CN105245300B - User number estimation method under a kind of underlay frequency spectrum shares mode - Google Patents

User number estimation method under a kind of underlay frequency spectrum shares mode Download PDF

Info

Publication number
CN105245300B
CN105245300B CN201510534901.3A CN201510534901A CN105245300B CN 105245300 B CN105245300 B CN 105245300B CN 201510534901 A CN201510534901 A CN 201510534901A CN 105245300 B CN105245300 B CN 105245300B
Authority
CN
China
Prior art keywords
mrow
mtr
mtd
msub
signal
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
CN201510534901.3A
Other languages
Chinese (zh)
Other versions
CN105245300A (en
Inventor
刘明骞
陈健
王江宏
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.)
Xidian University
Original Assignee
Xidian University
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 Xidian University filed Critical Xidian University
Priority to CN201510534901.3A priority Critical patent/CN105245300B/en
Publication of CN105245300A publication Critical patent/CN105245300A/en
Application granted granted Critical
Publication of CN105245300B publication Critical patent/CN105245300B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Noise Elimination (AREA)

Abstract

The invention discloses user number estimation method under a kind of underlay frequency spectrum shares mode, comprise the following steps：Dynamic deferred extension is carried out to the signal received；Singular value decomposition, the singular value matrix of signal after being expanded are carried out to the signal after extension, and extract the leading diagonal vector of singular value matrix；With the previous element in leading diagonal vector subtract close to next element, construction feature amount ch1；With the previous element in ch1 divided by close to next element, construction feature amount ch2；The maximum searched in ch2, estimate user's number under underlay frequency spectrum share modes.The present invention has good estimation performance to user's number under underlay frequency spectrum share modes under conditions of low signal-to-noise ratio and high spectrum Duplication.

Description

User number estimation method under a kind of underlay frequency spectrum shares mode
Technical field
The present invention relates to communication technical field, and in particular to user's number is estimated under a kind of underlay frequency spectrum shares mode Method, available for user's number estimation under low signal-to-noise ratio and high spectrum Duplication.
Background technology
In Modern wireless communication field, with the development of the communication technology, frequency spectrum resource becomes more and more nervous.Underlay Frequency spectrum share mode can make primary user and time same frequency band of users to share, be the effective way for solving frequency spectrum resource shortage problem Footpath.Under underlay modes, each subscriber signal forms time-frequency overlapped signal, and the estimation to user's number is follow-up each only The basis of vertical signal modulation parameter Estimation, modulation identification and mixed signal separation.Therefore, for underlay frequency spectrum shares Mode, the source number estimate method of research time-frequency overlapped signal have certain meaning and value.
At present, the research on the estimation of time-frequency overlapped signal number is less and all overlapping in low signal-to-noise ratio and high spectrum Poor-performing is estimated in the case of rate.Luan haiyan et al. pass through Cyclic Spectrum using the sparse characteristic of modulated signal Cyclic Spectrum The correlation of line is classified to spectral line, so as to estimate signal source number (Luan haiyan, Jiang hua, Liu xiaobao.Source Number Estimation in Single Channel Blind Source Separation [C] .CSIP, 2010：4445-4449.).Wu liang et al. are led to using the characteristic of wavelet transformation extraction Singular Point information The extraction to strong periodic signal is crossed to estimate information source number (Wu liang, Zhou huiqing, Jiang hua.Source Number Estimation Algorithm in Wavelet Domain for Single-Channel Mixed-Signal [C] .ICCP, 2012：1048-1051.).Open pure grade and multiple signals are constructed by truncated signal again, then construct fourth order cumulant Matrix, by analyzing this cumulant matrices, so as to estimate information source number (Zhang Chun, Yang Junan, the leaf under Colored Noise Single channel number of source algorithm for estimating [J] signal transactings, 2012,28 (7) under rich gauss heat source models background：994-999.).King Blue or green red grade utilizes the sparse characteristic of the quadravalence cyclic cumulants of modulated signal, by extracting discrete spectral line in cycle frequency domain to estimate The information source number of timing frequency overlapped signal (Wang Qinghong, Peng Hua, Wang Bin, rises common channel multi signal inspections of the ripple based on cyclic cumulants Survey and source number estimate algorithm [J] information engineering college journals, 2012,13 (2)：184-189.).Pang lihui etc. exist On the premise of modulated signal is periodic signal, signal is reconstructed, and it classified by the correlation of reconstruction signal, So as to estimate information source number (Pang lihui, Qi Conghui.A novel joint parameter estimation method for single-channel and time-frequency overlapped multi-component Signal [C] .CSQRWC, 2013：124-127.).
The content of the invention
In view of the shortcomings of the prior art, the present invention is intended to provide user under a kind of effective underlay frequency spectrum share modes Individual number estimation method, to improve under low signal-to-noise ratio and high spectrum Duplication, user's number under underlay modes is estimated Estimate performance.
To achieve these goals, the technical method that the present invention uses is as follows：
S1 carries out dynamic deferred extension to the signal x (t) received, so as to obtain signal y (t)；
S2 carries out singular value decomposition to y (t), obtains y (t) singular value matrix, and the master for extracting singular value matrix is diagonal Line vector
S3 is usedIn previous element subtract close to next element, construction feature amount ch1；
S4 with the previous element in ch1 divided by close to next element, construction feature amount ch2；
The position of maximum, as user's number under underlay modes in S5 search ch2.
It should be noted that the building method of the signal y (t) in step S1 is as follows：
The signal model of reception signal is expressed as：
Wherein, si(t) (i=1 ..., N) is the overlapping subscriber signal component of time-frequency, and N is the number of subscriber signal component, t For the time；N (t) is additive white Gaussian noise；AiFor subscriber signal component si(t) amplitude；aikFor modulated signal；pi(t) (i= 1 ..., N) be rolloff-factor α raised cosine shaping filter function, andTsiFor symbol Speed；fiFor carrier frequency；J is the representation of imaginary number, and meets j2=-1；Between subscriber signal component and subscriber signal Between component and noise independently of each other；
It is determined that delay τiIt can represent as follows with number of plies M, construction signal y (t)：
Wherein, τi(i=1 ..., M-1) and M system of selection are as follows：
First, define x (t) and y (t) coefficient correlation is as follows：
Wherein, cov (xtyt) for signal x (t) and y (t) cross covariance, D (xt) and D (yt) it is respectively x (t) and y (t) Auto-covariance；E(xt) and E (yt) be respectively x (t) and y (t) expectation；
It is any to choose τ0And gradually increase τi+1(i=0 ..., M-2), calculate yiAnd yi+1Coefficient correlationWhen First fitWhen, τ nowi+1It is exactly the delay of i+1 layer in y (t), wherein, δ1For coefficient correlationInstitute is right The threshold value answered, after signal is normalized, the empirical value of this threshold value is 0.8.
Y is calculated respectively0And yiThe coefficient correlation of (i=1 ..., M-1)I.e.：
WhenWhen, now yiCorresponding M is y (t) number of plies.Wherein, δ2For coefficient correlationCorresponding Threshold value, after signal is normalized, the empirical value of this threshold value is 0.1.
In τiIn M selection course, both should alternately, i.e., each τ for calculatingi, first calculate y0With yiCoefficient correlationWhenIt is unsatisfactory forWhen, increasing layer number M value, select next τi, closed until selecting Untill suitable M.
It should be noted that singular value leading diagonal vector is extracted in step S2Method it is as follows：
To each component signal y in y (t)i(i=0,1 ..., M-1) carries out N point samplings with identical sample frequency, Then y can be expressed as：
It is theoretical according to singular value decomposition, then have
Y=U ∑s VT
Wherein, U is M × M rank matrixes；V is N × N rank matrixes；∑ is the diagonal matrix of singular value composition.Remember subscriber signal The number of component is K, then ∑ is represented by
Wherein, κi(i=1 ..., K) is the singular value of subscriber signal component, and σ is the singular value of noise.In theory, ∑ The elements in a main diagonal meet：
λ1≥λ2≥…≥λK＞ λK+1=λK+2=...=λM
Due to the truncation effect of signal, its singular value is simultaneously unsatisfactory for above formula, but meets relation as follows with probability 1 Formula：
λ1＞ λ2＞ ... ＞ λK＞ λK+1＞ λK+2＞ ... ＞ λM
Extract element composition of vector on leading diagonalThen have：
It should be noted that the building method of the characteristic quantity ch1 in step S3 is as follows：
In step s 2, the leading diagonal vector of y (t) singular value matrix is extracted
Wherein, λ1＞ λ2＞ ... ＞ λM.Then ch1 building method is as follows：
ch1i=λi-λi+1(i=1,2 ..., M-1)
Ch1=[ch11, ch12..., ch1M-1]
Wherein, λiFor the leading diagonal vector extracted in step S2I-th of element.
It should be noted that the building method of the characteristic quantity ch2 in step S4 is as follows：
It is as follows according to the characteristic vector ch1 constructed in step S3, characteristic vector ch2 building method：
Ch2=[ch21, ch22..., ch2M-2]
Wherein, ch1iFor the characteristic quantity ch1 constructed in step S3 i-th of element.
It should be noted that source number estimate method is performed as follows in step S5：
Wherein, ch2iFor the characteristic quantity ch2 constructed in step S4 i-th of element, K is the user's number estimated.
Beneficial effect of the present invention is：For the time-frequency overlapped signal under underlay frequency spectrum modes, the present invention is in low letter Make an uproar than with high spectrum Duplication there is good user's number to estimate performance.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the present invention；
Fig. 2 is estimation accuracy of the present invention to the time-frequency overlapped signal of different modulating type under different signal to noise ratio.
Embodiment
Below with reference to accompanying drawing, the invention will be further described, it is necessary to which explanation, the present embodiment is with this technology side Premised on case, detailed embodiment and specific operating process are given, but protection scope of the present invention is not limited to this reality Apply example.
As shown in figure 1, the present invention is under underlay frequency spectrum share modes, one kind has low signal-to-noise ratio and high spectrum weight The subscriber signal number estimation method of folded rate, the described method comprises the following steps：
S1 carries out dynamic deferred extension to the signal x (t) received, so as to obtain signal y (t).
It should be noted that the building method of the signal y (t) in step S1 is as follows：
The signal model of reception signal is expressed as：
Wherein, si(t) (i=1 ..., N) is the overlapping subscriber signal component of time-frequency, and N is the number of subscriber signal component, t For the time；N (t) is additive white Gaussian noise；AiFor subscriber signal component si(t) amplitude；aikFor modulated signal；pi(t) (i= 1 ..., N) be rolloff-factor α raised cosine shaping filter function, andTsiFor symbol Speed；fiFor carrier frequency；J is the representation of imaginary number, and meets j2=-1；Between subscriber signal component and subscriber signal Between component and noise independently of each other；
The suitable delay τ of selectioniIt can represent as follows with number of plies M, construction signal y (t)：
Wherein, τi(i=1 ..., M-1) and M system of selection are as follows：
First, define x (t) and y (t) coefficient correlation is as follows：
Wherein, cov (xtyt) for signal x (t) and y (t) cross covariance, D (xt) and D (yt) it is respectively x (t) and y (t) Auto-covariance.
It is any to choose τ0And gradually increase τi+1(i=0 ..., M-2), calculate yiAnd yi+1Coefficient correlationWhen First fitWhen, τ nowi+1It is exactly the delay of i+1 layer in y (t), wherein, δ1For coefficient correlationInstitute is right The threshold value answered, after signal is normalized, the empirical value of this threshold value is 0.8.
Y is calculated respectively0And yiThe coefficient correlation of (i=1 ..., M-1)I.e.
WhenWhen, now yiCorresponding M is y (t) number of plies.Wherein, δ2For coefficient correlationCorresponding Threshold value, after signal is normalized, the empirical value of this threshold value is 0.1.
In τiIn M selection course, both should alternately, i.e., each τ for calculatingi, first calculate y0With yiCoefficient correlationWhenIt is unsatisfactory forWhen, increasing layer number M value, select next τi, closed until selecting Untill suitable M.
S2 carries out singular value decomposition to y (t), obtains y (t) singular value matrix, and the master for extracting singular value matrix is diagonal Line vector
It should be noted that singular value leading diagonal vector is extracted in step S2Method it is as follows：
To each component signal y in y (t)i(i=0,1 ..., M-1) carries out N point samplings with identical sample frequency, Then y can be expressed as：
It is theoretical according to singular value decomposition, then have
Y=U ∑s VT；
Wherein, U is M × M rank matrixes；V is N × N rank matrixes；∑ is the diagonal matrix of singular value composition, remembers subscriber signal The number of component is K, then ∑ is represented by
Wherein, κi(i=1 ..., K) is the singular value of subscriber signal component, and σ is the singular value of noise.In theory, ∑ The elements in a main diagonal meet：
λ1≥λ2≥…≥λK＞ λK+1=λK+2=...=λM；
Due to the truncation effect of signal, its singular value is simultaneously unsatisfactory for above formula, but meets relation as follows with probability 1 Formula：
λ1＞ λ2＞ ... ＞ λK＞ λK+1＞ λK+2＞ ... ＞ λM
Extract element composition of vector on leading diagonalThen have：
S3 passes throughIn previous element subtract close to next element, construction feature amount ch1；
It should be noted that the building method of the characteristic quantity ch1 in step S3 is as follows：
In step s 2, the leading diagonal vector of y (t) singular value matrix is extracted
Wherein, λ1＞ λ2＞ ... ＞ λM.Then ch1 building method is as follows：
ch1i=λi-λi+1(i=1,2 ..., M-1)；
Ch1=[ch11, ch12..., ch1M-1]；
Wherein, λiFor the leading diagonal vector extracted in step S2I-th of element.
S4 by the previous element in ch1 divided by close to next element, construction feature amount ch2；
It should be noted that the building method of the characteristic quantity ch2 in step S4 is as follows：
It is as follows according to the characteristic vector ch1 constructed in step S3, characteristic vector ch2 building method：
Ch2=[ch21, ch22..., ch2M-2]
Wherein, ch1iFor the characteristic quantity ch1 constructed in step S3 i-th of element.
The position of maximum, as user's number under underlay frequency spectrum shares mode in S5 search ch2.
It should be noted that source number estimate method is performed as follows in step S5：
Wherein, ch2iFor the characteristic quantity ch2 constructed in step S4 i-th of element, K is the user's number estimated.
For the performance of appraisal procedure, emulation experiment below uses the type of signal as any MPSK (M mixed two-by-two =2,4,8) 2000 Monte Carlo experiments are carried out and 16QAM signals, and.The evaluation criteria of detection is estimation accuracy.
In order to test the performance of the test statistics of this method, parameter setting is as follows：The rolling of raised cosine shaping filter function Factor alpha=0.35 drops；Chip rate Ts1=Ts2=1；Carrier frequency f1=4, f2=4.5；Sample frequency fs=20；Now, when The spectrum overlapping rate of frequency overlapped signal is 75%.Simulation result is as shown in Fig. 2 when SNR is -5dB, and the estimation of information source number is just True rate is still more than 80%.Thus the inventive method is illustrated under the conditions of low signal-to-noise ratio and high spectrum Duplication, to underlay User's number estimation under frequency spectrum share mode has preferably estimation performance.
For those skilled in the art, technical scheme that can be as described above and design, make other each Kind is corresponding to be changed and deforms, and all these change and deformed the protection model that should all belong to the claims in the present invention Within enclosing.

Claims (4)

  1. A kind of 1. user number estimation method under underlay frequency spectrum shares mode, it is characterised in that：Comprise the following steps：
    S1 carries out dynamic deferred extension to the signal x (t) received, so as to obtain signal y (t)；Dynamic deferred extension signal y (t) construction is carried out by the following method：
    It is determined that delay τi(i=1 ..., M-1) and number of plies M, the signal y (t) of construction represent as follows：
    <mrow> <mi>y</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <mrow> <mi>x</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <msub> <mi>&amp;tau;</mi> <mn>0</mn> </msub> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>x</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <msub> <mi>&amp;tau;</mi> <mn>1</mn> </msub> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>x</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>+</mo> <msub> <mi>&amp;tau;</mi> <mrow> <mi>M</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>=</mo> <mfenced open = "(" close = ")"> <mtable> <mtr> <mtd> <msub> <mi>y</mi> <mn>0</mn> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>y</mi> <mn>1</mn> </msub> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <mo>.</mo> </mtd> </mtr> <mtr> <mtd> <msub> <mi>y</mi> <mrow> <mi>M</mi> <mo>-</mo> <mn>1</mn> </mrow> </msub> </mtd> </mtr> </mtable> </mfenced> <mo>;</mo> </mrow>
    τi(i=1 ..., M-1) and M system of selection are as follows：
    It is any to choose τ0And gradually increase τi+1(i=0 ..., M-2), calculate yiAnd yi+1Coefficient correlationWhenFirst MeetWhen, τ nowi+1It is exactly the delay of i+1 layer in y (t)；Wherein, δ1For coefficient correlationCorresponding Threshold value；
    According to selected delay τi+1, y is calculated respectively0And yiThe coefficient correlation of (i=1 ..., M-1)WhenWhen, Now yiCorresponding M is the number of plies for constructing signal y (t)；Wherein, δ2For coefficient correlationCorresponding threshold value；
    S2 carries out singular value decomposition to y (t), obtains y (t) singular value matrix, and extract the leading diagonal of singular value matrix to Amount
    S3 is usedIn previous element subtract close to next element, construction feature amount ch1；
    S4 with the previous element in ch1 divided by close to next element, construction feature amount ch2；
    The position of maximum, as user's number under underlay modes in S5 search ch2.
  2. 2. user number estimation method under underlay frequency spectrum shares mode according to claim 1, it is characterised in that step Characteristic quantity ch1 in rapid S3 is constructed as follows：
    ch1i=λi-λi+1(i=1,2 ..., M-1)；
    Ch1=[ch11,ch12,…,ch1M-1]；
    Wherein, λiFor the leading diagonal vector extractedI-th of element.
  3. 3. user number estimation method under underlay frequency spectrum shares mode according to claim 1, it is characterised in that step Characteristic quantity ch2 in rapid S4 is constructed as follows：
    <mrow> <mi>c</mi> <mi>h</mi> <msub> <mn>2</mn> <mi>i</mi> </msub> <mo>=</mo> <mfrac> <mrow> <mi>c</mi> <mi>h</mi> <msub> <mn>1</mn> <mi>i</mi> </msub> </mrow> <mrow> <mi>c</mi> <mi>h</mi> <msub> <mn>1</mn> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> </mrow> </mfrac> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mo>...</mo> <mo>,</mo> <mi>M</mi> <mo>-</mo> <mn>2</mn> <mo>)</mo> </mrow> <mo>;</mo> </mrow>
    Ch2=[ch21,ch22,…,ch2M-2]；
    Wherein, ch1iFor the characteristic quantity ch1 constructed i-th of element.
  4. 4. user number estimation method under underlay frequency spectrum shares mode according to claim 1, it is characterised in that step The method that maximum value position in ch2 is searched in rapid S5 is performed as follows：
    <mrow> <mi>K</mi> <mo>=</mo> <munder> <mrow> <mi>arg</mi> <mi> </mi> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> <mi>i</mi> </munder> <mi>c</mi> <mi>h</mi> <mn>2</mn> <mo>=</mo> <munder> <mrow> <mi>arg</mi> <mi> </mi> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> <mi>i</mi> </munder> <mo>{</mo> <mi>c</mi> <mi>h</mi> <msub> <mn>2</mn> <mi>i</mi> </msub> <mo>,</mo> <mn>1</mn> <mo>&lt;</mo> <mi>i</mi> <mo>&lt;</mo> <mi>M</mi> <mo>-</mo> <mn>2</mn> <mo>}</mo> </mrow>
    Wherein, ch2iThe amount of being characterized ch2 i-th of element, K are the user's number estimated.
CN201510534901.3A 2015-08-27 2015-08-27 User number estimation method under a kind of underlay frequency spectrum shares mode Active CN105245300B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510534901.3A CN105245300B (en) 2015-08-27 2015-08-27 User number estimation method under a kind of underlay frequency spectrum shares mode

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510534901.3A CN105245300B (en) 2015-08-27 2015-08-27 User number estimation method under a kind of underlay frequency spectrum shares mode

Publications (2)

Publication Number Publication Date
CN105245300A CN105245300A (en) 2016-01-13
CN105245300B true CN105245300B (en) 2017-12-05

Family

ID=55042816

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510534901.3A Active CN105245300B (en) 2015-08-27 2015-08-27 User number estimation method under a kind of underlay frequency spectrum shares mode

Country Status (1)

Country Link
CN (1) CN105245300B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105978641B (en) * 2016-04-28 2018-11-02 西安电子科技大学 The signal-noise ratio estimation method of time-frequency overlapped signal in a kind of cognitive radio
CN106647534A (en) * 2017-01-23 2017-05-10 谢圣超 Ophthalmological information processing control system based on cloud computation

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1885741A (en) * 2006-06-26 2006-12-27 西安交通大学 Authorization user signal detecting method for cognitive radio system
KR20090034457A (en) * 2007-10-04 2009-04-08 한국과학기술원 Underlay transmission method in cognitive radio based wireless communication system
CN101800623A (en) * 2010-01-29 2010-08-11 华中科技大学 Throughput-maximized cognitive radio system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8379529B2 (en) * 2007-09-19 2013-02-19 Motorola Solutions, Inc. Distributed spectrum allocation scheme for cognitive radio

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1885741A (en) * 2006-06-26 2006-12-27 西安交通大学 Authorization user signal detecting method for cognitive radio system
KR20090034457A (en) * 2007-10-04 2009-04-08 한국과학기술원 Underlay transmission method in cognitive radio based wireless communication system
CN101800623A (en) * 2010-01-29 2010-08-11 华中科技大学 Throughput-maximized cognitive radio system

Also Published As

Publication number Publication date
CN105245300A (en) 2016-01-13

Similar Documents

Publication Publication Date Title
CN103220052B (en) A kind of method detecting frequency spectrum cavity-pocket in cognitive radio
CN105933257A (en) Signal-to-noise ration estimating method of time frequency overlapping signals under frequency spectrum sharing mode
CN103731380B (en) The method and device that a kind of time-frequency combination channel based on compressed sensing is estimated
CN108123774A (en) A kind of narrowband Internet of Things down link synchronisation method and its system
CN103795479B (en) A kind of cooperative frequency spectrum sensing method of feature based value
CN101986583B (en) Covariance matching-based multi-antenna spectrum sensing method
CN103199983A (en) N-order local area power model in side channel power analysis and application thereof
CN109919015A (en) A kind of radio-frequency fingerprint extraction and recognition methods based on more sampling convolutional neural networks
CN104660356B (en) A kind of collaborative frequency spectrum sensing method of half-blindness with reliable false-alarm performance
CN104333406B (en) For the coding/decoding method in multiple cell multi-user multi-aerial system
CN105245300B (en) User number estimation method under a kind of underlay frequency spectrum shares mode
CN106506415A (en) A kind of method of multiuser MIMO ofdm system channel estimation
CN109255874A (en) A kind of passage and number detection method based on general commercial WiFi equipment
CN105634544A (en) Blind despreading method of synchronous long code DS-CDMA (Direct Sequence-Code Division Multiple Access) signals
CN104618278A (en) Multi-user TDDM-BOC signal pseudo code rate estimating method based on spectral correlation
CN103780520B (en) Compression sensing narrow-band interference estimation method based on priori information assistance and device
CN105119670B (en) A kind of broadband cooperative frequency spectrum sensing method
CN103795481A (en) Cooperative spectrum sensing method based on free probability theory
CN106443718A (en) Measurement zero value non-error tracking system and method based on correlation peak correction under non-ideal channel
CN104219008B (en) A kind of broadband frequency spectrum detection method and device
CN104408027A (en) Underdetermined blind identification method based on general covariance and tensor decomposition
CN202918327U (en) Automatic classification system for digital modulation signal
CN102664835B (en) Multicell channel estimation method based on interference elimination route by route
CN102111228B (en) Cognitive radio frequency spectrum sensing method based on circulation symmetry
CN104363078B (en) The real orthogonal space time packet blind-identification method of under determined system based on robust Competition Clustering

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant