KR20100003112A - Apparatus and method for systematic constant amplitude precoding, and apparatus and method for decoding according to it - Google Patents

Abstract

PURPOSE: An apparatus and method for systematic constant amplitude precoding, and apparatus and method for decoding according to it are provided to preserve the prototype of the parity. CONSTITUTION: The user data Input part(40) inputs the user data vector to the generating matrix multiplier(42). The generating matrix multiplicator multiplies the generating matrix in the inputted data vector. The adder(43) adds up the surplus vector for the output of the generating matrix multiplicator and polarity reversal. The matrix convert converts(44) the first encoding result into the transposed matrix.

Description

Systematic constant amplitude precoding device and method thereof, and decoding device and method therefor {APPARATUS AND METHOD FOR SYSTEMATIC CONSTANT AMPLITUDE PRECODING, AND APPARATUS AND METHOD FOR DECODING ACCORDING TO IT}

The present invention relates to a systematic constant amplitude precoding apparatus and a method thereof, and a decoding apparatus and method according thereto, and more particularly, a final step after performing systematic encoding by introducing a generator matrix By spreading only, the encoding process can be simplified to facilitate implementation, and the parity bits can be kept circular even when the encoding stage (encoding order) is increased, so that the existing superior decoding scheme (for example, the LDPC scheme) at the receiving end can be achieved. And a systematic constant amplitude precoding apparatus and method thereof, and a decoding apparatus and method according thereto.

The present invention is derived from the research conducted as part of the high-tech urban technology development project of the Ministry of Land, Transport and Maritime Affairs [Task management number: 07 national land information C04, task name: intelligent national land information technology innovation project].

In March 2007, IEEE 802.15.4a adopted Impulse Ultra WideBand (UWB) as the Wireless Personal Area Networks (WPAN) physical layer technology. According to this standard, impulse UWB enables location recognition and low-speed data transfers of less than 1Mbps. Here, UWB is a wireless technology for transmitting a large amount of digital data over a wide spectrum frequency at low power over a short range.

In Korea, the Korea Electronics and Telecommunications Research Institute (ETRI) succeeded in developing a prototype system on a chip (SOC) based on this standard in October 2007, demonstrating distance measurement and demonstrating BPAM (Binary Pulse Amplitude Modulation) DS-UWB transmission technology. Has shown to be practical.

Based on these findings, the Ministry of Land, Transport and Maritime Affairs' Intelligent Homeland Information Division researches the application of impulse UWB technology as a technology that can transmit USN (Ubiquitous Sensor Network) data as well as indoor location recognition. When multiple users access the USN radio channel that provides low-speed data at the same time, a collision occurs. To avoid this (i.e. to solve the multiple access problem), the PPM (Pulse Position Modulation) UWB is used for each user. Different frames are allocated differently to be used exclusively, and in CDMA, orthogonal codes are assigned to spread and transmit signals. That is, in USN, multiple access is required. Impulse UWB has a constraint that does not allow the overlapping of signals occurring in multiple access, and it is divided into PPM or spreading code that allocates time slots differently to solve the multiple access problem. The use of DS-UWB has emerged.

Multi-code generates a high peak voltage when overlapping. As a method for reducing the peak-to-average power ratio (PAPR) problem, constant amplitude precoding has been studied, and the present invention provides such constant amplitude precoding. Is applied as a multiple access method of DS-UWB. In addition, constant amplitude precoding is also a product code and can be decoded by a low density parity check (LDPC) method.

Among the conventional techniques, Wada's precoding method is a method of making a constant amplitude by making a 4-bit combination so that the XOR value is "1" for a 3-bit input signal, and then spreading it by Hadamard Code. There is a technique proposed by Pseudo-Hadamard Matrix so that it can be extended to 64-bit code.

Since these two conventional methods use bit-replacement to spread a signal at each encoding stage by using Hadamard code or pseudo-hadamard code, the originality of parity bits is not preserved and encoded in one channel. The problem is that the spreading code must be used as many steps.

Before examining the prior art and its problems in more detail, the need for constant amplitude precoding will be discussed.

In order to make the M-ary signal of the PPM UWB, a value is represented according to M sub-interval positions in the time frame, BPAM represents a UWB pulse, +1 is positive phase, and -1 is reverse phase. As a result of the development of SOC chip of ETRI, UWB signal has been demonstrated that it is possible to transmit UWB signal in the normal phase and the reverse phase in the form of Gaussian impulse and synchronously receive it.

DS-UWB channel coding can be deployed based on the feasibility of implementing the synchronous BPAM UWB transmission / reception technology. In the USN environment, when multiple UWB sensor nodes are installed, access control such as software user addition and reduction should be enabled for efficient multi-access control, and channel division by code such as CDMA needs to be considered. Since constant amplitude precoding can be divided into channels by similar pseudo Hadamard codes, it is possible to control software multiple access using it.

As a result of efforts to use UWB for wireless digital communications, R.A., University of Sourthern California (USC). Scholtz's UWB multiple access method has been proposed and various studies have been attempted. Equation 1 below shows the multiple access method proposed by Scholtz.

를 할당받으며, 이 것을 '호핑 코드'(Hopping Code)라 한다. 이 호핑 코드

는 주기적인 유사 난수(Pseudorandom) 코드이다. 일반적으로 UWB 시스템은 가우스 함수를 2차 미분한 신호를 송신신호로 사용하며, 펄스 모양이나 변조방식은 다양하게 선택될 수 있다.In this manner, a multiple access signal consisting of uniformly arranged pulses is susceptible to damage by collisions in which multiple pulses are simultaneously received. To avoid such collisions in multiple connections, each link identified by k has a distinct pulse shift pattern.

Is assigned a 'hopping code' (Hopping Code). This hopping code

Is a periodic pseudorandom code. In general, the UWB system uses a second derivative of a Gaussian function as a transmission signal, and a pulse shape or a modulation method can be variously selected.

For the description of signal generation and wireless transmission, a Gaussian function is described as an example. Equation 2 below is a Gaussian function and can be made through a shaping filter.

[Equation 3] is a first derivative of the Gaussian function of [Equation 2], and is a signal that is differentiated through the transmission antenna.

Binary Pulse Amplitude Modulation (BPAM) The DS-UWB system uses a spreading code to classify users. The signal of the k-th user is expressed by Equation 4 below.

Here, c _n is a spreading code assigned to a user, d _j is bipolar data, and W (t) is a Gaussian pulse.

If the data set whose value is +1 or -1 occurs from four users at the same time and is overlapped in the load, s is s = {4, 2, 0, -2, -4}. The overlapping amplitude varies depending on the combination of.

However, since the impulse UWB system cannot distinguish the magnitude of the signal, it cannot be applied when the magnitude of the signal is variable. Therefore, in order to overcome this, it is necessary to use a method of encoding d _j by a constant amplitude precoding scheme and decoding it by LDPC decoding.

Next, a description will be given of conventional constant amplitude precoding in detail.

1 is a diagram illustrating a conventional amplitude precoding method of a conventional Wada.

을 만족하는 4 비트(bit)를 Hadamard Matrix로 확산시키면 신호 진폭(amplitude)이 일정하게 된다는 것을 발견하였다. 이러한 성질 을 이용하여 1단계 인코딩(10)은 3 비트 사용자 데이터

를 입력받아서 이들의 XNOR 값

를 덧붙여서 4비트 코드워드

를 형성한다(수학식 5 참조). 그리고 나서,

를 바이폴라(Bipolar) 신호로 변환한 후, 4x4 Hadamard Matrix H⁴로 확산시켜 송신신호 s를 만든다(수학식 5 참조).Wada

We found that spreading 4 bits satisfying the Hadamard matrix results in a constant signal amplitude. Using this property, one-step encoding (10) uses three-bit user data.

To get their XNOR values

Plus 4-bit codeword

Form (see Equation 5). Then the,

Is converted into a bipolar signal and then spread to the 4x4 Hadamard Matrix H ⁴ to produce the transmission signal s (see Equation 5).

 For example, if the user data is (1 1 1), you can get (1 1 1 -1) by adding "0" which is the 3 bit input XNOR value and converting it into bipolar signal, which is 4x4 Hadamard Matrix Spreading on each line of will result in a 4-bit message {(1 1 1 1), (1 -1 1 -1), (1 1 -1 -1), (-1 1 1 -1)} Simultaneous transmission results in a linear combination of signals (2, 2, 2, -2) resulting in a constant amplitude.

메시지와 이들을 아래의 [수학식 6]과 같이 중간과정에서 확산시키는 비트 교체(Bit-Replacement)를 이용해서

(12)를 구하여, 최종적으로는 16비트 메시지로 만든다(14). Second Order encoding (14)

By using messages and bit-replacement to spread them in the intermediate process as shown in Equation 6 below.

(12) is obtained and finally made into a 16-bit message (14).

In the Wada method as described above, a constant amplitude can be obtained through two-step encoding Q (16,9) conversion, but there is a problem in that it cannot proceed to a further encoding step.

2 is an explanatory diagram of a conventional constant amplitude multicode (CAMC) encoding method.

를 사용함으로써 3단계 인코딩, 즉 Q(64, 27) 이상을 가능하게 한다. 여기서, N은 코드 길이이다.In the conventional technology proposed to overcome the problems of the conventional Wada method as described above, the CAMC (Constant Amplitude Multi-Code) method is a pseudo-Hadamard matrix instead of a Hadamard matrix.

By using, three levels of encoding are possible, namely Q (64, 27) or more. Where N is the code length.

In the Pseudo-Hadamard Matrix, each row is orthogonal and there are four "1s" per row, so the user data is spread over four "1s" and (n-4) "0s".

The two conventional methods use bit-replacement to spread the Hadamard Code or Pseudo-Hadamard Code for each encoding step (12, 14, 20, 22), so that the originality of the parity bits is not preserved. There is a problem that different spreading codes are applied to each encoding stage in a channel.

The prior art as described above has a problem in that encoding is complicated by bit-replacement and parity bit primitives are not preserved. It is an object of the present invention to solve such a problem.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention which are not mentioned above can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Also, it will be readily appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

In order to solve the above object, the present invention is characterized by spreading only in the final step after performing systematic encoding by introducing a generator matrix (Generator Matrix).

More specifically, the present invention provides a constant amplitude precoding apparatus comprising: input means for receiving user data; An "odd parity-type encoding message" is generated using a generation matrix and a surplus vector for polarity inversion, wherein the primary encoding message is the user data to be encoded, and the encoding message of order N (N≥2) is (N -1) encoding means for targeting the secondary encoding message; Matrix transform means for generating a transpose matrix for the (N-1) th encoded message and inputting it to the encoding means; And spreading means for converting the N-th order encoded message output from the encoding means into a bipolar signal and then spreading it into a similar Hadamard matrix.

In addition, the present invention provides a constant amplitude precoding method comprising: an input step of receiving user data; An encoding step of generating an "odd parity-type encoded message" using a generation matrix and a surplus vector for polarity inversion; Converting the encoded message into a bipolar signal; And spreading the encoded message converted into the bipolar signal into a similar Hadamard matrix.

In addition, the present invention provides a decoding apparatus for systematic constant amplitude precoding, comprising: despreading means for despreading a constant amplitude precoded encoded signal into a similar Hadamard matrix and then converting it into a binary encoded message; Conversion means for converting the binary type encoded message into an even parity type encoded message; And decoding means for decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme.

In addition, the present invention provides a method for decoding a systematic constant amplitude precoded message, the method comprising: despreading a constant amplitude precoded encoded signal into a similar Hadamard matrix and then converting it into a binary encoded message; Converting the binary type encoded message into an even parity type encoded message; And a decoding step of decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme.

In addition, the present invention, the constant amplitude precoding system having a processor, an input function for receiving user data; An encoding function for generating an "odd parity-type encoding message" using a generation matrix and a surplus vector for polarity inversion; A conversion function for converting the encoded message into a bipolar signal; And a computer-readable recording medium having recorded thereon a program for realizing a spreading function for spreading the encoded message converted into the bipolar signal into a similar Hadamard matrix.

In addition, the present invention, in order to decode a systematic constant amplitude precoded message, in a decoding system having a processor, despread the constant amplitude precoded encoded signal into a similar Hadamard matrix, and then into a binary encoded message. Despreading function to transform; A conversion function for converting the binary type encoded message into an even parity type encoded message; And a computer readable recording medium having recorded thereon a program for realizing a decoding function of decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme.

As described above, since constant amplitude precoding is performed in a systematic encoding method using a generator matrix, higher order encoding may be performed through a simple encoding process without using a spreading code. It can be effective. In other words, the present invention has the effect of performing constant amplitude precoding in software without using complicated logic circuits.

In addition, the present invention does not use a bit-replacement method in performing constant amplitude precoding, thereby preserving the originality of the parity bit, thereby syndrome decoding and LDPC. There is an effect of facilitating the application of decoding.

In addition, since the systematic constant amplitude precoding result according to the present invention is a systematic block code, it is possible to multiply the DS-UWB channel by spreading it to Pseudo-Hadamard Code and to apply LDPC decoding. This can improve the performance of noise.

In short, the present invention provides a systematic encoding scheme by introducing a generator matrix to the Wada code in order to compensate for the problems of the existing technology in which the encoding is complicated due to bit-replacement and the parity bit prototype is not preserved. By performing the encoding and spreading only in the final stage, the parity bit can be maintained even if the encoding stage (encoding order) is increased, which causes the hard end to undergo a hard decision through a bit flipping process. And other well-known decoding methods such as LDPC decoding.

The above objects, features and advantages will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, whereby those skilled in the art may easily implement the technical idea of the present invention. There will be. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

3 is a block diagram of an embodiment of a multiple access impulse shock transmission / reception system according to the present invention. As shown in FIG. 3, a multiple access impulse shock transmission system 30 and a multiple access impulse shock reception system 34 are included. Is done.

First, the multiple access impulse shock transmission system 30 will be described. The multiple access impulse shock transmission system 30 according to the present invention comprises a constant amplitude precoding device 31 and a DS-UWB modulator 32.

As shown in FIG. 3, the constant amplitude precoding device 31 includes a precoding unit 311, a level converting unit 312, and a diffusion unit 313.

조건을 만족시키는 인코딩 메시지)를 생성한다. 그러면, 레벨 변환부(312)는 프리코딩부(311)에서 출력되는 이진 형태의 인코딩 메시지를 바이폴라 신호로 변환한다. 즉, "1"→"+1", "0"→"-1"로 변환한다.The precoding unit 311 receives the user data (information bits) and uses the generator matrix and the surplus vector R for polarity inversion to form an " odd parity type encoding message "

Generates an encoding message that satisfies the condition. Then, the level converter 312 converts the binary encoded message output from the precoding unit 311 into a bipolar signal. That is, it converts into "1"->"+1" and "0"->"-1".

Subsequently, the spreader 313 spreads the encoded data converted into the bipolar signal by the level converter 312 using a predetermined spreading code (for example, Pseudo-Hadamard matrix), and the DS-UWB modulator 303. Is modulated by the DS-UWB method and transmitted. Here, a detailed process of the precoding unit 311 will be described with reference to FIG. 4.

Next, the multiple access impulse shock receiving system 34 will be described. The multiple access impulse shock receiving system 32 according to the present invention comprises a DS-UWB demodulation unit 35 and a decoding device (decoding device corresponding to the constant amplitude precoding device) 36.

The demodulator 321 demodulates a signal received through a wireless channel and converts the signal into a digital bit string, and the despreader 361 converts the bit string output from the DS-UWB demodulator 35 into a spreading code on the transmitting side. Despread using the same spreading code. The level converter 362 converts a bipolar signal into binary data. That is, it converts from "+1" to "1" and "-1" to "0".

Thereafter, the decoder 363 applies a bit flipping and a parity check matrix to a parity message (a message having information bits and parity bits) output from the level converter 362. By performing the decoding, the original user data (information bits) are restored. The adder 3631 generates an even parity message by adding ΣR ^N to the output of the level converter 362. The LDPC decoding unit 3632 decodes the even parity message by applying the LDPC decoding scheme. Detailed description thereof will be described later.

4 is a detailed configuration diagram of an embodiment of the precoding unit of FIG. 3 according to the present invention.

In the conventional constant amplitude precoding method, when performing a second order or more encoding, bit-replacement, which is a spreading process, must be performed for each encoding step, but the systematic constant amplitude free method according to the present invention is performed. The coding method performs a systematic encoding using a generator matrix and then spreads it into a pseudo-hadadamard matrix.

을 만족하여 4번째 비트가 나머지 3 비트의 XNOR 연산결과와 같다.The Q (4,3) codeword of the systematic constant amplitude precoding scheme according to the present invention is {0001, 0010, 0100, 1000, 1110, 1101, 1011, 0111} in the same manner as the conventional Wada scheme and the CAMC scheme. This codeword is the first of constant amplitude code conditions.

Is satisfied, and the 4th bit is the same as the remaining 3 bits of XNOR operation result.

은 다음의 [수학식 8]과 같이 생성된다.Conventionally, bit replacement has been performed to generate (16, 9) codes. However, in the systematic constant amplitude precoding scheme according to the present invention, encoding is performed using a generator matrix. If the user data is u, the generator matrix G, the surplus vector R for polarity inversion, and the code length is N, the encoded message

Is generated as shown in Equation 8 below.

,

이다.here,

,

to be.

은 N-tuple column vector이고

행렬(matrix)에서 열 순서대로(column-by-column) 추출한다. 예를 들어, 1차 인코딩 메시지(1^st order

)와 2차 인코딩 메시지(2^nd order

)는 다음의 [수학식 9]와 같이 생성된다.

Is an N-tuple column vector

Extract column-by-column from matrix For example, the primary encoding message (1 ^st order

) And the second encoded message (2 ^nd order

) Is generated as shown in Equation 9 below.

Hereinafter, the constant amplitude precoding method as described above will be described with reference to FIG. 4.

= [1 0 1 1]을 생성한다. 도 4에서 생성행렬 곱셈부(42) 및 가산기(43)를 묶어서 인코딩부(41)라고 할 수 있다.First, a first stage (primary) encoding process, that is, Q (4, 3) will be described. When the user data input unit 40 inputs the user data vector [1 0 1] to the generation matrix multiplier 42, the generation matrix multiplier 42 generates the generation matrix G at the input [1 0 1]. Multiply (see Equation 8) to produce [1 0 1 0]. Thereafter, the adder 43 sums [1 0 1 0] output from the generation matrix multiplier 42 and the surplus vector R for polarity inversion [0 0 0 1].

= [1 0 1 1]. In FIG. 4, the generation matrix multiplier 42 and the adder 43 may be collectively referred to as an encoding unit 41.

Next, a two-step (secondary) encoding process, that is, a process of generating Q (16, 9) will be described. In the case of user data (1 0 1, 1 1 0, 0 0 1), iteratively encodes first-order for user vectors [1 0 1], [1 1 0], and [0 0 1] in units of 3 bits. When the process is performed, the encoding result is [1 0 1 1], [1 1 0 1], [0 0 1 0].

The matrix converter 44 converts each primary encoding result into a transpose matrix, stores the result, and outputs the result of the second encoding. The output is expressed by Equation 10 below.

The generation matrix multiplier 42 in the two-step encoding process multiplies the matrix output from the matrix converter 44 by the generation matrix G used in the first-order encoding process (see Equation 8). Equation 11] is calculated.

In this case, the surplus vector R for polarity inversion is as shown in Equation 12 below.

The adder 43 in the two-stage encoding process adds the matrix of [Equation 11] generated in the two-stage encoding process and the matrix of surplus vectors (Equation 12) for polarity inversion, thereby adding [Equation 13] and the result. Calculate.

를 1차원 행태로 표현하면, [1 0 1 1 1 1 0 1 0 0 1 0 1 0 1 1]와 같다.And, represented by the 4 × 4 matrix as shown in [Equation 13]

When expressed as a one-dimensional behavior, it is equal to [1 0 1 1 1 1 0 1 0 0 1 0 1 0 1 1].

By repeatedly performing the above method, the encoding order (step) can be extended, and the polarity inversion redundancy vector used in this process may use an extended value corresponding to the encoding order.

That is, the precoding unit 311 according to the present invention generates an "odd parity-type encoding message" using the generation matrix and the redundant vector for polarity inversion, but the primary encoding message is to encode the user data. The N (N≥2) th encoded message is characterized by targeting the (N-1) th encoded message.

은 다음의 [수학식 14]와 같이 바이폴라(bipolar) 신호로 변환된 후, Pseudo-Hadamard Code(

)(수학식 7 참조)에 의하여 확산된다.Message encoded through the process as shown in FIG.

Is converted into a bipolar signal as shown in [Equation 14], and then Pseudo-Hadamard Code (

(See Equation 7).

는 임펄스UWB 펄스성형필터(Pulse Shaping Filter)를 통하여 다음의 [수학식 15]와 같은 형태의 송신단 출력신호

로 변환된다.Send message as above

Transmitter output signal of the form as shown in [Equation 15] through Impulse UWB Pulse Shaping Filter

Is converted to.

5 is a diagram illustrating an embodiment of a method for LDPC decoding of systematic constant amplitude precoding according to the present invention.

First, an LDPC decoding method using a parity check matrix will be described.

과 수신신호

가 패리티 검사 이벤트를 만족시킬 조건확률

의 결합으로 표현된다.In the LDPC algorithm, the posterior probability (APP: A posteriori probability) is a pre-probability of a received signal as shown in Equation 16 below.

And received signal

Probability to satisfy parity check event

It is expressed as a combination of.

인 조건을 만들어 주는 4비트 홀수 패리티 코딩(4-bit odd parity coding) 방식이기 때문에 이를 4비트 짝수 패리티 코딩(4-bit even parity)로 변환하면 LDPC 알고리즘을 적용할 수 있다.Meanwhile, Wada's constant amplitude precoding scheme is a 4-bit unit.

Since it is a 4-bit odd parity coding method that creates an in condition, the LDPC algorithm can be applied by converting it to 4-bit even parity coding.

Since the generator matrix used in the present invention is recursively calculated according to the encoding level, the parity check matrix cannot be directly extracted from the generation matrix, and the parity check is performed from the encoded message pattern. Extract the matrix

인 관계가 있는 비트들의 위치에 "1"을 마킹(Marking) 함으로써 패리티 검사 행렬 H를 얻을 수 있다. 여기서, 패리티 검사 행렬 H의 각 행(Row)은 하나의

인 관계를 표현한다.Analyze the pattern of encoded messages to generate different

The parity check matrix H can be obtained by marking " 1 " at positions of bits having a relation of. Here, each row of the parity check matrix H is one

Express a relationship.

이다. (4,3) 코드의 패리티 검사 행렬은

이다. 아래의 [수학식 17]은 일정진폭 프리코딩의 패리티 검사 행렬의 일반적인 구조를 나타낸다.In the present invention, the parity check matrix H has the number of " 1 " belonging to each column, that is, the column weight equals the encoding level (step).

to be. The parity check matrix of the (4,3) code is

to be. Equation 17 below shows the general structure of the parity check matrix of constant amplitude precoding.

The simplified constant amplitude precoding scheme has (nk) independent rows and some extra rows of checks on checks. For example, H ₁₆ consists of seven independent rows and one check on check. The feature of the parity check matrix proposed by the present invention is that the row weight is fixed as "4" and the column weight is increased by the number of encoding levels. For example, H ₁₆ of level 2 encoding (two-level encoding) has a column weight of "2" and a row weight of "4" (see Equation 18), H ₆₄ of level 3 encoding (three-level encoding) has a column weight of “3” (see Equation 19).

가 추출된다. 홀수 패리티(odd parity)인

를 짝수 패리티(even parity)로 변환하기 위하여, 패리티 비트(parity bit)만 골라서 비트 극성을 바꿔줘야 한다(50). Multiply the received signal by the pseudo-adamadam matrix to despread the message.

Is extracted. Odd parity

In order to convert to even parity, it is necessary to change the bit polarity only by selecting a parity bit (50).

은 잉여 벡터(Remainder Vector) R과 같다는 성질을 이용한다. 수신 측에서는 이렇게 구한 잉여 벡터(Remainder Vector)를 수신 신호에 가산함으로써, 비트 플리핑(Bit Flipping)을 간편하게 수행할 수 있다. 이와 같은 비트 플리핑(Bit Flipping) 과정을 거치면 짝수 패리티(even parity) 메시지가 된다(50). 도 3에서 레벨 변환부(362)의 출력에 ΣR^N 을 더하면 자동적으로 비트 플리핑 과정이 수행되어 짝수 패리티(even parity) 메시지가 된다. 이렇게 짝수 패리티(even parity) 메시지로 변환되었기 때문에, 바로 기존의 다양한 LDPC 디코딩 방식(51)을 적용할 수 있는 것이다.However, if the encoding is performed evenly, there is no change in polarity of the parity bits. To simplify this parity bit polarity inversion, when all inputs are all zero inputs (u = 0)

Uses the property equal to the residual vector R. On the receiving side, bit flipping can be easily performed by adding the obtained residual vector to the received signal. This bit flipping process results in an even parity message (50). In FIG. 3, when ΣR ^N is added to the output of the level converter 362, a bit flipping process is automatically performed to form an even parity message. Since this is converted into an even parity message, various existing LDPC decoding schemes 51 can be applied.

In short, conventional methods cannot convert odd parity to even parity because the original parity bit is not preserved because bit-replacement is performed, but the present invention is a systematic encoding. Encoding preserves the original parity bit and converts the received odd parity message into an even parity message.

In the bipartite graph as shown in FIG. 5, R represents bit flipping, and serves to convert odd parity to even parity. Drawing a bipartite graph from the parity check matrix (generating), the girth forming the closed loop is " 8 ", which is generally stable if the girth is 6 or more. Rows and columns of the parity check matrix become check nodes and variable nodes, respectively. If the conventional LDPC decoding scheme 51 is simply arranged, it can be arranged in the following four steps, and a detailed description thereof will be omitted.

(1) Step 1: Initialization of bit nodes

(2) Step 2: Message Checks to Bits Node (Checks to bits)

(3) Step 3: Messages from bits node to check node (bits to check)

(4) Stop Criterion

Next, a hard decision decoding method using a syndrome will be described.

은 패리티 검사 행렬의 각 행

와 곱한 신드롬 벡터

를 이용하여 오류를 정정할 수도 있다. 예를 들어, H¹⁶ 패리티 검사 행렬을 살펴보면, 1 열(column)에 "1"이 2개 있어서 수신신호 r의 어느 한 비트가 에러이면 2개의 행(Row)에서 신드롬이 "0"이 아니게 되며, 해당되는 비트를 변환(invert)하여 오류를 정정할 수 있다.In some embodiments, the received signal

Is each row of the parity check matrix

Syndrome vector multiplied by

You can also correct the error using. For example, H ¹⁶ Looking at the parity check matrix, if there are two "1s" in one column and one bit of the received signal r is an error, the syndrome is not "0" in two rows, and the corresponding bits are converted. (invert) to correct the error.

In this way, a syndrome can be obtained by encoding order for one bit. As the encoding order increases, the error correction capability also increases.

(Iii) If the syndrome is obtained and is "0", since there is no error, the bit except for the parity bit of the received signal r is decoded to v.

들이 "0"이 아니면, 수신신호

는 오류비트이므로 수신 비트값을 바꿔준다.(Ii) The following is a syndrome of " 1 " in the same column j of the parity check matrix.

If they are not "0", the received signal

Since is an error bit, it changes the receive bit value.

(Iii) If the Parity Check Sum is not " 0 " and less than or equal to the maximum number of repetitions, step (ii) is repeated.

6 is an improved BER performance by LDPC decoding according to the present invention.

을 대입하고, 확산코드에 Pseudo Hadamard 코드를 대입한 후, Rayleigh 및 AWGN 환경에서 LDPC 디코딩 BER 성능을 시뮬레이션하였다. 이 시뮬레이션을 통하여 본 발명에 따른 체계적인 일정진폭 프리코딩 방식의 다중접속 임펄스 UWB 시스템(61 내지 63)은 BER 10^-3에서 비부호화(Uncoded) BPAM DS-UWB 시스템(60)에 비하여 2dB의 이득이 있음을 확인할 수 있다. User data encoded in user data to evaluate the performance of the present invention.

After substituting the Pyudo Hadamard code into the spreading code, we simulate the LDPC decoding BER performance in Rayleigh and AWGN environments. Through this simulation, the systematic constant amplitude precoding multiple access impulse UWB system 61 to 63 has a gain of 2 dB compared to the uncoded BPAM DS-UWB system 60 at BER 10 ^-3 . It can be confirmed.

On the other hand, the method of the present invention as described above can be written in a computer program. And the code and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the written program is stored in a computer-readable recording medium (information storage medium), and read and executed by a computer to implement the method of the present invention. The recording medium may include any type of computer readable recording medium.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

1 is an explanatory diagram for a conventional constant amplitude precoding method of Wada,

2 is an explanatory diagram of a conventional constant amplitude multicode (CAMC) encoding method;

3 is a block diagram of an embodiment of a multiple access impulse power transmission and reception system according to the present invention;

4 is a detailed configuration diagram of an embodiment of the precoding unit of FIG. 3 according to the present invention;

5 is a diagram for explaining an LDPC decoding method of systematic constant amplitude precoding according to the present invention;

6 is an improved BER performance by LDPC decoding according to the present invention.

* Explanation of symbols on the main parts of the drawings

30: Multiple Access Impulse Check Transmission System 31: Constant Amplitude Precoding Device

32: DS-UWB Modulator 33: Channel (Wireless Channel)

34: Multiple Access Impulse Check Receiver System 35: DS-UWB Demodulation Unit

36: decoding apparatus for constant amplitude precoding 311: precoding section 312, 362: level converting section 313: diffusion section

361: despreader 40: user data input

41: encoding unit 42: generation matrix multiplier

43: redundant vector adder for polarity inversion 44: matrix transform unit

Claims (18)

In the constant amplitude precoding device, Input means for receiving user data; An odd-parity-type encoded message is generated using a generation matrix and a surplus vector for polarity inversion. The primary encoded message is the user data to be encoded, and the encoded message of N (N≥2 natural number) difference is generated. Encoding means for targeting the (N-1) th encoded message; Matrix transform means for generating a transpose matrix for the (N-1) th encoded message and inputting it to the encoding means; And Spreading means for spreading the N-th order encoded message output from the encoding means into a bipolar signal and then spreading it into a similar Hadamard matrix Systematic constant amplitude precoding device comprising a. The method of claim 1, The generation matrix,

Systematic constant amplitude precoding device, characterized in that. The method of claim 1, The surplus vector for polarity reverse,

Systematic constant amplitude precoding device, characterized in that. The method of claim 1, The similar Hadamard matrix is

Systematic constant amplitude precoding device, characterized in that. In the constant amplitude precoding method, An input step of receiving user data; An encoding step of generating an "odd parity-type encoded message" using a generation matrix and a surplus vector for polarity inversion; Converting the encoded message into a bipolar signal; And Spreading the encoded message converted into the bipolar signal into a similar Hadamard matrix Systematic constant amplitude precoding method comprising a. The method of claim 5, wherein The encoding step, An "odd parity-type encoded message" is generated using the generation matrix and the redundant vector for polarity inversion, wherein the primary encoded message is the user data to be encoded, and the encoded message of N (N≥2) order A systematic constant amplitude precoding method, characterized by targeting (N-1) th encoded messages. The method of claim 6, The encoding step, A plurality of " odd parity type code words " are generated for the user data by using the generation matrix and the surplus vector for the first polarity inversion for each predetermined number of bits, and then the plurality of individually generated code words are combined. Generating one primary encoded message; And Generating a second or more encoded message using a previous order encoded message, the generation matrix, and a redundancy vector for the polarity inversion of the order Systematic constant amplitude precoding method comprising a. The method of claim 5, wherein The generation matrix,

Systematic constant amplitude precoding method characterized in that. The method of claim 8, The surplus vector for polarity reverse,

Systematic constant amplitude precoding method characterized in that. The method of claim 5, wherein The similar Hadamard matrix is

Systematic constant amplitude precoding method characterized in that. In the decoding device for systematic constant amplitude precoding, Despreading means for despreading the constant amplitude precoded encoded signal into a similar Hadamard matrix and then converting it into a binary encoded message; Conversion means for converting the binary type encoded message into an even parity type encoded message; And Decoding means for decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme Decoding apparatus comprising a. The method of claim 11, The conversion means, And an encoding message obtained by inputting all user data as " 0 " in a systematic constant amplitude precoding device to the binary encoding message output from the despreading means. The method of claim 11, The similar Hadamard matrix is

Decoding apparatus characterized in that. A method for decoding a systematic constant amplitude precoded message, Despreading the constant amplitude precoded encoded signal into a similar Hadamard matrix and then converting it into a binary encoded message; Converting the binary type encoded message into an even parity type encoded message; And A decoding step of decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme; Decoding method comprising a. The method of claim 14, The conversion step, And an encoding message obtained by inputting all user data as " 0 " The method of claim 14, The similar Hadamard matrix is

Decoding method characterized in that. In a constant amplitude precoding system with a processor, An input function for receiving user data; An encoding function for generating an "odd parity-type encoding message" using a generation matrix and a surplus vector for polarity inversion; A conversion function for converting the encoded message into a bipolar signal; And A spreading function for spreading the encoded message converted into the bipolar signal into a similar Hadamard matrix A computer-readable recording medium having recorded thereon a program for realizing this. In a decoding system having a processor for decoding a systematic constant amplitude precoded message, A despreading function that despreads the constant amplitude precoded encoded signal into a similar Hadamard matrix and then converts the encoded signal into a binary encoded message; A conversion function for converting the binary type encoded message into an even parity type encoded message; And A decoding function for decoding the converted even parity encoded message by applying a low density parity check (LDPC) decoding scheme A computer-readable recording medium having recorded thereon a program for realizing this.

Images