TWI791023B - Method and apparatus for encoding input data as polar code, decoding method and apparatus for decoding code word - Google Patents

Abstract

A method of encoding input data as a polar code includes generating unfrozen bits by adding at least one designated information bit to information bits which have been generated based on the input data, reordering the unfrozen bits and frozen bits by assigning the unfrozen bits to polarized sub-channels having higher reliability than the frozen bits having a value known to both of an encoder and a decoder, and generating a code word by polar-coding results of the reordering may be provided. The at least one designated information bit may have the value known to both the encoder and the decoder.

Description

Method and device for encoding input data into polar code, and decoding method Method and device for decoding codewords [Cross Reference to Related Application]

This application requests Korean Patent Application No. 10-2017-0071888 filed on June 8, 2017 and Korean Patent Application No. 10-2017-0181521 filed on December 27, 2017 at 35 U.S.C.§ 119, the disclosure of which is incorporated herein by reference in its entirety.

The inventive concept relates to data encoding and decoding methods, and more particularly to polar code-based data encoding and decoding methods.

In wireless communication systems, channel coding can be implemented to improve the reliability of data transmission. Polar codes, one of channel coding schemes, may achieve Shannon capacity with low encoding/decoding complexity. Polar codes use channel polarization, where the bit lanes (or sub-channels) observed in the input are polarized into good bit lanes and bad bit lanes to transmit information bits through good bit lanes while passing through bad bit lanes Transfers frozen bits with a value known to both the encoder and decoder (for example, zero).

The inventive concept provides a method and apparatus for polar encoding and decoding with improved efficiency by using predetermined information.

According to an aspect of the inventive concept, there is provided a method of encoding input data into a polar code, comprising: generating unfrozen bits by adding at least one designated information bit to information bits already generated based on the input data, by adding reordering the unfrozen bits and the frozen bits by assigning the unfrozen bits to more reliable polarized sub-channels than the frozen bits having values known to both the encoder and the decoder; The sorted polar-encoded results are generated to generate codewords in which at least one specified information bit may have a value known to both the encoder and the decoder.

According to another aspect of the inventive concept, there is provided an apparatus for encoding input data into a polar code, comprising: a memory configured to store computer-readable instructions; and one or more processors configured to execute a computer-readable Instructions readable such that the one or more processors are configured to: generate bits by reordering unfrozen bits and frozen bits having values known to both the encoder and the decoder based on the reliability of the polarized subchannels a sequence of bits, the unfrozen bits comprising at least one designated information bit and an information bit based on input data; and a codeword generated by polar encoding the bit sequence, wherein the at least one designated information bit has an encoder and A value known to the decoder.

According to another aspect of the inventive concept, there is provided a decoding method comprising: receiving a codeword generated by polar encoding a bit sequence; and performing the codeword generation by generating a list of L decoding paths if L is a positive integer List decoding in which the sequence of bits contains unfrozen bits containing input bits, cyclic redundancy check (CRC) bits, and at least one specified information bit and has both encoder and decoder frozen bits of known values, and wherein the performing includes terminating decoding of the codeword early based on a result of decoding at least one specified information bit.

1 is a block diagram illustrating a wireless communication system 100 including a base station (BS or eNB) 10 and a user equipment (UE) 20 according to an example embodiment of inventive concepts. The wireless communication system 100 may include (but not limited to) the 5th generation (5G) wireless system, the long term evolution (Long Term Evolution; LTE) system, the code division multiple access (CDMA) system, the global mobile communication (GSM) system, the wireless local area network (WLAN) system or any other wireless communication system.

The base station 10 may be a fixed station capable of communicating with UE 20 and/or other base stations and exchanging data and control information. For example, the base station 10 may be called a Node B, an evolved Node B (eNB), a magnetic zone, a bit point, a base transceiver system (Base Transceiver System; BTS), an access point (Access Point; AP), a middle Relay node, remote radio head (Remote Radio Head; RRH), radio unit (radio unit; RU), smaller cells, etc. In this disclosure, a base station 10 or a cell may refer to a base station controller (Base Station Controller; BSC) in code division multiple access, a node B in wideband code division multiple access (WCDMA), an eNB in long-term evolution, or a magnetic A number of areas or functions covered by a zone (bit point) and can contain various coverage areas (e.g. macrocells, macrocells, microcells, picocells, nanocells) as well as relay nodes, remote radio heads, radio units Or smaller cell communication range.

The UE 20 which is a wireless communication device may include various devices which may be fixed or mobile and which may communicate with the base station 10 to transmit and receive data and/or control information. For example, the user equipment 20 may be a terminal equipment, a mobile station (mobile station; MS), a mobile terminal (mobile terminal; MT), a user terminal (user terminal; UT), a subscriber station (subscriber station; SS) , wireless devices, handheld devices, and more.

The wireless communication network between the base station 10 and the user equipment 20 can support the communication of multiple users by sharing available network resources. For example, in a wireless communication network, information can be transmitted through various multiple access methods, such as code division multiple access (CDMA), frequency division multiple access (FDMA), time Division multiple access (time division multiple access; TDMA), orthogonal frequency division multiple access (orthogonal frequency division multiple access; OFDMA), single carrier frequency division multiple access (Single Carrier Frequency Division Multiple Access; SC-FDMA), OFDM-FDMA , OFDM-TDMA or OFDM-CDMA.

Referring to FIG. 1 , a base station 10 and a user equipment 20 may communicate with each other through a downlink (DL) 30 and an uplink (uplink; UI) 40 . For example, in a wireless system such as the Long Term Evolution system or the Long Term Evolution Advanced system, the downlink 30 and the uplink 40 may be controlled via a control channel (such as a Physical Downlink Control Channel (PDCCH), a physical control format Physical Control Format Indicator Channel (PCFICH), Physical Hybrid ARQ Indicator Channel (PHICH), Physical Uplink Control Channel (PUCCH) or Enhanced Physical Downlink Control Channel (Enhanced Physical Downlink Control Channel; EPDCCH)) transmits control information, and can transmit data through a data channel (such as a Physical Downlink Shared Channel (PDSCH) or a Physical Uplink Shared Channel (PUSCH)).

In this disclosure, transmitting and receiving signals through a physical control channel (such as a physical uplink control channel, a physical uplink shared channel, a physical downlink control channel, an enhanced physical downlink control channel, or a physical downlink shared channel) can be expressed as "transmitting and receiving physical Uplink Control Channel, Physical Uplink Shared Channel, Physical Downlink Control Channel, Enhanced Physical Downlink Control Channel, and Physical Downlink Shared Channel". Moreover, transmitting or receiving a physical downlink control channel or transmitting or receiving signals through a physical downlink control channel may include transmitting or receiving an enhanced physical downlink control channel or transmitting or receiving signals through an enhanced physical downlink control channel. That is to say, the physical downlink control channel may be a physical downlink control channel or an enhanced physical downlink control channel, and may include the physical downlink control channel and the enhanced physical downlink control channel.

Channel coding can be used to improve the reliability of data transmission through the downlink 30 and uplink 40 in the wireless communication system 100 . For example, in the wireless communication system 100, polar codes (or polar flags) can be used for channel coding, and the base station 10 and the UE 20 can include encoders and decoders for the polar codes, respectively. The polarity code may be based on the channel polarization by which bit-lanes (or sub-channels) observed in the input are polarized into good and bad bit-lanes. Thus, in the polar code, information bits based on the input data DIN can be assigned to good bit lanes, while frozen bits with values known to both encoder and decoder can be assigned to bad bits meta channel. As described below, according to some example embodiments of the inventive concept, at least one designated information (interchangeably referred to as predetermined information (PI)) bit having a predetermined or desired value may additionally be assigned to the polar code's Good bit channel. In some example embodiments, predetermined information bits may contain values known to both the encoder and the decoder (e.g., predetermined or desired fixed values and/or identification words of the user equipment 20), and may be assigned to Corresponds to the index of a good bit-lane in the sequence of bits SEQ that is the input to the encoder 16 determined based on the characteristics of the polar code.

Encoding and decoding devices and/or methods according to some example embodiments of the inventive concepts may enable early termination of polar code decoding operations by using specified information in the encoding and decoding of polar codes, and thus may be significant due to efficient early termination The load (eg, computing resources) of the decoding operation in the wireless communication system 100 is reduced. The encoding and decoding apparatuses and methods according to some other example embodiments of the inventive concept can obtain the desired false alarm rate (FAR) by using designated information without additional CRC bits, At the same time improved decoding performance is provided by dropping the decoding path early during the decoding operation.

Referring to FIG. 1 , a base station 10 may include a loop redundancy check processor 12 , a subchannel mapping unit 14 , an encoder 16 and a rate matching unit 18 . The UE 20 may include a polarity decoder 22 . In the following, an example is described in which the base station 10 encodes data during the process of transmitting a signal to the user equipment 20 through the downlink 30, and the user equipment 20 decodes the material during the process of receiving a signal from the base station 10 through the downlink 30 . However, example embodiments are not limited thereto. According to some example embodiments, the user equipment 20 can transmit signals to the base station 10 through the uplink 40 , and the base station 10 can receive signals from the user equipment 20 through the uplink 40 . For example, although not shown in FIG. 1 , base station 10 may include a polar decoder, and UE 20 may include a polar encoder. The elements of the base station 10 and the user equipment 20 may be implemented as hardware blocks implemented by logic synthesis etc. in some example embodiments, may be implemented as processors and software blocks implemented by the processors in some example embodiments, and In some example embodiments may be implemented as a combination of hardware blocks, processors and software blocks.

The loop redundancy check processor 12 can generate a loop redundancy check (CRC) bit of the input data DIN, and generate loop redundancy check data including the input data DIN and the loop redundancy check bits DCRC. For example, the loop redundancy check processor 12 can generate (K+J) bits of Loop redundancy check data DCRC (K and J are positive integers). In this disclosure, the cyclic redundancy check data DCRC may be referred to as information bits including the input bits of the input data DIN and the cyclic redundancy check bits. The loop redundancy check bit can be used to confirm whether there is an error in the data received in the user equipment 20, and the loop redundancy check processor 12 can generate loop redundancy in any manner required by the wireless communication system 100. The rest of the parity bits. In some example embodiments, the loop redundancy check bit may be masked using a Radio Network Temporary Identifier (RNTI), as described below with reference to FIG. 2 .

The sub-channel mapping unit 14 may generate a sequence of bits SEQ from the cycle redundancy check data DCRC, which is an input to the encoder 16 . In some example embodiments, the sub-channel mapping unit 14 can generate unfrozen bits by adding at least one predetermined information bit to the loop redundancy check data DCRC, and can generate unfrozen bits by comparing the frozen bits and the unfrozen bits Reorder to generate the bit sequence SEQ. For example, by adding J' bits of predetermined information bits to (K + J) bits of loop redundancy check data DCRC, the sub-channel mapping unit 14 can generate (K + J + J' ) bits of unfrozen bits and can generate (N - (K + J + J')) bits of frozen bits (J' and N are positive integers). Next, the sub-channel mapping unit 14 may generate a bit sequence SEQ of N bits by reordering the unfrozen bits and the frozen bits.

The bit sequence SEQ is the input to the encoder 16 which generates the polarity encoded codeword PCW. An index of the bit sequence SEQ may correspond to a polar sub-channel of a polar code. Therefore, the sub-channel mapping unit 14 can identify a good sub-channel (or a good bit-lane) and a bad sub-channel (or a bad bit-lane) based on the index, and can pass to the sub-channel containing the frozen bit and at least one predetermined information bit Unfrozen bits are assigned to polarized sub-channels with higher reliability than frozen bits to generate the bit sequence SEQ. Therefore, at least one predetermined information bit contained in the unfrozen bits can be assigned to the polarized sub-channel with relatively high reliability. Some examples of generating a sequence of bits SEQ from unfrozen bits and frozen bits will be described below with reference to FIGS. 6-9 .

產生N個位元的極性編碼碼字PCW

。 [等式1]

The encoder 16 may generate a polar encoding codeword PCW by processing the bit sequence SEQ. For example, the encoder 16 can start from the bit sequence SEQ of N bits as shown in [Equation 1] below

Generate a polar codeword PCW of N bits

. [equation 1]

可被稱為產生矩陣並且可以是

的n階克羅內克冪（Kronecker power）。舉例來說，圖4繪示根據

（即，N=8）的編碼器16的操作。根據[等式1]，隨著N增大，可以極化子通道，並且子通道中的每一者可被稱為極化子通道。從迴圈冗餘校驗資料DCRC產生極性編碼碼字PCW的子通道映射單元14和編碼器16可以統稱為極性編碼器。[Equation 1],

can be called the generating matrix and can be

The nth order Kronecker power (Kronecker power). For example, Figure 4 shows the

(ie, N=8) for the encoder 16 operation. According to [Equation 1], as N increases, the subchannels may be polarized, and each of the subchannels may be referred to as a polarized subchannel. The sub-channel mapping unit 14 and the encoder 16 that generate the polar code word PCW from the cycle redundancy check data DCRC may be collectively referred to as a polar encoder.

The rate matching unit 18 can generate the output data DOUT by performing rate matching on the polar code word PCW. The rate matching unit 18 can perform rate matching in a manner required by the wireless communication system 100 by weakening and/or shortening the polar code word PCW. The output data DOUT rate-matched by the rate-matching unit 18 can be converted by sequentially passing through a modulator, a mixer, a power amplifier, an antenna, etc., and can be transmitted to the UE 20 through the downlink 30 .

The UE 20 can receive signals transmitted from the base station 10 via the downlink 30 . For example, the received signal can be converted into data by passing through an antenna, a filter, a low noise amplifier, an analog-to-digital converter, etc., and the converted data can be sent to the polar decoder 22 as a polar code word PCW′. The polar decoder 22 can generate the decoded data DEC by decoding the polar code word PCW′, and thereby extract the information (eg, the input data DIN) sent by the base station 10 .

In some example embodiments, the polar decoder 22 may sequentially decode the polar code word PCW' bit by bit based on successive cancellation (SC) decoding. In some example embodiments, the polar decoder 22 may maintain a plurality of candidate decoding paths based on list decoding and decode the polar encoded codeword PCW' when discarding the decoding paths according to maximum likelihood. Successive cancellation list (SCL) decoding may be referred to as a combination of successive cancellation and list decoding. In some example embodiments, as described below with reference to FIG. 5B , methods such as the simplified successive cancellation (SSC) decoding method may be used by classifying nodes into predetermined (or alternatively, desired ) group to reduce the computational complexity. In some example embodiments, methods involving concatenated cyclic redundancy check codes and polar codes may be used to improve the performance of consecutive offset list decoding. It should be noted, however, that the above decoding methods are merely illustrative, and example embodiments of inventive concepts are not limited to the decoding methods described above.

According to an example embodiment of the inventive concept, the polar decoder 22 may be informed of the reservation (or alternatively, the required) value, and the result of decoding the predetermined information bits (eg, estimated predetermined information bits) may be used. Based on the estimated predetermined information bits, polar decoder 22 may terminate the decoding operation early in some example embodiments, and may use list compaction for list decoding in some example embodiments. Moreover, the estimated predetermined information bits can reduce the false alarm rate (FAR) by performing loop redundancy check function.

FIG. 2 is a block diagram illustrating an example of the structure of the downlink 30 of FIG. 1 of an example embodiment of the inventive concept. For example, FIG. 2 illustrates a transmission time interval (Transmission Time Interval; TTI) of the downlink 30 in FIG. 1 . In the wireless communication system 100, the data of the downlink 30 can be transmitted in TTI units, and a TTI can be defined as a time interval including a plurality of symbols (eg, OFDM symbols). As an example, a TTI in Long Term Evolution may be a subframe with a length of 1 ms, and a TTI in 5G may be a Scalable TTI. Hereinafter, FIG. 2 will be described with reference to FIG. 1 .

Referring to FIG. 2 , the transmission time interval 5 of the downlink 30 may include two time regions multiplexed by Time Division Multiplexing (TDM). For example, two time regions may contain a control region for transmission of a control channel (e.g., a physical downlink control channel or a physical hybrid ARQ indicator channel) and a data region for transmission of a shared channel (e.g., a physical downlink shared channel) area. For example, the control region may contain a number of symbols for the control channel, and the data region may contain the remaining symbols for the common channel.

The control field may contain information about the downlink 30 . For example, in the LTE, downlink control information (downlink control information; DCI) can be transmitted through the physical downlink control channel in the control region. The downlink control information may include information required by the UE 20 to communicate with the base station 10, such as hopping flags, resource block allocation, modulation coding scheme (Modulation Coding Scheme; MCS) and/or redundancy version (Redundancy Version; RV). The base station 10 can transmit multiple physical control channels in the control area for multiple UEs. For example, a LTE-PDCCH may be transmitted on an aggregate of one or more connected control channel elements (CCEs). A control channel element may be referred to as a logical allocation unit of a physical downlink control channel for providing a predetermined (or alternatively, required) coding rate according to the state of the radio channel, and may correspond to a plurality of resource element groups (resource element group; REG). The base station 10 can add the loop redundancy check to the control information. Loop redundancy checks can be masked based on owner or usage of the physical downstream control channel. For example, the base station 10 may generate a physical downlink control channel for the UE 20 by masking the unique identifier (eg, cell RNTI) of the UE 20 for the loop redundancy check.

The UE 20 can monitor multiple physical control channels. For example, the wireless communication system 100 may define a finite set of control channel element locations where a physical control channel may be located for the UE 20 . The finite set of control channel element locations may be the space in which the UE 20 may search its own physical control channel, and may be referred to as a search space (SS). In order to omit the auxiliary information for the location of the PCH and preserve resources for multiple UEs, the base station 10 may not provide the information about where the PCH is located in the control area to the UE 20 . Therefore, the UE 20 can find its own PCH by trying to decode the PCH candidates in TTI5. Through this blind decoding, UE 20 can identify the physical control channel to which the transmission is made. For example, in the long-term evolution, the user equipment 20 can unmask the physical downlink control channel in the form of a unique identifier (cell RNTI), and if there is no loop redundancy check error, the user equipment 20 can send all The above physical downlink control channel is identified as its own physical downlink control channel. In some example embodiments, the UE 20 may be required to perform blind decoding up to 60 times within one transmission time interval 5 .

The UE 20 may have a heavy decoding load due to blind decoding, which may require a high computing performance of the UE 20 or may degrade the overall performance of the UE 20 due to heavy decoding load. For example, if the wireless communication system 100 is 5G, the maximum number of blind decoding attempts of the user equipment 20 in a transmission time interval 5 may be further increased compared to the long-term evolution, and the cost of each attempt may need to be shortened. time. Therefore, it may be necessary to terminate the decoding operation of the physical control channel early in blind decoding. For example, in the case of 5G, blind decoding using polar codes may need to be terminated early due to the use of polar codes as the channel coding method.

FIG. 3 shows an example of a process in which downlink control information D31 is transmitted from the base station 10 of FIG. 1 to the UE 20 according to an example embodiment of the inventive concept. For example, FIG. 3 shows the process of transmitting the downlink control information D31 from the base station 10 to the UE 20 when the wireless communication system 100 is LTE. As shown in FIG. 3, the base station 10 may transmit the downlink control information D31 to the user equipment 20 by performing a series of operations S300 to S340, and the user equipment 20 may perform a series of operations S350 to S390 to generate an estimated downlink Road Control Information (DCI') D32. Hereinafter, FIG. 3 will be described with reference to FIG. 1 .

Referring to FIG. 3 , in operation S300 , an operation of inserting a loop redundancy check into the downlink control information D31 may be performed. In some example embodiments, the base station 10 may mask the unique identifier (eg, cell RNTI) of the UE 20 for the loop redundancy check.

In operation S310, channel encoding may be performed. For example, channel coding may include coding using polar codes, and as described above with reference to FIG. 1 , unfrozen bits and predetermined information bits may be added to the frozen bits of the downlink control information D31. Therefore, a polar encoding codeword can be generated. An example of operation S310 will be described with reference to FIGS. 6 to 10 .

In operation S320, rate matching may be performed. For example, base station 10 may perform rate matching by performing attenuation and/or shortening. Next, in operation S330, an operation of modulating the rate matching codeword may be performed. In operation S340, an operation of mapping a control channel element to a physical resource element (resource element; RE) may be performed. Therefore, the signal including the information of the downlink control information D31 can be transmitted to the UE 20 .

In operation S350, an operation of demapping physical resource elements to control channel elements may be performed. In operation S360, an operation of demodulating the control channel element may be performed. As described above, since the UE 20 may not be informed which CCH aggregation is used to receive the PDCCH, the UE 20 can demodulate each CCH aggregation. Next, in operation S370, rate matching may be performed on the demodulated material. Similarly, since the UE 20 may not be informed which DL control information payload size is used to receive the control information, the UE 20 may perform rate matching for each DL control information format.

In operation S380, channel decoding may be performed. For example, the polar decoder 22 included in the UE 20 may perform decoding on the rate-matched data based on, for example, continuous offset list decoding, and detect whether an error occurs through a round-robin redundancy check. In some example embodiments, during the decoding process, the decoding of the rate matching data may be terminated early according to the decoding result of predetermined information bits, and the decoding of other rate matching data may be continued or started. An example of operation S380 will be described below with reference to FIGS. 11 to 14 .

In operation S390, an operation of removing the loop redundancy check may be performed. For example, the UE 20 may detect its physical downlink control channel and obtain estimated downlink control information (DCI') D32 by removing the cyclic redundancy check from the decoded data in operation S380.

FIG. 4 illustrates an example of the operation of the encoder 16 of FIG. 1, according to an example embodiment of inventive concepts. As described above with reference to FIG. 1 , encoder 16 may perform channel encoding based on polar codes. For _example , FIG _. 4 illustrates generating 8 bits { x ₁ ,..., x ₈ } polar encoding codeword and receive 8 bits of {y ₁ ,..., y ₈ } data in the receiver (e.g., user equipment 20 in Fig. 1) instance.

以遞迴方式組合。舉例來說，圖4繪示組合通道

以及子通道

和

。組合通道

可以劃分成N個二進位輸入座標通道W

: X→Y^N × X^i-1 , 1

i

N，並且每一通道中的轉移概率可以如下文[等式2]中所示來定義。 [等式2]

A channel W N of N independent channels _W is combined by using a bit sequence of N bits as input: X ^N → Y ^N can be obtained from

Compose in a recursive manner. As an example, Figure 4 depicts the combined channel

and subchannels

and

. combined channel

Can be divided into N binary input coordinate channels W

: X→Y ^N × X ^i-1 , 1

i

N, and the transition probability in each channel can be defined as shown in [Equation 2] below. [equation 2]

The transition probability of [Equation 2] can be expressed in a recursive manner as shown in [Equation 3] below. [equation 3]

中的每一者的轉移概率可以如下文[等式4]中所示來計算。 [等式4]

In the Binary Erasure Channel (BEC), the transition probability according to [Equation 3] can be simply calculated. For example, when the deletion probability ε is 0.5, the channel shown in Fig. 4

The transition probability of each of can be calculated as shown in [Equation 4] below. [equation 4]

的轉移概率相對接近1，而對應於通道

的轉移概率相對接近0。在N增大的情況下，轉移概率中的每一者可以收斂到1或0，並且這種現象可以被稱為通道極化。可以將具有可變值的未凍結位元指配到良好通道，即，具有低轉移概率的位元通道，而將具有固定值的凍結位元指配到到不良通道，即具有高轉移概率的位元通道。舉例來說，在圖4的實例中，四個位元{u₄ , u₆ , u₇ , u₈ }可以是未凍結位元，而剩餘四個位元{u₁ , u₂ , u₃ , u₅ }可以是凍結位元。[Equation 4] indicates that the channel corresponding to

The transition probability of is relatively close to 1, while the corresponding channel

The transition probability is relatively close to 0. Each of the transition probabilities may converge to 1 or 0 as N increases, and this phenomenon may be referred to as channel polarization. Unfrozen bits with variable values can be assigned to good lanes, i.e., bit lanes with low transition probability, while frozen bits with fixed values are assigned to bad lanes, i.e., with high transition probability. bit channel. For example, in the example of Figure 4, four bits {u ₄ , u ₆ , u ₇ , u ₈ } could be unfrozen bits, while the remaining four bits {u ₁ , u ₂ , u ₃ , u ₅ } can be frozen bits.

As described above, at least one of the unfrozen bits {u ₄ , u ₆ , u ₇ , u ₈ } can be used as a predetermined information bit known to both the encoder and decoder. That is, the unfrozen bits {u ₄ , u ₆ , u ₇ , u ₈ } may contain information bits and at least one predetermined information bit, and the information bits may contain input bits and loop redundancy check bits. Thus, predetermined information bits having predetermined (or alternatively, desired) values can be assigned to unfrozen bits instead of frozen bits, and thus desired features can be obtained, whereby wireless communication systems (e.g., Efficiency of the wireless communication system 100 of FIG. 1 ).

5A and 5B illustrate an example of the operation of the polar decoder 22 of FIG. 1 according to example embodiments of inventive concepts. For example, FIG. 5A shows a grid of polar codes, where N=8, and FIG. 5B shows a binary tree structure of polar codes, where N=8. For example, FIGS. 5A and 5B may correspond to the encoder structure of FIG. 4 . As shown in FIGS. 5A and 5B , in general, sub-channels with low indexes may have relatively low reliability, and sub-channels with high indexes may have relatively high reliability.

表示，其中i和j分別表示網格的層級和階段（1

i

n+1，1

j

N）。

為

的對數似然比（log likelihood ratio；LLR），其可以如下文[等式5]中所示來計算。 [等式5]

Referring to Figure 5A, the estimated bits corresponding to variable nodes are given by

, where i and j represent the level and stage of the grid, respectively (1

i

n+1, 1

j

N).

for

The log likelihood ratio (LLR) of , which can be calculated as shown in [Equation 5] below. [equation 5]

可以表達為下文[等式6]。 [等式6]

Based on [Equation 5], the estimated bit sequence in the search space decoding

can be expressed as [Equation 6] below. [equation 6]

可以表達為下文[等式7]。 [等式7]

In [Equation 6],

can be expressed as [Equation 7] below. [equation 7]

的節點和其父節點可以由一個速率0節點替換，並且估計位元

和其父節點可以由一個速率1節點替換。由於二叉樹簡化，可以簡化連續抵消解碼。Referring to FIG. 5B , in the case of 2 ⁿ =N, consecutive offset decoding can be represented by a binary tree with depth n. For example, as shown in FIG. 5B , the grid of FIG. 5A may be represented by a binary tree with depth 3 and 2 ³ =8 leaf nodes. In a binary tree, each node can be classified according to its characteristics. For example, as shown in FIG. 5B , nodes in a binary tree can be classified as rate 0 nodes having only frozen bits as children, rate 1 nodes having only unfrozen bits as children, and rate 1 nodes having frozen bits as children. Rate R nodes with meta and unfrozen bits as child nodes. Subtrees containing nodes classified in the same way can be replaced by a single node of the same classification, and thus the binary tree can be simplified. For example, in Figure 5B, corresponding to the estimated bit

A node and its parent can be replaced by a rate 0 node, and the estimated bit

and its parent node can be replaced by a rate 1 node. Successive offset decoding can be simplified due to binary tree simplification.

FIG. 6 is a flowchart illustrating an example of operation S310 of FIG. 3 according to example embodiments of inventive concepts. As described above with reference to FIG. 3 , in operation S310 ′ of FIG. 6 , channel encoding may be performed based on polar codes. As shown in FIG. 6, operation S310' may include operations S312 and S314, and may include encoding the (K+J) bit data (for example, DCRC in FIG. 1 ) received from operation S300 of FIG. 3 to generate An N-bit codeword (eg, PCW in FIG. 1 ) and provides the N-bit codeword to operation S320 of FIG. 3 . For example, operation S310' of FIG. 6 may be performed by the sub-channel mapping unit 14 and the encoder 16 of FIG. 1 . Hereinafter, FIG. 6 will be described with reference to FIG. 1 .

Referring to FIG. 6, in operation S312, sub-channel mapping may be performed. For example, the sub-channel mapping unit 14 may base on the input data by adding predetermined information bits of J′ bits having a predetermined (or alternatively, desired) value to (K+J) bits of data DIN (or exported from input data DIN) to generate (K + J + J') bits of unfrozen bits, and by unfrozen bits and N- (K + J + J') bits The frozen bits of are reordered to produce a bit sequence of N bits. As described above, in a bit sequence, unfrozen bits can be assigned to subchannels (or bit lanes) with relatively high reliability, while frozen bits can be assigned to subchannels (or bit lanes) with relatively low reliability. Sexual sub-channels. Therefore, due to the relatively high reliability of sub-channels, predetermined information bits can be utilized in various ways (eg, early termination and/or list reduction) at later decoding operations (eg, S380 in FIG. 3 ).

In operation S314, polar encoding may be performed. For example, the encoder 16 may generate an N-bit polar encoding codeword by processing the N-bit bit sequence from the sub-channel mapping unit 14 .

FIG. 7 illustrates an example of the operation of operation S312 of FIG. 6 according to example embodiments of inventive concepts. As described above with reference to FIG. 6, in operation S312, an operation of mapping sub-channels may be performed. For example, the example of FIG. 7 may be performed by sub-channel mapping unit 14 of FIG. 1 . FIG. 7 will be described below with reference to FIGS. 1 and 6 .

Referring to FIG. 7, (K+J+J') bits of unfrozen bits can be generated by adding J' bits of predetermined information bits to (K+J) bits of predetermined information bits . Therefore, the first bit sequence SEQ1 of N bits can be generated by inserting N−(K+J+J′) bits of frozen bits. The first bit sequence SEQ1 may be aligned according to sub-channel reliability, and as shown in FIG. 7 , unfrozen bits may correspond to sub-channels with higher reliability than frozen bits.

The first bit sequence SEQ1 may be reordered into the second bit sequence SEQ2. That is, the first bit sequence SEQ1 can be reordered based on the indices of the polar subchannels of the polar code. For example, as shown in FIG. 7, in the second sequence of bits SEQ2, most of the frozen bits may correspond to lower indexes, and some of the frozen bits may correspond to higher indexes, and as with reference to FIGS. As depicted in FIGS. 5A and 5B , good bit-lanes and bad bit-lanes can be interleaved in the second bit sequence SEQ2. Thus, as shown in FIG. 7, the frozen and unfrozen bits may be reordered in a second bit sequence SEQ2, and the second bit sequence SEQ2 may be provided to an encoder (e.g., Encoder 16).

In some example embodiments, a cyclic redundancy check bit of J bits may have a high index in the second bit sequence SEQ2, as shown in FIG. 7 . In some example embodiments, the index of the second bit sequence SEQ2 may be assigned such that J′ bits of predetermined information bits included in the unfrozen bits are assigned to sub-channels with relatively high reliability. Also, in some example embodiments, the index of the second bit sequence SEQ2 may be assigned based on rate matching performed after encoding. An example of assigning an index of the second bit sequence SEQ2 to predetermined information bits will be described below with reference to FIGS. 8 and 9 .

In some example embodiments, the loop redundancy check bits of J bits may be interleaved with the input bits, and thus the loop redundancy check bits of J bits may be different as shown in FIG. 7 The mode is distributed in the second bit sequence SEQ2. Distributed loop redundancy check bits can be considered as auxiliary bits in polar codes. The predetermined information bits can be distributed in the second bit sequence SEQ2 as shown in FIG. 7 or can be assigned to sub-channels corresponding to high reliability as will be described with reference to FIG. 8, and thus the distributed predetermined information bits can perform Same or similar functionality as distributed loop redundancy check bits. That is to say, due to the distributed predetermined information bits, the same effect as the distributed cyclic redundancy check bits is produced. Also, predetermined information bits can be used to reduce the false alarm rate (FAR). For example, the false alarm rate may be about 2 to 16 based on 16 bits of cyclic redundancy check bits. At this time, in the case where the number of lists in list decoding is 8, the false alarm rate may increase to 2 to 16+3. In the case that the desired false alarm rate is 2 to 16, the loop redundancy check bit can be increased to 19 bits to meet this requirement. However, instead of increasing the number of bits of the cyclic redundancy check bits, in the case of adding 3 bits of predetermined information bits, the above-mentioned function due to the predetermined information bits can be obtained, and at the same time A false alarm rate of 2 to 16 can be maintained. That is to say, the false alarm rate can be determined based on the number of cyclic redundancy check bits and the number of predetermined information bits.

FIG. 8 shows an example of a binary tree of a polar code according to an example embodiment of the inventive concept. For example, the binary tree of FIG. 8 may have 16 leaf nodes, among which black filled nodes represent unfrozen bits, white filled nodes represent frozen bits, and X filled nodes represent shortened bits. Moreover, in the example of FIG. 8, the 6-bit unfrozen bits may include 3-bit input bits, 2-bit loop redundancy check bits, and 1-bit predetermined information bits (ie, K=3, J=2, J'=1). In the example of FIG. 8, the parent block size N may be 16, and the block size M may be 12.

In some example embodiments, the predetermined information bits may be indexed such that the predetermined information bits are substantially evenly distributed among the unfrozen bits. For example, the predetermined information bits can be distributed based on the weights of the generation matrix. Also, in some example embodiments, predetermined information bits may be evenly distributed and simultaneously assigned to sub-channels with relatively high reliability. For example, a subtree that is part of a binary tree of a polar code may have a leaf node corresponding to a power of 2, and the rightmost leaf node of FIG. 8 with the highest index among the leaf nodes may correspond to subchannel. As shown in FIG. 8 , among the subtrees including the leaf nodes of index 5 to index 8 , the leaf node of index 8 may correspond to a subchannel with high reliability. Similarly, as shown in Figure 8, the highest indexed leaf node in a subtree containing 4 leaf nodes may correspond to a subchannel with high reliability. Therefore, the highest index in the subtree with unfrozen bits as leaf nodes (eg, 8, 12 or 16 in FIG. 8 ) can be assigned to the predetermined information bit. In other words, as described above with reference to FIG. 5B, since the rate 1 node and the rate R node include unfrozen bits as leaf nodes, predetermined information bits can be assigned between the leaf nodes of the rate 1 node or the rate R node. The subchannel with the highest reliability in .

In some example embodiments, predetermined information bits may be assigned in consideration of rate matching. For example, as shown in FIG. 8 , in case the end of the sequence is deleted by rate matching shortening performed after encoding, predetermined information bits may not be assigned to the shortened region. That is to say, in the example of FIG. 8 , 16 can be excluded from the candidate indexes 8 , 12 and 16 of the predetermined information bits.

In some example embodiments, predetermined information bits may be assigned to have a low index among candidate indices in the sequence of bits. As described above, the predetermined information bits can be used to terminate decoding early in the receiver's decoder, and in the case of decoding methods such as continuous offset decoding, where the predetermined information bits are placed at the front (e.g., in In case the predetermined information bit has a low index), an early termination of decoding can be determined at an earlier time. Therefore, in the example of FIG. 8 , the predetermined information bit of 1 bit may have 8 of candidate indexes 8 , 12 and 16 . Moreover, not only the predetermined information bits of 2 bits or more, the predetermined information bits may sequentially have candidate indexes of rising powers.

可以表示第i個資訊位元，並且

可以表示第i個預定資訊位元。而且，

可以表示位元序列中的索引，例如，圖8中的

可以是6。在下文中，將參考圖8描述圖9。FIG. 9 is a dummy code illustrating an example of a method of assigning predetermined information bits according to an example embodiment of the inventive concept. In Figure 9,

can represent the i-th information bit, and

may represent the i-th predetermined information bit. and,

can represent an index in a sequence of bits, for example, in Figure 8

It can be 6. Hereinafter, FIG. 9 will be described with reference to FIG. 8 .

的位元序列中的索引可以大於

。也就是說，可以在列表發散結束後開始提早終止。舉例來說，在L是4的情況下，

，並且因此

可以是8。在高編碼速率下，由於多個輸入位元可以放置在母快的前方區域中，可能需要放置預定資訊位元以提高提早終止的效率，如下文所描述。In the case where L indicates the number of decoding lists, the i-th predetermined information bit

Indices in the bit sequence of can be greater than

. That is, early termination can start after the list divergence ends. For example, in the case where L is 4,

, and therefore

Can be 8. At high code rates, since multiple input bits may be placed in the front area of the bus, it may be necessary to place predetermined information bits to improve the efficiency of early termination, as described below.

可以與前一預定資訊位元e_i-1 間隔開

來指配。另一方面，在輸入位元的索引並不超過N/4的情況下，在17行和18行處，預定資訊位元

可以與前一預定資訊位元e_i-1 間隔開

來指配，並且可以基於輸入位元

的索引

以

增大k。舉例來說，我們假設母塊的大小確定為N=512，n=9，以及L=8。在

超過128的情況下，預定資訊位元可以指配有索引256 (=2⁸ )、384 (=2⁸ +2⁷ )以及448 (=2⁸ +2⁷ +2⁶ )。在一些實例實施例中，可以進一步優化預定資訊位元的索引，並且例如在確定極性碼的參數的情況下，預定資訊位元的索引可以容易地從極性碼的參數匯出。In some example embodiments, where the set of predetermined information bits of J′ bits is E, the set E may be obtained as shown in FIG. 9 . Referring to FIG. 9, at lines 11 and 12, initialization of variables may be performed. At line 13, the position of the input bit can be checked. In the case where the index of the input bit exceeds N/4, at lines 14 and 15, the predetermined information bit

May be spaced apart from the previous predetermined information bit e _i-1

to assign. On the other hand, in the case where the index of the input bit does not exceed N/4, at lines 17 and 18, predetermined information bits

May be spaced apart from the previous predetermined information bit e _i-1

to assign, and can be based on the input bit

index of

by

Increase k. For example, we assume that the size of the mother block is determined to be N=512, n=9, and L=8. exist

In the case of exceeding 128, the predetermined information bits can be assigned with indexes 256 (=2 ⁸ ), 384 (=2 ⁸ +2 ⁷ ) and 448 (=2 ⁸ +2 ⁷ +2 ⁶ ). In some example embodiments, the index of the predetermined information bit can be further optimized, and for example, in the case of determining the parameters of the polar code, the index of the predetermined information bit can be easily derived from the parameter of the polar code.

FIG. 10 illustrates a code tree of bits of a bit sequence according to levels in consecutive offset decoding according to an example embodiment of the inventive concept. For example, FIG. 10 shows an example of a code tree with N=8.

中的索引依序確定所述位元，並且對應於具有零值的凍結位元的位元

的解碼路徑可以如圖10中所示在不分支的情況下繼續到下一個位元（即下一個層級）。Continuous offset decoding can be represented as a process of determining a path in the code tree of the bits of the estimated bit sequence in the vertical direction according to the levels as shown in FIG. 10 . The bits of the estimated bit sequence can be sequentially determined from a high level to a low level in the code tree according to the index. For example, as described above with reference to FIG. 5B , based on the estimated bit sequence

The indices in sequentially identify the bits, and the bits that correspond to frozen bits with a value of zero

The decoding path for can continue to the next bit (ie, the next level) without branching as shown in FIG. 10 .

The nodes of the code tree may respectively have log-likelihood ratios. Candidate decoding paths can be added or discarded in each of the levels using multiple candidate decoding paths according to log-likelihood ratio values in manifest decoding. The operation of discarding the decoding path and no longer performing decoding through the decoding path may be referred to as list reduction. List pruning occurs to reduce decoding time. As will be described later, the predetermined information bits estimated during decoding can be used not only for early termination of decoding, but also for list decoding, thereby improving decoding performance due to shortened decoding time.

FIG. 11 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. As described above with reference to FIG. 3 , in operation S380a of FIG. 11 , polar code-based channel decoding may be performed. As shown in FIG. 11 , operation S380a may include a plurality of operations S381a to S386a, and may be performed by, for example, the polar decoder 22 of FIG. 1 . Hereinafter, operation S380a of FIG. 11 will be described with reference to the example code tree of FIG. 10 .

In operation S381a, an operation of selecting a node may be performed. For example, in the case of performing successive cancellation list (SCL) decoding, the node after the previously processed node in the candidate decoding path can be selected, the node of another candidate decoding path can be selected, and the newly added candidate decoding can be selected the node of the path.

In operation S382a, an operation of determining a node type may be performed. As described above with reference to FIG. 7, a node may correspond to a frozen bit or an unfrozen bit. Unfrozen bits may include information bits and predetermined information bits, and thus a node may correspond to one of frozen bits, information bits, and predetermined information bits. As shown in FIG. 11 , operation S385a may then be performed if the node corresponds to a frozen bit, then operation S384a may be performed if the node corresponds to an information bit, and then operation S384a may be performed when the node corresponds to a predetermined information bit In the case of , perform operation S383a.

In operation S383a, an operation of determining early termination may be performed. That is, in operation S382a, in the case that the node type corresponds to the predetermined information bit, early termination may be determined according to the decoding result of the predetermined information bit. In some example embodiments, early termination may be determined based on decoding results of predetermined information bits corresponding to a single bit. In some example embodiments, early termination may be determined based on decoding results of predetermined information bits corresponding to the plurality of bits. If early termination is determined, operation S380a may be terminated, and if early termination is not determined, operation S384a may be subsequently performed. An example of operation S383a will be described later with reference to FIG. 12 .

List reduction may be performed in operation S384a. For example, operation S384a may be omitted due to a known value (eg, zero) in the node corresponding to the frozen bit. In the case that the node type is unfrozen bit and no early termination is determined, manifest reduction may be performed (S384a). As described above with reference to FIG. 10 , list reduction may be performed at each node based on log-likelihood ratio values.

In operation S385a, an operation of determining whether the node corresponds to the final bit may be performed. In the case that the node does not correspond to the final bit, in operation S381a, the operation of selecting another node may be repeated, and in the case of the node corresponding to the final bit, in operation S386a, loop redundancy checking may be performed. test. The occurrence of an error can be detected based on the result of the loop redundancy check.

FIG. 12 is a flowchart illustrating an example of operation S383a of FIG. 11 according to example embodiments of inventive concepts. As described above with reference to FIG. 11 , in operation S383a' of FIG. 11 , an operation of determining early termination may be performed. As shown in FIG. 12 , operation S383a' may include operation S383_2 and operation S383_4, and FIG. 12 will be described below with reference to FIG. 11 .

In operation S383_2, an operation of updating the early termination condition may be performed. As shown in FIG. 12, in operation S382a of FIG. 11, since it is determined that the node corresponds to the predetermined information bit, the early termination condition may be updated according to the decoding result of the predetermined information bit (for example, estimated value of the predetermined information bit) . In some example embodiments, the early termination conditions may be updated for each of the decoding paths.

In operation S383_4, an operation of determining whether the list satisfies an early termination condition may be performed. That is to say, in the case that all the decoding paths included in the manifest satisfy the early termination condition, operation S380a of FIG. Next, in operation S384a of FIG. 11 , list reduction may be performed. That is, in the process of performing decoding in a state in which candidate decoding paths are maintained according to list decoding, early termination may be determined in a case where all candidate decoding paths satisfy early termination conditions.

FIG. 13 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. As described above with reference to FIG. 3 , in operation S380b of FIG. 13 , polar code-based channel decoding may be performed. For example, compared to operation S380a of FIG. 11, it may first be determined whether a node corresponds to a frozen bit, and then early termination may be determined. Where a node corresponds to a frozen bit, list reduction may be performed first. As shown in FIG. 13 , operation S380b may include a plurality of operations S381b to S387b, and may be performed by, for example, the polarity decoder 22 of the drawing. Hereinafter, redundant descriptions of FIGS. 13 and 11 will be omitted.

In operation S381b, an operation of selecting a node may be performed. Next, in operation S382b, an operation of determining whether a node corresponds to a frozen bit may be performed. As shown in FIG. 13, in the case that the node is a frozen bit, then operation S386b may be performed, and in the case that the node is not a frozen bit (for example, in the case that the node is an unfrozen bit), in operation In S383b, list reduction may be performed.

After the list reduction is performed, in operation S384b, an operation of determining whether a node corresponds to an information bit may be performed. If the node does not correspond to an information bit, that is, if the node corresponds to a predetermined information bit, then operation S385b may be performed, and if the node corresponds to an information bit, then operation S386b may be performed.

In operation S385b, an operation of determining whether to terminate early may be performed. For example, as described above with reference to FIG. 12 , it may be determined whether all candidate decoding paths satisfy the early termination condition, ie whether the decoding result of the predetermined information bit is different from the known value. If the early termination condition is satisfied, operation S380b may end, and if the early termination condition is not satisfied, operation S386b may be subsequently performed.

In operation S386b, an operation of determining whether the node corresponds to the final bit may be performed. In case the node does not correspond to the final bit, in operation S381b, an operation of selecting another node may then be performed, and in the case of the node corresponding to the final bit, in operation S387b, loop redundancy may be performed check.

FIG. 14 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. As described above with reference to FIG. 3 , in operation S380c of FIG. 14 , polar code-based channel decoding may be performed. In the example of FIG. 14, predetermined information bits can be used for list compaction. As shown in FIG. 14 , operation S380c may include multiple operations (operations S381c to S384c). Operation S380c may also include additional operations required to complete the decoding operation. Hereinafter, the description of FIG. 14 that is the same as the description of FIGS. 11 and 13 will be omitted.

In operation S381c, an operation of selecting a node may be performed. Next, in operation S382c, an operation of determining whether a node corresponds to a predetermined information bit may be performed. As shown in FIG. 14, in the case where the node is not a predetermined information bit (i.e., is a frozen bit or an information bit), subsequent operations for decoding can be performed, while in the case of a node that is a predetermined information bit , and then operation S383c can be performed.

In operation S383c, an operation of determining whether the estimated predetermined information bit has a predetermined (or alternatively, required) value may be performed. As discussed above, the predetermined information bit may have a predetermined (or alternatively, desired) value (e.g. a fixed value (e.g., zero)) at the encoder, or a receiver including the decoder (e.g., the receiver of Fig. 1 The unique value of the decoder 20). The predetermined (or alternatively, required) value of the predetermined information bit can be known in advance by the decoder. In the case where it is estimated that the predetermined information bit has a predetermined (or alternatively, required) value Next, subsequent operations for decoding can be performed. On the other hand, in the case where it is estimated that the predetermined information bit has a value different from the predetermined (or alternatively, required) value, the current decoding path can be discarded (operation S384c) That is, predetermined information bits can be used for list compaction in the decoder, and thus decoding performance can be improved.

FIG. 15 is an example block diagram of a wireless communication device 50 according to an example embodiment of the inventive concepts. As shown in FIG. 15 , the wireless communication device 50 may include an application specific integrated circuit (ASIC) 51, an application specific instruction set processor (ASIP) 53, a memory 55, a main processor 57 and memory 59. Two or more of the ASIC 51, the ASIP processor 53, and the main processor 57 may communicate with each other. At least two of ASIC 51 , ASIC processor 53 , memory 55 , main processor 57 and main memory 59 may be embedded in one chip.

The special-purpose instruction set processor 53 is an integrated circuit customized for a specific application and may support a special-purpose instruction set for a particular application and execute instructions contained in the instruction set. Memory 55 may be in communication with special purpose instruction set processor 53 and may store a number of instructions for execution by special purpose instruction set processor 53 as a non-volatile storage device. For example, the memory 55 may include (but not limited to) any type of memory that can be accessed by the special-purpose instruction set processor 53, such as random access memory (random access memory; RAM), read-only memory (Read Only Memory; ROM), tapes, floppy discs, optical discs, volatile memory, non-volatile memory, and combinations thereof.

The main processor 57 can control the wireless communication device 50 by executing a plurality of instructions. For example, the main processor 57 can control the ASIC 51 and the ASIC 53 , and can process data received over the wireless communication network or user input to the wireless communication device 50 . Main memory 59 may be in communication with main processor 57 and may store a number of instructions for execution by main processor 57 as a non-transitory memory device. By way of example, main memory 59 may include, but is not limited to, any type of memory accessible by main processor 57, such as random access memory (RAM), read only memory (ROM), magnetic tape, magnetic chips, CDs, volatile memory, non-volatile memory, and combinations thereof.

The above-described encoding and/or decoding methods according to example embodiments of inventive concepts may be performed by at least one of elements included in the wireless communication device 50 of FIG. 15 . In some example embodiments, at least one of the operations of the encoding and/or decoding methods described above may be implemented as a plurality of instructions stored in memory 55 . In some example embodiments, the special instruction set processor 53 may perform at least one of the operations of the encoding and/or decoding methods by executing a plurality of instructions stored in the memory 55 . In some example embodiments, at least one of the operations of the encoding and/or decoding methods may be implemented in a hardware block designed by logic synthesis or the like and incorporated in the dedicated integrated circuit 51 . In some example embodiments, at least one of the operations of the encoding and/or decoding methods may be implemented as a plurality of instructions stored in the main memory 59, and the main processor 57 may execute the instructions stored in the main memory 59 by executing A plurality of instructions to perform at least one of the operations of the encoding and/or decoding methods.

As described above, example embodiments have been disclosed in the drawings and specification. However, it should be understood that the terms are used only for the purpose of describing the technical idea of the inventive concept and are not used to limit the scope of the inventive concept defined in the claims. Accordingly, those of ordinary skill in the art will appreciate that various modifications and changes can be made to the disclosed example embodiments without departing from the scope of the inventive concepts. Therefore, the true protection scope of the concept of the present invention should be determined by the technical concept of the appended claims.

‧‧‧對數似然比PCW、PCW'‧‧‧極性編碼碼字

、

、

‧‧‧估計位元S300~S390、S310'、S380a~S386a、S380b~S387b、S383a'、S383_2、S383_4、S380c~S384c‧‧‧操作SEQ、SEQ1、SEQ2‧‧‧位元序列u₁~u₈、x₁~x₈、y₁~y₈‧‧‧位元W、W₈ ‧‧‧通道W₂、W₄‧‧‧子通道5‧‧‧Transmission Time Interval 10‧‧‧Base Station 100‧‧‧Wireless Communication System 12‧‧‧Loop Redundancy Check Processor 14‧‧‧Sub-channel Mapping Unit 16‧‧‧Encoder 18‧‧‧Rate Matching unit 20‧‧‧user equipment 22‧‧‧polarity decoder 30‧‧‧downlink 40‧‧‧uplink 50‧‧‧wireless communication device 51‧‧‧special integrated circuit 53‧‧‧dedicated Instruction set processor 55, 59‧‧‧memory 57‧‧‧main processor D31‧‧‧downlink control information D32‧‧‧estimated downlink control information DCRC‧‧‧loop redundancy check data DEC‧‧‧decoding data DIN‧‧‧input data DOUT‧‧‧output data

‧‧‧Log likelihood ratio PCW, PCW'‧‧‧Polarity codeword

,

,

‧‧‧Estimating bits S300~S390, S310', S380a~S386a, S380b~S387b, S383a', S383_2, S383_4, S380c~S384c‧‧‧Operating SEQ, SEQ1, SEQ2‧‧‧bit sequence u ₁ ~u ₈ , x ₁ ~x ₈ , y ₁ ~y ₈ ‧‧‧bit W, W ₈ ‧‧‧channel W ₂ , W ₄ ‧‧‧subchannel

FIG. 1 is a block diagram illustrating a wireless communication system including a base station and a user equipment according to an example embodiment of the inventive concept. FIG. 2 is a block diagram illustrating an example of a structure of a downlink (DL) of FIG. 1 according to an example embodiment of inventive concepts. FIG. 3 illustrates a process of transmitting downlink control information (downlink control information; DCI) from the base station (base station; BS) of FIG. 1 to the user equipment (user equipment; UE) according to an example embodiment of the inventive concept instance of . FIG. 4 illustrates an example of the operation of the encoder of FIG. 1 according to example embodiments of inventive concepts. 5A and 5B illustrate an example of the operation of the polar decoder of FIG. 1 according to example embodiments of inventive concepts. FIG. 6 is a flowchart illustrating an example of operation S310 of FIG. 3 according to example embodiments of inventive concepts. FIG. 7 illustrates an example of the operation of operation S312 of FIG. 6 according to example embodiments of inventive concepts. FIG. 8 shows an example of a binary tree of a polar code according to an example embodiment of the inventive concept. FIG. 9 is a virtual code illustrating an example of a method of assigning predefined information (PI) bits according to an example embodiment of the inventive concept. FIG. 10 illustrates a code tree of bits of a bit sequence according to levels in consecutive offset decoding according to an example embodiment of the inventive concept. FIG. 11 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. FIG. 12 is a flowchart illustrating an example of operation S383a of FIG. 11 according to example embodiments of inventive concepts. FIG. 13 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. FIG. 14 is a flowchart illustrating an example of operation S380 of FIG. 3 according to example embodiments of inventive concepts. FIG. 15 is an example block diagram illustrating a wireless communication device according to an example embodiment of the inventive concept.

10: base station

12: loop redundancy check processor

14: Sub-channel mapping unit

16: Encoder

18: Rate matching unit

20: User equipment

22: Polar decoder

30: Downlink

40: Uplink

100: Wireless communication system

DCRC: loop redundancy check data

DEC: decode data

DIN: input data

DOUT: output data

PCW, PCW': Polar code word

SEQ: bit sequence

Claims (23)

A method of encoding input data into a polar code, the method comprising: generating unfrozen bits by adding at least one specified information bit to information bits already generated based on the input data; by adding the unfrozen reordering said unfrozen bits and said frozen bits by assigning bits to a more reliable polarization sub-channel than frozen bits having a value known to both an encoder and a decoder; and A codeword is generated from the reordered polar encoding results, wherein the at least one specified information bit has the value known to both the encoder and the decoder. The method for encoding input data into a polar code as described in item 1 of the scope of patent application, wherein the polar code tree includes only the rate 1 node with the unfrozen bit as a child node and the frozen bit and the unfrozen bit Freezing both rate R nodes as child nodes, and wherein said reordering comprises assigning said at least one designated information bit into a leaf node in said rate 1 node or said rate R node Polarized sub-channels with highest reliability. The method of encoding input data into a polar code as described in claim 2, wherein said assigning includes assigning said at least one designated information bit to said rate 1 node or said rate R node At least one leaf node with the lowest index among the associated candidate indexes. The method for encoding input data into a polar code as described in item 2 of the scope of the patent application further includes: performing rate matching by weakening or shortening the codeword, wherein the assignment includes converting the at least one designated information bit Bits are assigned to polarized sub-channels among the polarized sub-channels that do not correspond to weakened bits or shortened bits in the rate matching. The method for encoding input data into a polar code as described in item 1 of the scope of the patent application, wherein the information bits include input bits and loop redundancy check bits, and wherein the reordering further includes converting all The input bits and the loop redundancy check bits are assigned to some of the polarization sub-channels to which the at least one designated information bit is not assigned. The method for encoding input data into a polar code as described in item 5 of the scope of patent application, wherein said assigning said input bit and said loop redundancy check bit comprises assigning said input in an interleaved manner bit and the loop redundancy check bit. The method for encoding input data into a polar code as described in item 5 of the scope of the patent application further includes: determining the at least one designated information bit according to the false alarm rate and the number of the loop redundancy check bits quantity. The method for encoding input data into a polar code as described in item 1 of the scope of the patent application further includes: transmitting the code word to the physical control channel of the wireless communication system, The transmission includes transmission of RRC information omitting the codeword. The method for encoding input data into a polar code as described in claim 1, wherein the at least one designated information bit has a fixed value or a unique identification code of the decoder. An apparatus for encoding input data into a polar code, the apparatus comprising: a memory configured to store computer readable instructions; and one or more processors configured to execute the computer readable instructions such that the The one or more processors are configured to: generate a sequence of bits by reordering unfrozen bits and frozen bits having values known to both the encoder and the decoder based on the reliability of the polarization subchannels, said unfrozen bits comprising at least one designated information bit and an information bit based on said input data; and generating a codeword by polar encoding said sequence of bits, wherein said unfrozen bits are assigned to said Polarized sub-channels with higher reliability than frozen bits. The device for encoding input data into a polar code as described in claim 10, wherein said one or more processors are further configured to receive said information bits and pass said at least one designated information bit added to the information bits to generate the unfrozen bits. The device for encoding input data into a polar code as described in item 10 of the scope of patent application, wherein the polar code tree includes only the unfrozen bit as a child node a rate 1 node and a rate R node having both said frozen bit and said unfrozen bit as child nodes, and wherein said one or more processors are further configured to convert said at least one designated information bit A cell is assigned to the polarization subchannel with the highest reliability among the rate 1 node or the rate R node's leaf nodes. The apparatus for encoding input data into a polar code according to claim 10, wherein the one or more processors are further configured to: generate a loop redundancy check from input bits of the input data bit, and generating said information bit comprising said input bit and said loop redundancy check bit, and assigning said input bit and said loop redundancy check bit to Some polarization sub-channels to which the at least one designated information bit is not assigned. The device for encoding input data into a polar code as described in claim 10, wherein said one or more processors are further configured to: weaken or shorten said codeword, and convert said at least one specified information Bits are assigned to polarization sub-channels among the polarization sub-channels that do not correspond to weakened bits or shortened bits. A method of decoding comprising: receiving a codeword produced by polar encoding a sequence of bits; and performing list decoding of the codeword by producing a list of L decoding paths where L is a positive integer, wherein the bits The sequence consists of input bits, loop redundancy check bits and Unfrozen bits of at least one designated information bit and frozen bits having values known to both an encoder and a decoder, and wherein said performing includes early terminating said The list decoding of codewords, and wherein the unfrozen bits are assigned to a more reliable polarization sub-channel than the frozen bits. The decoding method according to claim 15, wherein the execution includes: determining the type of the node based on each of the L decoding paths based on the level of the code tree; and responding to the at least one The type of the node specifying an information bit determines early termination based on whether a value decoded by the node agrees with a value of one of the at least one specified information bit. The decoding method according to claim 16, wherein said determining early termination comprises responding to a value obtained by decoding said at least one designated information bit for all said L decoding paths not being consistent with said at least The value of one of the specified information bits agrees to determine early termination of the list decoding of the codeword. The decoding method according to claim 16, wherein said unfrozen bits comprise a plurality of specified information bits, and wherein said determining early termination comprises: in response to passing decoding among said L decoding paths All values obtained by the plurality of designated information bits are related to the at least one Specifying a decoding path in which a coincidence of the value of one of the information bits does not occur determines early termination of the list decoding of the codeword. The decoding method according to claim 16, wherein said performing further includes performing list reduction in response to said type of said node being an input bit. The decoding method of claim 16, wherein said performing further includes performing list reduction in response to said type of said node being unfrozen bits. The decoding method described in claim 16 of the patent application, wherein the execution further includes discarding the decoding path, in which the decoded value corresponding to the at least one specified information bit and the at least one specified information The value for one of the bits is inconsistent. The decoding method of claim 15, wherein the performing further includes performing a loop redundancy check on the decoding path in response to reaching a final bit in the decoding path. An apparatus for decoding codewords, comprising: a memory configured to store computer-readable instructions; and one or more processors configured to execute the computer-readable instructions such that the One or more processors are configured to execute the decoding method according to claim 15 of the patent application.

Images