JP2011130152A - Multi-value encoder and multi-value encoding method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a multi-value encoder that avoids misrecognition of signal points (amplitude level) because of a transition shortage and a waveform distortion and performs high-efficient transmission without large-scale circuit. <P>SOLUTION: The multi-value encoder has a transition probability calculator which calculates the transition probability of bit patterns between adjacent time slots for a plurality of bit patterns that data is divided for every specified time slot, a mapping information generator which maps bit patterns to specified signal points so that the fluctuation of an amplitude level of multi-value signals becomes small based on the calculated transition probability and generates mapping information indicating the relationship between bit patterns and signal points, a packet generator which generates a packet that is composed of a header including mapping information and a data part including data, and a data modulator which modulates data included in the data part to specified signal points based on the mapping information. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a multi-level encoding device and a multi-level encoding method used for data communication between a transmitter and a receiver, and more specifically, a bit pattern of data to be transmitted / received at a predetermined signal point. The present invention relates to a multi-level encoding apparatus and a multi-level encoding method for mapping to the above.

  In recent years, when data is communicated between a transmitter and a receiver, multi-level coding has been introduced into various communication schemes in order to realize more efficient data communication. Multi-level coding is a method for increasing the transmission speed by mapping bit patterns composed of a plurality of bits to different signal points. If the multi-level number in multi-level encoding is increased, the number of bits allocated to one signal point can be increased. Thereby, transmission efficiency can be improved.

  However, when multi-level encoding is performed within a limited amplitude level range, the distance between signal points is reduced in inverse proportion to the multi-level number. For this reason, when decoding multi-value encoded data, a signal point (amplitude level) identification error occurs. In addition, when the amplitude level fluctuates suddenly, signal point (amplitude level) identification errors further occur due to shortage of signal waveform rise and fall (insufficient transition) and waveform distortion. It becomes easy to do. As a result, a transmission error occurs in data communication.

  Therefore, in Patent Document 1, it is proposed to limit the transition of the amplitude level to be subjected to multi-level encoding to 0 or an amplitude level adjacent in the vertical direction of the amplitude level. This avoids signal point (amplitude level) identification errors due to the effects of insufficient transition and waveform distortion described above. Further, the symbol error generated by limiting the transition of the amplitude level in this way is subjected to error correction processing in the subsequent stage.

JP-A-9-198737

  However, the above-described conventional technology requires a correction processing capability larger than a simple noise error correction processing capability in order to correct an error intentionally generated by limiting the transition of the amplitude level. In other words, the constraint length in the error correction circuit is increased, and as a result, the circuit scale is increased. On the other hand, if the coding rate is decreased by reducing the number of multi-levels in order to avoid identification errors of signal points (amplitude levels), the data transmission efficiency will be deteriorated.

  Therefore, an object of the present invention is to increase the signal scale (amplitude level) due to the influence of insufficient transition, waveform distortion, etc. without increasing the circuit scale by applying multi-level coding focusing on the transition of bit patterns in transmission data. It is to provide a multi-level encoding device and a multi-level encoding method that can avoid the identification error of the above and realize high-efficiency transmission.

  In order to achieve the above object, the multilevel encoding apparatus of the present invention modulates data contained in a packet communicated between a transmitter and a receiver into a multilevel signal having a plurality of predetermined amplitude levels. A transition probability calculation unit that calculates a transition probability of a bit pattern between adjacent time slots for a plurality of bit patterns obtained by decomposing data in predetermined time slot units, and a calculated transition A mapping information generating unit that maps a bit pattern to a predetermined signal point so as to reduce a variation in the amplitude level of the multilevel signal based on the probability, and generates mapping information indicating a correspondence relationship between the bit pattern and the signal point; A packet generation unit that generates a packet including a header portion including mapping information and a data portion including data; and mapping information And Zui, and a data modulator for modulating the data contained in the data unit into a predetermined signal point.

  A preferable mapping information generation unit maps bit patterns having a large transition probability to adjacent signal points in order.

  Furthermore, a preferable mapping information generation unit maps a bit pattern having a Hamming distance of 1 to an adjacent signal point.

Preferably, the multi-level encoding device of the present invention further includes an occurrence probability calculating unit that calculates an occurrence probability of a bit pattern included in the data, and the mapping information generating unit is an occurrence calculated by the occurrence probability calculating unit. The bit pattern having the maximum probability is mapped to the signal point having the minimum or maximum amplitude level.
Alternatively, preferably, the multi-level encoding device of the present invention further includes an occurrence probability calculation unit that calculates an occurrence probability of a bit pattern included in the data, and the mapping information generation unit generates the occurrence calculated by the occurrence probability calculation unit The bit pattern having the maximum probability is mapped to the signal point having the minimum amplitude level.

  Furthermore, a preferable mapping information generation unit calculates the average transmission power based on the occurrence probability calculated by the occurrence probability calculation unit and the generated mapping information.

  Furthermore, a preferable mapping information generating unit maps the bit pattern to a predetermined signal point so that the average transmission power is minimized based on the calculated average transmission power.

  Preferably, the header part includes average transmission power information indicating the calculated average transmission power.

  Further preferably, the multi-level encoding device of the present invention further includes a mapping table that stores in advance mapping information indicating a correspondence relationship between a bit pattern and a signal point, and a mapping number corresponding to the mapping information, The mapping information generation unit refers to the mapping table and uses the mapping number corresponding to the generated mapping information as mapping information.

In addition, a preferable transition probability calculating unit calculates a transition probability for each segment for segment data obtained by subdividing the data into a plurality of segments having a predetermined length, and the mapping information generating unit is based on the transition probability calculated for each segment. Thus, with respect to the segment data, mapping information is generated for each segment.
Alternatively, the preferable transition probability calculation unit calculates a transition probability for each segment for segment data obtained by subdividing the data into a plurality of segments having a predetermined length, and the occurrence probability calculation unit calculates the occurrence probability for each segment for the segment data. The mapping information generation unit calculates and generates mapping information for each segment of the segment data based on the transition probability and occurrence probability calculated for each segment.

  In order to achieve the above object, a multilevel decoding apparatus according to the present invention demodulates a multilevel signal having a plurality of predetermined amplitude levels included in a packet communicated between a transmitter and a receiver. A decoding device that sequentially maps bit patterns having a large bit pattern transition probability between adjacent time slots to adjacent signal points for a predetermined signal point and a plurality of bit patterns decomposed in predetermined time slot units. Mapping information indicating a correspondence relationship with the bit pattern is extracted from the packet, and a mapping analysis unit that analyzes the extracted mapping information and a data demodulation unit that demodulates the multilevel signal based on the mapping information are provided.

  Further preferably, the multi-level decoding device of the present invention is a bit pattern having a large transition probability of a bit pattern between adjacent time slots for a predetermined signal point and a plurality of bit patterns decomposed in units of a predetermined time slot. Is further provided with a mapping table that stores in advance mapping information indicating a correspondence relationship with bit patterns obtained by sequentially mapping signal points adjacent to each other and a mapping number corresponding to the mapping information, and the packet includes a mapping number as mapping information. The mapping analysis unit extracts mapping information corresponding to the mapping number included in the packet with reference to the mapping table, and the data demodulation unit extracts the multilevel signal based on the extracted mapping information. Is demodulated.

  Further preferably, the multilevel decoding device according to the present invention is configured such that the packet includes, for a plurality of bit patterns decomposed in predetermined time slot units, an average calculated based on a bit pattern occurrence probability and mapping information. Transmission power information is included, and further includes a DC drift amount estimation unit that generates a DC drift amount estimation value based on the average transmission power information, and the data demodulation unit is based on the DC drift amount estimation value. When demodulating the value signal, the threshold value in the amplitude level signal determination is adjusted, and the multi-value signal is demodulated based on the mapping information.

  In order to achieve the above object, a multilevel encoding method of the present invention is a multilevel encoding method executed by a multilevel encoding apparatus that modulates data into a multilevel signal having a plurality of predetermined amplitude levels. , A transition probability calculating step for calculating a transition probability of a bit pattern between adjacent time slots for a plurality of bit patterns obtained by decomposing data in a predetermined time slot unit, and based on the calculated transition probability, A mapping information generating step for mapping a bit pattern to a predetermined signal point so as to reduce the fluctuation of the amplitude level, and generating mapping information indicating a correspondence relationship between the bit pattern and the signal point, and a data portion based on the mapping information And a data modulation step of modulating the data included in the data into predetermined signal points.

  In order to achieve the above object, a multilevel decoding method of the present invention is a multilevel decoding method executed by a multilevel decoding device that demodulates a multilevel signal having a plurality of predetermined amplitude levels, And a bit pattern in which bit patterns having a large transition probability of bit patterns between adjacent time slots are sequentially mapped to adjacent signal points for a plurality of bit patterns decomposed in predetermined time slot units. A mapping analysis step for analyzing the mapping information shown, and a data demodulation step for demodulating the multilevel signal based on the mapping information are executed.

  The above-described multilevel encoding method and multilevel decoding method of the present invention are provided in the form of a program for causing a computer to execute a series of processing procedures. This program may be installed in a computer in a form recorded on a computer-readable recording medium.

  According to the multi-level encoding apparatus and multi-level encoding method of the present invention, by performing multi-level encoding focusing on the transition of bit patterns in transmission data, the circuit scale is not increased, lack of transition, waveform distortion, etc. This makes it possible to avoid signal point (amplitude level) identification errors due to the influence of the above and to achieve highly efficient transmission.

The figure which shows the multi-value encoding apparatus 100 which concerns on the 1st Embodiment of this invention. The figure which shows the procedure which maps a bit pattern to a predetermined | prescribed signal point based on the transition probability 151 calculated by the transition probability calculation part 101. The figure which shows the structure of the packet 153 The figure which shows the multi-value decoding apparatus 200 which concerns on the 1st Embodiment of this invention. The figure which shows the procedure which maps a bit pattern to a predetermined | prescribed signal point based on the transition probability 151 calculated by the transition probability calculation part 101, and the restriction | limiting of a Hamming distance. The figure which shows the multi-value encoding apparatus 300 which concerns on the 3rd Embodiment of this invention. The figure which shows the procedure which maps a bit pattern to a predetermined | prescribed signal point based on the transition probability 151 calculated by the transition probability calculation part 101, and the generation probability 351 calculated by the generation probability calculation part 301. Diagram showing how multi-level identification errors occur The figure which shows the procedure which calculates the average transmission power at the time of selecting each mapping candidate about two types of mapping candidates shown in FIG. The figure which shows the multi-value encoding apparatus 400 which concerns on the 4th Embodiment of this invention. A mapping table 401 for assigning bit patterns “00”, “01”, “10”, “11” to four amplitude levels (signal points) “0”, “1”, “2”, “3”. Illustration FIG. 11 is a diagram illustrating a mapping table 402 when a bit pattern mapped to adjacent signal points in the mapping table 401 illustrated in FIG. The figure which shows the multi-value decoding apparatus 500 which concerns on the 4th Embodiment of this invention. The figure which shows the structure of the packet which the packet generation part 103 in the multi-value encoding apparatus which concerns on the 5th Embodiment of this invention produces | generates. The figure which shows the structure of the packet which the packet generation part 103 in the multi-value encoding apparatus concerning the 6th Embodiment of this invention produces | generates. The figure which shows the multi-value decoding apparatus 600 which concerns on the 6th Embodiment of this invention.

Hereinafter, each embodiment of the present invention will be described with reference to the drawings.
<First Embodiment>
FIG. 1 is a diagram illustrating a multi-level encoding apparatus 100 according to the first embodiment of the present invention. In FIG. 1, the multi-level encoding apparatus 100 includes a transition probability calculation unit 101, a mapping information generation unit 102, a packet generation unit 103, and a data modulation unit 104.

  The transition probability calculation unit 101 calculates a transition probability 151 of a bit pattern between adjacent time slots for a plurality of bit patterns obtained by decomposing the transmission data 150 in units of time slots.

  Specifically, for example, the transmission data 150 is “00101110001010”, and bit patterns “00”, “01”, “10”, “11” are set to four amplitude levels (signal points) 0, 1, 2, and 3. The case of assigning will be described.

  First, the transmission data 150 “00101110001010” is disassembled in units of seven time slots, and bit patterns “00”, “10”, “11”, “10”, “00”, “10”, “10” each having two bits. "

  Next, the transition of the bit pattern between adjacent time slots is analyzed. Here, for the four types of transition source bit patterns “00”, “01”, “10”, and “11”, the transition destination bit patterns are “00”, “01”, “10”, and “11”, respectively. ”. That is, 16 transitions of bit patterns between adjacent time slots can occur.

  In the transmission data 150 “00101110001010”, the transition probability 151 of the bit pattern that actually occurs is calculated. Bit pattern transitions are “00” → “10”, “10” → “11”, “11” → “10”, “10” → “00”, “00” → “10”, “10” → Six of “10”. Then, the transition probability of “00” → “10” is 2/6, the transition probability of “10” → “11” is 1/6, the transition probability of “11” → “10” is 1/6, The transition probability of “10” → “00” is 1/6, and the transition probability of “10” → “10” is 1/6. In this way, the bit pattern transition probability 151 is derived.

  The mapping information generation unit 102 maps the bit pattern to a predetermined signal point based on the transition probability 151 calculated by the transition probability calculation unit 101, and generates mapping information 152. Here, the mapping information 152 is information indicating the correspondence between the bit pattern and the signal point, and is generated by the following procedure.

  FIG. 2 is a diagram illustrating a procedure for mapping a bit pattern to a predetermined signal point based on the transition probability 151 calculated by the transition probability calculation unit 101. FIG. 2A is a diagram illustrating the transition probability 151 calculated by the transition probability calculation unit 101. Here, as an example of the transition probability 151 calculated by the transition probability calculation unit 101, the transition probability from “00” to “00” is 5%, the transition probability from “00” to “01” is 16%, The transition probability from “00” to “10” is 11%, the transition probability from “00” to “11” is 8%,... 151 is shown in a matrix.

  Based on the transition probability 151 shown in FIG. 2 (1), the mapping information generation unit 102 assigns transition source and transition destination bit patterns to adjacent signal points in descending order of transition probability. FIG. 2 (2) is a diagram showing a procedure for mapping the transition source and transition destination bit patterns to adjacent signal points in descending order of transition probability.

  In the first step, bit patterns “00” and “01” are assigned to adjacent signal points for the transition “00” → “01” having the maximum transition probability. At this time, there are two types of mapping candidates: mapping assigned to “00”-“01” and mapping assigned to “01”-“00” in ascending order of amplitude level.

  In the second step, after following the mapping candidate obtained in the first step, the bit patterns “01” and “10” are adjacent to each other for the transition “01” → “10” having the second largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacent relationship in the mapping candidate obtained in the first step is not destroyed. When assigning bit patterns “01” and “10” to adjacent signal points, bit pattern adjacency relationships “00”-“01” and “01”-“00” in the mapping candidates obtained in the first step. Do not break. That is, two types of mapping candidates, ie, mappings assigned to “00”-“01”-“10” and mappings assigned to “10”-“01”-“00” are obtained in ascending order of amplitude level.

  In the third step, following the mapping candidate obtained in the second step, the bit patterns “00” and “10” are adjacent to each other for the transition “00” → “10” having the third largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacent relationship in the mapping candidate obtained in the second step is not destroyed. For bit patterns “00” and “10”, the bit pattern adjacency relationship in the mapping candidates obtained in the second step has already been determined. Therefore, in the third step, the mapping candidates “00”-“01”-“10” obtained in the second step without newly assigning the bit patterns “00” and “10” to the adjacent signal points are newly provided. ”And“ 10 ”-“ 01 ”-“ 00 ”.

  In the fourth step, after following the mapping candidate obtained in the third step, the bit patterns “01” and “11” are adjacent to each other for the transition “01” → “11” having the fourth largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacency relationship in the mapping candidates obtained in the third step is not destroyed. In the bit pattern adjacency relationship in the mapping candidate obtained in the third step, the upper and lower adjacent signal points have already been determined for the bit pattern “01”. Therefore, in the fourth step, the mapping candidates “00”-“01”-“10” obtained in the third step without newly assigning the bit patterns “01” and “11” to the adjacent signal points are newly provided. ”And“ 10 ”-“ 01 ”-“ 00 ”.

  In the fifth step, after following the mapping candidate obtained in the fourth step, the bit patterns “00” and “11” are adjacent to the transition “00” → “11” having the fifth highest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacency relationship in the mapping candidates obtained in the fourth step is not destroyed. When assigning bit patterns “00” and “11” to adjacent signal points, bit pattern adjacency relationships “00”-“01”-“10” and “10” in the mapping candidates obtained in the fourth step. -“01”-“00” is not broken. That is, 2 of the mapping assigned to “11”-“00”-“01”-“10” and the mapping assigned to “10”-“01”-“00”-“11” from the smallest amplitude level. Street mapping candidates are obtained.

  As described above, in the first to fifth steps, the mapping information generation unit 102 makes adjacent transition source and transition destination bit patterns in descending order of transition probability based on the transition probability 151 calculated by the transition probability calculation unit 101. Assign to the signal point to be used.

  In the first to fifth steps described above, two types of mapping candidates are obtained as a result, but the mapping information generation unit 102 uses one of the two types of mapping candidates to perform mapping information. 152 is generated. Here, any one of the two mapping candidates may be selected, but a selection criterion for selecting which one may be set. For example, in the third embodiment of the present invention to be described later, description is given regarding selecting the one with the smaller average transmission power among the two mapping candidates.

  In the first to fifth steps described above, when selecting a bit pattern having the same transition probability, for example, any one of the bit patterns may be selected based on a preset selection criterion. You may choose at random. Alternatively, mapping candidates may be further generated for each bit pattern having the same transition probability, and any one of the generated more mapping candidates may be selected.

  The packet generator 103 generates a packet 153 for data communication between the transmitter and the receiver. FIG. 3 is a diagram showing the configuration of the packet 153. In FIG. 3, the packet 153 includes a header portion 161 and a data portion 162.

  The header portion 161 includes a control information portion 163 that stores control information related to packet control, and a mapping information portion 164 that stores mapping information 152 generated by the mapping information generation portion 102. The configuration of the mapping information unit 164 includes, for example, a list of bit patterns to be assigned in ascending order of amplitude level. Specifically, when the bit patterns “11”, “00”, “01”, “10” are assigned to the amplitude levels 0, 1, 2, 3, respectively, the mapping information stored in the mapping information unit 164 152 becomes “11000110”.

  The data part 162 stores transmission data 150. Note that if information related to the data portion 162 is stored in the control information portion 163, the data stored in the data portion 162 may be a fixed length or a variable length.

  Based on the mapping information 152 generated by the mapping information generation unit 102, the data modulation unit 104 converts the bit pattern of the transmission data 150 stored in the data unit 162 of the packet 153 generated by the packet generation unit 103 to a predetermined signal point. Modulate to.

  Specifically, a case will be described in which bit patterns “11”, “00”, “01”, and “10” are assigned to amplitude levels 0, 1, 2, and 3 based on the mapping information 152, respectively. For example, if the transmission data 150 stored in the data portion 162 of the packet 153 is “10000111101011001”, the data modulation unit 104 modulates “10000111101011001” to “31002232”. Then, the multilevel encoding apparatus 100 outputs a multilevel modulation signal 154 obtained by modulating the transmission data 150 stored in the data part 162 of the packet 153.

  As described above, according to the multilevel encoding device 100 according to the first embodiment of the present invention, based on the transition probability 151 calculated by the transition probability calculation unit 101, the transition source and Transition destination bit patterns can be assigned to adjacent signal points. As a result, transitions with large amplitude level variations can be probabilistically reduced in the multilevel modulation signal, and the average transition amount can be kept small. Can avoid the signal point (amplitude level) identification error and achieve high-efficiency transmission.

  FIG. 4 is a diagram illustrating the multi-level decoding device 200 according to the first embodiment of the present invention. In FIG. 4, the multilevel decoding apparatus 200 decodes the multilevel modulation signal 154 output from the multilevel encoding apparatus 100, and includes a mapping analysis unit 201 and a data demodulation unit 202.

  The multilevel decoding apparatus 200 receives the multilevel modulation signal 154 transmitted from the multilevel encoding apparatus 100. Here, the multi-level modulation signal 154 has the configuration of the packet 153 shown in FIG. 3, and the transmission data 150 stored in the data section 162 is multi-level modulated as described above.

  The mapping analysis unit 201 extracts the mapping information 152 stored in the mapping information unit 164 from the received packet, and analyzes the extracted mapping information 152. Then, the multi-level decoding apparatus 200 grasps the mapping performed by the multi-level encoding apparatus 100.

  Data demodulating section 202 demodulates the data stored in data section 162 of the received packet based on mapping information 152 analyzed by mapping analyzing section 201 and outputs received data 155.

  Here, the data stored in the data part 162 of the packet received by the multilevel decoding device 200 is changed based on the transition probability 151 calculated by the transition probability calculation unit 101 in the multilevel encoding device 100. Mapping processing for assigning transition source and transition destination bit patterns to adjacent signal points in descending order of probability is performed. Therefore, when decoding the multilevel modulation signal 154 received from the multilevel encoding apparatus 100, the multilevel decoding apparatus 200 avoids signal point (amplitude level) identification errors due to insufficient transition and waveform distortion. I can do it.

  As described above, according to the multilevel encoding device 100 and the multilevel decoding device 200 according to the first embodiment of the present invention, a transition having a large amplitude level change is stochastically detected in the multilevel modulation signal. Since the average transition amount can be reduced and reduced, the circuit scale is not increased, signal point (amplitude level) identification errors due to effects such as insufficient transition and waveform distortion are avoided, and high-efficiency transmission is realized. be able to.

<Second Embodiment>
The configuration of the multi-level encoding apparatus according to the second embodiment of the present invention is the same as the configuration of the multi-level encoding apparatus 100 according to the first embodiment of the present invention shown in FIG. Is used. In the first embodiment of the present invention, the mapping information generation unit 102 assigns transition source and transition destination bit patterns to adjacent signal points in descending order of transition probability. However, in this embodiment, the mapping information generation unit 102 further assigns a bit pattern to a predetermined signal point by adding a restriction on the Hamming distance. In the present embodiment, differences from the first embodiment of the present invention will be described in detail.

  FIG. 5 is a diagram illustrating a procedure for mapping a bit pattern to a predetermined signal point based on the transition probability 151 calculated by the transition probability calculation unit 101 and the Hamming distance constraint. FIG. 5 (1) is a diagram showing the transition probability 151 calculated by the transition probability calculation unit 101, as in FIG. 2 (1). As in the first embodiment of the present invention, the mapping information generation unit 102 sets a constraint that further sets the Hamming distance to 1 in descending order of transition probability, based on the transition probability 151 shown in FIG. In addition, the transition source and transition destination bit patterns are assigned to adjacent signal points.

  The Hamming distance indicates the number of different values at corresponding bit positions of two bit patterns. For example, for the two bit patterns “01” and “10”, the values of the first bit and the second bit are different, so the Hamming distance is 2. For the four bit patterns “00”, “01”, “10”, and “11”, the combinations in which the Hamming distance is 1 are “00”-“01”, “00”-“10”, There are four types, “01”-“11” and “10”-“11”.

  FIG. 5B is a diagram illustrating a procedure for mapping the transition source and transition destination bit patterns to adjacent signal points by adding a constraint that the Hamming distance is 1 in order of increasing transition probability.

  In the first step, bit patterns “00” and “01” are assigned to adjacent signal points for the transition “00” → “01” having the maximum transition probability. Here, since the hamming distance between the bit patterns “00” and “01” is 1, mapping is assigned to “00”-“01” and assigned to “01”-“00” in ascending order of amplitude level. Two types of mapping candidates are obtained.

  In the second step, after following the mapping candidate obtained in the first step, the bit patterns “01” and “10” are adjacent to each other for the transition “01” → “10” having the second largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacent relationship in the mapping candidate obtained in the first step is not destroyed. When assigning bit patterns “01” and “10” to adjacent signal points, bit pattern adjacency relationships “00”-“01” and “01”-“00” in the mapping candidates obtained in the first step. Do not break. However, since the hamming distance between the bit patterns “01” and “10” is 2, the bit patterns “01” and “10” are newly obtained in the first step without being assigned to adjacent signal points. The mapping candidates “00”-“01” and “01”-“00” are taken over.

  In the third step, following the mapping candidate obtained in the second step, the bit patterns “00” and “10” are adjacent to each other for the transition “00” → “10” having the third largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacent relationship in the mapping candidate obtained in the second step is not destroyed. When assigning bit patterns “00” and “10” to adjacent signal points for bit patterns “00” and “10”, the bit pattern adjacency relationship “00” − in the mapping candidates obtained in the second step “01” and “01”-“00” are not broken. Further, since the hamming distance between the bit patterns “00” and “10” is 1, the mapping assigned to “10”-“00”-“01” in ascending order of the amplitude level and “01”-“00” "-" 10 "mapping candidates are obtained.

  In the fourth step, after following the mapping candidate obtained in the third step, the bit patterns “01” and “11” are adjacent to each other for the transition “01” → “11” having the fourth largest transition probability. Assign to signal points. Here, it is examined whether or not the bit pattern adjacency relationship in the mapping candidates obtained in the third step is not destroyed. When assigning bit patterns “01” and “11” to adjacent signal points for bit patterns “01” and “11”, the bit pattern adjacency relationship “10” − in the mapping candidates obtained in the third step Do not break “00”-“01” and “01”-“00”-“10”. Furthermore, since the hamming distance between the bit patterns “01” and “11” is 1, the mapping assigned to “10”-“00”-“01”-“11” in ascending order of the amplitude level, and “11 "-" 01 "-" 00 "-" 10 "and two types of mapping candidates are obtained.

  As described above, the mapping information generation unit 102 further sets the constraint that the Hamming distance is 1 in descending order of the transition probability based on the transition probability 151 calculated by the transition probability calculation unit 101 in the first to fourth steps. In addition, the transition source and transition destination bit patterns are assigned to adjacent signal points.

  As described above, according to the multi-level encoding device according to the second embodiment of the present invention, in the multi-level modulation signal, transitions with large amplitude level changes are stochastically reduced, and the average transition amount is reduced. Therefore, it is possible to avoid a signal point (amplitude level) identification error caused by insufficient transition, waveform distortion, and the like, and realize high-efficiency transmission without increasing the circuit scale.

  Furthermore, since the hamming distance of the bit pattern between adjacent signal points is 1 in a multi-level modulation signal, bit errors can be minimized with respect to identification errors of adjacent signal points (amplitude levels). . More specifically, when “01” and “10” having a Hamming distance of 2 are mapped to adjacent signal points, if an identification error occurs, a 2-bit error occurs. On the other hand, when mapping “00” and “10” having a Hamming distance of 1 to adjacent signal points, even if an identification error occurs, the error is only the first bit, and the bit error is minimized. Can be suppressed.

<Third Embodiment>
FIG. 6 is a diagram illustrating a multi-level encoding apparatus 300 according to the third embodiment of the present invention. In FIG. 6, the multi-level encoding apparatus 300 further includes an occurrence probability calculation unit 301 in addition to the multi-level encoding apparatus 100 according to the first embodiment of the present invention illustrated in FIG. 1. In FIG. 6, the same components as those shown in FIG. In the present embodiment, a mapping procedure using the occurrence probability calculation unit 301 having a configuration different from that of the first embodiment of the present invention will be described in detail.

  The occurrence probability calculation unit 301 calculates the occurrence probability of the bit pattern included in the transmission data 150.

  Specifically, for example, the transmission data 150 is “00101110001010”, and bit patterns “00”, “01”, “10”, “11” are set to four amplitude levels (signal points) 0, 1, 2, and 3. The case of assigning will be described.

  First, the transmission data 150 “00101110001010” is disassembled in units of seven time slots, and bit patterns “00”, “10”, “11”, “10”, “00”, “10”, “10” each having two bits. "

  Next, the bit pattern of each time slot is analyzed. Here, four types of bit patterns “00”, “01”, and “10” may occur.

  In transmission data 150 “00101110001010”, an actual bit pattern occurrence probability 351 is calculated. There are seven bit patterns, “00”, “10”, “11”, “10”, “00”, “10”, and “10”. Then, the occurrence probability of “00” is 2/7, the occurrence probability of “01” is 0/7, the occurrence probability of “10” is 4/7, and the occurrence probability of “11” is 1/7. . In this way, the bit pattern occurrence probability 351 is derived.

  The mapping information generation unit 102 maps a bit pattern to a predetermined signal point based on the transition probability 151 calculated by the transition probability calculation unit 101 and the occurrence probability 351 calculated by the occurrence probability calculation unit 301. Information 152 is generated. A procedure for generating the mapping information 152 will be described below.

  FIG. 7 is a diagram illustrating a procedure for mapping a bit pattern to a predetermined signal point based on the transition probability 151 calculated by the transition probability calculation unit 101 and the occurrence probability 351 calculated by the occurrence probability calculation unit 301. is there. FIG. 7A is a diagram illustrating the occurrence probability 351 calculated by the occurrence probability calculation unit 301. Here, as an example of the occurrence probability 351 calculated by the occurrence probability calculation unit 301, the occurrence probability of “00” is 40%, the occurrence probability of “01” is 30%, the occurrence probability of “10” is 20%, 11 "is 10%.

  Here, mapping of a bit pattern having a high occurrence probability will be described. FIG. 8 is a diagram illustrating a state in which a multi-level identification error occurs. In FIG. 8, transmission data is multilevel encoded into four values using signal points whose amplitude levels are 0, 1, 2, and 3. When decoding data that has been multi-value encoded into four values, there is a possibility that a signal point (amplitude level) identification error may occur due to the influence of noise generated in a receiver or the like.

  Specifically, for example, if the amplitude level of the signal point indicated by the signal level (1) is smaller than the first determination threshold due to the influence of noise, the adjacent signal level (0) is set. Identified. On the contrary, if the amplitude level of the signal point indicated by the signal level (1) becomes larger than the second determination threshold value due to the influence of noise, the signal point is identified as the adjacent signal level (2). As described above, the signal point indicated by the signal level (1) may be erroneously identified as both the upper and lower adjacent signal points due to the influence of noise. Similarly, the signal point indicated by the signal level (2) may be erroneously identified as the adjacent signal level (3) or the adjacent signal level (1).

  On the other hand, the signal point indicated by the signal level (0) is erroneously identified as the adjacent signal level (1) only when the amplitude level becomes larger than the first determination threshold due to the influence of noise. The That is, since the signal level indicated by the signal level (0) has the minimum amplitude level, it is not mistakenly identified as a signal point adjacent thereto. Similarly, since the signal point indicated by the signal level (3) has the maximum amplitude level, it is not mistakenly identified as an adjacent signal point.

  In this way, the bit pattern mapped to the signal point with the minimum or maximum amplitude level is less affected by noise generated in the receiver or the like than the bit pattern mapped to the signal point with the other amplitude level. Even in some cases, it is unlikely to be mistakenly identified.

  Therefore, the mapping information generation unit 102 maps the bit pattern having the maximum occurrence probability to the signal point having the minimum or maximum amplitude level, based on the occurrence probability 351 shown in FIG. Further, if the bit pattern with the highest occurrence probability is mapped to the signal point with the smallest amplitude level among the signal points with the smallest or largest amplitude level, the average transmission power can be reduced.

  FIG. 7 (2) is a diagram showing the transition probability 151 calculated by the transition probability calculation unit 101, as in FIG. 2 (1). FIG. 7 (3) maps the bit pattern having the maximum occurrence probability to the signal point having the minimum amplitude level based on the occurrence probability 351 shown in FIG. 7 (1). It is a figure which shows the procedure which maps a transition source and a transition destination bit pattern to an adjacent signal point in order with a large transition probability based on the shown transition probability.

  In the first step, the bit pattern “00” having the highest occurrence probability is assigned to the signal point having the minimum amplitude level. When selecting bit patterns having the same occurrence probability, for example, any bit pattern may be selected based on a preset selection criterion, or may be selected randomly. Alternatively, a mapping candidate may be generated for each bit pattern having the same transition probability, and any one of the generated mapping candidates may be selected.

  After the second step, as in the first embodiment of the present invention, based on the transition probability 151 shown in FIG. 7B, the transition source and transition destination bit patterns are adjacent in descending order of transition probability. Assign to signal points.

  In the second step, after following the mapping candidate obtained in the first step, for the transition “00” → “01” having the maximum transition probability, the bit patterns “00” and “01” are adjacent signals. Assign to a point. Accordingly, mapping candidates to be assigned to “00”-“01” are obtained in ascending order of amplitude level.

  In the third step, after following the mapping candidate obtained in the second step, the bit patterns “01” and “10” are adjacent to each other for the transition “01” → “10” having the second largest transition probability. Assign to signal points. Therefore, mapping candidates to be assigned to “00”-“01”-“10” are obtained in ascending order of amplitude level.

  In the fourth step, since only the bit pattern “11” is unassigned, it is automatically assigned to an unassigned signal point (amplitude level 3). Therefore, mapping candidates to be assigned to “00”-“01”-“10”-“11” are obtained in ascending order of amplitude level.

  As described above, the mapping information generation unit 102 maps the bit pattern having the maximum occurrence probability to the signal point having the minimum amplitude level based on the occurrence probability 351 in the first to fourth steps, and the transition probability 151. Based on the above, the transition source and transition destination bit patterns are assigned to adjacent signal points in descending order of transition probability.

  As described above, according to the multi-level encoding device 300 according to the third embodiment of the present invention, in the multi-level modulation signal, transitions with large amplitude level changes are stochastically reduced, and the average transition amount is reduced. Since it can be kept small, the circuit scale is not increased, signal point (amplitude level) identification errors due to insufficient transition and waveform distortion are avoided, and high-efficiency transmission can be realized.

  Furthermore, by mapping the bit pattern with the highest probability of occurrence to the signal point with the minimum or maximum amplitude level, the occurrence rate of multi-level identification errors in the multi-level modulation signal can be reduced.

  Further, the average transmission power can be reduced by mapping the bit pattern having the highest occurrence probability to the signal point having the minimum amplitude level among the signal points having the minimum or maximum amplitude level.

  In the present embodiment, in the mapping procedure shown in FIG. 7 (3), after the second step, as shown in the second embodiment of the present invention, a constraint that the Hamming distance is 1 is further added. The transition source and transition destination bit patterns may be assigned to adjacent signal points.

  In this embodiment, as a means for reducing the average transmission power, the bit pattern with the highest occurrence probability is mapped to the signal point with the smallest amplitude level among the signal points with the smallest or largest amplitude level. It was. However, it is possible to generate a plurality of mapping candidates, calculate an average transmission power when each mapping candidate is selected, and select a mapping candidate having a low average transmission power from among the plurality of generated mapping candidates. .

  Specifically, for example, two mapping candidates shown in the first embodiment of the present invention will be described. As shown in FIG. 2, in the first embodiment of the present invention, mappings assigned to “11”-“00”-“01”-“10” from the smallest amplitude level and “10”-“ Two mapping candidates with the mapping assigned to “01”-“00”-“11” were obtained. A procedure for calculating the average transmission power when each mapping candidate is selected and selecting a mapping candidate having a low average transmission power from the two mapping candidates will be described below.

  FIG. 9 is a diagram showing a procedure for calculating the average transmission power when each mapping candidate is selected for the two mapping candidates shown in FIG. The average transmission power is calculated by the sum of values obtained by multiplying the amplitude level to which the bit pattern is mapped and the occurrence probability of the bit pattern. Here, the transmission data is multi-value encoded into four values using signal points with amplitude levels of 0, 1, 2, and 3. Further, the occurrence probability 351 calculated by the occurrence probability calculation unit 301 is, as shown in FIG. 7A, the occurrence probability of “00” is 40%, the occurrence probability of “01” is 30%, and “10”. It is assumed that the occurrence probability of “11” is 20% and the occurrence probability of “11” is 10%.

First, in FIG. 9A, the average transmission power Wa when the mapping candidate A (“11” − “00” − “01” − “10”) is selected is calculated.
Average transmission power Wa = (0 × 0.1) + (1 × 0.4) + (2 × 0.3) + (3 × 0.2) = 1.6

Next, in FIG. 9B, the average transmission power Wb when the mapping candidate B (“10” − “01” − “00” − “11”) is selected is calculated.
Average transmission power Wb = (0 × 0.2) + (1 × 0.3) + (2 × 0.4) + (3 × 0.1) = 1.4

  Thus, the average transmission power Wb when the mapping candidate B is selected is smaller than the average transmission power Wa when the mapping candidate A is selected. Therefore, for the two mapping candidates shown in FIG. 2, if the mapping to be assigned to “10”-“01”-“00”-“11” is selected from the smallest amplitude level, the average transmission power is reduced. can do.

  In the above description, the average transmission power is calculated using the amplitude level to which the bit pattern is mapped. However, actually, the transmission voltage value (V), current value (A), and power value ( It is desirable to calculate using W) or the like.

<Fourth Embodiment>
FIG. 10 is a diagram showing a multi-level encoding apparatus 400 according to the fourth embodiment of the present invention. In FIG. 10, the multi-level encoding apparatus 400 further includes a mapping table 401 in addition to the multi-level encoding apparatus 100 according to the first embodiment of the present invention shown in FIG. 10, the same components as those shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is omitted. In the present embodiment, a mapping procedure using the mapping table 401 having a configuration different from that of the first embodiment of the present invention will be described in detail.

  FIG. 11 is a diagram showing a mapping table 401 when bit patterns “00”, “01”, “10”, and “11” are assigned to four amplitude levels (signal points) 0, 1, 2, and 3. . In FIG. 11, since the mapping table 401 assigns bit patterns to four amplitude levels, there are 24 (= 4!) Mapping methods. That is, the mapping table 401 stores in advance all types for assigning bit patterns to four amplitude levels.

  The mapping information generation unit 102 assigns a bit pattern to a predetermined signal point by the mapping procedure described in the first to third embodiments of the present invention. Thereafter, the mapping information generation unit 102 refers to the mapping table 401 and generates a mapping number corresponding to the assigned mapping as the mapping information 152.

  Specifically, for example, as shown in the first embodiment of the present invention, the mapping information generation unit 102 performs bit patterns “11” and “00” for amplitude levels 0, 1, 2, and 3, respectively. ”,“ 01 ”, and“ 10 ”will be described. In this case, the mapping information generation unit 102 refers to the mapping table 401 and generates the mapping number “18” as the mapping information 152.

  Then, the packet generation unit 103 generates the packet 153 by storing the mapping number “18” as the mapping information 152 in the mapping information unit 164 of the packet 153 shown in FIG.

Here, the mapping information 152 stored in the mapping information unit 164 shown in the first embodiment of the present invention is “11000110” and is 8-bit information. On the other hand, when the mapping number “18” is expressed in binary, “10010” is 5-bit information. That is, by using the mapping table 401 as the mapping information 152, the information amount of the mapping information 152 stored in the mapping information unit 164 can be reduced. In the case of the mapping table 401 shown in FIG. 11, the mapping number “0” to “23” has an information amount of 5 bits (2 ⁵ = 32 possible representations).

  In addition, as shown in the second embodiment of the present invention, the bit pattern mapped to the adjacent signal point is further stored in the mapping information unit 164 when the restriction of the Hamming distance is 1 is added. The information amount of the mapping information 152 can be reduced.

FIG. 12 is a diagram showing the mapping table 402 when a constraint that the Hamming distance is 1 is added to the bit pattern mapped to the adjacent signal point in the mapping table 401 shown in FIG. In FIG. 12, the mapping table 402 stores mapping types limited to eight types. That is, for the mapping numbers “0” to “7”, the information amount is 3 bits (2 ³ = 8 possible representations).

  As described above, according to the multilevel encoding device 400 according to the fourth embodiment of the present invention, the amount of information of the mapping information 152 stored in the mapping information unit 164 can be reduced, and thus the overhead is small. By using the packet 153, more excellent high-efficiency transmission can be realized.

  FIG. 13 is a diagram showing a multi-level decoding apparatus 500 according to the fourth embodiment of the present invention. In FIG. 13, the multi-level decoding device 500 further includes a mapping table 401 in addition to the multi-level decoding device 200 according to the first embodiment of the present invention shown in FIG. In FIG. 13, the same components as shown in FIG. 4 are denoted by the same reference numerals, and the description thereof is omitted. In the present embodiment, analysis of mapping information using the mapping table 401 having a configuration different from that of the first embodiment of the present invention will be described in detail.

  The mapping table 401 included in the multi-level decoding device 500 has the same contents as the mapping table 401 included in the multi-level encoding device 100 shown in FIG. Specifically, when the multilevel encoding apparatus 100 includes the mapping table 401 illustrated in FIG. 11, the multilevel decoding apparatus 500 includes the same mapping table 401, and the multilevel encoding apparatus 100 is illustrated in FIG. When the illustrated mapping table 402 is provided, the multilevel decoding apparatus 500 also includes the same mapping table 402.

  The mapping analysis unit 201 extracts the mapping information 152 stored in the mapping information unit 164 from the received packet, and analyzes the extracted mapping information 152. At this time, the mapping information 152 stored in the mapping information unit 164 is a mapping number. For this reason, the mapping analysis unit 201 refers to the mapping table 401, extracts mapping information corresponding to the mapping number, and grasps the mapping performed by the multi-level encoding device 100.

  Data demodulating section 202 demodulates the data stored in data section 162 of the received packet based on mapping information 152 analyzed by mapping analyzing section 201 and outputs received data 155.

  As described above, according to the multilevel encoding apparatus 400 and the multilevel decoding apparatus 500 according to the fourth embodiment of the present invention, the information amount of the mapping information 152 stored in the mapping information unit 164 is reduced. Therefore, superior high-efficiency transmission can be realized using the packet 153 with low overhead.

<Fifth Embodiment>
The configuration of the multi-level encoding apparatus according to the fifth embodiment of the present invention is the same as the configuration of the multi-level encoding apparatus 100 according to the first embodiment of the present invention shown in FIG. Is used. In the first embodiment of the present invention, the packet generation unit 103 generates the packet 153 shown in FIG. 3, but in this embodiment, the packet generation unit 103 generates a packet configured in segment units. To do. In the present embodiment, the configuration of a packet that is different from the configuration of the first embodiment of the present invention will be described in detail.

  First, the mapping information generation unit 102 subdivides the transmission data 150 into N segments (N is an integer of 2 or more) of a predetermined length, and the above-described first to first aspects of the present invention are applied to the segment unit segment data. The mapping process shown in the fourth embodiment is performed. That is, the mapping information generation unit 102 generates N pieces of mapping information corresponding to N pieces of segment data obtained by subdividing the transmission data 150 into segments.

  Then, the packet generation unit 103 generates a packet based on the N pieces of mapping information and the N pieces of segment data. FIG. 14 is a diagram illustrating a configuration of a packet generated by the packet generation unit 103 in the multi-level encoding device according to the fifth embodiment of the present invention. 14 (1) and (2), N segment data obtained by subdividing the transmission data 150 into segment units are a first data unit 162-1, a second data unit 162-2,. , And the Nth data portion 162-N. Also, N pieces of mapping information corresponding to N pieces of segment data stored in the first data portion 162-1, the second data portion 162-2,... And the Nth data portion 162-N. Are stored in the first mapping information section 164-1, the second mapping information section 164-2,..., And the Nth mapping information section 164-N, respectively.

  Based on the mapping information stored in the mapping information section, the data modulation section 104 modulates the bit pattern of the segment data stored in the corresponding data section to a predetermined signal point.

  Note that the configuration of the packet generated by the packet generation unit 103 is not limited to the configuration shown in FIGS. 14 (1) and (2). It may be a packet configuration in which the relationship between N segment data subdivided into segment units and N mapping information corresponding to the N segment data is secured.

  Since the transmission data 150 is basically composed of a random bit pattern, as the data length increases, the bit pattern transition probability 151 and the occurrence probability 351 all tend to approach equal values. In other words, by subdividing the transmission data 150, mapping processing according to the bias of the bit pattern transition probability 151 and / or occurrence probability 351 is performed.

  As described above, according to the multi-level encoding device according to the fifth embodiment of the present invention, the transmission data 150 included in one packet is subdivided into a plurality of segments, so that each segment is optimum. Mapping processing can be realized.

<Sixth Embodiment>
The configuration of the multi-level encoding apparatus according to the sixth embodiment of the present invention is the same as the configuration of the multi-level encoding apparatus 300 according to the third embodiment of the present invention shown in FIG. Is used. In the third embodiment of the present invention, the packet generation unit 103 generates the packet 153 shown in FIG. 3. However, in this embodiment, the packet generation unit 103 further includes a packet having an average transmission power information unit. Is generated. In the present embodiment, data communication using a packet including an average transmission power information unit will be described in detail.

  FIG. 15 is a diagram illustrating a configuration of a packet generated by the packet generation unit 103 in the multi-level encoding device according to the sixth embodiment of the present invention. 15 differs from the packet 153 shown in FIG. 3 in that the header portion 161 includes an average transmission power information portion 165.

  As shown in the third embodiment of the present invention, the mapping information generation unit 102 calculates the average transmission power based on the amplitude level to which the bit pattern is mapped and the occurrence probability of the bit pattern. Then, as illustrated in FIG. 15, the packet generation unit 103 generates a packet including information on the average transmission power calculated by the mapping information generation unit 102.

  FIG. 16 is a diagram showing a multi-level decoding device 600 according to the sixth embodiment of the present invention. In FIG. 16, the multilevel decoding apparatus 600 further includes a DC drift amount estimation unit 601 in addition to the multilevel decoding apparatus 200 according to the first embodiment of the present invention shown in FIG. In FIG. 16, the same components as those shown in FIG. 4 are denoted by the same reference numerals, and the description thereof is omitted. In the present embodiment, the DC drift amount estimation unit 601 having a configuration different from that of the first embodiment of the present invention will be described in detail.

  The DC drift amount estimation unit 601 acquires the average transmission power information included in the average transmission power information unit 165 of the packet illustrated in FIG. Then, the DC drift amount estimation unit 601 estimates a DC drift amount generated when demodulating data based on the acquired average transmission power information, and generates a DC drift amount estimated value 171.

  Based on the mapping information 152 analyzed by the mapping analysis unit 201, the data demodulation unit 202 biases the threshold in signal determination based on the DC drift amount estimation value 171 generated by the DC drift amount estimation unit 601. The data stored in the data portion 162 of the received packet is demodulated. That is, the DC component included in the transmission signal can be canceled, and the accuracy of multi-level decoding of the transmission data included in the data unit 162 can be improved.

  As described above, according to the multi-level decoding apparatus 600 according to the sixth embodiment of the present invention, more accurate multi-level decoding can be performed by biasing the threshold in signal determination based on the DC drift amount estimated value 171. Value decoding can be realized.

  In the first to sixth embodiments of the present invention, for a plurality of bit patterns obtained by decomposing data in units of time slots, a 2-bit bit pattern is modulated into a multilevel signal having a four-level amplitude level. However, it is not limited to this. For example, a 3-bit bit pattern may be modulated into a multilevel signal having an 8-level amplitude level. In addition, it is needless to say that the same effect can be obtained even in a configuration that modulates an arbitrary multilevel signal.

  The multi-level encoding apparatus, multi-level decoding apparatus, multi-level encoding method, and multi-level decoding method of the present invention are useful for highly efficient data transmission using multi-level encoding.

100, 300, 400 Multi-level encoding apparatus 101 Transition probability calculation unit 102 Mapping information generation unit 103 Packet generation unit 104 Data modulation unit 150 Transmission data 151 Transition probability 152 Mapping information 153 Packet 154 Multi-level modulation signal 155 Reception data 161 Header part 162, 162-1 to 162-N Data part 163 Control information part 164, 164-1 to 164-N Mapping information part 165 Average transmission power information part 171 DC drift amount estimated value 200, 500, 600 Multilevel decoding apparatus 201 Mapping analysis unit 202 Data demodulation unit 301 Occurrence probability calculation unit 351 Occurrence probability 401, 402 Mapping table 601 DC drift amount estimation unit

Claims (16)

A multi-level encoding device that modulates data contained in a packet communicated between a transmitter and a receiver into a multi-level signal having a plurality of predetermined amplitude levels,
For a plurality of bit patterns obtained by decomposing the data in predetermined time slot units, a transition probability calculating unit that calculates a transition probability of bit patterns between adjacent time slots;
Based on the calculated transition probability, mapping the bit pattern to a predetermined signal point so as to reduce the fluctuation of the amplitude level of the multilevel signal, and showing the correspondence between the bit pattern and the signal point A mapping information generator for generating information;
A packet generation unit that generates a packet including a header part including the mapping information and a data part including the data;
A multi-level encoding apparatus comprising: a data modulation unit that modulates data included in the data unit to the predetermined signal point based on the mapping information.   The multi-level encoding apparatus according to claim 1, wherein the mapping information generation unit sequentially maps the bit patterns having a large transition probability to adjacent signal points.   The multi-level encoding apparatus according to claim 1, wherein the mapping information generation unit maps the bit pattern having a Hamming distance of 1 to adjacent signal points. An occurrence probability calculation unit for calculating an occurrence probability of a bit pattern included in the data,
The mapping information generation unit maps a bit pattern having the maximum occurrence probability calculated by the occurrence probability calculation unit to a signal point having the minimum or maximum amplitude level. The multi-level encoding device according to any one of the above. An occurrence probability calculation unit for calculating an occurrence probability of a bit pattern included in the data,
The mapping information generation unit maps a bit pattern having the maximum occurrence probability calculated by the occurrence probability calculation unit to a signal point having the minimum amplitude level. A multi-level encoding device according to claim 1.   The mapping information generation unit calculates an average transmission power based on the occurrence probability calculated by the occurrence probability calculation unit and the generated mapping information, according to any one of claims 4 to 5, The multi-level encoding device described.   The mapping information generation unit, based on the calculated average transmission power, maps the bit pattern to a predetermined signal point so that the average transmission power is minimized. Multi-level encoding device.   The multi-level encoding apparatus according to claim 6, wherein the header part includes average transmission power information indicating the calculated average transmission power. A mapping table that stores in advance mapping information indicating a correspondence relationship between the bit pattern and the signal point, and a mapping number corresponding to the mapping information;
The multiple mapping information according to any one of claims 1 to 8, wherein the mapping information generation unit refers to the mapping table and uses a mapping number corresponding to the generated mapping information as the mapping information. Value encoding device. The transition probability calculation unit calculates the transition probability for each segment for segment data obtained by subdividing the data into a plurality of segments having a predetermined length.
The mapping information generation unit generates the mapping information for each of the segments for the segment data based on the transition probability calculated for each of the segments. The multi-level encoding device described. The transition probability calculation unit calculates the transition probability for each segment for segment data obtained by subdividing the data into a plurality of segments having a predetermined length.
The occurrence probability calculation unit calculates the occurrence probability for each segment for the segment data,
The mapping information generation unit generates the mapping information for each segment for the segment data based on the transition probability and occurrence probability calculated for each segment. The multi-level encoding device according to any one of the above. A multi-level decoding device for demodulating a multi-level signal having a plurality of predetermined amplitude levels included in a packet communicated between a transmitter and a receiver,
Correspondence between a predetermined signal point and a bit pattern in which a bit pattern having a large transition probability between adjacent time slots is sequentially mapped to adjacent signal points for a plurality of bit patterns decomposed in predetermined time slot units Mapping information indicating that the mapping information is extracted from the packet and the extracted mapping information is analyzed;
A multi-level decoding device comprising: a data demodulator that demodulates the multi-level signal based on the mapping information. A bit pattern in which bit patterns having a large transition probability of bit patterns between adjacent time slots are sequentially mapped to adjacent signal points with respect to the predetermined signal points and a plurality of bit patterns decomposed in units of the predetermined time slots. A mapping table that stores in advance mapping information indicating a correspondence relationship and a mapping number corresponding to the mapping information;
The packet includes a mapping number as the mapping information,
The mapping analysis unit refers to the mapping table and extracts mapping information corresponding to a mapping number included in the packet;
The multi-level decoding apparatus according to claim 12, wherein the data demodulator demodulates the multi-level signal based on the extracted mapping information. The packet includes average transmission power information calculated based on the bit pattern occurrence probability and the mapping information for a plurality of bit patterns decomposed in units of the predetermined time slot,
A DC drift amount estimation unit that generates a DC drift amount estimate based on the average transmission power information, further comprising:
When the multi-level signal is demodulated based on the estimated DC drift amount, the data demodulator adjusts a threshold value in the signal determination of the amplitude level, and determines the multi-level signal based on the mapping information. The multi-level decoding device according to claim 12, wherein the multi-level decoding device is demodulated. A multi-level encoding method executed by a multi-level encoding apparatus that modulates data into a multi-level signal having a predetermined plurality of amplitude levels,
A transition probability calculation step for calculating a transition probability of a bit pattern between adjacent time slots for a plurality of bit patterns obtained by decomposing the data in predetermined time slot units;
Based on the calculated transition probability, mapping the bit pattern to a predetermined signal point so as to reduce the fluctuation of the amplitude level of the multilevel signal, and showing the correspondence between the bit pattern and the signal point A mapping information generation step for generating information;
And a data modulation step of modulating the data to the predetermined signal point based on the mapping information. A multilevel decoding method executed by a multilevel decoding device that demodulates a multilevel signal having a plurality of predetermined amplitude levels,
Correspondence between a predetermined signal point and a bit pattern in which a bit pattern having a large transition probability between adjacent time slots is sequentially mapped to adjacent signal points for a plurality of bit patterns decomposed in predetermined time slot units A mapping analysis step for analyzing mapping information indicating
And a data demodulation step of demodulating the multilevel signal based on the mapping information.

Images