[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN103763234A - Information transmission method based on multi-way TPM signals - Google Patents

Information transmission method based on multi-way TPM signals Download PDF

Info

Publication number
CN103763234A
CN103763234A CN201410060120.0A CN201410060120A CN103763234A CN 103763234 A CN103763234 A CN 103763234A CN 201410060120 A CN201410060120 A CN 201410060120A CN 103763234 A CN103763234 A CN 103763234A
Authority
CN
China
Prior art keywords
signal
tpm
road
phase
signals
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.)
Pending
Application number
CN201410060120.0A
Other languages
Chinese (zh)
Inventor
王红星
赵培洪
毛忠阳
陈昭男
卢卫
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.)
Individual
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 CN201410060120.0A priority Critical patent/CN103763234A/en
Publication of CN103763234A publication Critical patent/CN103763234A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Digital Transmission Methods That Use Modulated Carrier Waves (AREA)

Abstract

The invention discloses an information transmission method based on multi-way TPM signals, and belongs to the field of transmission methods of digital communication. In the information transmission method based on the multi-way TPM signals, with regard to the multi-way TPM transmission signals, time domain features of the TPM signals are extracted according to the ICA technology, and emission signals are recovered through detection. Through the ICA technology, the transmitted multi-way TPM signals can be received and then analyzed and processed so that phase jump characters of the TPM signals can be extracted. Information carried in the TPM signals is included in a short-time phase jump, the characters can be captured through ICA time domain filtering, and accordingly the TPM signals can be recovered by setting a detection threshold without a frequency domain. By means of the information transmission method based on the multi-way TPM signals, a receiver structure of a TPM communication system is simplified, binary modulation information transmission performance is improved, and multi-way signal transmission of the TPM communication system is effectively achieved.

Description

A kind of information transferring method based on multi-channel Time phase modulated signal
Technical field
The invention belongs to the technical field of digital communication and transmission, specifically in multi-channel Time phase-modulation communication system, adopt a kind of information transferring method of independent component analysis technology.
Background technology
American scholar H.R.Walker has proposed super narrowband modulation method (UNB) in late 1980s, this modulation technique for effectively solve communication system intermediate frequency spectrum utilance problem opened up new thinking (H. R. Walker. Attain high bandwidth efficiency using VMSK modulation[J]. Microwaves & RF Magazine, pp. 173-185, Dec. 1997.).The super narrowband modulation method of Walker of take is basis, time phase modulator approach (Time-Phase Modulation, TPM), phase modulator approach during abbreviation, to utilize rectangular pulse to decline fast or rising edge characteristic is carried out a kind of carrier modulation technology of phase-modulation, the starting point of this modulation technique is: utilize rectangular pulse to decline fast or rising edge characteristic changes carrier phase fast, thereby avoid because phase place slowly changes the spread spectrum problem of bringing, be that carrier wave only changes constantly in SPA sudden phase anomalies, all the other times keep sinusoidal wave form.
Time phase modulation signal utilize phase hit moment point carry information, if a plurality of different phase hit moment point are set in the situation that frequency is identical, can realize multi-user to the sharing of frequency, during with single channel, phase modulation signal is compared, and can greatly improve the band efficiency of system.
At present, the design of phase modulation signal frequency domain filter while being limited by, scholars to time phase Modulation Signal Detection Method research all based on single channel, carry out.The people such as Wu Lenan propose to adopt geometrical characteristic filtering device to carry out filtering (Wu Lenan to time limit modulation signal, the geometrical characteristic filtering method of time-frequency-aliasing signal, Chinese invention patent, publication number: CN1877990A, 2006.12.13), in low signal-to-noise ratio situation, to detect performance unsatisfactory for the method; King's red magnitude studied based on Cyclic Spectrum characteristic and time-frequency distributions time phase modulation signal detection demodulating algorithm, but all just explored these class methods to time phase modulation signal demodulation feasibility, demodulation performance is not optimized, transmission problem (Wang Hongxing, Wang Hongli, the Mao Zhongyang of phase modulation signal simultaneously while also not considering multichannel, extended binary phase shift keying demodulating algorithm based on Cyclic Spectrum characteristic, electric wave science journal, 2010,25 (5): 434-439; Wang Hongxing, Wang Hongli, Mao Zhongyang, the EBPSK demodulating algorithm based on time-frequency distributions is analyzed, Jilin University's journal (engineering version), 2011,41 (5): 1491-1496).
In sum, from the spectral characteristic of TPM signal, if will according to time the signal bandwidth of modulating mutually come designing filter to acquire a certain degree of difficulty, meanwhile, early stage for time phase modulation signal the research of detection demodulation method be mainly confined to single channel transmission.The present invention mainly considers the communication problem of multichannel TPM signal, receiving terminal adopts independent component analysis technology (Independent Component Analysis, ICA) the SPA sudden phase anomalies feature of extracting TPM signal is come, thereby realizes the transmission of multichannel TPM signal.
Summary of the invention
For time the phase modulated filter design problem that exists, the object of the invention is to propose a kind of information transferring method based on multichannel TPM signal, phase hit feature when the method can effectively be obtained multichannel in phase modulation signal, then by Threshold detection, adjudicate code element " 0 " or " 1 " of transmitting, phase modulation signal while finally recovering multichannel.
For achieving the above object, basic ideas of the present invention are:
A plurality of users launch separate binary digital signal simultaneously, through out-of-date, after modulation, launch mutually separately, at receiving terminal, a plurality of transducers are set and receive signal, number of sensors should be no less than the time phase modulation signal way of transmitting, the observation signal that each transducer obtains is all aforementioned linear hybrid of phase modulation signal when a plurality of, observation signal is carried out to independent component analysis, thereby by all SPA sudden phase anomalies feature extractions of TPM signal out, realize the demodulation to multichannel TPM signal.
According to above-mentioned thinking, the specific implementation step of the information transferring method based on multichannel TPM signal is:
1) transmitting terminal produces nthe binary digital information that road is separate, nfor positive integer, and
Figure 444391DEST_PATH_IMAGE002
, when being carried out respectively, each road binary message modulates mutually, and its basic principle is, corresponding cycle of each binary data (be positive integer, and
Figure 127548DEST_PATH_IMAGE006
) the continuous sine wave of phase place, when sending " 1 ", at certain hour
Figure 29645DEST_PATH_IMAGE008
(
Figure 821889DEST_PATH_IMAGE008
be less than or equal to sinusoidal wave period) interior sine wave signal generation phase hit
Figure 114330DEST_PATH_IMAGE010
, when sending " 0 ", former sinusoidal signal does not change; The signal of not going the same way can have identical carrier frequency
Figure 33745DEST_PATH_IMAGE012
, symbol time
Figure 541081DEST_PATH_IMAGE014
and phase hit
Figure 192642DEST_PATH_IMAGE016
, also can be different, but the phase hit duration of the signal of not going the same way
Figure 300275DEST_PATH_IMAGE018
with phase hit initial time must have a difference at least, the
Figure 555862DEST_PATH_IMAGE022
the modulation signal on road is:
，
Figure 786303DEST_PATH_IMAGE026
(1)
Wherein,
Figure 781941DEST_PATH_IMAGE028
for the modulation waveform of binary data 0 correspondence,
Figure 762404DEST_PATH_IMAGE030
for the modulation waveform of binary data 1 correspondence, with
Figure 385464DEST_PATH_IMAGE030
expression formula be respectively:
Figure 285341DEST_PATH_IMAGE034
Wherein,
Figure 511923DEST_PATH_IMAGE014
be
Figure 753549DEST_PATH_IMAGE022
the element duration of road signal, be the carrier frequency of road signal,
Figure 148255DEST_PATH_IMAGE035
be
Figure 923182DEST_PATH_IMAGE022
phase hit when road signal sends " 1 ",
Figure 142942DEST_PATH_IMAGE036
be
Figure 640919DEST_PATH_IMAGE022
phase hit zero hour when road signal sends " 1 ",
Figure 209304DEST_PATH_IMAGE018
be phase hit duration when road signal sends " 1 ", each road modulated signal
Figure 705366DEST_PATH_IMAGE038
(
Figure 933216DEST_PATH_IMAGE039
) through antenna transmission, go out respectively;
2) receiving terminal setting mthe individual transducer that is positioned at diverse location,
Figure 921769DEST_PATH_IMAGE041
, the
Figure 235070DEST_PATH_IMAGE022
the signal that individual transducer receives is
Figure 295168DEST_PATH_IMAGE043
( j=1,2 ..., m),
Figure 502158DEST_PATH_IMAGE043
by each Lu Shixiang modulation signal mix and form,
Figure 478521DEST_PATH_IMAGE047
, .Thereby the ICA model of being set up by TPM signal and the unknown mixed coefficint of observation signal, transmitting is as follows:
Figure 403807DEST_PATH_IMAGE051
(2)
Dui Ge road signal carries out discrete sampling processing, and the ICA model shown in formula (2) can be reduced to the form of matrix:
Figure 501207DEST_PATH_IMAGE053
(3)
Wherein, ,
Figure 50055DEST_PATH_IMAGE057
,
Figure 293954DEST_PATH_IMAGE059
.
3) next need the observation signal to receiving xprocess, that is to say the ICA model based on multichannel TPM signal is analyzed, by hybrid matrix coefficient
Figure 749206DEST_PATH_IMAGE061
estimate, and original signal the most at last sestimate, this process is exactly independent component analysis technology, in other words ICA time-domain filtering.The SPA sudden phase anomalies feature of phase modulation signal while extracting by employing ICA time-domain filtering algorithm, this feature is that the form strengthening by amplitude embodies.These algorithms comprise the non-Gauss's method of maximization, Maximum Likelihood Estimation, the method for estimation of minimization mutual information, the ICA method of estimation based on tensor etc., and the present invention adopts first two method, is introduced below.
, non-Gauss's method maximizes.
The non-Gauss's algorithm that maximizes is that in ICA method of estimation, application is more a kind of, and the TPM communication system analysis that therefore the present invention is based on ICA mainly adopts the non-Gauss's algorithm of maximization to realize.Its basic thought comes from central-limit theorem: under certain condition, the distribution of a plurality of independently stochastic variable sums trends towards Gaussian Profile.Or undemanding saying: the distribution that two independent random variable sums form than any one in two original stochastic variables closer to Gaussian Profile.
In fact, the mixed variable that more levels off to Gaussian Profile can be regarded original isolated component as and through the rotation of certain angle, obtains respectively.Therefore, ICA estimates to carry out reverse consideration, the independent element that namely we will seek can be summed up as: searching makes to mix rear distribution and has the corresponding direction of rotation problem of maximum non-Gauss, each local maximum provides an independent element, after obtaining an independent element, then continue remaining variable to adopt the non-Gauss's method of maximization.
For seeking maximum non-Gauss's direction of rotation, must first to know how non-Gauss measures." kurtosis " is one of measure of non-Gauss.Kurtosis is the tired amount of a kind of quadravalence Gauss.Kurtosis is defined as:
Figure 462079DEST_PATH_IMAGE063
(4)
If suppose in above formula that variable has been zero-mean, and variance
Figure 90506DEST_PATH_IMAGE065
,
Figure 126595DEST_PATH_IMAGE067
, it is a kind of normalized form of quadravalence distance.For gaussian variable y, its quadravalence is apart from equaling
Figure 64333DEST_PATH_IMAGE069
, so the kurtosis of gaussian variable is zero.Most of non-Gaussian random variable, kurtosis value is zero; Kurtosis is more bordering on null value, and Gauss is stronger.Therefore,, for obeying the independent element of non-Gaussian Profile before mixing, make direction that the absolute value of kurtosis gets maximum just corresponding to the independent element before mixing.
While having several obvious very large measured value in observation sample, may cause kurtosis value to become very large, and these measured values may be mistake or irrelevant with problem.Therefore, kurtosis may be to open country value and responsive." negentropy " is another measure of non-Gauss.Random vector for observation
Figure 248190DEST_PATH_IMAGE071
, negentropy
Figure 617991DEST_PATH_IMAGE073
calculation expression be:
(5)
In practice, for the absolute value of maximize kurtosis or negentropy, can be from certain initial weighing vector w(setting arbitrarily) starts, according to available sample value
Figure 446587DEST_PATH_IMAGE077
, calculating is sent as an envoy to kurtosis absolute value increase the fastest direction, then by vector wforward this direction to.This thinking can realize with gradient method and expansion thereof.
Figure 274920DEST_PATH_IMAGE079
the gradient of kurtosis absolute value can calculate according to following formula:
Figure 347919DEST_PATH_IMAGE081
(6)
For the data of albefaction,
Figure 128924DEST_PATH_IMAGE083
.Suppose weighing vector wbe positioned at unit ball
Figure 287373DEST_PATH_IMAGE085
upper, so the renewal amount of each iterative process of weighing vector is:
Figure 998977DEST_PATH_IMAGE087
(7)
Wherein, in formula (7)
Figure 441328DEST_PATH_IMAGE089
refer to the vector on the left side
Figure 642503DEST_PATH_IMAGE091
have identical direction with the right gradient vector, but exist a positive scalar factor can adjust step-length, this coefficient is designated as
Figure 960352DEST_PATH_IMAGE093
.Obtaining new weighing vector
Figure 288696DEST_PATH_IMAGE095
after, also to carry out normalizing standardization to it:
Figure 336286DEST_PATH_IMAGE097
(8)
According to above analysis, the step of the non-Gauss's method that obtains maximizing is:
The 1st step is to observation data xcarrying out centralization, to make its average be 0;
Then the 2nd step carries out albefaction, obtains z.Albefaction process:
1. ask xcovariance
Figure 646045DEST_PATH_IMAGE099
;
2. to covariance carry out Eigenvalues Decomposition [D, E]=eig (
Figure 762774DEST_PATH_IMAGE101
), D is the diagonal matrix consisting of characteristic value, the descending arrangement of characteristic value
Figure 878498DEST_PATH_IMAGE103
, E is characteristic of correspondence vector;
3. try to achieve albefaction matrix:
Figure 531327DEST_PATH_IMAGE105
;
4. albefaction is vectorial:
Figure 11987DEST_PATH_IMAGE107
.
An initialization weighing vector with unit norm of the random selection of the 3rd step w;
The 4th step utilizes the gradient direction of kurtosis absolute value to weighing vector wupgrade, ,
Figure 535427DEST_PATH_IMAGE111
;
The 5th step is to the weighing vector obtaining wbe normalized,
Figure 495293DEST_PATH_IMAGE113
;
If the weight vector before and after the 6th step one time iteration meets
Figure 209171DEST_PATH_IMAGE115
, think kurtosis value convergence, proceed the 7th step, wherein,
Figure 801957DEST_PATH_IMAGE117
be a minimum, generally get otherwise,, get back to step 4 and proceed next iteration.
The 7th step by step after the convergence of the kurtosis value in rapid 6, estimates to obtain an isolated component
Figure 385440DEST_PATH_IMAGE079
, to observation data now xcarry out principal component analysis, corresponding row in deduct and estimate the isolated component obtain , obtaining observation data, the characteristic vector of the cross-correlation matrix of calculating, leaves out eigenvalue of maximum characteristic of correspondence vector, utilizes remaining characteristic vector to be weighted summation to observation data, thereby realizes the dimensionality reduction of observation data.
The 8th step forwards the 1st step to, and the observation data after dimensionality reduction is carried out to independent component analysis, estimates to obtain new isolated component; By said process repeatedly, when obtaining
Figure 927969DEST_PATH_IMAGE125
after individual isolated component, independent component analysis process is complete.
, Maximum Likelihood Estimation.
It is a kind of basic skills in statistical estimate field that maximum likelihood is estimated, the basic principle of its estimation: adopt those and make gained observed quantity have the estimated parameter value of maximum probability.While solving the ICA model based on multichannel TPM signal by the method, the log likelihood of first will deriving.
The simplified model of ICA in formula (3)
Figure 905283DEST_PATH_IMAGE053
, order
Figure 209226DEST_PATH_IMAGE127
,
Figure 749928DEST_PATH_IMAGE129
represent the to transmit probability density of TPM, ICA model can be write as:
Figure 127820DEST_PATH_IMAGE131
(9)
Above formula can also be expressed as matrix function with vector x:
Figure 503493DEST_PATH_IMAGE135
(10)
Because x is observed m, use respectively ,
Figure 142601DEST_PATH_IMAGE139
...,
Figure 727298DEST_PATH_IMAGE141
represent.Likelihood score can be by inciting somebody to action so mindividual dot density is estimated to multiply each other to obtain, and likelihood score L can be used as
Figure 677936DEST_PATH_IMAGE143
function:
Figure 622758DEST_PATH_IMAGE145
(11)
In actual use, generally use the logarithm of likelihood score, thereby turn to log likelihood and be:
Figure 286827DEST_PATH_IMAGE147
(12)
Next the natural gradient numerical algorithm of log likelihood needs to maximize.The random gradient of formula (11) is:
Figure 229375DEST_PATH_IMAGE149
(13)
Here,
Figure 729626DEST_PATH_IMAGE151
be one by the vector function of element, the empirical function of generally choosing is following two kinds:
Figure 415823DEST_PATH_IMAGE153
(14)
Thereby obtain maximum likelihood estimation algorithm according to above-mentioned function:
Figure 435862DEST_PATH_IMAGE155
(15)
In above formula, each step computing need to be inverted to matrix B, causes operation slowly, thereby can improve by natural gradient method, utilizes the Geometric structure principium of parameter space according to natural gradient, integrates exactly formula (14) the right is multiplied by
Figure 549312DEST_PATH_IMAGE157
, obtain as follows
Figure 536860DEST_PATH_IMAGE159
(16)
This algorithm also can see do things non-linear get relevant,
Figure 761168DEST_PATH_IMAGE161
time convergence, mean now for
Figure 399828DEST_PATH_IMAGE163
,
Figure 418600DEST_PATH_IMAGE165
with
Figure 893443DEST_PATH_IMAGE167
time incoherent, it is a kind of nonlinear extensions of common irrelevance condition.
The online random Natural Gradient Algorithm step that maximum likelihood is estimated:
The 1st step is carried out centralization to data, and to make its average be 0;
The 2nd step is selected an initialization (can generate at random) separation matrix b.Select
Figure 734492DEST_PATH_IMAGE169
(
Figure 916074DEST_PATH_IMAGE171
) each initial value,
Figure 433643DEST_PATH_IMAGE169
can choose at random, also can utilize prior information.Select learning rate
Figure 379471DEST_PATH_IMAGE173
with
Figure 7899DEST_PATH_IMAGE175
;
The 3rd step is calculated
Figure 122616DEST_PATH_IMAGE177
If the 4th step non-linear form is not determined in advance:
(a) upgrade
Figure 748770DEST_PATH_IMAGE178
:
Figure 932627DEST_PATH_IMAGE180
(b) if
Figure 637450DEST_PATH_IMAGE182
, by element nonlinear function, get so , otherwise adopt
Figure 652996DEST_PATH_IMAGE186
.
The 5th step is upgraded separation matrix:
Figure 74882DEST_PATH_IMAGE188
Wherein,
Figure 982795DEST_PATH_IMAGE190
;
If the 6th step not yet restrains, return to step 3.
4) utilize the non-Gauss's method of maximization or Maximum Likelihood Estimation to decompose and obtain after multiple signals, Dui Ge road signal carries out respectively demodulation, and the mode by threshold judgement obtains demodulating information " 0 " or " 1 ".
As for thresholding, set, according to actual conditions and laboratory data, obtain.The amplitude peak of supposing non-SPA sudden phase anomalies time period part signal is
Figure 55793DEST_PATH_IMAGE192
, in the vector that ICA estimates, maximum is
Figure 335333DEST_PATH_IMAGE194
,
Figure 493782DEST_PATH_IMAGE194
one is positioned signal phase mutation time section part, threshold value thresholding
Figure 18436DEST_PATH_IMAGE196
be set as
Figure 945940DEST_PATH_IMAGE198
(17)
The present invention about time phase modulation communication ICA time-domain filtering method, than existing about time phase modulating system filtering technique, have following outstanding substantive distinguishing features and remarkable advantage:
1) the present invention is directed to a kind of information transferring method based on multichannel TPM signal, especially when receiving terminal adopts independent component analysis interface differential technique to receive, phase modulation signal directly carries out filtering in time domain, be different from frequency domain filtering method used in prior art, thereby can simplify receiving terminal structure.
2) current, traditional TPM technology only limits to use single channel signal transmission, and the present invention by TPM technique extension to multiplexing, realized the communication of multichannel TPM signal, effectively improved system transmission rate, improved the band efficiency of TPM communication system.
3) what the object of the present invention's research was considered is multichannel (or multi-user) TPM signal, be different from prior art and still only rest on the filtering of a road TPM signal and the research of demodulation, the practical application of phase modulation communication system when therefore the present invention can directly facilitate, will produce very large economic worth and social benefit.
Accompanying drawing explanation
The system complete block diagram of phase modulation communication system based on ICA time-domain filtering when Fig. 1 is
When Fig. 2 is multichannel, phase modulation signal is through the theory diagram of ICA time-domain filtering
The time domain waveform figure of phase modulation signal during Tu3Wei tri-tunnel.
Fig. 4 is the observation signal waveform that receiving terminal obtains, noiseless.
Fig. 5, for without making an uproar the time domain plethysmographic signal of observation signal after ICA time-domain filtering, has obtained Xin tri-road signals after the processing of ICA method Jiang San road observation signal, the SPA sudden phase anomalies feature of phase modulation signal when wherein front two-way has extracted accurately.
Fig. 6 is the white Gaussian noise when adding, signal to noise ratio snr=10 o'clock, and receiving terminal obtains San road observation signal waveform.
Fig. 7 is the time domain plethysmographic signal of noisy observation signal after ICA time-domain filtering, signal to noise ratio snr=10;
Fig. 8 is the white Gaussian noise when adding, signal to noise ratio snr=5 o'clock, and receiving terminal obtains San road observation signal waveform.
Fig. 9 is the time domain plethysmographic signal of noisy observation signal after ICA time-domain filtering, signal to noise ratio snr=5.
Figure 10 be proposed by the invention about time modulate mutually the characteristic curve of error code figure to signal demodulation after ICA time-domain filtering.When this figure shows, modulate mutually multiple signals transmission and can realize correct demodulation after ICA time-domain filtering, thereby the multiplexing of phase modulation signal can realize time.
Figure 11 is the two-way TPM signal modulation signal time domain waveform figure that carrier frequency is different.
Figure 12 when transmitting as two paths of signals, the observation signal waveform of receiving terminal, signal to noise ratio snr=10.
Figure 13 is in two-way situation, the time domain plethysmographic signal of observation signal after ICA time-domain filtering, signal to noise ratio snr=10.
Embodiment
Below in conjunction with accompanying drawing, embodiments of the invention are described in further detail.
Time phase modulation signal ICA time-domain filtering, can specifically implement as follows:
1) produce multichannel binary digit code element, and modulate mutually while carrying out and produce multichannel TPM signal;
2) at receiving terminal, a plurality of receivers are set and receive observation signals, each observation signal during by multichannel phase modulation signal linear hybrid form;
3) observation signal receiving is normalized with albefaction and is processed, so that signal meets the precondition of independent component analysis;
4) pretreated observation signal is carried out to ICA time-domain filtering, the SPA sudden phase anomalies feature of phase modulation signal during extraction.
5) according to extract time phase modulation signal SPA sudden phase anomalies feature demodulation output multi-channel numeric code signal.
Embodiment 1:
Embodiment object: when phase modulation signal forms during Yi San road, phase modulation communication system is example, by demodulation output original signal after ICA time-domain filtering.
Embodiment step:
1) phase modulation signal when user launches three tunnels in the present embodiment
Figure 84798DEST_PATH_IMAGE200
,
Figure 979810DEST_PATH_IMAGE202
,
Figure 495105DEST_PATH_IMAGE204
do not establish frequency interval, carrier frequency is 20kHz, every bit is modulated by 20 sinusoidal carriers, code element transmission rate is 1000bit/s, and sample frequency is 200kHz, and phase hit initial time is made as respectively the 9th, the 13rd and the 18th carrier wave place, design parameter is as table 1, and the time domain waveform after modulation is shown in Fig. 3.
Figure 277116DEST_PATH_IMAGE205
2) phase modulation signal enters Gaussian white noise channel through antenna transmission during above-mentioned three tunnel, at receiving terminal, by three transducers, is caught and obtains three road observation signals
Figure 399924DEST_PATH_IMAGE207
,
Figure 470649DEST_PATH_IMAGE209
,
Figure 586372DEST_PATH_IMAGE211
, they by
Figure 737737DEST_PATH_IMAGE212
,
Figure 218397DEST_PATH_IMAGE213
,
Figure 838734DEST_PATH_IMAGE214
linear hybrid forms
Figure 243301DEST_PATH_IMAGE216
, j=1,2,3.Be subject to the impact of channel, surrounding environment, this mixing is random, and the present invention sets this matrix coefficient A in routine and is implementing this
Figure 999905DEST_PATH_IMAGE218
When not considering noise
Figure 917045DEST_PATH_IMAGE220
When considering noise
Figure 8367DEST_PATH_IMAGE222
Wherein,
Figure 465893DEST_PATH_IMAGE224
for white Gaussian noise, average is zero, and variance is 1.
Plus noise not, add signal to noise ratio be 10 and the signal to noise ratio white Gaussian noise that is 5 after, the observation signal that receiving terminal receives is respectively as Fig. 4, Fig. 6 and Fig. 8.
3) observation signal is carried out to ICA time-domain filtering, after being processed, observation signal obtained new signal, phase modulating signal phase sudden change part and not mutated part when new signal is, wherein the amplitude of SPA sudden phase anomalies part has obviously obtained enhancing, and the signal after ICA time-domain filtering is respectively as Fig. 5, Fig. 7 and Fig. 9.Figure 10 be proposed by the invention about time modulate mutually the characteristic curve of error code figure to signal demodulation after ICA time-domain filtering.
Embodiment 2:
Embodiment object: when phase modulation signal forms during the different two-way of the carrier frequency of take, phase modulation communication system is example, by demodulation output original signal after ICA time-domain filtering.
1) phase modulation signal while only considering in the present embodiment that two users launch 2 tunnel
Figure 93315DEST_PATH_IMAGE226
, code element transmission rate is 1000bit/s, carrier frequency is respectively 20kHz and 22kHz, every bit is modulated by 10 sinusoidal carriers, phase hit initial time is made as respectively the 5th and the 9th carrier wave place, adds noise SNR=10, and design parameter is as table 2, time domain waveform after modulation is shown in Figure 11, gets front 10 code elements in figure.
Figure 245127DEST_PATH_IMAGE229
2) phase modulation signal enters Gaussian white noise channel through antenna transmission during above-mentioned three tunnel, at receiving terminal, by three transducers, is caught and obtains three road observation signals
Figure 444027DEST_PATH_IMAGE231
,
Figure 424491DEST_PATH_IMAGE233
, they by
Figure 480171DEST_PATH_IMAGE234
,
Figure 234501DEST_PATH_IMAGE235
linear hybrid forms
Figure 784562DEST_PATH_IMAGE237
, j=1,2.Wherein, matrix coefficient is produced at random by Matlab, and matrix coefficient A is
Figure 370264DEST_PATH_IMAGE239
Affected by noise because transmit, so observation signal is
Figure 800108DEST_PATH_IMAGE241
Observation signal, as Figure 12, has wherein added the white Gaussian noise that signal to noise ratio is 10.
3) observation signal is carried out to ICA time-domain filtering, after observation signal independent component analysis is processed, obtained the new signal of two-way, as shown in figure 13.The SPA sudden phase anomalies part that wherein a road comprises all input signals, shown in the upper figure of Figure 13, the amplitude of SPA sudden phase anomalies part is enhanced; Another road output signal comprises the non-SPA sudden phase anomalies part of all input signals, shown in Figure 13 figure below.

Claims (4)

1. the information transferring method based on multi-channel Time phase modulated signal, at transmitting terminal, utilize a plurality of separate sinusoidal signal SPA sudden phase anomalies carry informations that are positioned at particular moment, receiving terminal adopts independent component analysis Dui Ge road signal to carry out separation, by threshold judgement method, carry out demodulation again, its concrete steps are:
Step 10), transmitting terminal produce nthe binary digital information that road is separate
Figure 2014100601200100001DEST_PATH_IMAGE001
, nfor positive integer, and
Figure 99335DEST_PATH_IMAGE002
, , each road information
Figure 397593DEST_PATH_IMAGE004
while carrying out respectively, modulate mutually, its basic principle is that each binary data is corresponding
Figure 2014100601200100001DEST_PATH_IMAGE005
the continuous sine wave of phase place in individual cycle,
Figure 722395DEST_PATH_IMAGE006
for positive integer, and
Figure 2014100601200100001DEST_PATH_IMAGE007
, when sending " 1 ", at certain hour
Figure 466098DEST_PATH_IMAGE008
interior sine wave signal generation phase hit
Figure 2014100601200100001DEST_PATH_IMAGE009
,
Figure 294376DEST_PATH_IMAGE010
be less than or equal to sinusoidal wave period, when sending " 0 ", former sinusoidal signal does not change; The signal of not going the same way can have identical carrier frequency
Figure 91431DEST_PATH_IMAGE011
, symbol time and phase hit
Figure 811442DEST_PATH_IMAGE013
, also can be different, but the saltus step duration of the signal of not going the same way
Figure 258342DEST_PATH_IMAGE014
with saltus step initial time
Figure 226298DEST_PATH_IMAGE015
must have a difference at least, the the modulation signal on road is:
，
Figure 774588DEST_PATH_IMAGE018
Wherein,
Figure 2014100601200100001DEST_PATH_IMAGE019
for the modulation waveform of binary data 0 correspondence,
Figure 913445DEST_PATH_IMAGE020
for the modulation waveform of binary data 1 correspondence, with
Figure 684272DEST_PATH_IMAGE020
expression formula be respectively:
Figure 2014100601200100001DEST_PATH_IMAGE021
Figure 105764DEST_PATH_IMAGE022
Wherein,
Figure 353206DEST_PATH_IMAGE012
be
Figure 423930DEST_PATH_IMAGE016
the element duration of road signal,
Figure 415020DEST_PATH_IMAGE011
be
Figure 254800DEST_PATH_IMAGE016
the carrier frequency of road signal,
Figure 2014100601200100001DEST_PATH_IMAGE023
be phase hit when road signal sends " 1 ",
Figure 401802DEST_PATH_IMAGE024
be
Figure 258900DEST_PATH_IMAGE016
phase hit zero hour when road signal sends " 1 ", be
Figure 542431DEST_PATH_IMAGE016
phase hit duration when road signal sends " 1 ", each road modulation signal
Figure 2014100601200100001DEST_PATH_IMAGE025
(
Figure 463113DEST_PATH_IMAGE026
) through antenna transmission, go out respectively;
Step 20), utilize receiving terminal mindividual transducer receives respectively TPM signal,
Figure 2014100601200100001DEST_PATH_IMAGE027
, the observation signal that each transducer receives is that the weighted linear of TPM signal is mixed
Figure 294541DEST_PATH_IMAGE028
,
Figure 2014100601200100001DEST_PATH_IMAGE029
, for the observation signal being comprised of TPM signal, employing is carried out independent element technology it is analyzed, its basic step is: first Jiang Ge road receives vectorial centralization, and its average is made zero, and utilizes albefaction to process the correlation of removing each road direction amount, set up non-Gauss's tolerance, from certain initial weighing vector wstart, the observation vector after albefaction is weighted and sues for peace and calculate its non-Gauss's tolerance, utilize the absolute value of non-Gauss's tolerance to increase the fastest direction, former weighing vector w is upgraded, iteration process, until non-Gauss's metric convergence;
Step 30), according to independent element, analyse the signal estimating
Figure 718700DEST_PATH_IMAGE030
, by the mode of Threshold detection, adjudicate and obtain transmit symbol " 0 " or " 1 ".
2. the information transferring method based on multi-channel Time phase modulated signal as claimed in claim 1, is characterized in that, described non-Gauss's tolerance utilizes kurtosis or negentropy or log likelihood to calculate, wherein, and random signal
Figure 2014100601200100001DEST_PATH_IMAGE031
kurtosis be:
Figure 744425DEST_PATH_IMAGE032
，
Random signal
Figure 2014100601200100001DEST_PATH_IMAGE033
negentropy be:
Figure 447676DEST_PATH_IMAGE034
，
Matrix
Figure 2014100601200100001DEST_PATH_IMAGE035
log likelihood
Figure 584260DEST_PATH_IMAGE036
for:
Figure 2014100601200100001DEST_PATH_IMAGE037
，
Wherein,
Figure 925242DEST_PATH_IMAGE035
for observing matrix
Figure 121868DEST_PATH_IMAGE038
estimation, , and , represent random vector
Figure 948231DEST_PATH_IMAGE042
probability density function.
3. the information transferring method based on multi-channel Time phase modulated signal as claimed in claim 1, is characterized in that, described non-Gauss's metric convergence refers to, the weight vector before and after an iteration meets
Figure 471617DEST_PATH_IMAGE043
, wherein,
Figure 839144DEST_PATH_IMAGE044
.
4. the information transferring method based on multi-channel Time phase modulated signal as claimed in claim 1, is characterized in that, described detection threshold
Figure 251409DEST_PATH_IMAGE045
meet:
Figure 792111DEST_PATH_IMAGE046
，
Wherein,
Figure 373265DEST_PATH_IMAGE047
the amplitude peak of the non-SPA sudden phase anomalies time period part signal of the one-channel signal obtaining for independent component analysis,
Figure 646115DEST_PATH_IMAGE048
the whole amplitude peak of the one-channel signal obtaining for independent component analysis.
CN201410060120.0A 2014-02-24 2014-02-24 Information transmission method based on multi-way TPM signals Pending CN103763234A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410060120.0A CN103763234A (en) 2014-02-24 2014-02-24 Information transmission method based on multi-way TPM signals

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410060120.0A CN103763234A (en) 2014-02-24 2014-02-24 Information transmission method based on multi-way TPM signals

Publications (1)

Publication Number Publication Date
CN103763234A true CN103763234A (en) 2014-04-30

Family

ID=50530396

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410060120.0A Pending CN103763234A (en) 2014-02-24 2014-02-24 Information transmission method based on multi-way TPM signals

Country Status (1)

Country Link
CN (1) CN103763234A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104113921A (en) * 2014-07-11 2014-10-22 山东科技大学 MIMO-EDM wireless communication system and method
CN109120266A (en) * 2018-09-05 2019-01-01 四川大学 A kind of efficient calculation matrix phase shift calibration method for DMWC
CN110289922A (en) * 2019-07-09 2019-09-27 南京航空航天大学 A kind of time-modulation and demodulation method
CN111682881A (en) * 2020-06-17 2020-09-18 北京润科通用技术有限公司 Communication reconnaissance simulation method and system suitable for multi-user signals

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094212A (en) * 2007-07-17 2007-12-26 东南大学 Multielement positional phase shift keying modulation and demodulation method
CN102739590A (en) * 2011-04-14 2012-10-17 苏州东奇信息科技有限公司 CP-EBPSK communication system for pseudorandom sequence phase modulation and communication method thereof
CN103139123A (en) * 2013-03-12 2013-06-05 苏州东奇信息科技有限公司 MPPSK demodulation method based on location information
CN103346989A (en) * 2013-05-06 2013-10-09 中国人民解放军重庆通信学院 Multi-path frequency hopping-based single-channel blind source separation anti-interference communication system and method thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094212A (en) * 2007-07-17 2007-12-26 东南大学 Multielement positional phase shift keying modulation and demodulation method
CN102739590A (en) * 2011-04-14 2012-10-17 苏州东奇信息科技有限公司 CP-EBPSK communication system for pseudorandom sequence phase modulation and communication method thereof
CN103139123A (en) * 2013-03-12 2013-06-05 苏州东奇信息科技有限公司 MPPSK demodulation method based on location information
CN103346989A (en) * 2013-05-06 2013-10-09 中国人民解放军重庆通信学院 Multi-path frequency hopping-based single-channel blind source separation anti-interference communication system and method thereof

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
于善海: "基于独立分量分析算法的CDMA多用户检测技术的研究", 《中国优秀硕士学位论文全文数据库》 *
程瑶: "基于独立分量分析的盲信号分离算法研究", 《中国优秀硕士学位论文全文数据库》 *
赵培洪等: "基于稀疏编码的TPM通信噪声抑制分析", 《电讯技术》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104113921A (en) * 2014-07-11 2014-10-22 山东科技大学 MIMO-EDM wireless communication system and method
CN104113921B (en) * 2014-07-11 2018-08-24 山东科技大学 A kind of MIMO-EDM wireless communication systems and method
CN109120266A (en) * 2018-09-05 2019-01-01 四川大学 A kind of efficient calculation matrix phase shift calibration method for DMWC
CN110289922A (en) * 2019-07-09 2019-09-27 南京航空航天大学 A kind of time-modulation and demodulation method
CN111682881A (en) * 2020-06-17 2020-09-18 北京润科通用技术有限公司 Communication reconnaissance simulation method and system suitable for multi-user signals
CN111682881B (en) * 2020-06-17 2021-12-24 北京润科通用技术有限公司 Communication reconnaissance simulation method and system suitable for multi-user signals

Similar Documents

Publication Publication Date Title
Liu et al. Biologically inspired covert underwater acoustic communication by mimicking dolphin whistles
US11581967B2 (en) Wireless channel scenario identification method and system
CN104168228B (en) Compressed sensing ultra-wideband channel method of estimation and system based on cluster position collection
CN103763234A (en) Information transmission method based on multi-way TPM signals
CN106936742A (en) Multi gear bit rate adaptive demodulation system and method based on neutral net
CN107317778B (en) BPSK modulating signal phase transition detection method based on 1D-CNN
Zhao et al. Federated meta-learning enhanced acoustic radio cooperative framework for ocean of things
CN104618278B (en) A kind of pseudo-code rate-estimation method of the multi-user TDDM BOC signal related based on spectrum
CN104168233A (en) DSSS/UQPSK signal pseudo code sequence estimation method based on characteristic decomposition and Messay algorithm
CN101594166B (en) Method and device for eliminating interference in signals in short distance wireless network
Liu et al. Jamming recognition based on feature fusion and convolutional neural network
Li et al. Compressive link acquisition in multiuser communications
CN103763233A (en) GSM external radiation radar signal processing method based on digital modulation and demodulation
CN104868962B (en) Frequency spectrum detecting method and device based on compressed sensing
CN104181508A (en) Threatening radar signal detection method based on compressed sensing
CN105812300B (en) Eliminate the long code DSSS blind symbol estimation method of information code jump
CN110059401B (en) OFDM system underwater sound channel impulse response reconstruction method
CN108777604B (en) Wavelength division multiple access ultra wide band multi-user detection method based on Gaussian mixture model clustering
CN113541726B (en) Code index spread spectrum underwater acoustic communication method based on recurrent neural network
CN101237250B (en) Frequency spreading wave blind estimation method based on odd value analysis
CN102571656B (en) Method and device for eliminating interference from signal in short-range wireless network
CN106301486B (en) Framing bit detector in near-field communication
Fu et al. The DPSK signal noncoherent demodulation receiver based on the Duffing oscillators array
CN110011745A (en) The estimation method of signal number in broader frequency spectrum perception based on DPMM
CN103973622A (en) Rapid MSK communication detection demodulation method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20140430

WD01 Invention patent application deemed withdrawn after publication