CN108768445A - A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters - Google Patents
A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters Download PDFInfo
- Publication number
- CN108768445A CN108768445A CN201810586596.6A CN201810586596A CN108768445A CN 108768445 A CN108768445 A CN 108768445A CN 201810586596 A CN201810586596 A CN 201810586596A CN 108768445 A CN108768445 A CN 108768445A
- Authority
- CN
- China
- Prior art keywords
- chaos
- spread spectrum
- volterra
- sef
- vector
- 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
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/69—Spread spectrum techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
- H04B1/69—Spread spectrum techniques
- H04B1/707—Spread spectrum techniques using direct sequence modulation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/001—Modulated-carrier systems using chaotic signals
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Complex Calculations (AREA)
Abstract
The invention belongs to field of communication technology, specifically a kind of chaos spread spectrum factor blind estimating method based on sef-adapting filter.The present invention is substantially that convergent Volterra sef-adapting filters weights can occur acutely to be mutated this characteristic when symbol overturn using chaos spread spectrum sequence, carries out the detection that symbol overturning is put, and then estimates spreading factor this parameter.The method of the present invention has very strong universality, for different types of chaos spread spectrum sequence, can accurate blind estimate go out its spreading factor.The present invention still has very high estimation performance under low signal-to-noise ratio.
Description
Technical field
The invention belongs to field of communication technology, specifically a kind of chaos based on Volterra sef-adapting filters expands
Frequency factor blind estimating method.
Background technology
Chaotic communication such as chaos Direct Sequence Spread Spectrum Communication (abbreviation chaos direct expansion), pseudo-random sequence is replaced with chaos sequence
As spreading code, have compared with common-path interference and robustness, before there is extremely strong application in military radar, military radio station
Scape.Only in the meaning, the research of non-cooperation chaotic Signals Processing is just of great significance, thus is paid attention to increasingly.
Simultaneously compared with cooperative communication, almost without any prior information when non-partner is the problems such as facing noise, unknown parameter, because
And the increasingly complex difficulty of signal extraction, thus confidentiality is extremely strong.
Presently, there are two kinds of chaos direct expansion to decode algorithm --- and DM-UKF decodes method and IMM-UKF decodes method.The former applies
Generalized synchronization system, the different values based on binary message symbol establish corresponding Filtering Model and are fitted the chaos direct expansion intercepted
Signal solves information symbol value indirectly by two norms of error of fitting come the current Filtering Model of ruling.The latter is by chaos
Direct sequence signal is modeled as the output of the two subsystems worked alternatively as a result, and the different values of information symbol determine corresponding son
The kinetics-driven equation of system.Work at present system is judged again by error of fitting, to extracts information symbols value.
But both the above algorithm is all to know that spreading factor length (being called spreading gain), algorithm could be realized in advance,
There is presently no the algorithms of special blind estimate spreading factor, therefore set forth herein a kind of based on Volterra sef-adapting filters
Chaotic signal spreading factor blind estimating method.
Invention content
The purpose of the present invention decodes problem aiming at above-mentioned chaos direct expansion, provides a kind of adaptive based on Volterra
The chaotic signal spreading factor blind estimating method of filter.The present invention is overturn in symbol using chaos spread spectrum sequence
When convergent Volterra sef-adapting filters weights can occur acutely to be mutated this characteristic, carry out symbol overturning point
Detection, and then estimate spreading factor this parameter.The method of the present invention has very strong universality, for different types of mixed
Ignorant frequency expansion sequence, can accurate blind estimate go out its spreading factor.In addition, the present invention still have under low signal-to-noise ratio it is very high
Estimate performance.
The technical solution adopted in the present invention is：
S1：If the chaos spread spectrum signal that some communication receiving device receives isIt is normalized：
Wherein N is data length.
S2：To xnIt carries out phase space reconfiguration and obtains vector Un。
Phase space reconstruction technique allows people that can rebuild a Complex Nonlinear System from an individual time series
Holonomic Dynamics.This reconstruction is not perfect reconstruction truly, that is, the phase space and system after rebuilding are intrinsic
Phase space be not duplicate, but as long as phase space reconfiguration is appropriate, then phase space after reconstructing and original mutually empty
Between topological structure be consistent.Specifically, one group of measured value can be obtained by observation for a nonlinear system：
xn, n=1,2 ... N can construct one group of d dimensional vector using this measured value:
Un=[x (n), x (n- τ), x (n-2 τ) ..., x (n- (d-1) τ)], n=(d-1) τ+1 ... N
If parameter τ, d selection is appropriate, then UnOriginal system can be described.τ is known as time delay, and d is known as Embedded dimensions.By xn
Construct UnProcess be known as phase space reconfiguration.Specific steps include：
Step 1：Correlation method calculating time delay τ.First calculate chaos sequence auto-correlation function：
When the image that auto-correlation function R (τ) changes with delay time, arrived when under auto-correlation function:R(τ0)=(1-e-1)
When (0) R, obtained τ0The as delay time of phase space reconstruction.
Step 2：False proximal point algorithm calculates Embedded dimensions d.
In d ties up phase space, each vector yi(d)=(x (i) ..., x (i+ (d-1) τ)), 1≤i≤n- (d-1) τ
All there are one the closest point y of Euclidean distancen(i,d)(d), distance is：
Ri(d)=| | yi(d)-yn(i,d)(d)||2
When the dimension of phase space becomes d+1 from d, the distance of the two points will change, and new distance is Ri(d+
If 1) Ri(d+1) compare Ri(d) much larger, then can it is thought that due in multi-dimension Chaos attractor two it is non-conterminous
Caused by becoming on spot projection to low-dimensional coordinate at adjacent 2 points, such point of proximity is false, is enabled
ai(d)=| | Ri(d+1)-Ri(d)||/Ri(d)
Work as Ri(d) when tending towards stability, i.e. ai(d) it is less than the threshold value a of setting0When, it is believed that chaos attractor is beaten completely
It opens, d at this time is Embedded dimensions.
Step 3：According to τ, the d for calculating gained, x is usednConstruct vector Un。
S3：By UnAs the input signal vector of Volterra auto-adaptive fir filters, calculated using RLS algorithm
Volterra sef-adapting filter weights H=[w1n,w2n,...wmn...wMn], wherein M is filter order, wmnIndicate adaptive
Answer the value of m-th of weights at the n moment of filter.The detailed step of RLS algorithm is as follows：
Step 1：Initialization,δ is small positive number,For unit matrix；Weight vector initial value
Step 2：For k=1,2,3 ... N, λ are the number close to 1, carry out following interative computation：
S4：Appoint in H and takes a filter weights sequence wm, make difference processing：
dk=wmk+1-wmk, k=1,2 ... N-1
S5：Search for dkIn trip point：Set a thresholding d0, traverse dk, k=1,2,3...N-1,
If there is dk> d0, then record trip point index k and preserve, after having traversed, the index of whole trip points can be obtained：
K=[k1,k2,...kl]
S6：Calculate the interval of adjacent index：
Di=ki+1-ki, i=1,2 ... l-1
S7：Spreading factor blind estimate value is all DiThe greatest common divisor of (i=1,2 ... l-1)：
G=gcd { D1,D2,...Dl-1}
Beneficial effects of the present invention are that method of the invention has very strong universality, and different types of chaos is expanded
Frequency sequence, can accurate blind estimate go out its spreading factor.The present invention still has very high estimation performance under low signal-to-noise ratio.
Description of the drawings
Fig. 1 is the flow chart of realization process of the present invention；
Fig. 2 is that the embodiment of the present invention 1 generates chaos sequence time-domain diagram；
Fig. 3 is the figure that Volterra sef-adapting filters weights change over time in the embodiment of the present invention 1, a~f weight ws1
~w6The figure changed over time.
Specific implementation mode
The present invention is described in detail below with reference to the accompanying drawings and embodiments, preferably so as to those skilled in the art
Understand the present invention.
Embodiment 1：
The purpose of the present embodiment is to prove that chaos spread spectrum sequence is convergent when symbol is overturn
Volterra sef-adapting filters weights can occur acutely to be mutated this characteristic, this characteristic is the core place of algorithm.
Emulation platform is matlab, generates one section of logistic chaos sequence, sequence length N=1000, superposition first
White Gaussian noise makes signal-to-noise ratio be SNR=30dB, and preceding 500 point data is constant, and latter 500 points are multiplied by -1, obtained sequence such as Fig. 2
It is shown, obtained data are normalized, phase space reconfiguration is then carried out, the vector for then obtaining phase space reconfiguration
UnAs the input vector of Volterra sef-adapting filters, filter weights H, the filtering emulated are calculated with RLS algorithm
Device weight w1~w6Shown in the figure changed over time such as Fig. 3 (a~f)
It can see by simulation result Fig. 3, each weight w has all restrained before 500 moment, and at 500 to 501
Quarter is mutated.Thus conclusion is demonstrated, to which further trip point can be carried out to any weights to chaos spread spectrum sequence
Detection, realizes the blind estimate of spreading factor.
Embodiment 2：
The purpose of the present embodiment be under different signal-to-noise ratio and spreading factor to algorithm blind estimate accuracy in the present invention
Emulation, the validity of algorithm is illustrated with this.
Emulation platform is matlab, generates one section of logistic chaos spread spectrum sequence, sequence length N=first
200000, spreading factor G=63, superposition white Gaussian noise makes signal-to-noise ratio be SNR=3dB, according to the step in technical solution
S1-S7 carries out 100 Monte Carlo simulation experiments.Then modification spread spectrum G, the value of Signal to Noise Ratio (SNR) repeat to test.Simulation result
Statistics is as shown in table 1.By the Simulation result data of table, it can be appreciated that this algorithm also has higher blind under low signal-to-noise ratio
Estimate that accuracy, signal-to-noise ratio 20dB or more can reach 100%.
The different signal-to-noise ratio of table 1 and the lower 100 Monte Carlo Experiment accuracy of spreading factor
Claims (1)
1. a kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters, which is characterized in that including with
Lower step：
S1, set some communication receiving device reception chaos spread spectrum signal asIt is normalized：
Wherein N is data length；
S2, to xnIt carries out phase space reconfiguration and obtains vector Un：
According to measured value xn, n=1,2 ... N constructs one group of d dimensional vector：
Un=[x (n), x (n- τ), x (n-2 τ) ..., x (n- (d-1) τ)], n=(d-1) τ+1 ... N
τ is time delay, and d is Embedded dimensions, by xnConstruct UnProcess be known as phase space reconfiguration, specific steps include：
S21, using correlation method calculating time delay τ, first calculate chaos sequence auto-correlation function：
It is arrived when under auto-correlation function:R(τ0)=(1-e-1) (0) R when, obtained τ0The as delay time of phase space reconstruction；
S22, false proximal point algorithm calculate Embedded dimensions d：
In d ties up phase space, each vector yi(d)=(x (i) ..., x (i+ (d-1) τ)), 1≤i≤n- (d-1) τ has one
The closest point y of a Euclidean distancen(i,d)(d), distance is：
Ri(d)=| | yi(d)-yn(i,d)(d)||2
When the dimension of phase space becomes d+1 from d, the distance of the two points will change, and new distance is Ri(d+1), if
Ri(d+1) compare Ri(d) it is much larger, then it is assumed that this is because two non-conterminous spot projections are sat to low-dimensional in multi-dimension Chaos attractor
Caused by putting on become adjacent 2 points, such point of proximity is false, is enabled：
ai(d)=| | Ri(d+1)-Ri(d)||/Ri(d)
Work as Ri(d) when tending towards stability, i.e. ai(d) it is less than the threshold value a of setting0When, it is believed that chaos attractor is opened completely, this
When d be Embedded dimensions；
S23, τ, d obtained by calculating, use xnConstruct vector Un；
S3, by UnAs the input signal vector of Volterra auto-adaptive fir filters, Volterra is calculated certainly using RLS algorithm
Adaptive filter weights H=[w1n,w2n,...wmn...wMn], wherein M is filter order, wmnIndicate sef-adapting filter
In m-th of weights at n moment, the detailed step of RLS algorithm is as follows：
S31, initialization,δ is small positive number,For unit matrix；Weight vector initial value
S32, it is the number close to 1 for k=1,2,3 ... N, λ, carries out following interative computation：
S4, in H appoint take a filter weights sequence wm, make difference processing：
dk=wmk+1-wmk, k=1,2 ... N-1
S5, search dkIn trip point：Set a thresholding d0, traverse dk, k=1,2,3...N-1, if there is dk> d0, then record
Trip point indexes k and preserves, and after having traversed, the index of whole trip points can be obtained：
K=[k1,k2,...kl]
S6, the interval for calculating adjacent index：
Di=ki+1-ki, i=1,2 ... l-1
S7, spreading factor blind estimate value are all DiThe greatest common divisor of (i=1,2 ... l-1)：
Gain=gcd { D1,D2,...Dl-1}。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810586596.6A CN108768445A (en) | 2018-06-05 | 2018-06-05 | A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810586596.6A CN108768445A (en) | 2018-06-05 | 2018-06-05 | A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108768445A true CN108768445A (en) | 2018-11-06 |
Family
ID=63999582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810586596.6A Pending CN108768445A (en) | 2018-06-05 | 2018-06-05 | A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108768445A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263968A (en) * | 2019-05-07 | 2019-09-20 | 深圳大学 | A kind of prediction technique of new time series |
| WO2020224112A1 (en) * | 2019-05-07 | 2020-11-12 | 深圳大学 | Time series forecasting method employing training series model |
| US20210006390A1 (en) * | 2018-03-16 | 2021-01-07 | Yanhua JIAO | Method for Generating Digital Quantum Chaotic Wavepacket Signals |
-
2018
- 2018-06-05 CN CN201810586596.6A patent/CN108768445A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210006390A1 (en) * | 2018-03-16 | 2021-01-07 | Yanhua JIAO | Method for Generating Digital Quantum Chaotic Wavepacket Signals |
| US11509452B2 (en) * | 2018-03-16 | 2022-11-22 | Yanhua JIAO | Method for generating digital quantum chaotic wavepacket signals |
| CN110263968A (en) * | 2019-05-07 | 2019-09-20 | 深圳大学 | A kind of prediction technique of new time series |
| WO2020224112A1 (en) * | 2019-05-07 | 2020-11-12 | 深圳大学 | Time series forecasting method employing training series model |
| WO2020224111A1 (en) * | 2019-05-07 | 2020-11-12 | 深圳大学 | Multivariate time series prediction method |
| CN110263968B (en) * | 2019-05-07 | 2021-04-30 | 深圳大学 | New time series prediction method and device, electronic equipment and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103326971B (en) | A kind of small region search method based on LTE system and device thereof | |
| CN108768445A (en) | A kind of chaos spread spectrum factor blind estimating method based on Volterra sef-adapting filters | |
| Cattivelli et al. | Distributed nonlinear Kalman filtering with applications to wireless localization | |
| TWI575900B (en) | Methods and devices for channel estimation and ofdm receiver | |
| CN109450486A (en) | Asynchronous while co-channel full duplex underwater sound communication system number self-interference cancellation method | |
| CN102664841B (en) | Method for SC-FDE (single carrier-frequency domain equalization) system low complexity RLS self-adaption channel estimation | |
| CN107306242B (en) | Carrier phase recovery method and device | |
| CN110166080B (en) | Coherent multi-carrier capturing method and device | |
| Bordin et al. | Consensus-based distributed particle filtering algorithms for cooperative blind equalization in receiver networks | |
| CN104506465A (en) | Power line communication channel estimation method | |
| Garg et al. | Gradient descent approach for secure localization in resource constrained wireless sensor networks | |
| CN109388778A (en) | A kind of iteration volume point Unscented kalman filtering method | |
| CN104158512A (en) | self-adaptive sparse system identification method based on impact-interference-resistance of independent activity factor | |
| CN102340472B (en) | Method for generating ZC sequence of frequency domain and random access method based on ZC sequence | |
| CN112865891B (en) | PUCCH signal detection method, apparatus, communication device and storage medium | |
| McGuire | Channel estimation for secret key generation | |
| CN104469837A (en) | Method and terminal for determining reference signal receiving power and system | |
| CN105262706A (en) | Method and device for estimation and compensation of frequency offset | |
| Wang et al. | Improved model-based channel tracking for underwater acoustic communications | |
| Zhang et al. | DiPLoc: Direct signal domain particle filtering for network localization | |
| KR101231122B1 (en) | Method and apparatus for carrier frequency offset estimation using sage technique in ofdma system | |
| Arroyo-Valles et al. | Distributed wideband spectrum sensing for cognitive radio networks | |
| CN101989864B (en) | Blind multi-access interference suppression method for DS/CDMA system | |
| Tsiligkaridis et al. | Distributed probabilistic bisection search using social learning | |
| Fattah et al. | An algorithm for ARMA model parameter estimation from noisy observations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |