CN109327285B - Physical layer network coding method based on serial cascade CPM - Google Patents

Abstract

In a wireless two-way relay channel, two source node sending ends both adopt the same SCCPM framework, and at the MAC stage of the two-way relay, a relay node performs relay detection on two paths of superposed SCCPM signals to obtain an exclusive OR value of data sent by the two source nodes; and in the BC stage of the bidirectional relay, the relay node performs SCCPM modulation on the estimated value and broadcasts the modulated value to the source node, the source node performs iterative decoding on the received single-path SCCPM signal, and then the estimated value of the opposite side information is obtained through XOR operation with the self information, so that the bidirectional relay is completed. The invention provides a wireless bidirectional relay transmission model based on the joint channel coding, CPM modulation and PNC of a serial cascade structure, and simulation results show that the invention further improves the coding gain while giving full play to the advantages of the CPM technology.

Description

Physical layer network coding method based on serial cascade CPM

Technical Field

The invention belongs to the technical field of communication, relates to a wireless relay network and physical layer network coding, and particularly relates to a physical layer network coding method based on serial cascade CPM.

Background

In 2006, Zhang Shengli et al introduced the idea of Network Coding into a wireless two-way relay Network and proposed a Physical-layer Network Coding (PNC) concept based on a Decode-and-Forward (DF) protocol (document [ 1]]) The DF is also referred to as Denoise and Forward (DNF), and this scheme greatly improves the network throughput by using the superimposed signal received by the relay node. PNC-based transport protocolThe wireless two-way relay channel model of (1) is shown in figure 1, the source node S₁And S₂Information exchange is carried out, but the information exchange is not in the coverage range of each node, the information needs to be forwarded through the relay node R, and all the nodes are configured with single antennas and work in a half-duplex mode. According to the PNC protocol, the relay adopts a decoding forwarding mechanism to map two paths of source node information arriving at the same time into corresponding XOR information and then broadcasts the XOR information, so that only two stages are needed for completing one information exchange: a Multiple Access (MAC) phase and a Broadcast (BC) phase. The specific process is as follows: first time slot, i.e. MAC stage, source node S₁And S₂Simultaneously sending data a and b to a relay node R, and obtaining network coding data a ^ b by the R through a certain signal processing technology according to the received superposed signals; the second time slot, the BC phase, R broadcasts the data a ≦ b to the two source nodes, the source node S₁And S₂And after receiving a ^ b, carrying out exclusive OR operation on the received a ^ b and the self data to obtain the required data. Compared with the traditional relay scheme, the PNC technology is adopted to double the network throughput.

In a wireless communication system, transmission errors are inevitably introduced into information transmission due to the fading characteristics of a channel and the influence of various interferences and noises, and channel coding is an important means for realizing reliable data transmission in a communication channel, so that the reliability of transmission can be enhanced while the throughput of a wireless network is improved by combining physical layer network coding and channel coding. Document [2] proposes a simplified scheme for joint design of convolutional codes and PNC; document [3] proposes a physical layer network coding mechanism based on TCM codes (Trellis Coded Modulation), which obtains higher coding gain; documents [4-5] and [6-8] respectively apply Turbo codes and LDPC codes to PNC systems, and a relay joint decoding scheme is designed. However, at present, the technical solutions of joint channel coding and physical layer network coding basically have independent modulation modes, and only conventional linear modulation, such as PAM, PSK, QAM, and the like, is adopted. Continuous Phase Modulation (CPM) is a memory nonlinear constant envelope Modulation technique, and has many advantages such as high signal spectrum efficiency and strong anti-amplitude fading capability compared with a conventional linear Modulation mode. At present, relevant documents introduce a CPM technology into a PNC-based bidirectional relay channel, so that the spectrum efficiency and the power efficiency of a system are improved. Document [9] presents a physical layer network coding scheme using binary continuous phase frequency shift keying with a modulation index of an integer, continuous phase frequency shift keying CPFSK being a full-response CPM with a shaped pulse of a rectangle; documents [10-11] respectively research a PNC system relay coherent detection algorithm based on a CPFSK modulation mode under the conditions of an Additive White Gaussian Noise (AWGN) channel and a Rayleigh fading channel; documents [12-13] propose a relay incoherent multi-symbol detection method based on binary full response CPM (CPM with memory length of 1) in a PNC system. In one-way communication, document [14] indicates that a CPM signal has a trellis coding characteristic, and can be designed jointly with channel coding, so as to further improve coding gain, and the performance of the CPM signal is superior to that of a system in which the CPM signal and the channel coding are designed separately. At present, the CPM literature related to coding mainly relates to CPM of convolutional code coding, CPM of trellis coding, and CPM adopting a Turbo structure, wherein a design scheme of a Turbo iteration idea is adopted: serial Concatenated CPM (SCCPM) is considered to be a better solution for improving power and bandwidth efficiency in a dynamic multipath environment (documents [15-16 ]).

Reference to the literature

[1]S.Zhang,S.C.Liew,and P.Lam,“Hot topic:physical layer network coding,”in Proc.ACM MobiCom,Los Angeles,CA,USA,pp.358-365,Sep.2006.

[2]D.To and J.Choi,“Convolutional codes in two-way relay networks with physical-layer network coding,”IEEE Transactions on Wireless Communications,vol.9,no.9,pp.2724-2729,Sep.2010.

[3] "a joint design of channel coding and physical layer network coding based on TCM", journal of electronics and informatics, volume 33, phase 11, page 2594 and 2599, 2011.

[4]S.Lu,Y.Li,and J.Cheng,“Low-complexity turbo decoding scheme for two-way relay network,”in Proc.International Conference on Wireless Communications and Signal Processing,pp.1-5,2010.

[5]D.To and J.Choi,“Reduced-state decoding in two-way relay networks with physical-layer network coding,”in Proc.Information Theory Workshop(ITW),pp.1-5,2010.

[6]A.Tanc,T.Duman,and C.Tepedelenlioglu,“Design of LDPC codes for two-way relay systems with physical-layer network coding,”IEEE Commun.Lett.,vol.17,no.12,pp.2356-2359,Dec.2013.

[7]X.Wu,C.Zhao,and X.You,“Joint LDPC and physical-layer network coding for asynchronous bi-directional relaying,”IEEE Journal on Selected Areas in Communications,vol.31,no.8,pp.1446-1454,Aug.2013.

[8]S.Sharifi,A.K.Tanc and T.M.Duman,“LDPC code design for the two user Gaussian multiple access channel”,IEEE Trans.Wireless Commun.,vol.15,no.4,pp.2833–2844,Apr.2016.

[9]M.C.Valenti,D.Torrieri,and T.Ferrett,“Noncoherent physical-layer network coding using binary CPFSK modulation,”in Proc.IEEE Military Commun.Conf.,(Boston,MA),pp.1-7,Oct.2009.

[10]N.Sha,Y.Gao,X.Yi,W.Li,and W.Yang,“Joint CPFSK modulation and physical-layer network coding in two-way relay channels,”IEICE Trans.Fundamentals,vol.E97-A,no.4,pp.1021-1023,Apr.2014.

[11]N.Sha,Y.Gao,M.Guo,S.Wang,and K.Xu,“Physical-layer network coding for fading bidirectional relay channels with M-CPFSK,”IEICE Trans.Fundamentals,vol.E101-A,no.6,pp.974-977,Jun.2018.

[12] Dang Xiaoyu, Liu Mei Tong, Li Bao Long, Li Qiang, noncoherent multi-symbol detection of continuous phase modulation signals in physical layer network coding, electronic and information newspaper, volume 38, phase 4, page 877-.

[13]X.Dang,Z.Liu,B.Li,and X.Yu,“Noncoherent multiple-symbol detector of binary CPFSK in physical-layer network coding,”IEEE Communications letters,vol.20,no.1,pp.81-84,Jan.2016.

[14]B.Rimoldi,“A decomposition approach to CPM,”IEEE Transactions on Information Theory,vol.34,no.3,pp.260-270,Mar.1988.

[15]P.Moqvist and T.Aulin,“Serially concatenated continuous phase modulation with iterative decoding,”IEEE Trans.on Communications,vol.49,no.11,pp.1901-1915,Nov.2001.

[16]T.Carter,S.Kim,and M.Johnson,“High throughput,power and spectrally efficient communications in dynamic multipath environments,”IEEE Military Communications Conference,vol.22,no.1,pp.61-66,2003.

[17]M.Xiao and T.Aulin,“Serially concatenated continuous phase modulation with symbol interleavers:performances,properties and design principles,”in Proc.IEEE Global Telecommunications Conference,pp.179-183,2004.

[18]S.Benedetto,D.Divsalar,G.Montorsi,and F.Pollara,“Serial concatenation of interleaved codes:performance analysis,design,and iterative decoding,”IEEE Trans.Inform.Theory,vol.44,no.3,pp.909-926,May 1998.

[19]J.Hagenauer,and P.Hoeher,“A Viterbi algorithm with soft-decoding outputs and its applications,”in Proc.IEEE Global Communications Conference,pp.1680-1686,1989.

[20]J.Hagenauer,P.Robertson,and L.Papke,“Iterative(Turbo)decoding of systematic convolutional codes with MAP and SOVA algorithms,”in Proc.ITG Conf on Source and Channel Codes,pp.1-9,1994.

Disclosure of Invention

The purpose of the invention is: the serial cascade continuous phase modulation SCCPM mode in the unidirectional communication is introduced into a wireless bidirectional relay channel, and meanwhile, a wireless bidirectional relay transmission model combining the advantages of SCCPM and PNC technologies is designed by utilizing the idea of physical layer network coding PNC, so that the aims of improving the spectrum efficiency, the power efficiency and the throughput of a wireless bidirectional relay system are fulfilled. Under the background, the technical problem to be solved in an important way is how the relay node performs PNC detection on the received two paths of superposed SCCPM signals to obtain an exclusive or value of data sent by two source nodes. The invention carries out deep research in the direction and provides a relay detection method for designing combined channel decoding, demodulation and PNC mapping at a relay node by utilizing an iterative decoding idea.

The technical scheme of the invention is as follows: in a wireless bidirectional relay channel, a relay node designs a relay detection scheme by using an iterative decoding idea to realize the joint design of channel coding, CPM modulation and physical layer network coding, wherein two source node sending ends all adopt the same SCCPM framework: the SCCPM signal is sent out by sequentially connecting a binary linear convolution encoder, an M system mapper, an M system symbol interleaver and an M system CPM modulator;

in the MAC stage of the bidirectional relay, the relay node performs relay detection on the two paths of superposed SCCPM signals to obtain an exclusive OR value of data sent by two source nodes: two SCCPM signals are approximately regarded as a combined serial cascade convolutional code SCCC for relay detection, wherein a convolutional encoder in the SCCPM signals is regarded as an outer code in an SCCC framework, CPM modulation and PNC serve as inner codes in the SCCC framework, a relay node adopts a decoding structure of the SCCC to carry out soft-input soft-output SISO iterative decoding on a received mixed signal r (t), in the decoding structure, a decoding algorithm of an SISO inner decoder utilizes a Rimo decomposition model of the CPM signals, two combined inclined phase state grid graphs represent the overlapped CPM signals, an output value of the SISO inner decoder is obtained through a soft-output Viterbi algorithm, the outer decoder is a standard SISO convolutional decoder, and after the last iterative decoding is finished, hard judgment is carried out on the output of the outer decoder, and then an estimated value of an exclusive OR value of data sent by a source node can be obtained;

in the BC stage of the bidirectional relay, the relay node performs binary convolutional coding, M-system mapping, M-system symbol interleaving and M-system CPM modulation on the estimated value, which are the same as those of the source node, and then broadcasts the result to the source node, the source node performs iterative decoding on the received single-path SCCPM signal, and then obtains the estimated value of the opposite side information through XOR operation with the information of the source node, so that the bidirectional relay is completed.

The invention researches the physical layer network coding scheme of the relay node, and the PNC detection scheme of the relay node is different because of different signal source signal patterns. The main work of the invention is as follows:

1. the invention provides a method for applying an SCCPM mode in unidirectional communication to a wireless bidirectional relay channel adopting physical layer network coding, and designs a physical layer network coding system (PNC-SCCPM) based on the SCCPM. Due to the utilization of the technical advantages of SCCPM and the adoption of high-order CPM modulation, power efficiency and bandwidth efficiency can be improved compared to the existing coded modulated PNC systems (documents [2-13 ]).

2. The invention designs a PNC detection scheme of the relay node. The relay detection scheme is a difficulty of system design of the invention, and the relay node needs to carry out joint detection on two paths of mixed SCCPM signals to obtain an exclusive OR value of data sent by two source nodes, so the SCCPM signal detection scheme of unidirectional communication cannot be directly adopted. Therefore, through analysis and research, the equivalent model of joint transmission of two signal source node SCCPM signals in an MAC stage is obtained firstly, so that a decoding structure and an iterative decoding process of a relay node are designed, and finally a decoding algorithm of an inner decoder in the decoding structure is designed in a key mode.

At present, the technical scheme related to joint channel coding and physical layer network coding basically has independent modulation modes, and only conventional linear modulation is adopted, but the introduction of the CPM modulation mode enables the joint design of the channel coding, the modulation and the physical layer network coding to be possible, and simultaneously, the coding characteristic of the CPM can be fully exerted, so that the spectral efficiency, the power efficiency and the error performance of a PNC system are further improved. The invention provides a wireless bidirectional relay transmission model based on the combination of channel coding, CPM modulation and PNC of a serial cascade structure, designs a decoding structure of a relay node and a CPM + PNC decoding algorithm, and simulation results show that the invention further improves coding gain while fully playing the advantages of the CPM technology.

Drawings

Fig. 1 is a physical layer network coding scheme in a bidirectional relaying channel.

FIG. 2 is a PNC-SCCPM system model of the present invention.

FIG. 3 shows the MAC stage signal transmission model and equivalent transformation according to the present invention.

FIG. 4 is a structure of iterative decoding of a relay node of the PNC-SCCPM system of the present invention.

Fig. 5 is a state trellis diagram of CPM signal according to an embodiment of the present invention (M-4, h-1/4, and L-1).

FIG. 6 is a diagram of a joint state transition relationship according to an embodiment of the present invention.

Fig. 7 shows a joint state transition path of two CPM signals (M-4, h-1/4, and L-1) according to an embodiment of the present invention.

FIG. 8 is a comparison of error code performance of PNC-SCCPM and PNC-CPM of the present invention.

FIG. 9 shows the error code performance comparison between PNC-SCCPM and PNC-CC-CPM according to the present invention.

Detailed Description

The invention introduces single-path SCCPM into a wireless bidirectional relay channel, provides a physical layer network coding system (PNC-SCCPM) based on the SCCPM, and designs a relay detection scheme combining channel decoding, demodulation and PNC mapping at a relay node by using an iterative decoding idea, thereby realizing the combined design of channel coding, CPM modulation and physical layer network coding. In order to improve the spectrum utilization of the system and effectively resist channel burst errors, multilevel CPM modulation and a symbol interleaver are adopted in the SCCPM architecture (document [17 ]). Different from the detection of the SCCPM signals of unidirectional communication, the relay node needs to carry out joint detection on the two superposed SCCPM signals to obtain the exclusive OR value of data sent by two source nodes. For the purpose, for two SCCPM architectures at the MAC stage, linear characteristics of linear Convolutional coding, symbol natural mapping and interleaving operations relative to xor operations are utilized to approximately view the two SCCPM architectures as a joint Serial Concatenated Convolutional Code (SCCC), where the Convolutional coder serves as an outer Code in the SCCC architecture, and the CPM modulation and the PNC serve as inner codes in the SCCC architecture, so that the relay node can perform Soft Input Soft Output (SISO) iterative decoding by using a decoding structure similar to the SCCC. In the designed decoding structure, the outer decoder can be regarded as a standard SISO convolutional decoder, and the related decoding algorithm can be referred to documents [17-18 ]; the decoding algorithm of the SISO inner decoder is the core problem to be solved, a Rimo decomposition model (document [14]) of a CPM signal is utilized, two paths of combined inclined phase state grid graphs are used for representing the overlapped CPM signal, and the output value of the SISO inner decoder is obtained through a soft output Viterbi algorithm. The invention researches a design scheme for effectively combining the channel coding, the CPM modulation and the physical layer network coding from the perspective of combining coding and decoding, and further improves the spectrum efficiency, the power efficiency and the error performance of the wireless bidirectional relay system.

Considering a wireless bidirectional relay channel, fig. 2 shows a physical layer network coding system model based on serial concatenated CPM. Two source nodes S₁And S₂The distance of the system exceeds a reliable communication range, information interaction is required by means of a relay node R, three nodes are provided with single antennas and work in a half-duplex mode, the same SCCPM framework, namely a binary linear convolutional encoder, an M-system mapper, an M-system symbol interleaver and an M-system CPM modulator, is adopted by the transmitting end, the system is supposed to be completely synchronous, the signal transmitting power is equal, and the channel is an AWGN channel.

In the MAC phase, let b be assumed_i＝{b_i,0,b_i,1,b_i,2… as source S_iI ∈ {1,2} of a bit information sequence, the element b of which_i,nE {0,1}, n-0, 1,2, … are statistically independent and equally distributed with equal probability. Obtaining binary code word sequence c after binary linear convolution coder_i＝{c_i,0,c_i,1,c_i,2… } and then becomes a symbol sequence d after passing through the M-ary mapper_i＝{d_i,0,d_i,1,d_i,2… }, wherein the element d_i,nE {0,1,2, …, M-1}, where a natural mapping mechanism is employed. d_iAfter passing through M-system symbol interleaver, becomes symbol information sequence u_i＝{u_i,0,u_i,1,u_i,2… }, wherein the element u_i,_nBelongs to {0,1,2, …, M-1}, and then is modulated by M-system CPM to obtain a signal s_i(t) of (d). Source node S₁And S₂Simultaneous transmission of s₁(t) and s₂(t) to the relay node R, the signal received by R can be represented as

r(t)＝s₁(t)+s₂(t)+n_R(t) (1)

Wherein n is_R(t) AWGN term at Relay R, mean 0, variance σ²Bilateral power spectral density of N₀/2. Under the PNC protocol, the relay node needs to detect the received mixed signal to obtain the xor form of the source information, that is, the relay node needs to detect the received mixed signal to obtain the xor form of the source information

Defining a bit sequence m ═ m₀,m₁,m₂… }, wherein the element m_nE {0,1} is two source bit data b₁,_nAnd b_2,nIs mathematically defined as

For convenience of expression, m is represented herein as

I.e. the element in m is b₁And b₂And carrying out XOR operation on the elements at the corresponding positions to obtain a result. The relay R carries out PNC detection on the received mixed signal R (t) so as to obtain an estimated value of m

Namely, it is

In the BC stage, the relay node R detects the output bit sequence

Carrying out binary convolution coding, M-system mapping, M-system symbol interleaving and M-system CPM modulation which are the same as those of the source node, and then broadcasting the binary convolution coding, the M-system mapping, the M-system symbol interleaving and the M-system CPM modulation to the source node S_iConventional iterative SCCPM decoding schemes are applied to a received one-way SCCPM signal (see documents [17-18]]) Then the pair can be obtained

Is estimated value of

Then obtaining an estimate of the other party's information by an XOR operation with the own information, i.e.

The relay detection scheme design of the present invention is explained below.

According to the analysis, in the whole information interaction process, how the relay node detects the bidirectional relay system combining channel coding, CPM modulation and PNC to obtain the XOR information of the two information sources in the MAC stage

Is a key problem to be solved, and the following studies are focused.

1. Equivalent model for MAC stage signal transmission

Defining C as a mapping function of the convolutional encoder, i.e. C_i＝C(b_i) I ∈ {1,2 }; defining F as the mapping function of the M-ary mapper, i.e. d_i＝F(c_i) (ii) a Defining Π as the operating function of the M-ary symbol interleaver, i.e. u_i＝Π(d_i) Then enter the symbol information sequence u of the CPM modulator_iCan be expressed as

u_i＝Π(F(C(b_i))) (2)

Source node S_iTransmitted bit information sequence b_iCan be expressed as

b_i＝C^-1(F^-1(Π^-1(u_i))) (3)

Wherein, C^-1Is the inverse function of C, F^-1Is an inverse function of F, Π^-1Which is the inverse function of pi. By utilizing the linear characteristics of the C function, the F function and the pi function relative to the exclusive-OR operation, the method can further obtain

And

in relation to each other, i.e.

Therefore, two paths of code modulated SCCPM signals can be considered jointly, namely, two paths of code modulated SCCPM architectures are approximately considered as one path of joint SCCC architecture, wherein a binary convolution encoder serves as an outer code in the SCCC architecture, and an input bit information sequence is

CPM modulation and PNC serve as inner codes in SCCC architecture, and input symbol information sequence is

An equivalent relation diagram of joint transmission of two source node signals at the MAC stage is obtained as shown in fig. 3.

2. Relay SISO iterative decoding structure design

The two-path signal joint transmission equivalent model given in figure 3 can be regarded as an SCCC architecture,

an input bit information sequence of an outer encoder (i.e. a convolutional encoder),

is the input symbol sequence of the inner encoder (CPM + PNC encoder). In the MAC stage, the relay node is tasked with obtaining an input bit information sequence from a received signal r (t)

Therefore, according to the equivalent model, the relay node can adopt a decoding structure similar to SCCC to carry out SISO iterative decoding to obtain

Is estimated value of

The corresponding decoding structure is shown in fig. 4, where I represents input and O represents output; the CPM + PNC decoder is used as SISO inner decoder in SCCC decoding structure for detecting received signal r (t)

A log-likelihood ratio of; the convolutional code decoder is used as SISO outer decoder in SCCC decoding architecture and aims to obtain

The log likelihood ratio of (c).

The specific detection process is as follows: the inputs of the SISO intra decoder (CPM + PNC decoder) are a received signal r (t) and symbol information

Is a priori known to

And at the first iteration

Is set to 0. SISO inner decoder output updated symbol information

Log likelihood ratio of

Is obtained after symbol de-interleaving

Log likelihood ratio of

And as input to a SISO outer decoder, and

is a priori knowledge log likelihood ratio

Due to b_iAre statistically independently equally distributed and equally probable, and thus

The element probabilities in (1) are also equal, i.e.

The SISO outer decoder outputs updated bit information through a standard SISO convolutional code decoding algorithm

And code character number

Log-likelihood ratio of (i.e. of)

And

the interleaved symbols are fed to SISO inner decoder as input symbols

Is a priori known to

This completes one decoding. And repeating the above process to realize iterative decoding. Finally output to SISO outer decoder

Hard decision is carried out to obtain

Is estimated value of

3. CPM + PNC decoding algorithm design

It can be seen that the core of the whole decoding process is the SISO decoding module, the decoding algorithm needs to be designed by the coding structure, since the outer decoder is a standard convolutional decoder, the related SISO convolutional code decoding algorithm can be referred to in the references [17-18]]Therefore, the following mainly analyzes the decoding algorithm of the SISO inner decoder, referred to as CPM + PNC decoding algorithm, in order to obtain symbol information

Log likelihood ratio of

According to the document [14]]Given a tilted phase model (also called Rimo decomposition model) of CPM signal, for an input M-ary symbol sequence u ═ u { (u) }₀,u₁,…,u_n,…}，u_nE {0,1, …, M-1}, and the tilt phase ψ (t, u) of the output CPM signal after CPM modulation can be expressed as

Wherein, T is CPM symbol period, T is tau + nT, and is not less than 0 and not more than tau<T; l is the memory length; q (-) is the integral of the phase-shaped pulse of duration LT; w (τ) is an independent item independent of data; h is the modulation index and is assumed rationalThe number h can be expressed as K/P, where K and P are relatively prime positive integers, where P is assumed to be M, so that the occurrence of error events with a weight of 1 in the SCCPM system can be avoided, because for a CPM signal received through an AWGN channel, an error event with a weight of 1 has a great influence on the performance of the serial cascade system (see document [15 ])]). It can be seen from equation (6) that the CPM signal state is determined by the phase state at time t ═ nT

P and associated State { u_n-1,…,u_n-L+1Co-decisional, usually denoted as σ_n＝[u_n-1,…,u_n-L+1,v_n]. Fig. 5 shows a CPM signal state trellis diagram when M is 4, h is 1/4, and L is 1.

Under the condition of not considering noise, the signal r (t) received by the PNC-SCCPM system relay is two paths of CPM signals s₁(t,u₁) And s₂(t,u₂) And (3) superposition. For two-path superposed CPM signals, we can represent with a joint state trellis diagram. The joint state transition relationship between the nT ≦ T ≦ (n +1) T interval is shown in FIG. 6, where (σ ≦ T ≦ n +1)_1,n,σ_2,n) Two paths of CPM signals s at the moment of representing t-nT₁(t,u₁) And s₂(t,u₂) In (c), (d) combined state (u)_1,n,u_2,n) A joint input symbol representing time t-nT, (σ)_1,n+1,σ_2,n+1) And (n +1) represents the joint state of the two CPM signals at the time T. Obviously, in the joint input of symbols (u)_1,n,u_2,n) Under the action, the joint state of the two paths of superposed signals is represented by (sigma)_1,n,σ_2,n) To (sigma)_1,n+1,σ_2,n+1) And (5) transferring.

Assuming that the CPM signals (M4, h 1/4, and L1) output by the two source nodes have an initial phase of zero, the joint state transition paths of the two CPM signals are shown in fig. 7 for a set of jointly input symbol sequences { (1,0), (1,1), (1,2), (1,3), (2,0), (2,1), (2,2), (2,3) }.

In order to simplify the expression, the above explanations are combined

Then a priori information at the input of the SISO inner decoder

Abbreviated as Λ (x; I) for decoding the output symbol information

The log-likelihood ratio of (a) can be abbreviated as Λ (x; O). Taking the detection length as N, x ═ x₀,x₁,…,x_N-1}, then the elements therein

N is equal to {0,1, … N-1 }. The CPM + PNC decoding algorithm of SISO inner decoder is to obtain the log-likelihood ratio Λ (x; O) ═ Λ (x) of x₀；O),Λ(x₁；O),…,Λ(x_N-1(ii) a O) }. To calculate the element Λ (x) therein_n(ii) a O), it is necessary to observe the received signal for the time interval 0. ltoreq. t.ltoreq.NT and evaluate the conditional probability P (r (t) x) of all possible XOR information sequences x. For M-ary CPM, x has M^NA combination, wherein one combination can be expressed as

j∈{0,1,…,M^N-1}, then log-likelihood ratio Λ (x)_n(ii) a O) is represented by:

wherein k ∈ {1,2, …, M-1}, P (x)^(j)(ii) a I) Is x^(j)Can be obtained from Λ (x; I), and the element Λ (x) in the first decoding_n(ii) a I) 0, when decoding subsequently, Λ (x; I) the values of (a) are output by the SISO outer decoder and provided after symbol interleaving. Conditional probability P (r (t) | x^(j)) Expressed as:

wherein the content of the first and second substances,

and

are each u₁And u₂Possible values of (2). It is obvious that we can use the previously analyzed joint state trellis diagram of two-path CPM signal through the soft output Viterbi algorithm (documents [19-20 ]]) To obtain Λ (x)_n(ii) a O), thereby obtaining Λ (x; o) that is

Finally, the effect of the scheme of the invention is verified through error code performance simulation.

And simulating the relay error code performance of a physical layer network coding system (PNC-SCCPM) based on serial concatenated CPM designed by the method under the AWGN channel condition. In the simulation, CPM modulation with M-4, h-1/4, and L-1 is used, and Bit Error Rate (BER) of relay detection is used as a performance analysis indicator, and the abscissa is bit signal-to-noise ratio (SNR).

1. Comparison with PNC-CPM System

In order to embody the error code performance advantage of the PNC-SCCPM system provided by the invention, the error code performance is compared with that of an uncoded CPM-based physical layer network coding system (PNC-CPM), wherein the PNC-SCCPM system adopts a (13,17) convolutional code with a code rate of 1/2, and the symbol interleaving length is 1024. Because the PNC-SCCPM system adopts an iterative decoding mode to optimize the error code performance, the iteration times are respectively 3 and 6 in the simulation to verify the influence of the iteration times on the system performance, and the simulation result is shown in FIG. 8. Obviously, under the condition of small signal-to-noise ratio (SNR <1.5dB), the relay error code performance of the PNC-SCCPM system introducing the channel coding is slightly inferior to that of the PNC-CPM, but the error probability of the PNC-SCCPM is sharply reduced along with the increase of the signal-to-noise ratio and is far superior to that of the PNC-CPM. In addition, the error code performance of PNC-SCCPM is obviously improved along with the increase of the iteration number.

2. Compared with the PNC-CPM system based on the convolutional coding

The PNC-SCCPM system provided by the invention utilizes the coding characteristic of CPM to jointly design the channel coding, CPM modulation and PNC, and in order to embody the advantages of the joint coding and decoding of the three, the PNC-SCCPM system only provides the system error code performance comparison of the joint channel coding and the PNC (expressed by PNC-CC-CPM). In the PNC-CC-CPM scheme, the source node S_iFor input information sequence b_iPerforming convolutional coding and mapping, performing CPM modulation, performing CPM joint demodulation and demapping on the received superposed signals by the relay, and performing convolutional code decoding once to obtain the exclusive or value of the original information

The whole process considers CPM modulation and demodulation independently, and is not designed jointly with channel coding. In the simulation, both systems adopt (13,17) convolutional codes with code rate of 1/2, PNC-SCCPM relay decoding is iterated for 6 times, in addition, in order to reflect the influence of interleaving length on PNC-SCCPM performance, the interleaving length of symbols is respectively 256 and 1024, and the simulation result is shown in FIG. 9. Obviously, the PNC-SCCPM system jointly designed by the channel coding, the CPM modulation and the PNC further obtains coding gain, improves the error code performance, and obviously improves the performance along with the increase of the interleaving length.

Claims (5)

1. A physical layer network coding method based on serial cascade CPM is characterized in that in a wireless bidirectional relay channel, a relay node designs a relay detection scheme by utilizing an iterative decoding idea to realize the joint design of channel coding, CPM modulation and physical layer network coding, wherein two source node sending ends all adopt the same SCCPM framework: the SCCPM signal is sent out by sequentially connecting a binary linear convolution encoder, an M system mapper, an M system symbol interleaver and an M system CPM modulator;

in the MAC stage of the bidirectional relay, the relay node performs relay detection on the two paths of superposed SCCPM signals to obtain an exclusive OR value of data sent by two source nodes: two SCCPM signals are approximately regarded as a combined serial cascade convolutional code SCCC for relay detection, wherein a convolutional encoder in the SCCPM signals is regarded as an outer code in an SCCC framework, CPM modulation and PNC serve as inner codes in the SCCC framework, a relay node adopts a decoding structure of the SCCC to carry out soft-input soft-output SISO iterative decoding on a received mixed signal r (t), in the decoding structure, a decoding algorithm of an SISO inner decoder utilizes a Rimo decomposition model of the CPM signals, two combined inclined phase state grid graphs represent the overlapped CPM signals, an output value of the SISO inner decoder is obtained through a soft-output Viterbi algorithm, the outer decoder is a standard SISO convolutional decoder, and after the last iterative decoding is finished, hard judgment is carried out on the output of the outer decoder, and then an estimated value of an exclusive OR value of data sent by a source node can be obtained;

in the BC stage of the bidirectional relay, the relay node performs binary convolutional coding, M-system mapping, M-system symbol interleaving and M-system CPM modulation on the estimated value, which are the same as those of the source node, and then broadcasts the result to the source node, the source node performs iterative decoding on the received single-path SCCPM signal, and then obtains the estimated value of the opposite side information through XOR operation with the information of the source node, so that the bidirectional relay is completed.

2. The method as claimed in claim 1, wherein the CPM + PNC decoder is used as SISO inner decoder in SCCC decoding structure to obtain symbol information

Log likelihood ratio of

u_iRepresenting the sequence of symbol information, u, entering the CPM modulator_i＝{u_i,0,u_i,1,u_i,2,…}，i∈{1,2}，

Is the exclusive or value of the corresponding symbol information sequence of the two superimposed SCCPM signals,

order to

I denotes input, O denotes output, SISO inner decoder input prior information

Abbreviated Λ (x; I), log-likelihood ratio of decoded output

Abbreviated as Λ (x; O), with detection length N, CPM symbol period T, x ═ x₀,x₁,…,x_N-1}，

Observe the received signal in the time interval of 0 ≦ t ≦ NT and evaluate the conditional probability P (r (t) | x) of all possible XOR information sequences x, x having M for M-ary CPM modulation^NA seed combination of

j∈{0,1,…,M^N-1}, then log-likelihood ratio Λ (x)_n(ii) a O) is:

wherein k ∈ {1,2, …, M-1}, P (x)^(j)(ii) a I) Is x^(j)Is obtained from Λ (x; I), and the element Λ (x) in the first decoding_n(ii) a I) 0, when decoding subsequently, Λ (x; I) the value of (c) is provided after the output of SISO outer decoder and symbol interleaving, the conditional probability P (r (t) x^(j)) Comprises the following steps:

wherein the content of the first and second substances,

and

are each u₁And u₂The possible values of (a) to (b),

for possible values of two CPM signals, σ²Representing variance, obtaining Lambda (x) through a soft output Viterbi algorithm by a combined state grid diagram of two paths of CPM signals_n(ii) a O), thereby obtaining Λ (x; o) that is

3. The physical layer network coding method based on serial concatenated CPM as claimed in claim 1, wherein the MAC stage and BC stage of the bidirectional relay are specifically:

in the MAC stage, let b_i＝{b_i,0,b_i,1,b_i,2… as a source node S_iI ∈ {1,2} of a bit information sequence, the element b of which_i,nE {0,1}, n is 0,1,2, …, statistically independent and equally distributed and equally probable, and a binary code word sequence c is obtained after passing through a binary linear convolution encoder_i＝{c_i,0,c_i,1,c_i,2… } and then becomes a symbol sequence d after passing through the M-ary mapper_i＝{d_i,0,d_i,1,d_i,2… }, wherein the element d_i,nE {0,1,2, …, M-1}, where a natural mapping mechanism, d, is employed_iAfter passing through M-system symbol interleaver, becomes symbol information sequence u_i＝{u_i,0,u_i,1,u_i,2… }, wherein the element u_i,nBelongs to {0,1,2, …, M-1}, and then is subjected to M-system CPMModulating to obtain a signal s_i(t), Source node S₁And S₂Simultaneous transmission of s₁(t) and s₂(t) to the relay node R, the signal received by R is represented as

r(t)＝s₁(t)+s₂(t)+n_R(t) (1)

Wherein n is_R(t) is the AWGN term at the relay node R, the mean is 0, and the variance is σ²Bilateral power spectral density of N₀And/2, under the PNC protocol, the relay node detects the received mixed signal to obtain the exclusive or form of the source information, namely

Defining a bit sequence m ═ m₀,m₁,m₂,…,m_n… }, wherein the element m_nE {0,1} is two source bit data b_1,nAnd b_2,nIs mathematically defined as

Denote m as

The relay node R carries out PNC detection on the received mixed signal R (t), thereby obtaining an estimated value of m

Namely, it is

In the BC stage, the relay node R detects the output bit sequence

Carrying out binary convolution coding, M-system mapping, M-system symbol interleaving and M-system CPM modulation which are the same as those of the source node, and then broadcasting the binary convolution coding, the M-system mapping, the M-system symbol interleaving and the M-system CPM modulation to the source node S_iUsing conventional SC for received single SCCPM signalThe CPM iterative decoding scheme can obtain pairs

Is estimated value of

Then obtaining the estimation value of the other source node information of the bidirectional relay through the XOR operation with the self information, namely

4. A physical layer network coding method based on serial concatenated CPM as claimed in claim 1,2 or 3, wherein the sending of SCCPM signals by the source node is specifically:

defining C as the mapping function of the convolutional encoder to obtain a binary codeword sequence C_i＝C(b_i)，i∈{1,2}，b_i＝{b_i,0,b_i,1,b_i,2… as a source node S_iThe bit information sequence of (a); defining F as mapping function of M-system mapper, i.e. symbol sequence d obtained after mapping_i＝F(c_i) (ii) a Defining pi as the operating function of the M-ary symbol interleaver, d_iAfter passing through a symbol interleaver, becomes a symbol information sequence u_i＝Π(d_i) Then enter the symbol information sequence u of the CPM modulator_iExpressed as:

u_i＝Π(F(C(b_i))) (2)

then the source node S_iTransmitted bit information sequence b_iIs composed of

b_i＝C^-1(F^-1(Π^-1(u_i))) (3)

Wherein, C^-1Is the inverse function of C, F^-1Is an inverse function of F, Π^-1Obtaining an inverse function of pi by using linear characteristics of the C function, the F function and the pi function relative to an exclusive-or operation

And

in relation to each other, i.e.

Two SCCPM coding modulation architectures are approximately regarded as one combined SCCC architecture: bit information sequence

Obtained by a convolution encoder

The convolutional encoder is used as an outer code in an SCCC framework and then obtained by an M-ary mapper

Through a symbol interleaver

Finally, CPM modulation and PNC serve as inner codes in SCCC architecture, and the input symbol information sequence is

Outputting a joint signal s of the source nodes₁(t)+s₂And (t), adding the combined signal and the AWGN term at the relay node R, namely obtaining the mixed signal received by the relay node.

5. Physical layer network coding based on serially concatenated CPM according to claim 1 or 2The method is characterized in that SISO iterative decoding specifically comprises the following steps: u. of_iFor symbol information sequences entering a CPM modulator, b_iFor the source node bit information sequence, i ∈ {1,2}, b_iObtaining binary code word sequence c after binary linear convolution coder_iThen, the symbol sequence is converted into a symbol sequence d after passing through an M-system mapper_iThe CPM + PNC decoder is used as SISO inner decoder in SCCC decoding structure, and the received signal r (t) is detected to obtain

Log likelihood ratio of

The convolutional code decoder is used as SISO outer decoder in SCCC decoding structure to obtain

Log likelihood ratio of

I denotes an input, O denotes an output,

the input of SISO inner decoder is received signal r (t) and sign information

Is a priori known to

Set to 0 at the first iteration, the SISO inner decoder outputs updated symbol information

Log likelihood ratio of

Is obtained after symbol de-interleaving

Log likelihood ratio of

And

is a priori knowledge log likelihood ratio

As input to the SISO outer decoder, the SISO outer decoder outputs updated bit information

And code character number

Log likelihood ratio of

And

the interleaved symbols are fed to SISO inner decoder as input symbols

Is a priori known to

Thus, one-time decoding is completed, iterative decoding is realized by repeating the above process, and finally output to SISO outer decoder

Hard decision is carried out to obtain

Is estimated value of

Images