阅读说明：本技术 似然度生成装置 (Likelihood generation device ) 是由 土肥庆亮 于 2017-06-27 设计创作，主要内容包括：似然度生成装置具备：相似性检测部(101),其被输入调制方式选择信号(10)和接收值(11),根据调制方式选择信号(10)检测接收值(11)中包含的各个信息比特之间的似然度的相似性,输出指定应检索的似然度数据的似然度选择信号(13)和指定针对似然度数据的运算的运算选择信号(14)；似然度表参照部(102),其登记表示各个信息比特的似然度的似然度分布中的与其它小区域不重复的小区域作为似然度数据,从似然度表提取基于似然度选择信号(13)的似然度数据；以及相似性处理部(103),其对提取出的似然度数据进行由运算选择信号(14)指定的运算,由此求出似然度分布的整体。(The likelihood generation device includes: a similarity detection unit (101) to which a modulation scheme selection signal (10) and a reception value (11) are input, which detects the similarity of likelihoods between information bits included in the reception value (11) on the basis of the modulation scheme selection signal (10), and which outputs a likelihood selection signal (13) that specifies likelihood data to be searched and an operation selection signal (14) that specifies an operation for the likelihood data; a likelihood table reference unit (102) which registers, as likelihood data, small regions that do not overlap with other small regions in a likelihood distribution indicating the likelihood of each information bit, and extracts likelihood data based on a likelihood selection signal (13) from a likelihood table; and a similarity processing unit (103) that performs an operation specified by the operation selection signal (14) on the extracted likelihood data to thereby obtain the entire likelihood distribution.)

1. A likelihood generation device that generates a likelihood of each information bit included in a reception value modulated by an arbitrary modulation scheme, the likelihood generation device comprising:

a similarity detection unit which receives a modulation scheme selection signal for specifying a modulation scheme and the received value, detects similarity of likelihoods between the information bits included in the received value based on the modulation scheme selection signal, and outputs, as a detection result, a likelihood selection signal for specifying likelihood data to be searched in a likelihood table and an operation selection signal for specifying a type of operation to be performed on the likelihood data;

a likelihood table reference unit that divides the entire likelihood distribution indicating the likelihood of each of the information bits into a plurality of small regions, stores the likelihood table in which only small regions having a likelihood that does not overlap with other small regions are registered as likelihood data, and extracts the likelihood data specified by the likelihood selection signal from the likelihood table; and

and a similarity processing unit configured to perform the operation specified by the operation selection signal on the likelihood data extracted by the likelihood table reference unit, thereby obtaining and outputting the entire likelihood distribution.

2. A likelihood generation apparatus according to claim 1, wherein,

the similarity detection unit stores in advance an input/output table defining in advance a relationship between a section of the reception value and the likelihood selection signal and the operation selection signal,

the similarity detection unit extracts the likelihood selection signal and the operation selection signal corresponding to the section of the received value from the input/output table when the section of the received value is input.

3. A likelihood generation device that generates a likelihood of each information bit included in a reception value modulated by an arbitrary modulation scheme, the likelihood generation device comprising:

a similarity detection unit which receives a modulation scheme selection signal for specifying a modulation scheme and the received value, detects similarity of likelihoods between the information bits included in the received value based on the modulation scheme selection signal, and outputs, as a detection result, a 1 st likelihood selection signal for provisionally specifying likelihood data to be searched in the likelihood table and an operation selection signal for specifying a type of operation to be performed on the 1 st likelihood selection signal;

a similarity processing unit that performs the operation specified by the operation selection signal on the 1 st likelihood selection signal and the reception value output from the similarity detection unit, thereby outputting a 2 nd likelihood selection signal that finally specifies likelihood data to be searched in the likelihood table; and

a likelihood table reference unit that divides the entire likelihood distribution indicating the likelihood of each of the information bits into a plurality of small regions, stores the likelihood table in which only small regions having a likelihood that does not overlap with other small regions are registered as likelihood data, and extracts the likelihood data specified by the 2 nd likelihood selection signal from the likelihood table.

4. A likelihood generation apparatus according to claim 3, wherein,

the similarity detection unit stores in advance an input/output table defining in advance a relationship between a section of the reception value and the 1 st likelihood selection signal and the operation selection signal,

when the reception value is input to the section, the similarity detection unit extracts the 1 st likelihood selection signal and the operation selection signal corresponding to the section of the reception value from the input/output table.

Technical Field

The present invention relates to a likelihood generation device.

Background

In recent years, in order to increase the capacity of an optical communication system and improve the frequency utilization efficiency, multi-level modulation and demodulation techniques such as QPSK (Quadrature Phase-shift Keying), 8QAM (Quadrature amplitude modulation), and 16QAM have been used.

In the multi-level modulation and demodulation techniques such as QPSK, 8QAM, and 16QAM, which have been conventionally used for DPSK (Differential Phase-Shift Keying), etc., a higher signal-to-Noise ratio (snr) is required to achieve the same transmission distance and signal speed because the intervals between signal points are short.

In order to compensate for the SNR deficiency, an error correction decoding method in which soft decision decoding is combined is generally used for a strong error correction code such as an LDPC (Low Density Parity Check) code. In soft decision decoding, likelihood indicating the certainty of a bit assigned to a signal point is calculated from the position of the received signal point, and error correction is performed using the likelihood.

The calculation algorithm for calculating the likelihood differs depending on the modulation scheme. For example, in a wireless system, modulation schemes such as QPSK, 8QAM, 16QAM, 64QAM, FSK (Frequency Shift Keying), BPSK (Binary Phase Shift Keying) and the like are often used, but likelihood calculation algorithms differ according to these modulation schemes.

On the other hand, as a conventional technique, a likelihood generation circuit for switching to another modulation scheme by rotating the phase of a received signal, a circuit simplification by limiting the range of likelihood generation, and the like have been proposed.

For example, in patent document 1, a received symbol likelihood generation circuit for modulating 8QAM received symbols into QPSK by phase rotation is used. This technique described in patent document 1 is a technique for reducing the circuit scale by utilizing the symmetry of likelihood distribution, and corresponds to 8QAM as a modulation method.

In general, when calculating the likelihood of a signal point, the likelihood of a signal point is calculated from information of a constellation that maps a signal point of a digitally modulated signal onto a two-dimensional plane formed by 4 quadrants 1 to 4.

Patent document 2 proposes a technique for generating likelihood only in the first quadrant by rotating the phase of a received signal in order to reduce the circuit scale. In this technique described in patent document 2, only QAM is applied as a modulation method.

Disclosure of Invention

Problems to be solved by the invention

As described above, the conventional techniques described in patent documents 1 and 2 have a problem that the target modulation scheme is limited. The reason why the modulation scheme must be defined is that the characteristics of the likelihood distribution differ depending on the modulation scheme.

Next, the likelihood distributions of QPSK and 32QAM will be described with reference to fig. 1A to 2E. Fig. 1A and 1B show examples of likelihood distributions #1 and #2 of QPSK, respectively, and fig. 2A to 2E show examples of likelihood distributions #1, #2, #3, #4, and #5 of 32 QAM. As shown in these figures, the likelihood distribution represents the likelihood with respect to a received symbol composed of an I component and a Q component, the I component being taken on the horizontal axis and the Q component being taken on the vertical axis, and the lightness of color. Here, a Log-likelihood ratio (LLR) is shown as a likelihood. Specifically, the likelihood distribution means that the likelihood is closer to positive infinity as the color is closer to white, and the likelihood is closer to negative infinity as the color is closer to black. The line in the distribution is a contour line formed by connecting portions having the same likelihood.

One of the large differences in the likelihood distributions due to the modulation scheme is that, as shown in fig. 1A and 1B, 2 likelihood values per symbol are generated from the received signal modulated to QPSK, but as shown in fig. 2A to 2E, 5 likelihood values per symbol are generated from the received signal modulated to 32 QAM. Each likelihood distribution represents a different situation. For example, the likelihood distribution of QPSK varies only in a certain dimension as shown in fig. 1A and 1B. If the likelihood distribution #1 is present, the change is only in the horizontal direction, and if the likelihood distribution #2 is present, the change is only in the vertical direction. In both likelihood distribution #1 and likelihood distribution #2, the interval between contours is substantially constant. On the other hand, as is clear from fig. 2A to 2E, the likelihood distribution of 32QAM is different from the likelihood distribution of QPSK, although it has regularity in each of the distributions #1 to # 5. As described above, although the likelihood distribution has regularity in each modulation scheme, the regularity differs for each modulation scheme, and thus likelihood calculation processing cannot be handled by 1 circuit.

The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a likelihood generating device capable of generating a likelihood with a minimum table capacity without limiting a modulation scheme and reducing a circuit scale.

Means for solving the problems

The present invention provides a likelihood generation device that generates a likelihood of each information bit included in a reception value modulated by an arbitrary modulation scheme, the likelihood generation device including: a similarity detection unit which receives a modulation scheme selection signal for specifying a modulation scheme and the received value, detects similarity of likelihoods between the information bits included in the received value based on the modulation scheme selection signal, and outputs, as a detection result, a likelihood selection signal for specifying likelihood data to be searched in a likelihood table and an operation selection signal for specifying a type of operation to be performed on the likelihood data; a likelihood table reference unit that divides the entire likelihood distribution indicating the likelihood of each of the information bits into a plurality of small regions, stores the likelihood table in which only small regions having a likelihood that does not overlap with other small regions are registered as likelihood data, and extracts the likelihood data specified by the likelihood selection signal from the likelihood table; and a similarity processing unit configured to perform the operation specified by the operation selection signal on the likelihood data extracted by the likelihood table reference unit, thereby obtaining and outputting the entire likelihood distribution.

Effects of the invention

According to the likelihood generating device of the present invention, by performing preprocessing for detecting the feature of the likelihood distribution for each modulation scheme before calculating the likelihood, the likelihood can be generated with the minimum table capacity without limiting the modulation scheme, and therefore, the circuit scale can be reduced.

Drawings

Fig. 1A is a diagram showing an example of likelihood distribution #1 of QPSK.

Fig. 1B is a diagram showing an example of likelihood distribution #2 of QPSK.

Fig. 2A is a diagram showing an example of likelihood distribution #1 of 32 QAM.

Fig. 2B is a diagram showing an example of likelihood distribution #2 of 32 QAM.

Fig. 2C is a diagram showing an example of likelihood distribution #3 of 32 QAM.

Fig. 2D is a diagram showing an example of likelihood distribution #4 of 32 QAM.

Fig. 2E is a diagram showing an example of likelihood distribution #5 of 32 QAM.

Fig. 3 is a block diagram showing the configuration of a likelihood generating device according to embodiment 1 of the present invention.

Fig. 4A is a diagram showing an example of the entire original likelihood distribution including small regions stored in the likelihood table in the likelihood generating device according to embodiment 1 of the present invention.

Fig. 4B is a diagram showing a small region in the original likelihood distribution of fig. 4A stored in the likelihood table in the likelihood generating device according to embodiment 1 of the present invention.

Fig. 5 is a diagram showing an example of input and output of the similarity detection unit in the likelihood generation device according to embodiment 1 of the present invention.

Fig. 6 is a block diagram showing the configuration of a likelihood generating device according to embodiment 2 of the present invention.

Fig. 7 is a block diagram showing a hardware configuration of a likelihood generating device according to embodiment 3 of the present invention.

Fig. 8 is a block diagram showing the configuration of a likelihood generating device according to embodiment 3 of the present invention.

Detailed Description

Next, an embodiment of the likelihood generating device according to the present invention will be described with reference to the drawings.