CN101900601B - Method for identifying direct sound in complex multi-path underwater sound environment - Google Patents
Method for identifying direct sound in complex multi-path underwater sound environment Download PDFInfo
- Publication number
- CN101900601B CN101900601B CN2010101389381A CN201010138938A CN101900601B CN 101900601 B CN101900601 B CN 101900601B CN 2010101389381 A CN2010101389381 A CN 2010101389381A CN 201010138938 A CN201010138938 A CN 201010138938A CN 101900601 B CN101900601 B CN 101900601B
- Authority
- CN
- China
- Prior art keywords
- threshold
- energy
- direct sound
- width
- thresholding
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention provides a method for identifying direct sound in a complex multi-path underwater sound environment, which comprises the steps of: (1) initializing a basic parameter which comprises a large width threshold, a small width threshold, a minimum energy threshold, a concave point compensating threshold and a time delay interval threshold; (2) computing an input signal envelope, wherein a self-adaptive Notch filter is adopted for narrowband filtering and self-adaptive iteration is used for estimating the signal envelope; (3) estimating background noise energy by using a recursive filter; determining an energy threshold by using a self-adaptive threshold and a minimum protective mechanism; (4) detecting energy and adopting a concave point compensating technique; and (5) identifying direct sound, namely judging by using double widths and time intervals. The method can be used for effectively enhancing the identifying rate of a direct sound signal in an environment polluted by reflected sound, effectively removing the interference of sharp pulses in background noise, reducing the detecting thresholds of a processor, and enhancing the universality and the stability.
Description
Technical field
The invention belongs to Underwater Acoustics Engineering, Signal and Information Processing field, be specifically related to the direct sound wave discrimination method under the environment of the many ways of the complicated underwater sound.
Background technology
The ocean is as the media of underwater signal transmission, is equivalent to carry out transmitting the wave filter of conversion.Because the existence of sea seabed interface and the unevenness of seawater medium; The underwateracoustic channel presents many ways characteristic; The acoustical signal of sending from sound source arrives acceptance point along various approach, and they interfere with each other stack, produces complex spatial interference figure and filtering characteristic.Receive the interference of sea, seabed one or many reflected sound; And the influence of relative motion; The amplitude of acceptance point place direct sound wave signal, frequency, phase place random variation; The serious distortion of waveform generation greatly differs from each other with former transmitting, and this has brought great difficulty for the reliable detection of direct sound wave and accurate identification.Therefore, the direct sound wave identification under the underwater sound environment of much more complicated way becomes one of significant challenge that Underwater Acoustics Engineering and signal Processing field face.
At present, the direct sound wave identification depends on many way inhibition technology more, and that uses always comprises diversity technique, spread spectrum technic, channel equalization technique and channel error correction coding etc.The starting point that each method suppresses many ways is different, and preceding two kinds of technology are through reducing the correlativity of multi-path signals; Channel equalization technique is the wave form distortion that causes through the many ways of compensate for channel effect, and the channel Error Correction of Coding is to use the redundant data that adds in the emission information to carry out error correction.In recent years, occurred mirror technology time reversal again, it is to utilize its space-focusing and Time Compression characteristic, the many ways spread signal that receives is compressed in time and is similar to former transmitting.
Said method is conceived to reduce the correlativity of multi-path signals or improve signal waveform through the pre-service of the waveform optimization that transmits, coding structure design and reception signal more.These technology all are from suppressing the angles on many ways, and do not make full use of the characteristic of multi-path signals.
Summary of the invention
The object of the present invention is to provide a kind of identifying direct sound in complex multi-path underwater sound environment method based on the multi-path signals signature analysis.
The objective of the invention is to realize like this, mainly comprise the steps:
(1) initialization basic parameter, said basic parameter comprise big width threshold, little width threshold, least energy thresholding, depression points compensation thresholding, time delay spacing thresholding;
(2) calculate the input signal envelope, adopt adaptive N otch wave filter to do narrow-band filtering, and utilize its adaptive iteration power estimated signal envelope;
(3) the estimating background noise comprising energy is confirmed energy threshold, and noise energy estimates to adopt the single order regressive filter, and energy threshold adopts adaptive threshold and minimum protection mechanism to choose;
(4) energy measuring, and adopt the depression points compensation technique;
(5) dual width judgement and time interval judgement are promptly adopted in direct sound wave identification.
The present invention can also comprise:
1, the concrete grammar of dual width judgement of said employing and time interval judgement is: the width threshold of the wide device that reflects is made as two kinds of sizes; During the width judgement,, be called broad pulse and burst pulse respectively no matter all temporarily keep pulse information greater than big width threshold or greater than little width threshold; Carry out the delay inequality judgement then, for broad pulse, if the individual pulsewidth that transmits of one of which has burst pulse within the time, think that then the two is a complete direct sound wave signal, last impulse time delay is the direct sound wave due in, otherwise broad pulse is a direct sound wave; For burst pulse,, think that then former and later two burst pulses form a complete direct sound wave, otherwise think that last burst pulse is a false-alarm, when prepulse is new burst pulse if its previous pulsewidth that transmits has burst pulse near the time.
2, said energy threshold is selected adaptive threshold for use, promptly obtains through real-time estimating background noise comprising intensity and according to the N-P criterion; Choose a least energy thresholding simultaneously, when adaptive threshold during greater than this thresholding, energy threshold is taken from the adaptation thresholding, otherwise, get the least energy thresholding.
3, described depression points compensation technique; Main contents comprise: receive noise effect; Signal energy the localized indentation trapping spot occurs in the part sampling instant; When doing energy judgement with the width counting, if less than M sampled point only energy threshold still carry out width totally, wherein M is that depression points compensates number.
Energy of the present invention/width joint-detection thought; Its main contents are: adopt energy detector and the wide device of mirror to carry out cascading judgement; Promptly have only when a signal energy N continuous time-sampling point surpasses energy threshold simultaneously and think that just signal is effective; Wherein N is the wide device width threshold of mirror, and its setting is relevant with factors such as energy threshold, signal pulsewidth, wave filter broadening, the relative broadenings of Doppler.The point that this detection thought can effectively be removed in the ground unrest dashes interference, reduces the processor detection threshold.
Anti-many way direct sound wave identification strategies of the present invention; Its core concept is to direct sound wave and reflected sound stack; Particularly under the situation of the two reverse stack, the direct sound wave signal divides, the forward and backward edge after the time domain waveform distortion; Its time interval is relevant with the pulsewidth that transmits, and the direct sound wave identification adopts dual width and front and back along delay inequality cascading judgement mechanism.Main contents are: the width threshold of the wide device that reflects is made as two kinds of sizes, and big width threshold is decided according to the pulse width that transmits, and little width threshold is decided according to the rule of direct sound wave, reflected sound delay inequality.During the width judgement,, be called broad pulse and burst pulse respectively no matter all temporarily keep pulse information greater than big width threshold or greater than little width threshold.Carry out the delay inequality judgement then, for broad pulse, pulsewidth has burst pulse within the time if one of tool neighborhood transmits, and thinks that then the two is a complete direct sound wave signal, and last impulse time delay is the direct sound wave due in, otherwise broad pulse is a direct sound wave; For burst pulse,, think that then former and later two burst pulses form a complete direct sound wave, otherwise think that last burst pulse is a false-alarm, when prepulse is new burst pulse if its previous pulsewidth that transmits has burst pulse near the time.
Adaptive threshold of the present invention and minimum threshold protection mechanism; Its main contents are: for adapting to the variation of neighbourhood noise under the different operating modes; Improve the input ability, energy threshold is selected adaptive threshold for use, promptly obtains through real-time estimating background noise comprising intensity and according to the N-P criterion.Simultaneously, for guaranteeing the stability of system, choose a least energy thresholding, when adaptive threshold during greater than this thresholding, energy threshold is taken from the adaptation thresholding, otherwise, get the least energy thresholding.This method is in Project Realization, and is very effective when especially there is round-off error in computation process.
Depression points compensation technique of the present invention, its main contents comprise: receive noise effect, signal energy the localized indentation trapping spot occurs in the part sampling instant; When these points are relatively adjudicated with energy threshold; Only thresholding causes thinking by mistake to be that spike pulse is disturbed, and occurs failing to report.For fear of the generation of this situation, doing width when counting, if less than M sampled point only energy threshold still carry out width accumulative total, wherein M is that depression points compensates number.
Description of drawings
Fig. 1 is certain sea area actual measurement velocity of sound profile figure;
Fig. 2 is many ways, diverse location place channel characteristics;
Fig. 3 arrives at acceptance point delay inequality change curve for different distance place direct sound wave and reflected sound;
Fig. 4 is typical direct sound wave and reflected sound stack time domain waveform figure;
Fig. 5 adaptive narrow-band wave filter theory diagram;
Fig. 6 is typical direct sound wave and reflected sound superposed signal envelope diagram;
Fig. 7 energy threshold is chosen logical flow chart;
Fig. 8 estimates for ground unrest and energy threshold is chosen the result;
Fig. 9 is the real-time processing flow chart of direct sound wave identification expert system.
Embodiment
For example the present invention is done description in more detail below in conjunction with accompanying drawing:
Basic parameter transmits: CW pulse, pulse width 10ms, centre frequency 5kHz; Transmitting arrives acceptance point through underwater acoustic channel, and channel velocity of sound distribution characteristics is as shown in Figure 1.The launching site position is (0,0,5) m, and the acceptance point degree of depth is 30m, and the two horizontal range changes from 0m～2km.Different distance position many ways channel characteristics and direct sound wave and reflected sound delay inequality such as Fig. 2, shown in Figure 3.Fig. 4 has provided that direct sound wave receives the time domain waveform (normalization amplitude) after reflected sound disturbs under the channel of typical many ways.
To above-mentioned condition, provide the expert system treatment step of this specific embodiment:
(1.1) expert system basic parameter initialization mainly comprises big width threshold, little width threshold, least energy thresholding, depression points compensation thresholding, time delay spacing thresholding etc.
1, big width threshold Width_L=8ms;
2, little width threshold Width_S=0.5ms;
3, least energy thresholding Amp_S=0.2;
4, depression points compensation thresholding Notch_N=5;
5, time delay spacing thresholding Delay_L=10.5ms, Delay_S=9.5ms.
(1.2) calculate the input signal envelope.Adopt adaptive N otch wave filter to do narrow-band filtering and envelope estimation, the current demand signal energy is an envelope square.The wave filter theory diagram is as shown in Figure 5.
Among the figure, d (k)=s (k)+n (k) is expectation signal time sequence, k=1, and 2 ..., L is the time sequence number, and s (k) is for receiving acoustical signal, and n (k) is a ground unrest; x
s(k) and x
c(k) be the input of two-way orthogonal reference.u
c(k), u
s(k) be iteration power; ε (k) is residual error output; Y (k) is signal output.
If the study step-length of this wave filter is constant μ, centre frequency ω
0, the amplitude of orthogonal reference input is A, then two-way is expressed as with reference to input:
x
c(k)＝Acos(ω
0k)，x
s(k)＝Asin(ω
0k) (1)
So, the system self-adaption iterative process can be expressed as:
y(k)＝u
c(k)x
c(k)+u
s(k)x
s(k) (2)
ε(k)＝d(k)-y(k) (3)
u
c(k+1)＝u
c(k)+με(k)x
c(k) (4)
u
s(k+1)＝u
s(k)+με(k)x
s(k) (5)
The input signal envelope does
Filter parameter is selected:
Centre frequency ω
0Be 5kHz;
Bandwidth is 200Hz;
Orthogonal reference input range A is 1.
Typical case's acoustical signal envelope is as shown in Figure 6.
(1.3) estimating background noise comprising energy, energy threshold adopt adaptive threshold and minimum protection mechanism to choose.
The ground unrest Energy Estimation adopts the single order regressive filter to make long-time integration.This method calculated amount is little, has function time memory, is convenient to real-time implementation.Recurrence formula is following:
Wherein:
---k sampling instant background noise levels estimated value;
M---time constant is got 2048 points;
S (k)---k sampling instant input noise intensity.
Usually; Energy threshold is set to 9 times ground unrest energy, but in order to guarantee the stability of system, just current demand signal intensity is only participated in the estimation of ground unrest less than energy threshold the time; And when adaptive threshold during less than the least energy thresholding, energy threshold is got the least energy thresholding.Concrete logic sees Fig. 7 for details.
It is as shown in Figure 8 that ground unrest estimation and energy threshold are chosen the result.
(1.4) energy judgement.The current time signal energy of step (1.2) acquisition and the energy threshold of step (1.3) acquisition are compared, if greater than energy threshold, then the width counter adds 1, changes step (1.2) then over to and carries out the processing of next sampled point.Otherwise, change step (1.5) over to.
(1.5) dual width judgement.After signal energy is less than energy threshold, at first, then be judged to broad pulse greater than Width_L with width counter and Width_L ratio; Less than then comparing with Width_S again; If greater than the just temporary transient pulse information of preserving of Width_S, be defined as burst pulse, change step (1.6) then over to; Otherwise directly be judged to glitch, and to the associated counter zero clearing.
(1.6) time interval judgement.After detecting effective impulse; If its front has burst pulse to occur, for being the situation of broad pulse when prepulse, whether the delay inequality of then judging wide, burst pulse is less than the minimum interval thresholding; Less than thinking that then the two is a complete direct sound wave; The burst pulse time delay is the due in of direct sound wave signal, otherwise thinks that burst pulse is a false-alarm, and broad pulse is a direct sound wave.For working as prepulse is the situation of burst pulse; Judge when time interval of prepulse and last burst pulse whether between maximum, minimum interval thresholding; Satisfy this condition and can judge that then last burst pulse is effective; Whole signal is a complete direct sound wave, otherwise thinks that last burst pulse is a false-alarm, when prepulse is new burst pulse.
Direct sound wave identification expert system process flow diagram is as shown in Figure 9.
What should explain at last is; Above embodiment appearance is in order to describe technical scheme of the present invention rather than the present technique method is limited; The present invention can extend to other modification, variation, application and embodiment on using, and therefore thinks that all such modifications, variation, application, embodiment are in spirit of the present invention and teachings.
Claims (1)
1. identifying direct sound in complex multi-path underwater sound environment method is characterized in that:
(1) initialization basic parameter, said basic parameter comprise big width threshold, little width threshold, least energy thresholding, depression points compensation thresholding, time delay spacing thresholding;
(2) calculate the input signal envelope, adopt adaptive N otch wave filter to do narrow-band filtering, and utilize its adaptive iteration power estimated signal envelope;
(3) the estimating background noise comprising energy is confirmed energy threshold, and noise energy estimates to adopt the single order regressive filter, and energy threshold adopts adaptive threshold and minimum protection mechanism to choose; It is to obtain through real-time estimating background noise comprising intensity and according to the N-P criterion that said energy threshold is selected adaptive threshold for use; Choose a least energy thresholding simultaneously, when adaptive threshold during greater than this thresholding, energy threshold is taken from the adaptation thresholding, otherwise, get the least energy thresholding;
(4) energy measuring, and adopt the depression points compensation technique; Promptly when doing energy judgement with the width counting, if less than M sampled point only energy threshold still carry out width totally, wherein M is that depression points compensates number;
(5) dual width judgement and time interval judgement are promptly adopted in direct sound wave identification, and concrete grammar is: the width threshold of the wide device that reflects is made as two kinds of sizes; In width when judgement, no matter still all temporarily keep pulse information greater than little width threshold greater than big width threshold, be called broad pulse greater than big width threshold, greater than little width threshold, be called burst pulse less than width threshold greatly; Carry out the delay inequality judgement then, for broad pulse, if the individual pulsewidth that transmits of one of which has burst pulse within the time, think that then the two is a complete direct sound wave signal, last impulse time delay is the direct sound wave due in, otherwise broad pulse is a direct sound wave; For burst pulse,, think that then former and later two burst pulses form a complete direct sound wave, otherwise think that last burst pulse is a false-alarm, when prepulse is new burst pulse if its previous pulsewidth that transmits has burst pulse near the time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101389381A CN101900601B (en) | 2010-04-02 | 2010-04-02 | Method for identifying direct sound in complex multi-path underwater sound environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101389381A CN101900601B (en) | 2010-04-02 | 2010-04-02 | Method for identifying direct sound in complex multi-path underwater sound environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101900601A CN101900601A (en) | 2010-12-01 |
| CN101900601B true CN101900601B (en) | 2012-02-01 |
Family
ID=43226335
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101389381A Expired - Fee Related CN101900601B (en) | 2010-04-02 | 2010-04-02 | Method for identifying direct sound in complex multi-path underwater sound environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101900601B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102636785B (en) * | 2012-04-06 | 2013-08-28 | 哈尔滨工程大学 | Submarine target three-dimensional positioning method |
| CN102636773B (en) * | 2012-04-06 | 2013-08-28 | 哈尔滨工程大学 | Single-element range ambiguity resistant method based on channel multipath characteristic |
| CN103152299B (en) * | 2013-03-04 | 2016-03-09 | 哈尔滨工程大学 | A kind of strong interference suppression method being applicable to cooperative work of offshore multi-acoustic system |
| CN104535989B (en) * | 2015-01-09 | 2017-06-20 | 哈尔滨工程大学 | The acoustical signal decomposition method of way more than shallow sea |
| CN106546955B (en) * | 2016-10-14 | 2017-10-03 | 中国船舶工业系统工程研究院 | A kind of method that anti multi path interference applied to marine geophysical prospecting obtains the acoustical signal through time |
| CN107656243A (en) * | 2017-08-25 | 2018-02-02 | 天津大学 | Combine DOA/TOA oceans multi-path environment localization method in inhomogeneous medium |
| CN109506767B (en) * | 2018-10-24 | 2020-12-08 | 西北工业大学 | Real-time detection method for sound field abnormity caused by underwater invasion target |
| CN109687915A (en) * | 2018-12-28 | 2019-04-26 | 哈尔滨工程大学 | A kind of method of unknown parameter underwater acoustic pulse signal detection |
| CN110687197B (en) * | 2019-08-26 | 2021-12-24 | 天津大学 | Self-adaptive spacecraft fragment collision positioning method |
| CN110826216B (en) * | 2019-11-01 | 2022-09-16 | 哈尔滨工程大学 | Decision tree-based underwater direct sound selection method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0453293A3 (en) * | 1990-04-19 | 1992-12-02 | Matsushita Electric Industrial Co., Ltd. | Sound field variable apparatus |
| RU2116632C1 (en) * | 1997-08-26 | 1998-07-27 | Александр Сергеевич Мирсков | Method of directed measurement of acoustic signals of sound source ( versions ) |
| CN1971649A (en) * | 2006-12-01 | 2007-05-30 | 哈尔滨工程大学 | Positioning and communicating integral buoy |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5024613B2 (en) * | 2007-08-08 | 2012-09-12 | 清水建設株式会社 | Sound field analyzer |
-
2010
- 2010-04-02 CN CN2010101389381A patent/CN101900601B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0453293A3 (en) * | 1990-04-19 | 1992-12-02 | Matsushita Electric Industrial Co., Ltd. | Sound field variable apparatus |
| RU2116632C1 (en) * | 1997-08-26 | 1998-07-27 | Александр Сергеевич Мирсков | Method of directed measurement of acoustic signals of sound source ( versions ) |
| CN1971649A (en) * | 2006-12-01 | 2007-05-30 | 哈尔滨工程大学 | Positioning and communicating integral buoy |
Non-Patent Citations (4)
| Title |
|---|
| JP特开2009-42043A 2009.02.26 |
| 付进等.界面反射对定位系统性能影响及应对策略研究.《兵工学报》.2009,第30卷(第1期),全文. * |
| 刘文海等.水下被动定位声纳显示及航迹跟踪方法.《声学技术》.2010,第29卷(第1期),全文. * |
| 生雪莉等.垂直阵时反被动定位方法及其改进技术研究.《声学技术》.2009,第28卷(第2期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101900601A (en) | 2010-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101900601B (en) | Method for identifying direct sound in complex multi-path underwater sound environment | |
| Preisig et al. | Surface wave focusing and acoustic communications in the surf zone | |
| CN105634543B (en) | A kind of anti-narrowband interference method and system | |
| CN101533091B (en) | Space-time two-dimensional narrow band barrage jamming method | |
| CN101179283A (en) | Method of detecting and suppressing pulse noise | |
| CN106936514B (en) | Underwater acoustic preamble signal detection method based on energy concentration under sparse channel | |
| CN102353952A (en) | Line spectrum detection method by coherent accumulation of frequency domains | |
| CN102183741B (en) | Method for realizing high frame rate unambiguous positioning in long baseline asynchronous acoustic positioning system | |
| CN106879068A (en) | The arrival time method of estimation of signal under a kind of strong multi-path environment | |
| Huang et al. | GPS signal detection under multiplicative and additive noise | |
| CN105527614A (en) | LFM pulse radar anti-DRFM-interference method based on moving target detection | |
| Su et al. | Fast estimation of underwater acoustic multipath channel based on LFM signal | |
| CN102571671B (en) | Modified smoothed pseudo Wigner-Ville distribution-based (MSPWVD-based) blind estimation method for pseudo code sequence of pseudo-random Bi-phase code-linear frequency modulation (PRBC-LFM) composite signal | |
| Cui et al. | Timing estimation of multiple hyperbolic frequency‐modulated signals based on multicarrier underwater acoustic communication | |
| CN105356907A (en) | Underwater acoustic communication method based on cyclic shift energy detection of time reversal mirror | |
| US7239580B2 (en) | Noise adaptive sonar signal processor | |
| CA2528598A1 (en) | Estimation of background noise and its effect on sonar range estimation | |
| Shimura et al. | Demonstration of time-reversal communication combined with spread spectrum at the range of 900 km in deep ocean | |
| CN110673118A (en) | Active sonar single-frequency pulse train waveform design and detection algorithm | |
| FR3028960A1 (en) | TELEMETRY METHOD, TELEMETRY DEVICE, LOCATION DEVICE, AND LOCATION METHOD | |
| Mao et al. | Phase-modulated waveform design for the target detection in the presence of signal-dependent clutter | |
| CN103152299B (en) | A kind of strong interference suppression method being applicable to cooperative work of offshore multi-acoustic system | |
| Chen et al. | Research on FRFT-PPM underwater acoustic communication system | |
| CN113138372B (en) | Radar target detection method based on improved M/N detector in multipath environment | |
| CN103901470A (en) | Seabed settled layer reflection strength and time delay estimation 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120201 Termination date: 20170402 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |