repetition rate as described above.

  According to the present embodiment described above, the EXIT analyzing unit 16 on the receiver side analyzes the EXIT characteristics of the demodulator in the propagation path whose characteristics change every moment, such as mobile transmission, and the optimum code configuration searching unit 17 An encoder configuration suitable for the demodulator EXIT characteristic is searched. The searched encoder configuration is fed back, and by configuring the encoder based on the encoder configuration obtained from the feedback information, it is possible to realize highly efficient transmission that can fully utilize the channel capacity of the propagation path.

  Although the invention made by the present inventor has been specifically described based on the embodiments and examples, the present invention is not limited to the above-described embodiments and examples, and various modifications can be made. Not too long.

DESCRIPTION OF SYMBOLS 1 ... Variable encoder 2 ... Interleaver 3 ... Mapping part 4 ... Transmission part 5 ... Circulator 6 ... Antenna 7 ... Reception part 8 ... Demodulation part 9 ... Antenna 10 circulator 11 receiving unit 12 demodulating unit 13 deinterleaver 14 decoder 15 interleaver 16 EXIT analyzing unit 17 optimum Code configuration search unit 18 ... modulation unit 19 ... transmission unit BS ... base station MS ... mobile station

Claims (3)

A wireless communication system comprising a base station and a mobile station,
The base station
A variable encoder that performs encoding using a repetition code with a variable repetition rate;
A mapping unit for converting the output signal of the variable encoder into an IQ plane;
A transmitter that modulates and transmits a mapping signal from the mapping unit;
With
The mobile station
A receiver for receiving a transmission signal from the base station;
A demodulator that demodulates the result received by the receiver and calculates a likelihood for the code bit;
A decoder that repeatedly decodes the result demodulated by the demodulator;
Repetitive processing means for performing repetitive processing while propagating information between the demodulator and the decoder;
An EXIT analysis unit that calculates a mutual information amount of a demodulation result for each iterative process of the iterative processing unit;
An optimum code configuration search unit that calculates a repetition rate based on the EXIT analysis result of the EXIT analysis unit;
Transmitting means for modulating and transmitting the repetition rate searched by the optimum code configuration search unit as transmission information;
With
The base station further includes:
A receiver for receiving a transmission signal from the mobile station;
A demodulator that demodulates the received signal from the receiver of the base station, demodulates the modulation of the transmitter of the mobile station, and restores the repetition rate transmitted from the mobile station;
With
The variable encoder is a wireless communication system that performs repetitive encoding based on a repetition rate restored by the demodulation unit. In claim 1,
The base station further includes a first interleaver that rearranges the order of the output signals of the variable encoder,
The mobile station repetitive processing means comprises:
A deinterleaver for returning the order rearranged by the first interleaver;
A second interleaver that rearranges in the same pattern as the first interleaver;
With
The demodulator calculates a log likelihood ratio of the code bit and outputs it as a demodulator external log likelihood ratio;
The deinterleaver generates a decoder prior log likelihood ratio by reordering the demodulator external log likelihood ratio;
The decoder performs decoding on a repetitive code based on the decoder prior log likelihood ratio and the code configuration from the optimal code configuration search unit, and calculates a decoder external log likelihood ratio,
The second interleaver reorders the decoder outer log likelihood ratio to generate a demodulator prior log likelihood ratio;
The EXIT analysis unit calculates a decoder mutual information amount based on the decoder prior log likelihood ratio,
The optimal code configuration search unit is a wireless communication system that calculates a repetition rate based on the decoder mutual information amount. In claim 2,
The optimum code configuration search unit includes:
A decoder EXIT calculation unit for calculating the EXIT characteristic of the decoder corresponding to each repetition rate while changing the repetition rate;
A minimum EXIT area search unit for calculating an optimum repetition rate based on the EXIT characteristic of the decoder and the amount of information of the decoder;
With
The minimum EXIT area search unit calculates an EXIT characteristic of the demodulator based on the decoder information amount, the EXIT characteristic of the decoder is lower than the EXIT characteristic of the demodulator, and the EXIT characteristic of the demodulator A wireless communication system that searches for an optimum repetition rate that minimizes an area with the EXIT characteristic of the decoder.

Images