CN115720129A - Information transmission method and system for polar coding continuous phase modulation - Google Patents

Abstract

The invention discloses an information transmission method and system of polar coding continuous phase modulation, wherein the method comprises the following steps: carrying out cyclic redundancy check coding and polarization coding on the information stream; interleaving and mapping the code words obtained by coding; and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission. The invention combines the CPM modulation technology and Polar code encoding and decoding technology, ensures higher error code performance of the system and lower receiving end complexity, and can be used in underwater environment with severe communication environment.

Description

Information transmission method and system for polar coding continuous phase modulation

Technical Field

The present invention relates to the field of communications technologies, and in particular, to an information transmission method and system for polar coding continuous phase modulation.

Background

The polarization code is based on a digital signal processing technology called channel polarization, and introduces correlation through channel division and combination, when the number of channels participating in channel polarization is enough, the reliability of the obtained channel can generate polarization phenomenon, and the more the number of channels is, the more obvious the channel polarization is, thereby the coding structure can be optimized. Polar codes have proven to achieve channel capacity when the code length is sufficiently large, and in practical transmission, the channel code length is limited, so the maximum achievable rate is lost.

The most basic decoding scheme of the polarization code is SC algorithm, theoretically, when the code length is sufficiently long, the channel capacity can be achieved, but the error code performance of the polarization code cannot achieve the ideal effect in the actual situation. Because the decoding algorithm of the polar code can be regarded as a process of searching the path on the polar code tree, the performance of the system can be effectively improved by introducing the breadth-first traversal improved path searching algorithm, namely a Serial Cancellation List (SCL) algorithm. After that, an optimized algorithm is derived, namely a decoding scheme (namely CA-SCL) after cyclic redundancy check and Polar code cascade connection, the communication performance can be further improved by carrying out construction optimization on the CRC-Polar code, and meanwhile, the complexity can be ensured to be unchanged.

Based on the channel polarization phenomenon, channels need to be allocated according to the reliability of each polarization channel to set the frozen bits, and the general methods for calculating the channel reliability include traditional Density Evolution (DE) and gaussian approximation scheme (GA), which has smaller complexity and smaller performance gap compared with the former.

The CPM modulation related to the system belongs to constant envelope modulation, and a power amplifier at a transmitting end can stably work in a nonlinear amplification area, so that nonlinear distortion is avoided and higher power efficiency is obtained. The traditional CPM modulation receiving end has higher complexity, and the complexity of the system can be greatly reduced while the system performance is ensured to be nearly unchanged by adopting the scheme of carrying out matched demodulation on the main components of the received signals by adopting the Laurent decomposition. Suppose an M-aryThe debugging index of the CPM signal is h, the memory length is L, and the pulse integral waveform q (t) is an integral of a cosine pulse or a gaussian pulse, so that the corresponding Laurent decomposition formula is as follows:

wherein Q =2 ^L-1 ，P＝log ₂ M，g _K (t) is a component waveform obtained by Laurent decomposition, namely a PAM signal, K is the serial number of the component waveform, a _K,n The coefficient corresponding to the Kth component waveform at time n is the information sequence { alpha } _n } of the complex symbols, commonly referred to in the literature as pseudosymbols (pseudosymbols), it is particularly noted that the component waveforms are real waveforms and the corresponding pseudosymbols are complex numbers.

Since Laurent decomposition decomposes the original CPM signal into N component pulses, and these component pulses differ in duration and amplitude, this also results in very uneven distribution of energy for these pulses. Statistically, when the modulation order M or the memory length L is small, the energy ratio of the M-1 pulses before the energy is about 90% or more of the total energy, which also means that a PAM signal with large energy can be used to approximate the CPM signal in the scheme, thereby further simplifying the processing of the signal by the receiver.

Under the condition of the prior experience information and the like, according to the maximum likelihood criterion, under the condition that the correlation metric between the received signal and the decision signal at the receiving end is maximum, the information sequence corresponding to the decision signal is the result. Through Laurent decomposition, different kinds of Laurent pulses of the CPM signal are received and matched with a received signal, x _K,n ＝∫r(t)g _K (t-nT)dt，x _K,n And combining the matched output response with the pseudo symbol to obtain branch measurement, forming a trellis diagram structure (trellis structure) of the CPM signal, and demodulating by using a BCJR algorithm. Thus, the complexity of the demodulation algorithm is related to the number of states of the trellis structure. When the CPM modulation index is h = q/p (p, q are co-prime integers), the number of states is pM if all Laurent pulses of the CPM signal are matched ^L-1 At this time, for optimum demodulation, the increase of the number of states leads to a complexThe degree of clutter increases exponentially. When only the first M-1 pulses are matched, the number of states is only p, the complexity is low, but the demodulation precision is reduced, so that the subsequent error code performance is reduced.

Linear modulation dependent coded modulation dependent schemes are many, but few have non-linear modulation as an entry point. The traditional non-linear coding modulation scheme mostly adopts RS codes and serial concatenation of convolutional codes to improve the error code performance, and some cases that the coding modes such as LDPC codes and turbo codes are also combined with non-linear modulation appear at present.

In summary, the disadvantages of the prior art are: the traditional coding schemes combining RS codes, convolutional codes and the like with nonlinear coding have poor error code performance, and decoding ends of LDPC codes, turbo codes and the like are complex; in addition, the complexity of the conventional CPM receiving-end detector is high, although the idea of demodulating by using the decomposition of CPM signals already exists, these schemes are only directed to binary CPM, or only use the first M-1 order principal component for matching, when the modulation index and the memory length are increased, only use of the M-1 order principal component may cause the error code performance to be degraded due to insufficient energy ratio, and a more flexible pulse matching mechanism is lacking.

Disclosure of Invention

In view of this, the present invention provides an information transmission method and system for Polar coding continuous phase modulation, which combines a CPM modulation technique with a Polar code encoding and decoding technique, thereby ensuring higher error code performance of the system and lower complexity of a receiving end, and being applicable to an underwater environment with a severe communication environment.

In view of the above object, the present invention provides an information transmission method of polar coding continuous phase modulation, comprising:

carrying out cyclic redundancy check coding and polarization coding on the information stream;

carrying out bit interleaving on the code words obtained by encoding and mapping binary system to multilevel system;

and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

Further, the method further comprises:

after receiving the CPM signal transmitted through the channel, demodulating the CPM signal by adopting a log-MAP algorithm;

and after de-interleaving the soft information obtained by demodulation, decoding by adopting a serial offset list assisted by cyclic redundancy check.

The present invention also provides an information transmission system of polar coding continuous phase modulation, comprising:

the CRC cascade polarization code encoder is used for carrying out cyclic redundancy check encoding and polarization encoding on the information stream;

the bit interleaver is used for carrying out bit interleaving on the code words obtained by coding and mapping binary system to multilevel system;

and the continuous phase modulator is used for sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

Further, the system further comprises:

a continuous phase demodulator for demodulating the CPM signal transmitted through the channel by using a log-MAP algorithm;

a deinterleaver for deinterleaving the soft information obtained by the demodulation;

and the decoder is used for decoding the soft information after de-interleaving by adopting a serial offset list assisted by cyclic redundancy check.

The invention also provides a signal sending end, comprising:

the CRC cascade polarization code encoder is used for carrying out cyclic redundancy check encoding and polarization encoding on the information stream;

the bit interleaver is used for performing bit interleaving on the code words obtained by encoding and mapping binary system to multilevel system;

and the continuous phase modulator is used for sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

The present invention also provides a signal receiving terminal, including:

a continuous phase demodulator for demodulating the CPM signal transmitted through the channel by using a log-MAP algorithm;

a deinterleaver for deinterleaving the soft information obtained by the demodulation;

and the decoder is used for decoding the soft information after the de-interleaving by adopting a serial offset list assisted by cyclic redundancy check.

In the technical scheme of the invention, the information flow is subjected to cyclic redundancy check coding and polarization coding; interleaving and mapping the code words obtained by coding; and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission. The invention uses CRC (Cyclic Redundancy Check) and Polar code cascade connection, can effectively improve the performance of Polar code under short code, and compared with LDPC and Turbo code, the invention has low complexity and high reliability of Polar code encoding and decoding, thereby ensuring higher error code performance of the system and lower complexity of a receiving end, and can be used in underwater environment with severe communication environment.

Furthermore, in the technical scheme of the invention, the optimized method in the Gaussian approximation algorithm is utilized in the estimation of the reliability of the polarization channel in the polarization coding process

Solving the log-likelihood ratio of each sub-channel by an expression AGA-2 and a likelihood ratio iterative algorithm; because the AGA-2 approximate expression does not generate great performance loss when the code length is longer, the complexity is much lower than that of a density evolution method, and an inverse function is easier to obtain, the calculation of channel reliability estimation can be simplified.

Further, in the technical solution of the present invention, the precision parameter D of continuous phase modulation may be determined by selecting a minimum value of values of D corresponding to matching pulse energy ratios higher than a set threshold value according to different matching pulse energy ratios corresponding to different precision parameters D under the condition that a modulation order, a modulation index, a memory length L, and a pulse waveform are determined, and taking the minimum value as the finally determined precision parameter D of continuous phase modulation; therefore, when the modulation order M or the memory length L is large, compared with the fixed precision parameter D value in the conventional continuous phase modulation technique, the error performance degradation caused by insufficient pulse energy ratio due to the adoption of only the main component pulse matching (D = 1) can be avoided, and the high complexity of matching (D = L) by adopting all the component pulses can be avoided.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic structural diagram of an information transmission system with polar coding and continuous phase modulation according to an embodiment of the present invention;

fig. 2 is a flowchart of an information transmission method of polar coding continuous phase modulation according to an embodiment of the present invention;

FIG. 3 is a diagram illustrating an expected simulation result of a sub-channel log-likelihood ratio using a Gaussian approximation algorithm according to an embodiment of the present invention;

fig. 4 is a schematic diagram of an internal structure of a source matched filter bank according to an embodiment of the present invention;

fig. 5a is a schematic state transition diagram of a trellis structure according to an embodiment of the present invention;

fig. 5b is a schematic diagram of a state transition trellis diagram of a trellis structure according to an embodiment of the present invention;

fig. 6 is a flowchart of a method for detecting a CPM signal by using a log-MAP algorithm according to an embodiment of the present invention;

FIG. 7 is a flowchart of a decoding method of the CA-SCL algorithm according to an embodiment of the present invention;

fig. 8 is a schematic diagram of experiments and simulation provided in the embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.

It is to be noted that technical terms or scientific terms used in the embodiments of the present invention should have the ordinary meanings as understood by those having ordinary skill in the art to which the present disclosure belongs, unless otherwise defined. The use of "first," "second," and similar terms in this disclosure is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.

The invention provides an information transmission method of polarization coding continuous phase modulation, which carries out cyclic redundancy check coding and polarization coding on an information stream; interleaving and mapping the code words obtained by coding; and sending the CPM (continuous Phase Modulation) signal obtained by carrying out continuous Phase Modulation on the mapped symbol into a channel for transmission. The invention uses CRC (Cyclic Redundancy Check) and Polar code cascade connection, can effectively improve the performance of Polar code under short code, and compared with LDPC and Turbo code, the invention has low complexity and high reliability of Polar code encoding and decoding, thereby ensuring higher error code performance of the system and lower complexity of a receiving end, and can be used in underwater environment with severe communication environment.

Moreover, the technical scheme of the invention combines the CPM modulation technology with Polar (polarization) code coding and decoding technology; the CPM modulation enables the system to have the characteristics of constant envelope, low out-of-band leakage and high channel utilization rate, for a signal receiving end, the complexity of the receiving end is related to the precision parameters defined in the invention, and in practical application, the trade-off compromise can be carried out between the error code performance and the complexity of the receiving end according to requirements.

The technical solution of the embodiments of the present invention is described in detail below with reference to the accompanying drawings.

The information transmission system of polar coding continuous phase modulation provided by the embodiment of the invention has a structure shown in fig. 1, and comprises: a signal transmitting end 101 and a signal receiving end 102;

the signal transmitting end 101 includes: a CRC concatenated polar code encoder 111, a bit interleaver 112, a continuous phase modulator 113;

the receiving end 102 of the signal includes: a continuous phase demodulator 121, a deinterleaver 122, and a decoder 123.

Based on the information transmission system of the polar coding continuous phase modulation, the information transmission method of the polar coding continuous phase modulation provided by the embodiment of the present invention has a flow as shown in fig. 2, and includes the following steps:

step S201: and carrying out cyclic redundancy check coding and polarization coding on the information stream.

In this step, cyclic redundancy check coding and polarization coding are performed on the information stream in the CRC concatenated polarization code encoder 111 of the signal transmitting end 101: grouping an input information stream into a sequence consisting of K information bits, performing Cyclic Redundancy Check (CRC) coding on the information sequence, adding m CRC bits to obtain a code length K (K = K + m), and finally passing the code through a polarization code encoder with a code rate R = K/N to obtain a polarization code with a code length N.

The specific process of polarization encoding comprises the following steps: and (3) estimating the reliability of the polarization channel, mixing bits and constructing a generating matrix. Specifically, a gaussian approximation algorithm may be adopted when estimating the reliability of the polarization channel; the application of the gaussian approximation algorithm is most widespread and most practical and the computational complexity is small compared to density evolution: according to the set signal-to-noise ratio initialized by the channel model and the modulation scheme, the channel likelihood value is solved according to the signal-to-noise ratio, and the optimized signal-to-noise ratio in the Gaussian approximation algorithm is utilized

The expression AGA-2 and the likelihood ratio iterative algorithm work out the log likelihood ratio of each sub-channel, and the Babbitt parameters and log likelihoods of N sub-channels can be further obtainedThe reliability of the polarization channel is reflected to a certain extent by the ratio or the babbitt parameter, that is, the reliability of the sub-channel can be estimated according to the log-likelihood ratio of each sub-channel; and further carrying out bit mixing: sorting channels according to the reliability of the channels, setting K sub-channels with high reliability as information transmission bits, setting the rest sub-channels as freezing bits, filling the information sequence on the information transmission bits to form an information sequence to be transmitted

Thereafter, a generator matrix is constructed, i.e. an N × N matrix G is generated _N ＝B _N F _N . Wherein B is _N Is a bit reverse order permutation matrix, completes the bit reverse order operation,

representation matrix

Log of ₂ Results of Kronecker product N times, i.e. log was performed ₂ N times

Performing recursive operation; finally, the information sequence to be transmitted is generated according to the generating matrix

Mapping to channel coded bits

Namely that

The most critical part is the estimation part of the polarized channel reliability, and the gaussian approximation construction scheme adopted in the scheme is introduced as follows:

assuming that the probability density function of the channel-demodulated received symbol log-likelihood ratios satisfies symmetry, and for a binary codeword, the further channel is approximated as a binary additionThe probability density function of log-likelihood ratio (LLR) obtained by soft bit demodulation of white Gaussian noise channel can be represented by Gaussian distribution with mean value of m and variance of 2m, and the equivalent signal-to-noise ratio (SNR) is

Assuming a channel-side log-likelihood ratio of

In the channel polarization process of the N long polarization codes, N polarization channels with length 1 are polarized into 1 polarization channel with length N from the channel side (each bit is regarded as one sub-channel), and the expectation of the log likelihood ratio of each sub-channel for gaussian approximation can be calculated by the following recursive formula 1 and formula 2:

wherein,

represents the log-likelihood ratio of the ith polarized channel in the polarized code with length n in the channel polarization process,

representing the log-likelihood ratio of the channel hypothesis, can be considered as the log-likelihood ratio of a polarized channel of length 1.

Due to the function in equation 1

The expression is more complicated, and the scheme adopts optimization

Approximate expression (AGA-2) approachSimilarly, the expression:

the AGA-2 approximate expression does not generate great performance loss when the code length is longer, the complexity is much lower than that of a density evolution method, and the inverse function is easier to obtain. By recursively averaging the polarized channels, the information bits can be selected by sorting them in descending order according to the expectation of log-likelihood ratio, taking the top k-bit subchannel indices as information bits.

Examples are: selecting N =1024, the code rate is 1/2,

the approximate expression selects an AGA-2 approximate expression, and the log likelihood ratio expectation simulation result using the Gaussian approximation algorithm is shown in FIG. 3. From this, the following conclusions can be drawn:

(1) When the channel index is smaller, the expected value of the corresponding channel log-likelihood ratio is smaller, and vice versa;

(2) When the channel index is in the middle value, the expected value of the corresponding channel log-likelihood ratio is in a step-up trend;

(3) In selecting information bits, the information bits may be sorted in descending order as desired by log-likelihood ratio, taking the top k-bit subchannel indices as information bits.

Step S202: and carrying out bit interleaving on the code words obtained by encoding and mapping binary system to multilevel system.

In this step, the bit interleaver 112 in the signal sending end 101 performs bit interleaving on the code word obtained by encoding, and mapping from binary to multilevel;

preferably, for the subsequent implementation of the BCJR algorithm, the bit interleaver 112 in this step may further set the codeword structure as follows: bit interleaving and multilevel mapping are carried out on the coded binary bits, and the obtained multilevel symbols are

Wherein a is _i ∈{-M+1,-M+3,……，M-3，M-1}，N _M Is the post-mapping symbol length

N is the code length of the polar code, L symbols are respectively added before and after the multilevel symbol, and the multilevel transmission sequence can be expressed as

Wherein, the first L symbols are all-M +1, so that the initial state can be started from a zero state; similarly, the last L symbols may cause the initial state to end with a zero state.

Step S203: and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

In this step, the continuous phase modulator 113 in the sending end 101 of the signal sends the CPM signal obtained by performing continuous phase modulation on the mapped symbol to the channel for transmission;

during the nth symbol period, the phase state of CPM can be expressed as the following equation 3:

in formula 3, α is the multilevel symbol sequence, h is a modulation index, a positive integer L is a memory length, and T is a symbol period, that is, a duration of a multilevel symbol of CPM; q (t) is a pulse integration function, which is a pulse shaping function g _cpm Integration of (t), generally g _cpm The waveforms of (t) are square wave pulses, raised cosine pulses, and gaussian pulses.

Corresponding to the sending end, the invention adopts CRC-Polar and CPM combined decoding demodulation at the receiving end, thereby greatly reducing the complexity of the receiving end and improving the error code performance of the receiving end. Firstly, demodulating a CPM signal by using a log-MAP (log Maximum a posteriori) algorithm, transmitting soft information obtained by demodulation to a CRC-Polar decoding module, and decoding by using a cyclic redundancy check-assisted serial cancellation list (CA-SCL) decoding algorithm at a decoding end, as described in the following steps S204 and S206.

Step S204: after receiving the CPM signal transmitted through the channel, demodulating the CPM signal by adopting a log-MAP algorithm.

In this step, after receiving the CPM signal transmitted through the channel, the continuous phase demodulator 121 in the signal receiving end 102 demodulates the CPM signal by using a log-MAP algorithm; the continuous phase demodulator 121 includes a matched filter bank and a trellis structure.

Specifically, a trellis structure and a filter bank demodulated by a receiving end are constructed based on the modulation parameters of the CPM signal. Before receiving signals, pre-initializing front and rear measurement, assuming the input signal prior information and the like, and decomposing the received CPM signals into superposition of a plurality of PAM signals by a matched filter bank; converting the symbols of the detected PAM signal into branch measurement through a trellis structure, finally combining and calculating the likelihood value of the symbols through iterative front and back measurement, finally converting the multi-system symbol likelihood value into the likelihood ratio of binary bits, and finally taking the likelihood ratio of the binary bits as the calculated soft information.

In the specific details of the CPM demodulation scheme related to the invention, a log-MAP algorithm and Laurent decomposition are adopted to demodulate CPM signals.

As shown in fig. 4, the matched filter bank based on multilevel lorentt Decomposition (source Decomposition) at the receiving end constructed based on CPM parameters is called as source matched filter bank for short, and the matched filter bank utilizes the source Decomposition principle of the CPM signal to decompose the CPM signal into superposition of several PAM (Pulse Amplitude Modulation) signals (Pulse signals).

Before demodulation, a trellis structure of the receiving end may be constructed according to the CPM parameters, and the following describes a structure of the receiving end and a construction manner of a state variable:

suppose that the state variable of the CPM demodulator at the nth time is defined as

Wherein,

estimated symbol representing state versus time i

Is indicative of the state of inclination of the optical element,

representative state corresponding accumulated phase

The sharp corner number on the symbol represents the inclined state, and the relation between the inclined symbol and the original symbol is

mod denotes the remainder operation, p denotes the denominator of the CPM modulation factor (conditioning factor h = q/p, q and p are a pair of coprime integers). D represents the precision parameter defined by the invention (1 ≦ D ≦ L). The number of states at one time is pM ^D-1 And (4) respectively.

The state variable transition process may be carried out using a state transition recursion:

(or is represented by

To represent I(s) _n ,s _n+1 ) The representation represents the state s from n time _n Go to state s _n+1 Corresponding to the estimated n time instants

Therefore, the state transition relation of the trellis structure of the receiving end can be constructed. Referring to fig. 5a, the state transition manner of the state variables of the trellis structure of the receiving end is described in the form of a finite state machine. Examples are: as shown in fig. 5b, when M =4 and d =1, the state transition trellis diagram represents.

The value range of the precision parameter D is an integer (including 1 and L) between 1 and the memory length L, which represents that the filter bank at the receiving end selects all Laurent pulses with the duration longer than L-D for receiving and matching, the approximate precision of the CPM signal at the receiving end is reflected, and the state variable of the corresponding demodulation end can also change along with the change of D.

As a preferred embodiment, in the technical scheme of the present invention, the precision parameter D may be selected according to an energy ratio of matched pulses in a filter bank:

under the condition that the modulation order, the modulation index, the memory length L and the pulse waveform of the CPM signal are determined, sequentially taking integer values from 1 to L for D, and sequentially calculating corresponding matched pulse energy ratio;

selecting a matching pulse energy ratio higher than a set energy ratio threshold value, and further taking the minimum value of the values of D corresponding to the selected matching pulse energy ratio as a finally determined precision parameter D;

the energy ratio of the matched pulses specifically refers to the energy ratio of the energy of the matched pulses in the Laurent matched filter bank to the energy of all the Laurent pulses.

Empirically, the energy share threshold may be set to 98.70%; examples are: as shown in table 1 below, it shows the energy ratio in the filter bank when different precision parameters D are selected (the number in the right-hand bracket of D in the table represents the number of all pulses whose duration is greater than L-D, i.e. the number of pulses corresponding to the filter bank) under different memory lengths L (the number in the right-hand bracket of L in the table represents the number of all pulses) when the modulation order M =4, the modulation index h is 1/4, and the waveform of the shaping pulse function is Raised Cosine (RC):

TABLE 1

D＝1(3)

D＝2(12)

D＝3(48)

D＝4(192)

D＝5(768)

D＝6(3072)

L＝1(3)

1

L＝2(12)

99.96％

1

L＝3(48)

98.66％

99.99％

1

L＝4(192)

94.12％

99.93％

100.00％

1

L＝5(768)

86.80％

99.30％

99.99％

100.00％

1

L＝6(3072)

78.01％

97.33％

99.90％

99.99％

100.00％

1

According to the selection rule of the precision parameter mentioned in the present invention, in the above example, when the memory length L =1 (the total number of pulses is 3), the precision parameter should be selected as D =1 (the number of pulses of the corresponding matched filter bank is 3); when the memory length is L =2 (the total number of pulses is 12), the precision parameter is selected to be D =1 (the number of pulses of the corresponding matched filter bank is 3); when the memory length is L =3 (the total number of pulses is 48), the precision parameter is selected to be D =2 (the number of pulses of the corresponding matched filter bank is 12); when the memory length is L =4 (the total number of pulses is 192), the precision parameter is selected to be D =2 (the number of pulses of the corresponding matched filter bank is 12); when the memory length is L =5 (the total number of pulses is 768), the precision parameter is selected to be D =2 (the number of pulses of the corresponding matched filter bank is 12); when the memory length is L =6 (total number of pulses is 3072), the precision parameter is selected to be D =3 (corresponding to a number of pulses of 48 matched filter bank).

In summary, the precision parameter D of the continuous phase modulation may be determined according to the modulation parameter of the continuous phase modulation and the set energy threshold; when the modulation order M or the memory length L is larger, compared with the conventional continuous phase modulation technique in which the precision parameter D is a fixed value, the method can avoid the decrease in error code performance caused by insufficient pulse energy occupation ratio due to the adoption of only the main component pulse matching (D = 1), and can also avoid the high complexity of matching (D = L) by adopting all the component pulses, wherein the complexity is mainly embodied in the state variable s of the receiving end _n The number of states and the number of matched pulses in the corresponding matched filter bank.

Since the CPM is known in the beginning and end states and the state transition is known, the continuous phase demodulator 121 in this step adopts the log-MAP algorithm to perform a specific process for detecting the CPM signal, as shown in fig. 6, which includes the following sub-steps:

substep S601: initializing;

in the sub-step, initializing a forward measurement A and a backward measurement B of the state variable of the trellis structure: due to the fact that in step S202 there is a precessionL symbols are added before and after the symbol making, so that the head state and the tail state are zero states, namely s = [0,0, … … 0]) Hence initialized Forward metric A at time 0 ₀ (s ₀ ＝[0,0,……，0])＝0、A ₀ (s ₀ ≠[0,0,……，0]) = - ∞, initialize backward metric B at end time _end (s _end ＝[0,0,……，0])＝0，B _end (s _end ≠[0,0,……，0])＝-∞。

Substep S602: calculating branch metrics, i.e. likelihood values of state transitions;

specifically, the MAP algorithm obtains the probability of the corresponding estimated symbol according to the transition probability of the state, and since the baseband signal is randomly generated, the probabilities are equal, and for M-ary, the prior probabilities of the input symbols of trellis structure are all p (α) _i ) =1/M, then the a posteriori probability of the estimated symbol at time n is equal to the likelihood of the state transition, i.e.:

Γ _n (s _n ,s _n+1 )＝logp(r _n (t)|s _n ,I(s _n ,s _n+1 ) Equation 4

In formula 4, Γ _n (s _n ,s _n+1 ) A branch metric expression representing n time, corresponding to n time state s _n State s to time n +1 _n+1 The transition probability of (2); i(s) _n ,s _n+1 ) The representation represents the state s from n time _n Go to state s _n+1 Corresponding to the estimated input symbols

It can also be said that conditions for state transition; p (r) _n (t)|s _n ,I(s _n ,s _n+1 ) Represents a known state transition (state s from time n) _n State s to time n +1 _n+1 ) In the process of (2), the received observation symbol r _n (t) probability.

Since maximizing the log-likelihood function equivalently maximizes the correlation value, i.e. for the estimated sequence

Has a correlation value of

Where r (t) is the received sequence,

the sequence being an estimated sequence

CPM Signal of time, (·) ^* It is indicated that the conjugate operation is taken,

indicating the operation of the real part. Approximate expression of Laurent decomposition of s (t)

Bringing into the above-mentioned correlation value

Expression, which can be further derived to obtain a correlation value based on Luartent pulse at n time, and is proportional to the expression gamma representing the branch metric at n time _n (s _n ,s _n+1 ) Namely:

in the formula 5, a _K,n Is the pseudo-symbol of the K-th Laurent pulse at time n, which is associated with the state transition(s) _n ,s _n+1 ) One-to-one correspondence is realized; { x _K,n Is the result of the K-th real pulse in the matched filter bank passed by CPM signal r (t) received by the receiving end, x _K,n ＝∫r(t)g _K (t-nT)dt，N _D The number of matched filters in the filter bank, i.e. the number of pulses of Laurent with a duration greater than L-D, N _D ＝(2 ^D-1 ) ^P (2 ^P -1)，P＝log ₂ And M. Since the Laurent pulses are all real pulses, no conjugate has to be taken.

For the K-th time of Laurent matched filter bank nPseudo-symbol a of a pulse _K,n For pulses with duration greater than L-D, the corresponding pseudo-symbol can be obtained by the following equations 6 and 7:

wherein, a _K,n Pseudo-symbol representing the K-th pulse of the Laurent matched filter bank in the nth symbol period, P = log ₂ M, M, D are the modulation order and precision parameter of the CPM signal, respectively, and an M-ary CPM modulation signal can be regarded as the result of multiplying P binary CPM signals (correspondingly, a multilevel Laurent pulse is the result of multiplying P binary CPM signal Laurent pulses; a pseudo symbol corresponding to a multilevel Laurent pulse is the result of multiplying P binary CPM signal pseudo symbols);

decomposing a corresponding k-th pseudo symbol at the nth time for the source of the l (l is more than or equal to 0 and less than or equal to P-1) binary CPM signal; according to a multilevel Laurent decomposition algorithm, a multilevel Laurent pulse set { g } _K (t) } is divided into Q ^P Group (Q = 2) ^L-1 ) For convenience, a mapping relation of (j, m) → K, i.e., the jth group (0. Ltoreq. J. Ltoreq.Q), is constructed ^P The mth pulse of-1) corresponds to the kth multilevel Laurent pulse, { d } _j,l Denotes the j (0. Ltoreq. J. Ltoreq.Q) ^P -1) the serial number of the l (l is more than or equal to 0 and less than or equal to P-1) binary Luartent pulse corresponding to the group of multi-system pulses;

denotes the j (0. Ltoreq. J. Ltoreq.Q) ^P -1) the m-th pulse of the group

All can be obtained by a multilevel Luartent decomposition algorithm; { gamma. } was prepared from a mixture of two or more of the above-mentioned compounds _n,l Denotes the corresponding estimated symbol of the l (0 ≦ l ≦ P-1) binary CPM signal in the nth symbol period, and the scaling relationship between it and the estimated symbol of the multilevel CPM signal is

{β _k,i The sequence after binary expansion is carried out on integer k according to a formula

To find, i represents the number of digits of the integer k after binary expansion.

Substep S603: based on the forward measurement, backward measurement and branch measurement (the posterior probability of pseudo symbols) of iterative computation, finally combining and computing the likelihood value of the multi-system code element;

specifically, the forward metric a is iteratively calculated according to equation 8 below _n (s _n )：

In the formula 8, the first and second groups of the compound,

and max ^* The operation can be iteratively applied using the Jacobian formula ln (e) ^x +e ^y )＝max(x,y)+ln(1+e ^-|x-y| ) Calculating; xi(s) denotes a state backward connected to the state, xi ^-1 (s) represents a state forward connection with the state.

Iteratively calculating the backward metric B according to equation 9 below _n (s _n )：

The log-likelihood values of the multilevel symbols are obtained according to the following equation 10:

max in the log-MAP algorithm may be used in cases where error performance is acceptable ^* The (-) operation is replaced by a Max (-) operation, so that the computational complexity can be greatly reduced, namely the Max-Log-MAP algorithm.

Substep S604: converting the likelihood probability of the multi-system symbol into a binary bit likelihood ratio to obtain demodulated soft information;

in particular, let a multilevel (oblique) sign

The corresponding binary bit sequence is

From likelihood values of symbols under multiple systems

Calculating the log-likelihood ratio of the binary bit of the corresponding binary symbol sequence to obtain a binary estimation symbol sequence as the demodulated soft information, as shown in formula 11:

in particular, when the modulation order is 2, the corresponding log-likelihood ratio is

Step S205: de-interleaving soft information obtained by adopting a log-MAP algorithm and Laurent decomposition;

in this step, the deinterleaver 122 in the signal receiving end 102 deinterleaves the soft information obtained by using the log-MAP algorithm and the Laurent decomposition to obtain LLR values of the binary bit streams of all the received sequences, that is, soft information of the received sequences

i =1,2, … … N, N is the length of the polar code.

The invention de-interleaves the soft information obtained by adopting a log-MAP algorithm and a Laurent decomposition and then uses the soft information as the input of a CRC-Polar code decoding end. And a cyclic redundancy check assisted serial cancellation list (CA-SCL, CRC-aid SCL) is adopted as a decoding algorithm at a decoding end. The decoding process of the CA-SCL algorithm is as described in step S206.

Step S206: and decoding the soft information subjected to de-interleaving by adopting a serial offset list (CA-SCL) algorithm assisted by cyclic redundancy check, and removing cyclic redundancy check bits to obtain a recovered information stream.

In this step, the decoder 123 in the receiving end 102 of the signal decodes the soft information that is deinterleaved by using the serial cancellation list assisted by the cyclic redundancy check: determining values of elements in the serial offset list according to a binary bit sequence in the soft information; and after obtaining a candidate path list in the serial offset list based on an SCL algorithm, performing CRC (cyclic redundancy check) on the candidate paths according to the sequence of path metrics from large to small to obtain a decoding result.

Specifically, according to the log-likelihood ratio of the binary bit sequence obtained in the above steps, the maximum number of candidate paths (the size of the list) is set to be L _list The decoding process is shown in fig. 7, and includes the following steps:

step S701: initializing;

in this step, first, L is defined ⁽ⁱ⁾ Writing a null sequence into an initial serial offset list for a candidate path set corresponding to the ith layer of an SCL decoding code tree, namely initializing L ⁽⁰⁾ = phi and initializes its corresponding path metric value to 0;

step S702: path expansion: determining a candidate path in the serial cancellation list according to a binary estimation symbol sequence in the soft information;

specifically, through the ith layer, the serial cancellation list L is paired ^(i-1 ) All candidate paths within

By adding the element v _i All candidate paths obtained by path expansion of =0 or 1 are recorded in the table L ⁽ⁱ⁾ Performing the following steps; wherein, { v _i }v _i E {0,1} corresponds to the binary estimated symbol sequence in the soft information obtained in step S604

In particular, if

Then corresponds to v _i =0; if it is

Then corresponds to v _i ＝1；

Represents from v ₁ To v _i-1 The sequence of (a);

recording all the obtained candidate paths in a table

Utilizing soft information of received sequences according to SCL algorithm

Calculate and record all of the lists

Path metric value of

Indicating a certain path from the root node of the SCL algorithm decoding code tree to the current i-th layer

Corresponding to a known received sequence of

Next, the estimated sequence is

Logarithm of the posterior probability of time.

Step S703: path competition;

specifically, if the list of successive cancellation L ⁽ⁱ⁾ Is less than or equal to the set number L _list If yes, all the candidate paths are reserved; otherwise, L with the maximum path metric value is reserved _list A candidate path, and from L ⁽ⁱ⁾ The remaining candidate paths are deleted.

Step S704: judging whether to terminate the judgment; specifically, if the path does not extend to the leaf node, i.e. if i < N, then i is increased by 1 and then the step S702 is executed; if i ≧ N, the termination decision goes to step S705. Wherein, N is the length of the polarization code.

Step S705: get list L ^(N) The candidate path having the largest path metric value.

Specifically, the list L is acquired ^(N) The candidate path with the maximum path metric value is used as decoding output; l is ^(N) The candidate path with the largest path metric value is

Preferably, the decoding assistance can also be performed through the cyclic redundancy check of the following step S706:

step S706: after obtaining the candidate path list by SCL algorithm, the list L is compared ^(N) Performing CRC on the medium candidate paths according to the order of the path metrics from large to small, and taking decoding sequences corresponding to the candidate paths passing the CRC as decoding results;

specifically, according to the sequence of the path metric from big to small, CRC check is carried out on the decoding sequence corresponding to the candidate path;

for the current candidate path for CRC, if the obtained check bits are all 0, the candidate path is a correct decoding path, and the corresponding decoding bit sequence is output to complete decoding;

if the obtained check bits are not all 0, the check fails; selecting the next candidate path with the largest path metric in the candidate path list to perform CRC (cyclic redundancy check) check until check bits are all 0, and outputting a corresponding decoding bit sequence;

if the check results of all the obtained candidate paths are not correct, outputting a decoding sequence corresponding to the candidate path with the largest path metric value in the list as a decoding result; the output results of CA-SCL and SCL decoding are the same.

The invention has been carried out with many experiments and simulations, the following details are given for the results:

the simulation parameters are shown in table 2 below:

TABLE 2

CPM parameters	Configuration of
		Modulation index	0.5
Modulation order	2
		Over-sampling rate	4
Normalized zone	0.3
		Data rate	1kHz
Channel with a plurality of channels	AWGN, signal-to-noise ratio SNR (dB)

Referring to fig. 8, in AWGN channel, when the code length is 512 and the code rate is 0.5, it can be seen that the error performance is significantly improved compared to the conventional RS code, when the memory length L =1 and the block error rate is 10 ^-4 And compared with the RS code, the technical scheme of the invention has 2.6dB performance gain.

In the technical scheme of the invention, the information flow is subjected to cyclic redundancy check coding and polarization coding; interleaving and mapping the code words obtained by coding; and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, for storing information may be implemented in any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device.

Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the idea of the invention, also features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.

In addition, well known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown within the provided figures for simplicity of illustration and discussion, and so as not to obscure the invention. Furthermore, devices may be shown in block diagram form in order to avoid obscuring the invention, and also in view of the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the present invention is to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the invention, it should be apparent to one skilled in the art that the invention can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.

While the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures, such as Dynamic RAM (DRAM), may use the discussed embodiments.

The embodiments of the invention are intended to embrace all such alternatives, modifications and variances that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements and the like that may be made without departing from the spirit and principles of the invention are intended to be included within the scope of the invention.

Claims (11)

1. An information transmission method of polar coding continuous phase modulation, comprising:

carrying out cyclic redundancy check coding and polarization coding on the information stream;

carrying out bit interleaving on the code words obtained by coding and mapping from binary system to multilevel system;

and sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

2. The method of claim 1, further comprising:

after receiving continuous phase modulation CPM signals transmitted through a channel, demodulating the CPM signals by adopting a log-MAP algorithm;

and after de-interleaving the soft information obtained by demodulation, decoding by adopting a serial offset list assisted by cyclic redundancy check.

3. The method according to claim 1, wherein the estimation of the reliability of the polarization channel in the polarization coding process specifically adopts a gaussian approximation algorithm, and specifically comprises:

the expectation of the log-likelihood ratio for each subchannel is calculated according to equation 1 below:

wherein,

represents the log-likelihood ratio of the ith polarized channel in the polarized code with length n in the channel polarization process,

represents the log-likelihood ratio of the channel hypothesis, which can be regarded as the log-likelihood ratio of the polarized channel with the length of 1;

approximating the above formula by the following approximate expression AGA-2

And estimating the reliability of the sub-channel according to the calculated log-likelihood ratio of the sub-channel.

4. The method according to claim 3, wherein the receiving the CPM signal transmitted through the channel and then demodulating the CPM signal by using a log-MAP algorithm specifically comprises:

decomposing a received CPM signal into superposition of a plurality of PAM pulses based on a Lauren matched filter bank;

determining an input symbol of a trellis structure according to the pseudo symbol of the PAM pulse, and further calculating the posterior probability of the pseudo symbol based on a state transition mode of a state variable of the trellis structure;

calculating the log-likelihood value of the input symbol according to the forward measurement and the backward measurement of the iterative state variable and the posterior probability of the pseudo symbol;

and converting the log-likelihood value of the input symbol into a binary bit likelihood ratio to obtain demodulated soft information.

5. The method of claim 4, wherein the pseudo-symbol is calculated according to the following equations 6 and 7:

wherein, a _K,n Pseudo-symbol representing the K-th pulse of the Laurent matched filter bank in the nth symbol period, P = log ₂ M, M, D are respectively the modulation order and the precision parameter of the CPM signal;

decomposing a corresponding kth pseudo symbol at the nth time for the source of the ith binary CPM signal; { d) _j,l Indicating the serial number of the l binary Luartent pulse corresponding to the j group of multilevel pulses;

representing the time delay of the l binary Laurent pulse corresponding to the m pulse of the j group;

all can be obtained by a multilevel Luartent decomposition algorithm; { gamma. } was prepared from a mixture of two or more of the above-mentioned compounds _n,l Denotes an estimated symbol of the l-th binary CPM signal in the n-th symbol period.

6. The method of claim 5, wherein D is determined according to the following method:

under the condition that the modulation order, the modulation index, the memory length L and the pulse waveform of the CPM signal are determined, sequentially taking integer values from 1 to L for D, and sequentially calculating corresponding matched pulse energy ratio;

selecting a matching pulse energy ratio higher than a set threshold value, and further taking the minimum value of values of D corresponding to the selected matching pulse energy ratio as a finally determined precision parameter D;

the energy ratio of the matched pulses specifically refers to the energy ratio of the energy of the matched pulses in the Laurent matched filter bank to the energy of all the Laurent pulses.

7. The method according to claim 4, wherein the deinterleaving the demodulated soft information and then decoding the demodulated soft information by using a cyclic redundancy check-assisted serial cancellation list includes:

determining a candidate path in the serial cancellation list according to a binary estimation symbol sequence in the soft information;

and after obtaining a candidate path list in the serial offset list based on an SCL algorithm, performing CRC (cyclic redundancy check) on the candidate paths according to the sequence of path metrics from large to small to obtain a decoding result.

8. A polar coded continuous phase modulated information transmission system, comprising:

the CRC cascade polarization code encoder is used for carrying out cyclic redundancy check encoding and polarization encoding on the information stream;

the bit interleaver is used for performing bit interleaving on the code words obtained by encoding and mapping binary system to multilevel system;

and the continuous phase modulator is used for sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

9. The system of claim 8, further comprising:

a continuous phase demodulator for demodulating the CPM signal transmitted through the channel by using a log-MAP algorithm;

a deinterleaver for deinterleaving the soft information obtained by the demodulation;

and the decoder is used for decoding the soft information after de-interleaving by adopting a serial offset list assisted by cyclic redundancy check.

10. A transmitting end of a signal, comprising:

the CRC cascade polarization code encoder is used for carrying out cyclic redundancy check encoding and polarization encoding on the information stream;

the bit interleaver is used for carrying out bit interleaving on the code words obtained by coding and mapping binary system to multilevel system;

and the continuous phase modulator is used for sending the CPM signal obtained by carrying out continuous phase modulation on the mapped symbol into a channel for transmission.

11. A receiving end of a signal, comprising:

a continuous phase demodulator for demodulating the CPM signal transmitted through the channel by using a log-MAP algorithm;

a deinterleaver for deinterleaving the soft information obtained by the demodulation;

and the decoder is used for decoding the soft information after de-interleaving by adopting a serial offset list assisted by cyclic redundancy check.

Images