CN102255697B - Distributed physical layer network coding modulation method for multiple access channel in wireless network - Google Patents

Distributed physical layer network coding modulation method for multiple access channel in wireless network Download PDF

Info

Publication number
CN102255697B
CN102255697B CN201110235422.3A CN201110235422A CN102255697B CN 102255697 B CN102255697 B CN 102255697B CN 201110235422 A CN201110235422 A CN 201110235422A CN 102255697 B CN102255697 B CN 102255697B
Authority
CN
China
Prior art keywords
mobile terminal
via node
cooperation
signal waveform
source mobile
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201110235422.3A
Other languages
Chinese (zh)
Other versions
CN102255697A (en
Inventor
林子怀
刘扬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin Membrane Technology Industrial Environment Protection & Technology Co Ltd
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN201110235422.3A priority Critical patent/CN102255697B/en
Publication of CN102255697A publication Critical patent/CN102255697A/en
Application granted granted Critical
Publication of CN102255697B publication Critical patent/CN102255697B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Error Detection And Correction (AREA)

Abstract

The invention discloses a distributed physical layer network coding modulation method for a multiple access channel in a wireless network. The method is implemented in a distributed physical layer network coding modulation way by utilizing a punctured convolutional coding digital phase modulation technology. In the method, a theoretical limit for a distributed system is deduced at the same time. A simulation result shows that the theoretical limit is in accordance with a simulation result value; and when a signal to noise ratio is increased, the simulation result is asymptotic to a theoretical limit value, thereby verifying the accuracy of a deduced theoretical value.

Description

The distributed physical-layer network coding modulator approach of multiple access channel in wireless network
Technical field
The present invention relates to cellular mobile network technology, be specifically related to be applicable to coding and the modulator approach of the distributed physical layer network of multiple access channel in wireless network.
Background technology
Distributed coding, as a kind of special chnnel coding cooperative communication network development mode, has attracted a large amount of researchers' concern.The structure theory of distributed coding has been applied to traditional channel coding, as document [B. Zhao and M. C. Valenti, " Distributed turbo codes:towards the capacity of the relay channel, " IEEE VTC ' 03-Fall, vol. 1, pp. 322 – 326, Oct. 2003.] disclosed Distributed T urbo code, document [S. Yiu, R. Schober and L. Lampe, " Distributed space-time block coding, " IEEE Trans. Commun., vol.54, no.7, pp. 1195 – 2006, Jul. 2006.] disclosed distributed space-time code, document [A. Chakrabarti, A. Baynast, A. Sabharwal and B. Aazhang, " Low density parity check codes for the relay channel, " IEEE JSAC, vol. 25, no. 2, pp. 280 – 291, Feb. 2007.] disclosed distributed low-density checksum (LDPC) code.Experimental result shows, above-mentioned distributed coding scheme can improve a little to the reliability of the radio channel transmission of point.
Above-mentioned distributed coding scheme is for small-scale clean culture junction network, and information is by list/multi-hop wireless junction network, from single source node, to single destination node, transmits.For example, under the scene shown in Fig. 1, a source mobile terminal (MT) is cooperated with other mobile terminals (MT), and by a via node (RN), to base station, (BS) transmits data.The traditional approach that transmits this information is by route, and the packet that receives is simply stored and forwarded to via node to destination.
Document [R. Ahlswede, N. Cai, S. Y. R. Li, and R.W. Yeung, " Network Information Flow, " IEEE Trans. Inform. Theory, vol. 46, no. 4, pp. 1204 – 1216, July 2000.] propose a kind of network code (NC) method and substituted routing algorithm.In network code, via node can be encoded the packet of receiving from a plurality of sources mobile terminal.Information after coding is sent to destination subsequently.Experimental result shows, compares with traditional routing algorithm, and network code can improve network capacity and throughput.Because research and design (DNCC) theory of this distributed network passage code just starts recently, many problems that run in the design and implementation process of distributed coding are not all also resolved.
Summary of the invention
The object of the present invention is to provide the distributed physical-layer network coding modulator approach of multiple access channel in a kind of brand-new wireless network.The method, based on puncture convolution code, adopts collaboration type uplink distributed coding code.The present invention has also analyzed the performance of system on this basis, derives the error rate theory analysis upper limit of the present invention.
Technical scheme of the present invention is as follows:
In wireless network, a distributed physical-layer network coding modulator approach for multiple access channel, comprises the steps:
1) source mobile terminal is encoded to transmitted codewords by punctured convolutional code device by packets of information, is then broadcast to via node and cooperation mobile terminal;
2) via node is the signal waveform from source mobile terminal that receives with from the signal waveform of cooperation mobile terminal, and separated according to mark space, then alternately series connection generates new signal waveform;
3) the new signal waveform forming is by separating mediation decoding at the decoder of via node, output needle is the estimation code word with the combined coding device of cooperation mobile terminal to source mobile terminal;
4) the recursive system convolution coder that described estimation code word is admitted to via node carries out network code and carries out memoryless modulation, is then sent to base station;
5) base station receives and from the signal waveform of via node transmitting, first does soft demodulation and then decode.
Described step 2), in, if via node can be correct the data decode of source mobile terminal, the mobile terminal that need not cooperate carries out cooperation transmission; If via node can not be correct the data decode of source mobile terminal, via node will send the instruction that requires transmission identical data to cooperation mobile terminal, cooperation mobile terminal is received after instruction after the data decode from source mobile terminal, utilize different punctured convolutional code devices to encode, then send to via node.
In described step 3), the decoder of via node is the decoder for the combined coding device of source mobile terminal and cooperation mobile terminal.
In described step 5), soft demodulation is to allow the signal waveform receiving through one group of waveform matched filter, the decoder that the value of exporting from this group filter is sent to base station carries out decoding, and the decoder of described base station is for the decoder that is distributed in the super grid coder of source mobile terminal and the cooperation punctured convolutional code device of terminal and the recursive system convolution coder of via node composition.
Useful technique effect of the present invention is:
The present invention is directed to the ul transmissions scheme that wireless network proposes a kind of collaboration type, the present invention adopts distributed physical-layer network coding modulation system, utilizes punctured convolutional code digital phase modulation technology to realize.The present invention has derived a theoretical limit for this distributed system simultaneously.Simulation result shows that this theoretical derivation limit meets simulation result value, when signal to noise ratio increases, and simulation result asymptotic theory limiting value, thereby the accuracy of the theoretical value that checking is derived.
Accompanying drawing explanation
Fig. 1 is with a source mobile terminal, a cooperation mobile terminal, the uplink system block diagram of a via node and a base station.
Fig. 2 is the simulation result of bit error rate of the present invention and the comparison diagram of the theory analysis upper limit.
Embodiment
Below by embodiment, invention is described further.
What the present invention considered is the ul transmissions system in radio honeycomb mobile network.Referring to Fig. 1, Fig. 1 shows with a source mobile terminal MT(A), a cooperation mobile terminal MT(B), a via node RN(R) and the uplink system block diagram of a base station BS (C).In this system, there is a source mobile terminal MT to send packet to base station BS by via node RN.Herein, the present invention is only usingd cellular mobile network as embodiment, and in fact the present invention can be applied in all wireless networks, such as wireless sense network, ad hoc(are point-to-point), WLAN (wireless local area network) etc., should not regard restriction as.
The cooperation transmission stage
Source mobile terminal can cooperate to carry out with other mobile terminals transfer of data.According to document [K.Doppler and M.Xiao, " Innovative concepts in peer-to-peer and network coding; " Tech. Rep. D1.3, CELTIC Telecommunication Solutions, 2009.], suppose to have direct link between source mobile terminal MT and cooperation mobile terminal MT, but there is no direct link between base station BS and mobile terminal.The transmission of a packet can be divided into two time slots.At first time slot, source mobile terminal MT first utilizes punctured convolutional code device to encode to information, then to via node RN and the packet cooperating after mobile terminal MT broadcast code.Via node RN and cooperation mobile terminal MT start decoding.When cooperation mobile terminal MT is correctly after the decoding of the information from source mobile terminal MT, recycle different punctured convolutional code devices this information of having decoded is encoded again.In the time of needs, the code word from cooperation mobile terminal, will send to via node RN at the second time slot.As shown in Figure 1, mobile terminal MT in source is used punctured convolutional code device C 1, and cooperation mobile terminal MT is used punctured convolutional code device C 2.
If via node RN can be correct the data decode of source mobile terminal MT, via node RN, just the modulation of the data message recompile from source mobile terminal MT after decoding, is then transmitted to base station BS so.In this case, the source mobile terminal MT cooperation transmission of mobile terminal MT that need not cooperate.If via node RN can not be correct the data decode of source mobile terminal MT, via node RN will send the instruction that requires transmission identical data to cooperation mobile terminal MT so, after cooperation mobile terminal MT receives instruction, the data from source mobile terminal MT can be encoded and is then sent to via node RN with different encoders.
Suppose to there is identical code check k/n at each encoder of source mobile terminal and cooperation mobile terminal.For a system with a source mobile terminal and k-1 cooperation mobile terminal, the unicode rate of k combined coding device will be 1/n so.The convolution coder of supposing all mobile terminals has identical as document [J. G. Proakis, Digital Communication, McGraw-Hill, New York, third edition, 1995.] described constraint length (constraint length).According to document [A. J. Viterbi and L. K. Omura, Principles of Digital Communication and Coding, McGraw-Hill, New York, 1979.], the convolution coder that is 1/n for code check, generator polynomial matrix
Figure 429852DEST_PATH_IMAGE001
, here
Figure 2011102354223100002DEST_PATH_IMAGE002
time parameter
Figure 2011102354223100002DEST_PATH_IMAGE004
belong to set { 0,1}.High code check puncture convolution code (PCC) can obtain by the Binary Convolutional Code of other (parent) 1/n of its upper level of puncture.By using (n * p) puncture matrix, P mat, wherein p is the puncture cycle, realizes the operation that punctures some coded identifications.Here establish s and be transmitted bit number altogether while the puncture cycle being p, PCC encoding rate is r=p/s.
The via node stage
Via node RN aloft mixes the signal waveform from source mobile terminal MT of receiving with from the signal waveform of cooperation mobile terminal MT, then combined decoding.This concrete air mix operating process is as follows: via node RN is separated according to mark space two signal waveforms, and then alternately series connection generates new waveform, and this new waveform is by separating mediation decoding at the decoder of via node RN.This method is different from traditional physical layer network situation the method for the signal waveform stack receiving.The decoder of via node RN for be the combined coding device C of source mobile terminal MT and cooperation mobile terminal MT 1and C 2.That this decoder is exported is the combined coding device C for source mobile terminal MT and cooperation mobile terminal MT 1and C 2estimation code word.A recursive system convolution (RSC) encoder that then this estimation code word is fed to via node RN carries out network code and carries out memoryless modulation.Here saidly by RSC, carry out network code and refer to two data streams that come from source mobile terminal MT and cooperation mobile terminal MT are mixed by RSC, form a new data flow, different with XOR (XOR) operation in legacy network coding, the present invention adopts RSC to carry out network code operation.Rsc encoder and memoryless modulator can be seen as a digital coding Continuous Phase Modulation (CPM) device.At via node RN, a vector is inputted and generated to rsc encoder using an output symbol from associating mobile terminal decoder as one, and memoryless modulator will produce the waveform of a transmission with this vector.
The base station stage
Code word after decoding is through recursive system convolution (RSC) encoder and carry out after memoryless modulation, is sent to base station BS (being also known as destination node).Base station BS receives first to be done soft demodulation (Soft demodulation) and then starts decoding from the signal waveform of relaying node RN transmitting, here said soft demodulation refers to and allows the signal waveform receiving through one group of waveform matched filter, from the value of this group filter output, be sent to decoder and carry out decoding, this decoder for be to be distributed in the super grid coder that source mobile terminal MT and the punctured convolutional code device of cooperation terminal MT and the rsc encoder of via node RN form.
Below will provide one in the situation that the channel condition of additive white Gaussian noise (AWGN) and employing maximum likelihood detect the systematic function theoretical upper limit of (MLSD) algorithm.
The theory analysis upper limit
At via node, be distributed in source mobile terminal and the cooperation punctured convolutional code device of mobile terminal and the RSC of via node and formed a super grid coder.When definition discrete time is j, the state of super grid coder
Figure 2096DEST_PATH_IMAGE005
for
Figure 2011102354223100002DEST_PATH_IMAGE006
, here
Figure 272671DEST_PATH_IMAGE007
with
Figure 2011102354223100002DEST_PATH_IMAGE008
when being illustrated respectively in discrete time and being j, the state of Joint Distribution convolution coder (JDCE) and the state of RSC.For a Joint Distribution convolution coder with m element, and one with the reasonable binary system total regression CPM system of irreducible modulation index h=K/P again, and the state of super trellis encoder adds up to
Figure 962409DEST_PATH_IMAGE009
.The input of associating terminal encoder has determined state conversion
Figure 2011102354223100002DEST_PATH_IMAGE010
.The incoming symbol that also have rsc encoder relevant to this transfer process and an average vector, this average vector is by the CPM waveform of transmission is obtained through a complex set of filter process, these filters match with transmitting.Foregoing can be referring to document [Z.Lin, Joint Source-Channel Coding using Trellis Coded CPM, Ph.D Thesis, Chalmers University of Technology, Gothenburg, Sweden, Jan. 2006, http://www.ce.chalmers.se/TCT.].
In the present invention, suppose that source node and cooperative node have identical transmitting power.If E bfor information bit power, N 0/ 2 is the power spectral density of two-sided additive white Gaussian noise.The wrong probability of the memoryless information source sequence of a puncture trellis coding CPM system, will follow following theorem 1.
Theorem 1: detect under (MLSD) and the condition of source information piece without endless at maximum likelihood, the upper limit of the bit mistake code check of the digital phase modulation CPM system of a distributed punctured convolutional code that has a discrete memoryless equiprobability distributed digital loop source sequence provides the upper limit by following formula:
Figure 2011102354223100002DEST_PATH_IMAGE012
（1）
Here
Figure 944850DEST_PATH_IMAGE013
minimum normalized square of Euclidean distance (NSED),
Figure 2011102354223100002DEST_PATH_IMAGE014
it is dummy variable, according to document [T. Aulin, " Symbol Error Probability Bounds for Coherently Viterbi Detected Continuous PhaseModulated Signals; " IEEE Trans. Commun., vol. COM-29, no. 11, pp. 1707 – 1715, Nov. 1981.], r is the encoding rate of PRCC trellis coding, r=p/s.Average transmission equation is:
Figure 19116DEST_PATH_IMAGE015
（2）
Here
Figure 2011102354223100002DEST_PATH_IMAGE016
that while being j since the time, state is
Figure 2011102354223100002DEST_PATH_IMAGE018
, when NSED is d 2, length is
Figure 563361DEST_PATH_IMAGE019
and have
Figure 2011102354223100002DEST_PATH_IMAGE020
the quantity of all error events of individual mismark.Q function is defined as follows:
Figure 432091DEST_PATH_IMAGE021
Below provide the proof of theorem 1.
Figure 2011102354223100002DEST_PATH_IMAGE022
being designated as from discrete time is total bit mistake that j starts, all error events produce.With a state arbitrarily, state s for example, , wherein S is state space.If
Figure 2011102354223100002DEST_PATH_IMAGE024
for mistake starts when the time is j, initial condition is s, and length is that a l normalizing square Euclidean distance is (NSED) d 2.
Figure 611848DEST_PATH_IMAGE020
for by error event the mismark bringing, initial condition and sequence pair
Figure 627394DEST_PATH_IMAGE025
can express this error event completely
Figure 682069DEST_PATH_IMAGE024
.Here
Figure 19509DEST_PATH_IMAGE027
with
Figure 2011102354223100002DEST_PATH_IMAGE028
it is the data sequence of two reconstructions.
For a punctured convolutional code and Full response continuous phase modulation (CPM) system, with an error event
Figure 2011102354223100002DEST_PATH_IMAGE030
relevant NSED d 2may be calculated
Figure 538346DEST_PATH_IMAGE031
(6)
Here T is-symbol _ time interval, and , here
Figure 2011102354223100002DEST_PATH_IMAGE034
it is poor phase state.
Figure 2011102354223100002DEST_PATH_IMAGE036
the operator of mould x, phase response, referring to document [J. B. Anderson, T. Aulin, and C. E. Sundberg, Digital Phase Modulation, Plenum Press, New York, 1986].
The desired value of the wrong bitrate being caused by error event when initial time j as shown in the formula:
Figure 2011102354223100002DEST_PATH_IMAGE038
(7)
Here with
Figure 2011102354223100002DEST_PATH_IMAGE040
two random vectors, their result space be the reconstruction signal sequence that likely starts when the time is j and, be when state is s, there is (NSED) d 2, length is the error event quantity of the total mismark of l while being τ, the desired value of formula (7) be since the time be all error events after j.
The encoding rate of supposing upper level (parent) convolution coder is k/ (k+1).Have like this 2 kindividual branch road enters and leaves each state.For the digital source sequence of a discrete memoryless equal probability, all take that state s starts, to have the sequence that length is l be that same probability is contingent, releases following formula:
Figure 2011102354223100002DEST_PATH_IMAGE042
(8)
Conditional probability
Figure 446479DEST_PATH_IMAGE043
on be limited to
Figure 2011102354223100002DEST_PATH_IMAGE044
,
Here Q equation is defined as
Figure 272483DEST_PATH_IMAGE045
,
Prove as follows: establish
Figure 2011102354223100002DEST_PATH_IMAGE046
be all in initial time, be j, state is that s, length are sequence
Figure 382839DEST_PATH_IMAGE047
set.Note
Figure 2011102354223100002DEST_PATH_IMAGE048
for set
Figure 200753DEST_PATH_IMAGE046
base.When
Figure 451737DEST_PATH_IMAGE049
time, in set
Figure 872354DEST_PATH_IMAGE046
in only have two sequences
Figure 407241DEST_PATH_IMAGE051
with
Figure 2011102354223100002DEST_PATH_IMAGE052
.
Figure 327880DEST_PATH_IMAGE052
decision region be half of signal space,
Figure 315428DEST_PATH_IMAGE053
provided accurately the error rate , d here 2to transmit
Figure 415102DEST_PATH_IMAGE055
with
Figure 2011102354223100002DEST_PATH_IMAGE056
between NSED.Therefore, , when
Figure 2011102354223100002DEST_PATH_IMAGE058
,
Figure 449365DEST_PATH_IMAGE052
determining area between be just less than half that determines space, draw thus
Figure 861892DEST_PATH_IMAGE059
Based on the above results and formula (8), consider all error events and use associating upper limit technology, according to document [A. J. Viterbi and L. K. Omura, Principles of Digital Communication and Coding, McGraw-Hill, New York, 1979.], can set
Figure 2011102354223100002DEST_PATH_IMAGE060
on be limited to:
Figure 765257DEST_PATH_IMAGE061
(9)
For further reducing the computational complexity of (9), it is noted that the CPM for a coding, some states are of equal value.Of equal value the meaning when two error events here
Figure 2011102354223100002DEST_PATH_IMAGE062
with when the time is j, start, state is simultaneously
Figure 2011102354223100002DEST_PATH_IMAGE064
with
Figure 74196DEST_PATH_IMAGE065
, respectively by identical sequence array
Figure 2011102354223100002DEST_PATH_IMAGE066
during generation, they are equal to.The grade is here both the mismark that error event has identical length, equal NSED and can produce equal number.If two states
Figure 787068DEST_PATH_IMAGE064
with
Figure 228545DEST_PATH_IMAGE065
be equal to,
Figure 530213DEST_PATH_IMAGE067
.
According to above proof, the state of all punctured convolutional code CPM, when the time is j, has punctured convolutional code device equal state, identical. in individual coding total regression CPM state, only have 2 mindividual state is unequal mutually.Therefore, (9) can be designated as again:
Figure 2011102354223100002DEST_PATH_IMAGE068
(10)
For an independence, with minute other discrete memoryless source sequence, encoder can start at any s munder lattice type state one of in individual.Because become when puncture is periodicity, also depend on time j and cycle variation.The state of a CPM coding when the time is j
Figure 283176DEST_PATH_IMAGE005
equal 2 mamong individual, the probability of one of different conditions is
(11)
Suppose that source sequence is endless, work as x, y>0 is used inequality, according to document [J. B. Anderson, T. Aulin, and C. E. Sundberg, Digital Phase Modulation, Plenum Press, New York, 1986.] , make simultaneously
Figure 564433DEST_PATH_IMAGE071
, the symbol error rate upper limit can be decided to be
Figure 2011102354223100002DEST_PATH_IMAGE072
(12)
Here p is the puncture cycle of mentioning at second portion, and j=0 refers to that puncture starts.
The left side of inequality (12) can be calculated by generating function.Will be designated as at different conditions
Figure 686290DEST_PATH_IMAGE018
under generating function
Figure 467295DEST_PATH_IMAGE075
. be shown below:
Figure DEST_PATH_IMAGE076
（13）
Wherein
Figure 150398DEST_PATH_IMAGE014
dummy variable [14].Substitution above formula, equation (12) can be designated as:
Figure 77902DEST_PATH_IMAGE077
（14）
The average generating function here
Figure DEST_PATH_IMAGE078
Figure 29809DEST_PATH_IMAGE079
（15）
Arrive this, theorem 1 proof finishes.
By the state diagram of application product, according to document [E. Biglieri, " High-level modulation and coding for nonlinear satellite channels; " IEEE Trans. Commun., vol. COM-32, pp. 616 – 626, May 1984.] and document [J. Shi and R. D.Wesel, " Efficient computation of trellis code generating function, " IEEE Trans. Commun., vol. 52, no. 2, pp. 219 – 227, Feb. 2004.], can obtain transmission equation .One is j in the time product state can be defined as
Figure DEST_PATH_IMAGE080
, here
Figure 738319DEST_PATH_IMAGE005
the encoding state of CPM coded system, it is decoded state.State conversion process
Figure DEST_PATH_IMAGE082
just be expressed as
Figure 801112DEST_PATH_IMAGE083
（3）
Figure 871836DEST_PATH_IMAGE085
with represent respectively mismark quantity and NSED. indicate NSED
Figure 613659DEST_PATH_IMAGE086
the quantity in path,
Figure 515756DEST_PATH_IMAGE085
it is the mismark in this state conversion.
For the CPM of puncture trellis coding, the error rate only depends on the output of punctured convolutional code device.In other words, it is independent of the state pair of CPM
Figure 996415DEST_PATH_IMAGE089
.In addition, according to document [J. B. Anderson, T. Aulin, and C. E. Sundberg, Digital Phase Modulation, Plenum Press, New York, 1986.], NSED d 2the state difference that only depends on CPM .Product state has just reduced like this.For the CPM system of total regression, the product state of minimizing is
Figure 164223DEST_PATH_IMAGE091
, here
Figure DEST_PATH_IMAGE092
it is phase difference state.All product states of CPM system like this with the CPM coding of total regression are exactly
Figure 834370DEST_PATH_IMAGE093
.
The product state of CPM coding can be divided into initial condition, transmission state and done state.The initial condition of a product state is defined as the beginning of error event, when done state is defined as the end of error event.The condition of initial condition and done state is respectively
Figure DEST_PATH_IMAGE094
with other states are transition status.
With represent from an initial condition
Figure 163217DEST_PATH_IMAGE018
when the time is j, one step is to the state-transition process of transmission state, and Bj represents the transition process to done state by transmission state, Cj be illustrated in the time while being j transmission state one step arrive the process of transmission state.
Figure 371475DEST_PATH_IMAGE097
expression is from an initial condition
Figure 982585DEST_PATH_IMAGE018
one step arrives the transition process of done state.According to document [J. B. Anderson, T. Aulin, and C. E. Sundberg, Digital Phase Modulation, Plenum Press, New York, 1986.], transmission equation can calculate with following formula:
Figure DEST_PATH_IMAGE098
（4）
Here 1 is a vector that all elements is 1, and I represents unit matrix.
Equation (1) can be expressed as again:
（5）
Here
Figure DEST_PATH_IMAGE100
From formula (5), can find out minimum NSED
Figure 753183DEST_PATH_IMAGE013
with
Figure DEST_PATH_IMAGE101
(have
Figure 82533DEST_PATH_IMAGE013
error event quantity) performance of the asymptotic symbol error rate of fixing system certainly, it is larger,
Figure 560099DEST_PATH_IMAGE101
less, the progressive symbol error rate performance of system is better.In the present invention, the design standard of best puncture matrix is according to whether producing maximum
Figure 819042DEST_PATH_IMAGE013
realize.
Simulation result
The present invention under Gauss's additive white noise passage emulation the system of a plurality of employing different distributions formulas codings.At first systematic, the encoder of source mobile terminal is that 1 parent nonsystematic convolutional encoding obtains by puncture code check, and the generator polynomial of this parent convolutional encoding is
Figure DEST_PATH_IMAGE103
.The generator polynomial that the encoder of cooperation mobile terminal is is 1 by puncture rate is
Figure DEST_PATH_IMAGE105
parent nonsystematic convolutional encoding obtain.Therefore,, from via node, the combined coding device of source node and cooperation mobile terminal is to be by the generator polynomial of 1/2 rate of of puncturing
Figure DEST_PATH_IMAGE107
parent nonsystematic convolutional code generate.Table 1 has been listed the difference puncture matrix Pmat of different system, and puncture matrix Pmat is by all possible puncture pattern of exhaustive search, and select can provide maximum least square Euclidean distance d 2 minpuncture matrix.In table 1, puncture matrix provides with octadic form, for example (3,5) 0the puncture matrix form of representative is [011; 101] 0 be wherein, the position of the symbol that is punctured accordingly.The punctured convolutional code of using in emulation is that the generator polynomial by 1/2 rate that punctures is
Figure 261787DEST_PATH_IMAGE107
with parent nonsystematic convolutional code generate.In table 1, also listed file names with the d of different system 2 min, and observe symbolic number N b, N bthe CPM system that refers to coding reaches the required observation symbol quantity of minimum euclidean distance, and foregoing is referring to document [J. B. Anderson, T. Aulin, and C. E. Sundberg, Digital Phase Modulation, Plenum Press, New York, 1986.].
Table 1
Figure DEST_PATH_IMAGE110
In emulation, the value of corresponding each signal to noise ratio, adopts the block of information that 100 length are 8000 bits to carry out emulation.Fig. 2 has shown simulation result and the theoretical upper limit of the bit error rate (BER) of distributed network code modulation system of the present invention.Can see that simulation result meets the theoretical upper limit of derivation, when signal to noise ratio increases, simulation result approaches theoretical upper limit gradually.
Fig. 2 is simultaneously displayed on identical E b/ N 0condition under, adopt the bit error rate performance of parent convolution code to be always better than the convolution code that punctures accordingly.For example,, at E b/ N 0in the time of=8, the BER performance of parent convolution code (rate is 1/2) is better than corresponding puncture convolution code.Again for example, for producing multinomial, be father's convolution code and the puncture convolution code of its corresponding 3/4 rate, at ordinate bit error rate, equal 10 -4time, relatively with the solid line (parent convolution code) of circle with the solid line (the puncture convolution code of 3/4 rate) of No. x, with the solid line (parent convolution code) of circle at BER=10 -4time SNR=8dB, and with the solid line (the puncture convolution code of 3/4 rate) of x symbol at BER=10 -4time SNR=12dB, the performance gain of parent convolution code is probably 4 decibels of left and right.
Above-described is only the preferred embodiment of the present invention, the invention is not restricted to above embodiment.Be appreciated that the oher improvements and changes that those skilled in the art directly derive or associate without departing from the basic idea of the present invention, within all should thinking and being included in protection scope of the present invention.

Claims (2)

1. a distributed physical-layer network coding modulator approach for multiple access channel in wireless network, is characterized in that, comprises the steps:
1) source mobile terminal is encoded to transmitted codewords by punctured convolutional code device by packets of information, is then broadcast to via node and cooperation mobile terminal;
2) via node is the signal waveform from source mobile terminal that receives with from the signal waveform of cooperation mobile terminal, and separated according to mark space, then alternately series connection generates new signal waveform;
If via node can be correct the data decode of source mobile terminal, the mobile terminal that need not cooperate carries out cooperation transmission; If via node can not be correct the data decode of source mobile terminal, via node will send the instruction that requires transmission identical data to cooperation mobile terminal, cooperation mobile terminal is received after instruction after the data decode from source mobile terminal, utilize different punctured convolutional code devices to encode, then send to via node;
3) the new signal waveform forming is by separating mediation decoding at the decoder of via node, output needle is the estimation code word with the combined coding device of cooperation mobile terminal to source mobile terminal;
4) the recursive system convolution coder that described estimation code word is admitted to via node carries out network code and carries out memoryless modulation, is then sent to base station;
5) base station receives and from the signal waveform of via node transmitting, first does soft demodulation and then decode.
2. the distributed physical-layer network coding modulator approach of multiple access channel in wireless network according to claim 1, it is characterized in that: in described step 5), soft demodulation is to allow the signal waveform receiving through one group of waveform matched filter, the decoder that the value of exporting from this group filter is sent to base station carries out decoding, and the decoder of described base station is for the decoder that is distributed in the super grid coder of source mobile terminal and the cooperation punctured convolutional code device of terminal and the recursive system convolution coder of via node composition.
CN201110235422.3A 2011-08-17 2011-08-17 Distributed physical layer network coding modulation method for multiple access channel in wireless network Active CN102255697B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201110235422.3A CN102255697B (en) 2011-08-17 2011-08-17 Distributed physical layer network coding modulation method for multiple access channel in wireless network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110235422.3A CN102255697B (en) 2011-08-17 2011-08-17 Distributed physical layer network coding modulation method for multiple access channel in wireless network

Publications (2)

Publication Number Publication Date
CN102255697A CN102255697A (en) 2011-11-23
CN102255697B true CN102255697B (en) 2014-04-30

Family

ID=44982704

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201110235422.3A Active CN102255697B (en) 2011-08-17 2011-08-17 Distributed physical layer network coding modulation method for multiple access channel in wireless network

Country Status (1)

Country Link
CN (1) CN102255697B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103580809A (en) * 2013-11-22 2014-02-12 北京科技大学 Multi-hop cooperative transmission method based on multi-source distributed network
CN107835064B (en) * 2017-10-19 2020-09-29 北京科技大学 Physical layer network coding transmission method based on 1-bit feedback and lattice division

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101814974A (en) * 2010-03-12 2010-08-25 重庆大学 Transmission method combined with network coding and channel decoding based on adaptive modulation
CN101867451A (en) * 2010-06-02 2010-10-20 北京邮电大学 Multi-user network coding communication method with high-speed parallel encoding and decoding structure

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU2001269125A1 (en) * 2000-07-17 2002-01-30 Koninklijke Philips Electronics N.V. Coding of a data stream

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101814974A (en) * 2010-03-12 2010-08-25 重庆大学 Transmission method combined with network coding and channel decoding based on adaptive modulation
CN101867451A (en) * 2010-06-02 2010-10-20 北京邮电大学 Multi-user network coding communication method with high-speed parallel encoding and decoding structure

Also Published As

Publication number Publication date
CN102255697A (en) 2011-11-23

Similar Documents

Publication Publication Date Title
CN101867451B (en) Multi-user network coding communication method with high-speed parallel encoding and decoding structure
CN101534146B (en) Data relaying apparatus, communication apparatus, and data relaying method
Vigato et al. Joint discovery in synchronous wireless networks
Zhan et al. A novel error correction mechanism for energy-efficient cyber-physical systems in smart building
He et al. A lower bound analysis of Hamming distortion for a binary CEO problem with joint source-channel coding
CN102724021A (en) Collaborative transmission method based on distributed interweaved and group encoding
Attar et al. Bit and Packet Error Rate evaluations for Half-Cycle stage cooperation on 6G wireless networks
Guo et al. Jointly optimized design of distributed Reed–Solomon codes by proper selection in relay
CN102255697B (en) Distributed physical layer network coding modulation method for multiple access channel in wireless network
Zhang et al. Soft network coding in wireless two-way relay channels
Hernaez et al. On the design of a novel joint network-channel coding scheme for the multiple access relay channel
Attar et al. Network coding hard and soft decision behavior over the physical payer using PUMTC
CN102244561A (en) Relay transmission method applied to multiple access relay network
CN102355323A (en) Non-rate LT coding-based method for distributed network channel coding of wireless sensor network
CN102291207B (en) Wireless distributed network channel coding modulation method
Ong et al. Optimal routing for decode-forward in cooperative wireless networks
Ma et al. Novel nested convolutional lattice codes for multi-way relaying systems over fading channels
Zhan et al. A turbo-BICM based scheme for joint network coding and channel coding
Shirvanimoghaddam et al. User cooperation via rateless coding
Del Ser et al. On combining distributed joint source-channel-network coding and turbo equalization in multiple access relay networks
Kim et al. Network error correction from matrix network coding
Tianyi et al. Research on pdcch channel in 5g nr system
Noh et al. Optimum cooperation of the cooperative coding scheme for frequency division half-duplex relay channels
Demirdogen et al. FEC driven network coding based pollution attack defense in cognitive radio networks
Khan et al. Enhanced XOR network coding scheme with OFDM for two-way relay networks

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20180606

Address after: 300000 Jimei Industrial Park, Xiqing economic and Technological Development Zone, Tianjin, 11A

Patentee after: Tianjin membrane technology industrial environment protection & Technology Co., Ltd.

Address before: 214000 Room 303, 8 storey building, 2188 industrial design garden, Taihu West Road, Wuxi, Jiangsu.

Co-patentee before: Liu Yang

Patentee before: Lin Zihuai