Feature of Echo Envelope Fluctuation and Its Application in the Discrimination of Underwater Real Echo and Synthetic Echo
Feature of Echo Envelope Fluctuation and Its Application in the Discrimination of Underwater Real...
Chen, Yunfei;Li, Sheng;Jia, Bing;Li, Guijuan;Wang, Zhenshan
2018-08-09 00:00:00
applied sciences Article Feature of Echo Envelope Fluctuation and Its Application in the Discrimination of Underwater Real Echo and Synthetic Echo 1 , 2 , 1 2 2 2 Yunfei Chen *, Sheng Li , Bing Jia , Guijuan Li and Zhenshan Wang School of Naval Architecture, Dalian University of Technology, Dalian 116024, China; shengli@dlut.edu.cn Science and Technology on Underwater Test and Control Laboratory, Dalian 116013, China; jbrobin@sina.com (B.J.); 18641193261@163.com (G.L.); wangzhenshanmobile@163.com (Z.W.) * Correspondence: yunfeidlut@163.com; Tel.: +86-0411-8267-7308 (ext. 803) Received: 3 June 2018; Accepted: 7 August 2018; Published: 9 August 2018 Abstract: Discriminating a real underwater target echo from a synthetic echo is a key challenge to identifying an underwater target. The structure of an echo envelope contains information which closely relates to the physical parameters of the underwater target, and the characterization and extraction of echo features are problematic issues for active sonar target classification. In this study, firstly, the high-frequency envelope fluctuation of a complex underwater target echo was analyzed, the envelope fluctuation was characterized by the envelope fluctuation intensity, and a characterization model was established. The features of a benchmark model echo were extracted and analyzed by theoretical simulation and sea testing of a scaled model, and the result shows that the envelope fluctuation intensity varies with carrier frequency and azimuth of incident signal, but the echo envelope fluctuation of the synthetic target echo does not present these features. Then, based on the characteristics of echo envelope fluctuation, a novel method was developed for active sonar discrimination of a real underwater target echo from the synthetic echo. Through a sea experiment, the real target echo and synthetic echo were classified by their different echo envelope fluctuations, and the feasibility of the method was verified. Keywords: active sonar; underwater; envelope; envelope fluctuation intensity; synthetic echo; echo discrimination 1. Introduction An active acoustic decoy which can simulate a target echo provides very important countermeasures to active acoustic torpedoes and active sonar. Basically, the active acoustic decoy comprises several monostatic hydrophones, and once the hydrophones receive the incident sonar pulse, the monostatic hydrophones transmit an acoustic wave with a certain phase, magnitude, Doppler frequency shift, time delay, and signal broadening that characterize the target echo [1–4]. This kind of active acoustic decoy can simulate a series of an echo’s characteristic parameters, such as the relative positions of the main echo highlights, relative magnitude of echo highlights, and scale features of the target, and it is a great challenge for active sonar target classification. Many algorithms have been developed for underwater acoustic target tracking [5–8], but discriminating the real underwater target echo from the synthetic echo is still a key challenge to identifying an underwater target. The echo of an underwater target is the incident signal modulated by material, structure, and shape parameters, so the echo envelope structure is the characterization of the target’s geometric scattering and elastic scattering in a time domain [9–12]. Therefore, the envelope structure of an underwater echo contains the information which closely relates to the physical parameters of the target. In a high-frequency acoustic scattering wave, the echo’s energy is dominated Appl. Sci. 2018, 8, 1329; doi:10.3390/app8081329 www.mdpi.com/journal/applsci Appl. Sci. 2018, 8, 1329 2 of 17 by a geometrical scattering wave, the geometrical scattering highlights are always present in the high-amplitude fluctuation of an echo envelope, while the elastic scattering highlights are present in the fast and weak fluctuation of the envelope [13]. For the high-amplitude fluctuation of an echo envelope, the envelope’s structural features can be characterized by the number of echo highlights, relative magnitude, and relative spacing [14–18]. A statistical quantitative model of echo highlights has been established for a complex target, which quantitatively characterizes the echo highlights [19]. For the weak fluctuation of the echo envelope, which contains the elastic scattering, for an actual echo signal, the features of weak fluctuation echo envelope are hard to extract and use [13]. In previous work, echo envelope fluctuation features were studied in a frequency domain, and the envelope modulation rate of underwater target’s scattering signal is characterized. Theoretical and experimental research has indicated that the maximum envelope modulation frequency of a complex target increases when the incident frequency increases [20]. Based on the maximum envelope modulation frequency, a method for discriminating the real target from the synthetic target was proposed and tested [21]. The maximum envelope modulation frequency was obtained by using a threshold for the strong influence of low-frequency envelope modulation. The extraction of maximum frequency is always affected by the threshold, and the correct choice of the threshold is often not easy for actual sea data. Consequently, in order to deeply understand the mechanism of echo envelope modulation and to properly describe the echo envelope fluctuation, the authors studied the features of echo envelope fluctuation in a time domain, and they proposed a novel method for the discrimination of a real underwater target echo from a synthetic echo. Compared with previous work, the strong influence of a low-frequency echo envelope fluctuation was reduced, the fine structure of a high fluctuation envelope was enhanced, and the processing of the features extracted from a high-frequency modulation was straightforward. The major contributions of this paper are as follows: This paper focuses on the characterization and extraction of echo fluctuation features from an underwater complex target. Based on practical engineering applications, it presents a method for discrimination of a real echo from a synthetic echo underwater. The high-frequency fluctuation of an echo envelope was characterized by the echo envelope fluctuation intensity, and a model of echo envelope fluctuation intensity was established. Results from simulation and real sea experiments of a benchmark model and synthetic echo are provided in detail. The feasibility of the proposed method for discriminating between a real target echo and a synthetic echo was verified by real sea experiment. The method proposed in this paper has low processing complexity and provides a new and valuable insight into the classification of underwater real and synthetic echoes. The rest of this paper is organized as follows. Section 2 describes the theoretical analysis of the high-frequency envelope fluctuation of a complex underwater target echo, and the characterization model we established is presented. Section 3 presents the extraction of echo envelope fluctuation features from a scaled benchmark model and synthetic echo with different frequencies and pulse lengths, and the simulation and experimental results are discussed in detail. Section 4 describes the proposed method of real echo and synthetic echo discrimination. Section 5 consists of sea experiments, data processing results, and discussion of estimation performance. A summary of the proposed method is provided in Section 6. 2. Characterization of Echo Envelope Fluctuation Features 2.1. Relation of Echo Envelope Fluctuation with Carrier Frequency of Incident Pulse Normally, based on the echo highlight model, an echo is considered the superposition of echo highlights and background. In this study, the concept underlying the echo highlight model has been Appl. Sci. 2018, 8, 1329 3 of 17 generalized: the echo is the supposition of highlights from scatterings on the target, and the scatterings are equally spaced according to the carrier frequency of the incident pulse. For the linear dimension, the complex echo can be expressed by [20] y(t, q) = a (q) x(t). exp[j(2p f (i 1)t )] (1) i q i=1 where q is the azimuth of the incident signal; a (q) is the strength coefficient of the echo highlight, assumed equal spaced distribution of scatter on the target; x(t) is the incident signal; N is the number of echo highlights; t is the difference in time between the echo highlight and the start of the echo signal, i.e., L cos(q) t = (2) (N 1)C where L is the length of the target, C is the sound speed in water. Assuming equal strength coefficients of echo highlights, namely a (q) = a(q), the frequency domain expression of (1) is as below [21] s( f , q) = a(q).X( f ). å exp(j2p f (i 1)t ) (3) i=1 = a(q).X( f ).h( f , q) where h( f , q) = exp(j2p f (i 1)t ) i=1 (4) sin(p f Nt ) sin(p f t ) In (3) and (4), h( f , q) receives the maximum value when f t = m, (m = 1, 2, 3, . . .). The echo envelope forms dips and peaks correspondingly. As per the theoretical expression above, the spacing of echo highlights relates to the carrier frequency, and when the carrier frequency increases, a higher frequency modulation of the echo envelope forms, presenting more detailed information on the target. It also implies that when the carrier frequency changes, the characteristics of the echo envelope modulation change accordingly, representing the features of a real complex target. 2.2. Characterization of Underwater Target Echo Envelope Fluctuation Intensity A high-frequency fluctuation of the echo envelope represents the fine features of an underwater complex target echo, but the energy of a low-frequency fluctuation of the echo envelope dominates the whole echo envelope. Normally, extracting the features from a high-frequency fluctuation of the echo envelope is not easy. The maximum fluctuation frequency could be used to characterize the high-frequency envelope fluctuation and its features extracted in the frequency domain [20], but the maximum envelope modulation frequency is obtained by using a threshold, and correctly choosing the threshold is difficult for actual sea data. In this work, the extraction of features from a high-frequency fluctuation of the echo envelope was carried out in the time domain. The intensity of the magnitude of the envelope signal fluctuates with time, and it can be characterized by a differential coefficient, through differentiation of the echo envelope. The low-frequency envelope fluctuation is suppressed, equivalent to high-pass filtering, and the high-frequency envelope fluctuation is enhanced. For convenient analysis in the time domain, Equation (1) is expressed as y (t, q) = a (q). exp[j2p f (t t )] (5) 1 å i i i=1 t = (i 1)t (6) i q Appl. Sci. 2018, 8, 1329 4 of 17 Deriving the formula, with time (t) as a variable, the echo envelope fluctuation intensity is y (t, q) = a (q) j2p f exp[j2p f (t t )] (7) 2 å i i i=1 y (t, q) = 2p f a (q) exp[j2p f (t t )] (8) 2 å i i i=1 y (t, q) = 2p fjy (t, q)j (9) 2 1 From this expression, the echo envelope fluctuation intensity strongly relates to the carrier frequency of the incident pulse, and the echo envelope fluctuation intensity varies with the carrier frequency accordingly. The envelope fluctuation intensity is a time-changing value, and it cannot be used directly for discriminating between different echo signals. In this paper, the feature is characterized by two parameters, one being fluctuation intensity K(q) = jk j (10) n=1 k = X(n) X(n 1) n = 1, 2, 3 . . . . . . M (11) where X is the discrete series of y (t, q), k is the difference between adjacent elements of X, and M is the data length. K(q) is called the fluctuation intensity, which characterizes the whole fluctuation intensity of the echo envelope. Another parameter is the standard deviation (STD) of echo envelope magnitude fluctuation intensity, " # S(q) = (k E) (12) n=1 where E = k(n) (13) n=1 where S(q) characterizes the inconsistency of the fluctuation intensity of the envelope that varies with time. 3. Extraction of Echo Envelope Fluctuation Features 3.1. Simulation Study 3.1.1. Simulation Study of Benchmark Based on the analysis above, the envelope fluctuation intensity was studied through simulation, where the target studied is the benchmark model [22] scaled by a ratio of 1:20, the length is 3 m, and the scaled model is made of steel. The incident pulse is a linear frequency modulation (LFM) signal, the frequencies are 10–20 kHz and 20–40 kHz, and the pulse lengths are 1 ms and 3 ms. The method of echo simulation is called the frequency indirect echo simulation method. The acoustic scattering of the target can be considered a linear system or target scattering channel, the incident signal is the system input, and the echo is the system output that is transferred by the target system and underwater sound channel. In this study, the transfer function of the target was calculated using special software that is based on the planar element method (PEM)—a numerical model that converts an integral calculation to arithmetic calculation [23,24]. The software can rapidly calculate sonar echo characteristics from sonar targets of various shapes [25]. Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 17 The method of echo simulation is called the frequency indirect echo simulation method. The The method of echo simulation is called the frequency indirect echo simulation method. The acoustic scattering of the target can be considered a linear system or target scattering channel, the acoustic scattering of the target can be considered a linear system or target scattering channel, the incident signal is the system input, and the echo is the system output that is transferred by the target incident signal is the system input, and the echo is the system output that is transferred by the target system and underwater sound channel. In this study, the transfer function of the target was calculated system and underwater sound channel. In this study, the transfer function of the target was calculated using special software that is based on the planar element method (PEM)—a numerical model that Appl. Sci. 2018, 8, 1329 5 of 17 using special software that is based on the planar element method (PEM)—a numerical model that converts an integral calculation to arithmetic calculation [23,24]. The software can rapidly calculate converts an integral calculation to arithmetic calculation [23,24]. The software can rapidly calculate sonar echo characteristics from sonar targets of various shapes [25]. sonar echo characteristics from sonar targets of various shapes [25]. Figures 1–3 are the simulated echo envelope fluctuation features. Figures 1a and 2a are the Figures 1–3 are the simulated echo envelope fluctuation features. Figures 1a and 2a are the echo Figures 1–3 are the simulated echo envelope fluctuation features. Figures 1a and 2a are the echo echo envelope fluctuation intensities with different pulse lengths, and Figures 1b and 2b are the envelope fluctuation intensities with different pulse lengths, and Figures 1b and 2b are the envelope fluctuation intensities with different pulse lengths, and Figures 1b and 2b are the corresponding STDs of echo envelope fluctuation with different pulse lengths. corresponding STDs of echo envelope fluctuation with different pulse lengths. corresponding STDs of echo envelope fluctuation with different pulse lengths. The simulation results show that the fluctuation intensity and its standard deviation change The simulation results show that the fluctuation intensity and its standard deviation change The simulation results show that the fluctuation intensity and its standard deviation change dramatically with the azimuth, with the minimum intensity on the beam, and the maximum on dramatically with the azimuth, with the minimum intensity on the beam, and the maximum on the dramatically with the azimuth, with the minimum intensity on the beam, and the maximum on the the keel-line. The cause is that the time difference of the echo highlight in the keel-line is greater keel-line. The cause is that the time difference of the echo highlight in the keel-line is greater than that keel-line. The cause is that the time difference of the echo highlight in the keel-line is greater than that than that in the beam direction; correspondingly, the phase difference is larger than that causing the in the beam direction; correspondingly, the phase difference is larger than that causing the strong in the beam direction; correspondingly, the phase difference is larger than that causing the strong strong fluctuation of the echo envelope. However, in the beam direction, the phases of different echo fluctuation of the echo envelope. However, in the beam direction, the phases of different echo fluctuation of the echo envelope. However, in the beam direction, the phases of different echo highlights are nearly the same, and the envelope is flatter than that in other azimuths. highlights are nearly the same, and the envelope is flatter than that in other azimuths. highlights are nearly the same, and the envelope is flatter than that in other azimuths. Figure 3 compares the fluctuation intensity features of the echo envelope for different frequencies, Figure 3 compares the fluctuation intensity features of the echo envelope for different Figure 3 compares the fluctuation intensity features of the echo envelope for different 10–20 kHz and 20–40 kHz, with 1 ms pulse lengths. Figure 3a is the comparison of fluctuation intensity, frequencies, 10–20 kHz and 20–40 kHz, with 1 ms pulse lengths. Figure 3a is the comparison of frequencies, 10–20 kHz and 20–40 kHz, with 1 ms pulse lengths. Figure 3a is the comparison of and Figure 3b is the comparison of STDs of echo envelope fluctuation intensity, correspondingly. fluctuation intensity, and Figure 3b is the comparison of STDs of echo envelope fluctuation intensity, fluctuation intensity, and Figure 3b is the comparison of STDs of echo envelope fluctuation intensity, The simulated results show that the echo envelope fluctuation intensity greatly depends on the carrier correspondingly. The simulated results show that the echo envelope fluctuation intensity greatly correspondingly. The simulated results show that the echo envelope fluctuation intensity greatly frequency: when the carrier frequency increases, the echo envelope fluctuation intensity increases depends on the carrier frequency: when the carrier frequency increases, the echo envelope fluctuation depends on the carrier frequency: when the carrier frequency increases, the echo envelope fluctuation correspondingly, which is consistent with the theoretical analysis. intensity increases correspondingly, which is consistent with the theoretical analysis. intensity increases correspondingly, which is consistent with the theoretical analysis. 0.9 0.9 0.9 0.9 0.8 0.8 0.8 0.8 0.7 0.7 0.7 0.7 0.6 0.6 0.6 0.6 0.5 0.5 0.5 0.5 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) Azimuth(deg) Azimuth(deg) (a) (b) (a) (b) Figu Figure re 1. 1. ((a a)) Intensity Intensity of of echo echo envelope envelope fluctuation fluctuatioafter n after normalization, normalizatiothe n, the frequency frequen is cy 20–40 is 20kHz –40 kHz and Figure 1. (a) Intensity of echo envelope fluctuation after normalization, the frequency is 20–40 kHz the pulse length is 1 ms; (b) STD of echo envelope fluctuation after normalization. and the pulse length is 1 ms; (b) STD of echo envelope fluctuation after normalization. and the pulse length is 1 ms; (b) STD of echo envelope fluctuation after normalization. 0.9 0.9 0.8 0.8 0.8 0.8 0.7 0.7 0.6 0.6 0.6 0.6 0.5 0.5 0.4 0.4 0.4 0.4 0.3 0.3 0.2 0.2 0.2 0.2 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) Azimuth(deg) Azimuth(deg) (a) (b) (a) (b) Figure 2. (a) Intensity of echo envelope fluctuation after normalization, the frequency is 20–40 kHz and the pulse length is 3 ms; (b) STD of echo envelope fluctuation after normalization. Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 17 Figure 2. (a) Intensity of echo envelope fluctuation after normalization, the frequency is 20–40 kHz Figure 2. (a) Intensity of echo envelope fluctuation after normalization, the frequency is 20–40 kHz Appl. Sci. 2018, 8, 1329 6 of 17 and the pulse length is 3 ms; (b) STD of echo envelope fluctuation after normalization. and the pulse length is 3 ms; (b) STD of echo envelope fluctuation after normalization. -3 -3 -4 -4 x 10 x 10 x 10 x 10 10-20kHz 11 00 -2 -2 00 kk H H zz 10-20kHz 22 .5 .5 20-40kHz 22 00 -4 -4 00 kk H H zz 20-40kHz 11 00 1.5 1.5 0.5 0.5 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) Azimuth(deg) Azimuth(deg) (a) (b) (a) (b) Figure 3. (a) Fluctuation intensity comparison of echo envelopes with different frequencies; (b) STD Figu Figu re re 33 . .( a (a ) )F F luct luct uatio uatio n n in in tensity tensity co co mpar mpar iso iso n n oo f f ech ech oo envel envel oo pes pes wi wi th th differ differ ent ent frequen frequen cie cie s; s; (b (b ) ) STD STD comparison of echo envelope fluctuation intensity with different frequencies. co co m m par par iso iso n n oo f echo f echo en en vel vel oo pe pe fluct fluct uati uati oo n n in in tensity w tensity w ith ith different fre different fre quencies. quencies. 3.1.2. Simulation Study of Synthetic Echo 3 3.1.2. .1.2. Si Simulation mulation Stud Study y o of f S Synthetic ynthetic E Echo cho The The echo echo o o f f an an active active acous acous ti ti c c de de cc oo y y wa wa s s si si mul mul at at ed ed b b y y sev sev er er al al mono mono sta sta ti ti c c hydro hydro p p hones, hones, wh wh ere ere The echo of an active acoustic decoy was simulated by several monostatic hydrophones, where the the number of monostatic hydrophones represents the number of main echo highlights. In this study, the number numbof er monostatic of monostati hydr c hydro ophones phones repr rep esents resents the the number number of of main main echo echo highlights. highlights. IIn n this thisstudy, study, the the syntheti syntheti c c echo echo wa wa s s si si m m ul ul at at ed ed bas bas ed ed o o n n the the sta sta ti ti sti sti cal cal fe fe at at ures ures oo f f ech ech o o hi hi ghli ghli ght ght s s [1 [1 99 ]. ]. The The numb numb er er the synthetic echo was simulated based on the statistical features of echo highlights [19]. The number of highlights is six; the incident pulse is the LFM signal; the frequencies are 10–20 kHz, 20–40 kHz, o of f hi highlights ghlights iis s si six; x; the the iincident ncident pul pulse se iis s the the LFM LFM si signal; gnal; the thefr frequenci equencies es are are110–20 0–20 kHz kHz, , 220–40 0–40 kHz kHz, , an an d d 44 00 –– 88 00 kHz; an kHz; an d d tt he pul he pul se l se l ength ength s s are 1 are 1 ms a ms a nd 3 nd 3 m m s. s. and 40–80 kHz; and the pulse lengths are 1 ms and 3 ms. Figure 4 shows the results of the synthetic echo. Figure 4a compares the fluctuation intensity Fi Figur gure e 4 4 show shows s tthe he resul results ts o of f the the sy synthetic nthetic echo echo. . Fi Figur gure e 4 4a a co compar mpares es tthe he fl fluctuation uctuation iintensity ntensity features of the echo envelope for different frequencies, and Figure 4b compares the STD of the echo features of the echo envelope for different frequencies, and Figure 4b compares the STD of the echo features of the echo envelope for different frequencies, and Figure 4b compares the STD of the echo envelope fluctuation intensity, correspondingly. The simulation results of 3 ms pulse length are en envelope velope ffluctuation luctuation iintensity ntensity, , co corr rre espondingly spondingly. . The Thesi simulation mulation resul results ts oof f 33 ms ms pul pulse se llength ength are are consistent with Figure 4. From the results, we can see that the fluctuation intensity and its STD remain consistent with Figure 4. From the results, we can see that the fluctuation intensity and its STD remain consistent with Figure 4. From the results, we can see that the fluctuation intensity and its STD almost stable when the carrier frequencies change. This greatly differs from the results of the almo remain st stab almost le wh stable en the when carri the er carrier frequenci frequencies es change. change. This This greatl gry eatly diffe dif rs fers from from the the resul results ts oof f the the benchmark model because the synthetic echo is the superposition of signals from several monostatic benchmark model because the synthetic echo is the superposition of signals from several monostatic benchmark model because the synthetic echo is the superposition of signals from several monostatic hydrophones. The low-frequency modulation features of the echo envelope can be simulated, such hydro hydrophones. phones. T The he l low-fr ow-frequency equency modulation modulation featur features es of of the the echo echo envelope envelope can can be bsimulated, e simulated such , sucas h as the echo highlights, but the high-frequency modulation features cannot be simulated, such as echo as the echo highlights, but the high-frequency modulation features cannot be simulated, such as echo the echo highlights, but the high-frequency modulation features cannot be simulated, such as echo en en vv el el oo pe pe ff luctua luctua ti ti oo n n intens intens ity. ity. The The hi hi gh gh -- fr fr eque eque n n cy cy mod mod ul ul at at io io n n fe fe at at ures ures are are caus caus ed ed bb y y bb oo dy dy envelope fluctuation intensity. The high-frequency modulation features are caused by body scattering scattering and elastic scattering. scattering and elastic scattering. and elastic scattering. 0.08 0.08 0.06 0.06 00 .0 .0 77 00 .0 .0 55 0.06 0.06 0.04 0.04 0.05 0.05 00 .0 .0 44 0.03 0.03 10-20kHz 10-20kHz 10-20kHz 10-20kHz 0.03 0.03 20-40kHz 20-40kHz 20-40kHz 20-40kHz 00 .0 .0 22 40-80kHz 40-80kHz 40-80kHz 40-80kHz 00 .0 .0 22 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth Azimuth (de (de gg ) ) (( aa )) (( b b )) Figure Figure 4. 4. ( (a a) ) Fluctu Fluctuation ation intensity intensity compar comparison ison of of a a sy synthetic nthetic echo echo envel envelope ope wi with th differ different ent fre frequencies; quencies; Figure 4. (a) Fluctuation intensity comparison of a synthetic echo envelope with different frequencies; ((b b)) STD STD comparison comparison of of synthetic synthetic echo echo envelope envelope fluctuation fluctuation intensity intensity w with ith differen different t f frequencies. requencies. (b) STD comparison of synthetic echo envelope fluctuation intensity with different frequencies. Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 17 3.2. Experiment Study of Benchmark Model and Rock 3.2.1. Experiment Configuration For comparison with the simulation results, broadband acoustic scattering testing was Appl. Sci. 2018, 8, 1329 7 of 17 conducted with the scaled benchmark model and with rock. Figure 5 shows the experiment configuration: Figure 5a is the schematic diagram of the experimental layout, Figure 5b is the photo 3.2. Experiment Study of Benchmark Model and Rock of the sea experiment, and Figure 5c is the photo of the rock that was tested. The target and monostatic sonar were mounted on the experimental ship, and the target was 3.2.1. Experiment Configuration mounted by two thin ropes on the rotator, by which the azimuth was changed from 0 degree to 180 degree For s. comparison The depth owith f the the target simulation and mono results, static brsona oadband r was acoustic 5 m descattering ep in the t w esting ater, was and conducted the range between the target and the sonar was 21 m, which satisfies the testing requirements for far field with the scaled benchmark model and with rock. Figure 5 shows the experiment configuration: acousti Figurec 5asca istteri the ng. schematic The tran diagram smitting of the beam experimental was mainta layout, ined o Figur n the e 5b beis nch the mark photo to of dimi the ni sea sh interference from reverberation. experiment, and Figure 5c is the photo of the rock that was tested. (a) (b) (c) Figu Figure re 5. 5. Co Configuration nfiguration oof f tthe he b benchmark enchmark se sea a exper experiment. iment. (a () a )The The schema schematic tic dia diagram gram o of f the the experimental layout; (b) photo of the sea experiment; (c) photo of the rock that was tested. experimental layout; (b) photo of the sea experiment; (c) photo of the rock that was tested. A continuous wave (CW) signal and a linear frequency modulation wave signal were tested; the The target and monostatic sonar were mounted on the experimental ship, and the target was parameters were as listed in Table 1. mounted by two thin ropes on the rotator, by which the azimuth was changed from 0 degree to 180 degrees. The depth of the target and monostatic sonar was 5 m deep in the water, and the range Table 1. Parameters of the incident signal. between the target and the sonar was 21 m, which satisfies the testing requirements for far field acoustic scattering. The transmitting beam was maintained on the benchmark to diminish interference Signal Form Frequency Pulse Length Target from reverberation. CW 30 kHz, 60 kHz 1 ms, 3 ms, 10 ms Benchmark model A continuous wave (CW) signal and a linear frequency modulation wave signal were te ted; LFM 20–40 kHz, 40–80 kHz 1 ms, 3 ms, 10 ms Benchmark model the parameters were as listed in Table 1. 43–47 kHz 1 ms Rock 71–74 kHz 1 ms Rock Table 1. Parameters of the incident signal. 3.2.2. Experimental Results Signal Form Frequency Pulse Length Target The procedure for processing the test data for the echo envelope fluctuation feature of the CW 30 kHz, 60 kHz 1 ms, 3 ms, 10 ms Benchmark model benchmark model is: LFM 20–40 kHz, 40–80 kHz 1 ms, 3 ms, 10 ms Benchmark model 43–47 kHz 1 ms Rock 71–74 kHz 1 ms Rock Appl. Sci. 2018, 8, 1329 8 of 17 3.2.2. Experimental Results The procedure for processing the test data for the echo envelope fluctuation feature of the Appl. Sci. 2018, 8, x FOR PEER REVIEW 8 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 8 of 17 benchmark model is: a. Filter the data by a bandpass filter to minimize the interference; a. Filter the data by a bandpass filter to minimize the interference; a. Filter the data by a bandpass filter to minimize the interference; b. Calculate the envelope of the echo signal; b. Calculate the envelope of the echo signal; b. Calculate the envelope of the echo signal; c. Normalize the echo envelope for magnitude consistency; c. Normalize the echo envelope for magnitude consistency; c. Normalize the echo envelope for magnitude consistency; d. Differentiate the echo envelope; d. Differentiate the echo envelope; d. Differentiate the echo envelope; e. Calculate the fluctuation intensity and the standard deviation; e. Calculate the fluctuation intensity and the standard deviation; e. Calculate the fluctuation intensity and the standard deviation; f. Process the echo in whole azimuths. f. Process the echo in whole azimuths. f. Fi Pr gure 6a ocess the is the echo echo in whole envelope azimuths. with LFM 20–40 kHz at a certain azimuth, and the pulse length is Figure 6a is the echo envelope with LFM 20–40 kHz at a certain azimuth, and the pulse length is 1 ms; Figure 6b is the echo envelope fluctuation after differentiation; Figure 7a is the echo envelope 1 ms; Figure 6b is the echo envelope fluctuation after differentiation; Figure 7a is the echo envelope Figure 6a is the echo envelope with LFM 20–40 kHz at a certain azimuth, and the pulse length is with LFM 40–80 kHz, the pulse length is 1 ms; and Figure 7b is the echo envelope fluctuation after with LFM 40–80 kHz, the pulse length is 1 ms; and Figure 7b is the echo envelope fluctuation after 1 ms; Figure 6b is the echo envelope fluctuation after differentiation; Figure 7a is the echo envelope differentiation. Results show that, after differentiation, the energy of geometric scattering has differentiation. Results show that, after differentiation, the energy of geometric scattering has with LFM 40–80 kHz, the pulse length is 1 ms; and Figure 7b is the echo envelope fluctuation after decreased, and the fine structure of the high fluctuation envelope has been enhanced. decreased, and the fine structure of the high fluctuation envelope has been enhanced. differentiation. Results show that, after differentiation, the energy of geometric scattering has decreased, Then, the echo data in whole azimuths were processed, and the echo envelope fluctuation Then, the echo data in whole azimuths were processed, and the echo envelope fluctuation and the fine structure of the high fluctuation envelope has been enhanced. features with varied carrier frequency, pulse length, and azimuth were analyzed. features with varied carrier frequency, pulse length, and azimuth were analyzed. Then, the echo data in whole azimuths were processed, and the echo envelope fluctuation features with varied carrier frequency, pulse length, and azimuth were analyzed. A. Fluctuation Intensity A. Fluctuation Intensity 0.03 0.03 0.9 0.9 0.02 0.8 0.02 0.8 0.7 0.7 0.01 0.01 0.6 0.6 0.5 0.5 0.4 0.4 -0.01 0.3 -0.01 0.3 0.2 0.2 -0.02 -0.02 0.1 0.1 -0.03 0 1 2 3 4 5 6 -0.03 0 1 2 3 4 5 6 0 1 2 3 4 5 6 0 1 2 3 4 5 6 -3 Time(s) -3 Time(s) -3 Time(s) x 10 -3 Time(s) x 10 x 10 x 10 (a) (b) (a) (b) Figure 6. (a) Echo envelope with LFM 20–40 kHz frequency; (b) gradient of the echo envelope from (a). Figure 6. (a) Echo envelope with LFM 20–40 kHz frequency; (b) gradient of the echo envelope from (a). Figure 6. (a) Echo envelope with LFM 20–40 kHz frequency; (b) gradient of the echo envelope from (a). 1 0.08 1 0.08 0.9 0.06 0.9 0.06 0.8 0.8 0.04 0.04 0.7 0.7 0.02 0.6 0.02 0.6 0.5 0 0.5 0 0.4 0.4 -0.02 -0.02 0.3 0.3 -0.04 -0.04 0.2 0.2 -0.06 0.1 -0.06 0.1 0 -0.08 0 -0.08 0 1 2 3 4 5 6 0 1 2 3 4 5 6 0 1 2 3 4 5 6 0 1 2 3 4 5 6 -3 -3 Time(s) Time(s) -3 -3 Time(s) x 10 Time(s) x 10 x 10 x 10 (a) (b) (a) (b) Figure Figure 7. 7. ( (a a) ) E Echo cho e envelope nvelope w with ith L LFM FM 4 40–80 0–80 k kHz Hz fr frequency; equency; ( (b b) ) g gradient radient of of t the he e echo cho e envelope nvelope f fr rom om ( (a a). ). Figure 7. (a) Echo envelope with LFM 40–80 kHz frequency; (b) gradient of the echo envelope from (a). Figure 8 is the comparison of fluctuation intensity for experimental and simulated results for a Figure 8 is the comparison of fluctuation intensity for experimental and simulated results for a frequency of LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 8 shows that frequency of LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 8 shows that the tendencies of the experimental and simulated results are fundamentally similar; the value of the the tendencies of the experimental and simulated results are fundamentally similar; the value of the fluctuation intensity is at the minimum on the beam, and at the maximum on the keel-line. fluctuation intensity is at the minimum on the beam, and at the maximum on the keel-line. Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, x FOR PEER REVIEW 9 of 17 Figure 9 is the fluctuation intensity results of the scaled benchmark model with different carrier frequencies: the tested signal from Figure 9a is the broadband signal, and Figure 9b is a single frequency signal. From the results, we can see that the echo envelope fluctuation intensity greatly depends on the carrier frequency; when the carrier frequency increases, the echo envelope fluctuation Appl. Sci. 2018, 8, 1329 9 of 17 intensity increases correspondingly. According to the theoretical analysis, when the carrier frequency increases, the time resolution of the incident pulse increases, the echo envelope presents a finer fluctuation, and the fluctuation intensity of the envelope increases correspondingly. This feature A. Fluctuation Intensity notably represents the influence of target shape and structure on the echo envelope. Appl. Sci. 2018, 8, x FOR PEER REVIEW 9 of 17 Figure 8 is the comparison of fluctuation intensity for experimental and simulated results for a From the comparison of Figure 9a,b, we can see that the echo envelope fluctuation intensity of frequency of LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 8 shows that the narrowband signal does not change markedly compared to that of the broadband. This is because Figure 9 is the fluctuation intensity results of the scaled benchmark model with different carrier the tendencies of the experimental and simulated results are fundamentally similar; the value of the the time resolution of the broadband signal is greater than that of the narrowband signal, thus, the frequencies: the tested signal from Figure 9a is the broadband signal, and Figure 9b is a single fluctuation intensity is at the minimum on the beam, and at the maximum on the keel-line. envelope structure presents a higher frequency fluctuation. frequency signal. From the results, we can see that the echo envelope fluctuation intensity greatly depends on the carrier frequency; when the carrier frequency increases, the echo envelope fluctuation intensity increases correspondingly. According to the theoretical analysis, when the carrier frequency Theoretical Theoretical increases, the time resolution of the Tes inci ted dent pulse increases, the echo env Tes el te o dpe presents a finer 0.8 0.8 fluctuation, and the fluctuation intensity of the envelope increases correspondingly. This feature notably represents the influence of target shape and structure on the echo envelope. 0.6 From the comparison of Figure 9a,b, we can see that the echo envelope fluctuation intensity of 0.6 the narrowband signal does not change markedly compared to that of the broadband. This is because 0.4 the time resolution of the broadband signal is greater than that of the narrowband signal, thus, the 0.4 envelope structure presents a higher frequency fluctuation. 0.2 1 1 0.2 Theoretical Theoretical 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Tested Tested Azimuth(deg) Azimuth(deg) 0.8 0.8 (a) (b) 0.6 Fig Figure ure 8 8. . Theo Theor retical etical anand d test tested ed echo echo envelo envelope pe fluctuatio fluctuation n intensity intensity with with the frequen the frequency cy of LFM 2 of LFM 0–40 0.6 20–40 kHz. (a) The pulse length is 1 ms; (b) the pulse length is 3 ms. kHz. (a) The pulse length is 1 ms; (b) the pulse length is 3 ms. 0.4 0.4 Figure 9 is the fluctuation intensity results of the scaled benchmark model with different carrier 30kHz 40-80kHz 0.9 frequencies: the tested signal from Figure 9a is the broadband signal, and Figure 9b is a single frequency 60kHz 0.2 20-40kHz 0.8 signal. From the results, we can see that the echo envelope fluctuation intensity greatly depends on 0.2 0.8 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 the carrier frequency; when the carrier frequency increases, the echo envelope fluctuation intensity 0.7 Azimuth(deg) Azimuth(deg) increases correspondingly. According to the theor etical analysis, when the carrier frequency increases, 0.6 0.6 (a) (b) the time resolution of the incident pulse increases, the echo envelope presents a finer fluctuation, and the fluctuation intensity of the envelope increases corr 0.5espondingly. This feature notably represents Figure 8. Theoretical and tested echo envelope fluctuation intensity with the frequency of LFM 20–40 0.4 the influence of target shape and structure on the echo envelope. kHz. (a) The pulse length is 1 ms; (b) the pulse length is 3 ms. 0.4 0.3 0.2 30kHz 40-80kHz 0.2 0.9 60kHz 0 20 40 60 80 100 201 -4 20 0kH 1z 40 160 180 0 20 40 60 80 100 120 140 160 180 0.8 0.8 Azimuth(deg) Azimuth(deg) (a) (b) 0.7 0.6 Figure 9. (a) Fluctuation intensity comparison of the wideband echo envelope with different 0.6 frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the CW echo envelope 0.5 with different frequencies, the pulse length is 3 ms. 0.4 0.4 Figure 10a,b is the echo fluctuation intensities of the CW and LFM signals, with pulse lengths of 0.3 0.2 1, 3, and 10 ms, respectively. The results show that the magnitude of the fluctuation intensity changes 0.2 as the pulse length changes; when the pulse length increases, the magnitude of the echo fluctuation 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 intensity decreases. Thi As zim is ub theca (deg use ) a smaller pulse length has a grea Ater zimut tih me (degre ) solution, the echo envelope presents finer fluctuation, and, correspondingly, the fluctuation intensity of the envelope (a) (b) Figure 9. (a) Fluctuation intensity comparison of the wideband echo envelope with different Figure 9. (a) Fluctuation intensity comparison of the wideband echo envelope with different frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the CW echo envelope frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the CW echo envelope with different frequencies, the pulse length is 3 ms. with different frequencies, the pulse length is 3 ms. Figure 10a,b is the echo fluctuation intensities of the CW and LFM signals, with pulse lengths of 1, 3, and 10 ms, respectively. The results show that the magnitude of the fluctuation intensity changes as the pulse length changes; when the pulse length increases, the magnitude of the echo fluctuation intensity decreases. This is because a smaller pulse length has a greater time resolution, the echo envelope presents finer fluctuation, and, correspondingly, the fluctuation intensity of the envelope Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, 1329 10 of 17 From the comparison of Figure 9a,b, we can see that the echo envelope fluctuation intensity of the narrowband signal does not change markedly compared to that of the broadband. This is because the time resolution of the broadband signal is greater than that of the narrowband signal, thus, the envelope structure presents a higher frequency fluctuation. Figure 10a,b is the echo fluctuation intensities of the CW and LFM signals, with pulse lengths of 1, 3, and 10 ms, respectively. The results show that the magnitude of the fluctuation intensity changes as the pulse length changes; when the pulse length increases, the magnitude of the echo fluctuation intensity decreases. This is because a smaller pulse length has a greater time resolution, Appl. Sci. 2018, 8, x FOR PEER REVIEW 10 of 17 the echo envelope presents finer fluctuation, and, correspondingly, the fluctuation intensity of the Appl. Sci. 2018, 8, x FOR PEER REVIEW 10 of 17 envelope increases. This feature is consistent with the theoretical analysis. Also, as shown in Figure 9, increases. This feature is consistent with the theoretical analysis. Also, as shown in Figure 9, the the wideband signal provides better time resolution than that of the narrowband signal. Then, increases. This feature is consistent with the theoretical analysis. Also, as shown in Figure 9, the wideband signal provides better time resolution than that of the narrowband signal. Then, the the envelope structure presents a higher frequency fluctuation, so the wideband pulse presents clearer wideband signal provides better time resolution than that of the narrowband signal. Then, the envelope structure presents a higher frequency fluctuation, so the wideband pulse presents clearer envelope structure presents a higher frequency fluctuation, so the wideband pulse presents clearer results resul than ts than the the narr na owband rrowbansignal d signawith l with the s the same ame p pulse ulse llength. ength. results than the narrowband signal with the same pulse length. 1.2 10ms 10ms 1.2 3 1m 0m s s 1.2 3 1m 0m s s 1 3m ms s 1.2 1ms 3ms 1 1 1ms 1ms 0.8 0.8 0.8 0.8 0.6 0.6 0.6 0.6 0.4 0.4 0.4 0.4 0.2 0.2 0.2 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0.2 Azimuth(deg) Azimuth(deg) 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) (a) (b) Figure 10. (a) Fluctuation intensity of the CW echo envelope with different pulse lengths; (b) Figure 10. (a) Fluctuation intensity of the CW echo envelope with different pulse lengths; (b) fluctuation fFig luct ure uati 10 on . (in a)tensity of t Fluctuatio he n in wideb tensity and o echo f the en CW velo echo pe w envelo ith dipe fferwi ent pulse th differen leng t ths. pulse lengths; (b) intensity of the wideband echo envelope with different pulse lengths. fluctuation intensity of the wideband echo envelope with different pulse lengths. B. Standard Deviation of Fluctuation Intensity B. Standard Deviation of Fluctuation Intensity B. Standard Deviation of Fluctuation Intensity Figure 11 compares the STDs of the theoretical and experimental echo envelope fluctuation Figure 11 compares the STDs of the theoretical and experimental echo envelope fluctuation Figure 11 compares the STDs of the theoretical and experimental echo envelope fluctuation intensities with LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 11 shows intensities with LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 11 shows intensities with LFM 20–40 kHz and pulse lengths of 1 ms and 3 ms, respectively. Figure 11 shows that the tendency of the experimental and simulated results are fundamentally similar: the STD of that the tendency of the experimental and simulated results are fundamentally similar: the STD of the that the tendency of the experimental and simulated results are fundamentally similar: the STD of the fluctuation intensity is minimum on the beam, and maximum on the keel-line. fluctuation intensity is minimum on the beam, and maximum on the keel-line. the fluctuation intensity is minimum on the beam, and maximum on the keel-line. Theoretical 1 Theoretical 0.9 0.9 T Te hs eto erd etical T Te hs eto erd etical 0.9 0.9 0.8 Tested 0.8 Tested 0.8 0.8 0.7 0.7 0.7 0.7 0.6 0.6 0.6 0.5 0.6 0.5 0.5 0.4 0.5 0.4 0.4 0.3 0.4 0.3 0.3 0.2 0.3 0.2 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) (a) (b) Figure 11. STD of theoretical and tested echo envelope fluctuation intensity with the frequency of Figure 11. STD of theoretical and tested echo envelope fluctuation intensity with the frequency of LFM LFM Figure 20– 11. 40 k STD Hz. o (f a) T theo he retical pulse len and gth i test sed 1 ms echo ; (b envelo ) the pulse l pe fluct eng uatio th isn 3 ms. intensity with the frequency of LFM 20–40 kHz. (a) The pulse length is 1 ms; (b) the pulse length is 3 ms. 20–40 kHz. (a) The pulse length is 1 ms; (b) the pulse length is 3 ms. Figure 12 is the STD results of fluctuation intensity of the scaled benchmark model, varying with Figure 12 is the STD results of fluctuation intensity of the scaled benchmark model, varying with azimuth: Figure 12a shows the results of the broadband signal, and Figure 12b shows the results of azimuth: Figure 12a shows the results of the broadband signal, and Figure 12b shows the results of the narrowband signal. The results of the two kinds of tested signals show that the standard deviation the narrowband signal. The results of the two kinds of tested signals show that the standard deviation of the fluctuation intensity depends on the carrier frequency; it increases with increasing carrier of the fluctuation intensity depends on the carrier frequency; it increases with increasing carrier frequency, and it is more notable for the broadband signal—similar to the fluctuation intensity results. frequency, and it is more notable for the broadband signal—similar to the fluctuation intensity results. Compared with the fluctuation intensity of the narrowband signal, the narrowband signal’s standard Compared with the fluctuation intensity of the narrowband signal, the narrowband signal’s standard Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude RelR ate iv la et iMa ve g Ma nitg unit deude Appl. Sci. 2018, 8, 1329 11 of 17 Figure 12 is the STD results of fluctuation intensity of the scaled benchmark model, varying with azimuth: Figure 12a shows the results of the broadband signal, and Figure 12b shows the results of the narrowband signal. The results of the two kinds of tested signals show that the standard deviation of the fluctuation intensity depends on the carrier frequency; it increases with increasing carrier frequency, and it is more notable for the broadband signal—similar to the fluctuation intensity Appl. Sci. 2018, 8, x FOR PEER REVIEW 11 of 17 results. Compared with the fluctuation intensity of the narrowband signal, the narrowband signal’s Appl. Sci. 2018, 8, x FOR PEER REVIEW 11 of 17 standard deviation of fluctuation intensity changes with carrier frequency, and these changes are deviation of fluctuation intensity changes with carrier frequency, and these changes are relatively relatively weak. The resolution performance of this feature is not notable. weak. The resolution performance of this feature is not notable. deviation of fluctuation intensity changes with carrier frequency, and these changes are relatively weak. The resolution performance of this feature is not notable. 40-80kHz 30kHz 0.9 20-40kHz 60kHz 40-80kHz 30kHz 0.8 0.9 20-40kHz 60kHz 0.7 0.8 0.8 0.6 0.7 0.8 0.5 0.6 0.6 0.4 0.5 0.6 0.3 0.4 0.4 0.2 0.3 0.4 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 0.2 Azimuth(deg) Azimuth(deg) 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 (a) (b) Azimuth(deg) Azimuth(deg) Figure 12. STD of echo( en a) velope fluctuation intensity with different frequen (b cies ) , the pulse length is 3 Figure 12. STD of echo envelope fluctuation intensity with different frequencies, the pulse length is ms. (a) STD comparison of wideband echo envelope fluctuation intensity with different frequencies; 3 ms. (a) STD comparison of wideband echo envelope fluctuation intensity with different frequencies; Figure 12. STD of echo envelope fluctuation intensity with different frequencies, the pulse length is 3 (b) STD comparison of CW echo envelope fluctuation intensity with different frequencies. (b) STD comparison of CW echo envelope fluctuation intensity with different frequencies. ms. (a) STD comparison of wideband echo envelope fluctuation intensity with different frequencies; (b) STD comparison of CW echo envelope fluctuation intensity with different frequencies. C. Envelope Fluctuation Features of Rock C. Envelope Fluctuation Features of Rock C. Env Fo el r o fur pe ther Fluctua study, w tion Feat e teste ures d t o he f Rock rock wi th frequencies of 43–47 kHz and 71–74 kHz, and a pulse For further study, we tested the rock with frequencies of 43–47 kHz and 71–74 kHz, and a pulse length of 1 ms. Figure 13a is the fluctuation intensity comparison of the rock echo envelope with For further study, we tested the rock with frequencies of 43–47 kHz and 71–74 kHz, and a pulse length of 1 ms. Figure 13a is the fluctuation intensity comparison of the rock echo envelope with different frequencies; Figure 13b is the STD comparison of the rock echo envelope fluctuation length of 1 ms. Figure 13a is the fluctuation intensity comparison of the rock echo envelope with different frequencies; Figure 13b is the STD comparison of the rock echo envelope fluctuation intensity intensity with different frequencies. The results of the test with the rock show that the echo envelope different frequencies; Figure 13b is the STD comparison of the rock echo envelope fluctuation with different frequencies. The results of the test with the rock show that the echo envelope fluctuation fluctuation intensity and its STD increase correspondingly when the carrier frequency increases, intensity with different frequencies. The results of the test with the rock show that the echo envelope which further verifies the theoretical analysis. intensity and its STD increase correspondingly when the carrier frequency increases, which further fluctuation intensity and its STD increase correspondingly when the carrier frequency increases, verifies the theoretical analysis. which further verifies the theoretical analysis. -3 -3 x 10 x 10 12 -3 -3 43-47kHz 43-47kHz x 10 x 10 71-74kHz 12 71-74kHz 43-47kHz 43-47kHz 71-74kHz 12 71-74kHz 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 (a) (b) Azimuth(deg) Azimuth(deg) Figure 13. (a) Fluctuatio (a) n intensity comparison of the rock echo envelope wi(th b)differ ent frequencies; (b) STD comparison of the rock echo envelope fluctuation intensity with different frequencies. Figure Figu 13. re 1 (a 3.) (Fluctuation a) Fluctuation inte inte nsity nsity comparison comparison o of f the therr oock ck echo echo envelope envelope with with differ difent ferent frequencies; frequencies; (b) STD comparison of the rock echo envelope fluctuation intensity with different frequencies. (b) STD comparison of the rock echo envelope fluctuation intensity with different frequencies. 4. Method of Real Echo and Synthetic Echo Discrimination 4. Me No thod of rmally, Rea the l Echo a syntheti nd c Synt echo he th ti at c Ech is si o mul Dis at crimina ed by an tion active sonar decoy is synthesized by the signal that is transmitted from several monostatic hydrophones. The number of hydrophones is Normally, the synthetic echo that is simulated by an active sonar decoy is synthesized by the determined by the number of predominant echo highlights, which are characterized by the relative signal that is transmitted from several monostatic hydrophones. The number of hydrophones is spacing of the hydrophones and the relative magnitude of the transmitted signal. For a real sonar determined by the number of predominant echo highlights, which are characterized by the relative spacing of the hydrophones and the relative magnitude of the transmitted signal. For a real sonar Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude RelatR ive ela M tiv ag en M itu ad ge nitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, 1329 12 of 17 4. Method of Real Echo and Synthetic Echo Discrimination Normally, the synthetic echo that is simulated by an active sonar decoy is synthesized by the signal that is transmitted from several monostatic hydrophones. The number of hydrophones is determined by the number of predominant echo highlights, which are characterized by the relative spacing of the hydrophones and the relative magnitude of the transmitted signal. For a real sonar decoy, Appl. Sci. 2018, 8, x FOR PEER REVIEW 12 of 17 the number of hydrophones is fixed, the high-frequency echo envelope modulation characteristics decoy, the number of hydrophones is fixed, the high-frequency echo envelope modulation (such as echo envelope fluctuation intensity) cannot be simulated, and the echo envelope fluctuation characteristics (such as echo envelope fluctuation intensity) cannot be simulated, and the echo intensity changes with frequencies. envelope fluctuation intensity changes with frequencies. Based on these echo envelope fluctuation characteristics, a novel method was developed for the Based on these echo envelope fluctuation characteristics, a novel method was developed for the discrimination of real underwater target echoes from synthetic echoes. The procedure is as below: discrimination of real underwater target echoes from synthetic echoes. The procedure is as below: a. The active sonar transmits two different frequency pulses with frequency f and f (LFM signal 1 2 Appl. Sci. 2018, 8, x FOR PEER REVIEW 12 of 17 a. The active sonar transmits two different frequency pulses with frequency f and f (LFM 1 2 is preferred), and the difference will be notable; decoy, the number of hydrophones is fixed, the high-frequency echo envelope modulation signal is preferred), and the difference will be notable; b. Filter the input data by a bandpass filter to minimize the interference; characteristics (such as echo envelope fluctuation intensity) cannot be simulated, and the echo b. Filter the input data by a bandpass filter to minimize the interference; c. Processing the envelope data for the two different frequency echoes; envelope fluctuation intensity changes with frequencies. c. Processing the envelope data for the two different frequency echoes; d. Extract Based the o envelope n these echo en fluctuat velope ion fluctua featur tion es cha for racteri thesti two cs, a n dif ov fer el met enthod frequency was develechoes, oped for the such as echo d. Extract the envelope fluctuation features for the two different frequency echoes, such as echo discrimination of real underwater target echoes from synthetic echoes. The procedure is as below: envelope fluctuation intensity; envelope fluctuation intensity; a. The active sonar transmits two different frequency pulses with frequency f and f (LFM 1 2 e. Compare the fluctuation features, and then discriminate between the real underwater target e. Compare the fluctuation features, and then discriminate between the real underwater target signal is preferred), and the difference will be notable; echo and the synthetic echo. echo and the s b. Fi yntheti lter the i c e nput data cho. by a bandpass filter to minimize the interference; Figure 14 is the proposed procedure for discriminating a real underwater target echo from a c. Processing the envelope data for the two different frequency echoes; Figure 14 is the proposed procedure for discriminating a real underwater target echo from a synthetic echo. d. Extract the envelope fluctuation features for the two different frequency echoes, such as echo synthetic echo. envelope fluctuation intensity; e. Compare the fluctuation features, and then discriminate between the real underwater target s (t) echo and the synthetic echo. Figure 14 is the proposed procedure for discriminating a real underwater target echo from a synthetic echo. s (t) s (t ) 2 K s (t ) 2 K Figure 14. Procedure for discriminating a real underwater target echo from a synthetic echo. Figure 14. Procedure for discriminating a real underwater target echo from a synthetic echo. 5. Experimental 5. Experimental Results Results Figure 14. Procedure for discriminating a real underwater target echo from a synthetic echo. In order to further verify the feasibility of the proposed method, the research team carried out a In order to further verify the feasibility of the proposed method, the research team carried out a 5. Experimental Results sea experiment in the sea near Dalian, China. The sea area is at a depth of 30–40 m, and it has fine sea experiment in the sea near Dalian, China. The sea area is at a depth of 30–40 m, and it has fine In order to further verify the feasibility of the proposed method, the research team carried out a sediment. Figure 15 is the configuration of the synthetic echo test. The echo was simulated by two sediment. Figure 15 is the configuration of the synthetic echo test. The echo was simulated by two sea experiment in the sea near Dalian, China. The sea area is at a depth of 30–40 m, and it has fine monostatic hydrophones, where the spacing between the two hydrophones was 1.4 m. By rotating monostatic hydrophones, where the spacing between the two hydrophones was 1.4 m. By rotating sediment. Figure 15 is the configuration of the synthetic echo test. The echo was simulated by two the pole that mounted the two hydrophones, the azimuth of the echo was changed from 0 degree to the pole that mono mounted static hydro the phones, two wh hydr ere ophones, the spacing the between azimuth the two ofhydro the echo phones was was changed 1.4 m. By ro frta om ting 0 degree to 180 degrees. The receiver and transmitter were mounted on the experimental ship. The receiver and the pole that mounted the two hydrophones, the azimuth of the echo was changed from 0 degree to 180 degrees. The receiver and transmitter were mounted on the experimental ship. The receiver and transmitter were 5 m deep in the water. The 20–40 kHz and 40–80 kHz LFM wideband signals were 180 degrees. The receiver and transmitter were mounted on the experimental ship. The receiver and transmitter were 5 m deep in the water. The 20–40 kHz and 40–80 kHz LFM wideband signals were tested usi transmi ng pul tter se l were engths o 5 m def ep 1 ms an in the w d at3 er. ms. The 20–40 kHz and 40–80 kHz LFM wideband signals were tested using pulse lengths of 1 ms and 3 ms. tested using pulse lengths of 1 ms and 3 ms. Figure 15. Configuration of the synthetic echo test. Figure 15. Configuration of the synthetic echo test. Figure 15. Configuration of the synthetic echo test. Figure 16 is the synthesized echo envelope with LFM 20–40 kHz and an azimuth of 45 degrees, and the results after differentiation. Figure 17 is the simulated echo envelope with LFM 40–80 kHz Figure 16 is the synthesized echo envelope with LFM 20–40 kHz and an azimuth of 45 degrees, and an azimuth of 45 degrees, and the results after differentiation. From the comparison of Figures and the results after differentiation. Figure 17 is the simulated echo envelope with LFM 40–80 kHz and an azimuth of 45 degrees, and the results after differentiation. From the comparison of Figures Appl. Sci. 2018, 8, 1329 13 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 13 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 13 of 17 Figure 16 is the synthesized echo envelope with LFM 20–40 kHz and an azimuth of 45 degrees, and the results after differentiation. Figure 17 is the simulated echo envelope with LFM 40–80 kHz and 16a and 17a, we can see that the shape and structure of the echo envelope remains almost the same 16a and 17a, we can see that the shape and structure of the echo envelope remains almost the same an azimuth of 45 degrees, and the results after differentiation. From the comparison of Figures 16a and when the carrier frequency changes, and the gradient of the echo envelope differs slightly. For the when the carrier frequency changes, and the gradient of the echo envelope differs slightly. For the 17a, we can see that the shape and structure of the echo envelope remains almost the same when the real target echo, from the comparison of Figures 6a and 7a, the shape and structure of the echo real target echo, from the comparison of Figures 6a and 7a, the shape and structure of the echo carrier frequency changes, and the gradient of the echo envelope differs slightly. For the real target envelope differ notably when the carrier frequency changes, as does the gradient of the echo envelope. envelope differ notably when the carrier frequency changes, as does the gradient of the echo envelope. echo, from the comparison of Figures 6a and 7a, the shape and structure of the echo envelope differ The echo in all azimuths was processed, and the echo envelope fluctuation features for the two The echo in all azimuths was processed, and the echo envelope fluctuation features for the two notably when the carrier frequency changes, as does the gradient of the echo envelope. different frequencies were extracted. different frequencies were extracted. 0.02 0.02 0.1 0.1 0.01 0.01 0.08 0.08 0.06 0.06 0 0.04 0.04 -0.01 -0.01 0.02 0.02 -0.02 -0.02 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 -3 Time(s) -3 Time(s) -3 Time(s) -3 x 10 Time(s) x 10 x 10 x 10 (a) (b) (a) (b) Figure 16. (a) Synthetic echo envelope with LFM 20–40 kHz frequency; (b) gradient of the synthetic Figure 16. (a) Synthetic echo envelope with LFM 20–40 kHz frequency; (b) gradient of the synthetic Figure 16. (a) Synthetic echo envelope with LFM 20–40 kHz frequency; (b) gradient of the synthetic echo envelope from (a). echo envelope from (a). echo envelope from (a). 0.02 0.02 0.03 0.03 0.025 0.01 0.025 0.01 0.02 0.02 0.015 0.015 0.01 0.01 -0.01 -0.01 0.005 0.005 -0.02 -0.02 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 0.5 1 1.5 2 2.5 3 3.5 -3 -3 Time(s) Time(s) -3 Time(s) -3 x 10 Time(s) x 10 x 10 x 10 (a) (b) (a) (b) Figure 17. (a) A synthetic echo envelope with LFM 40–80 kHz frequency; (b) gradient of the synthetic Figure 17. (a) A synthetic echo envelope with LFM 40–80 kHz frequency; (b) gradient of the synthetic Figure 17. (a) A synthetic echo envelope with LFM 40–80 kHz frequency; (b) gradient of the synthetic echo envelope from (a). echo echo en envelope velope fr from om ( (a a). ). Figure 18 is the fluctuation intensity of the real target echo and the synthetic echo of the Figure 18 is the fluctuation intensity of the real target echo and the synthetic echo of the The echo in all azimuths was processed, and the echo envelope fluctuation features for the two wideband echo envelope for different frequencies, with a pulse length of 1 ms. Figure 18a is the wideband echo envelope for different frequencies, with a pulse length of 1 ms. Figure 18a is the different frequencies were extracted. fluctuation intensity comparison of the real target’s wideband echo envelope for different frequencies; fluctuation intensity comparison of the real target’s wideband echo envelope for different frequencies; Figure 18 is the fluctuation intensity of the real target echo and the synthetic echo of the wideband Figure 18b is the fluctuation intensity comparison of the synthetic wideband echo envelope for Figure 18b is the fluctuation intensity comparison of the synthetic wideband echo envelope for echo envelope for different frequencies, with a pulse length of 1 ms. Figure 18a is the fluctuation different frequencies. different frequencies. intensity comparison of the real target’s wideband echo envelope for different frequencies; Figure 18b is Figure 19 is the standard deviation of the fluctuation intensity of the real target echo and the Figure 19 is the standard deviation of the fluctuation intensity of the real target echo and the the fluctuation intensity comparison of the synthetic wideband echo envelope for different frequencies. synthetic echo of the wideband echo envelope for different frequencies, where the pulse length is 1 synthetic echo of the wideband echo envelope for different frequencies, where the pulse length is 1 ms. Figure 19a compares the standard deviation of the fluctuation intensity of the real target’s ms. Figure 19a compares the standard deviation of the fluctuation intensity of the real target’s wideband echo envelope for different frequencies; Figure 19b compares the standard deviation of the wideband echo envelope for different frequencies; Figure 19b compares the standard deviation of the fluctuation intensity of the synthetic wideband echo envelope for different frequencies. fluctuation intensity of the synthetic wideband echo envelope for different frequencies. R R ela ela tiv tiv e e M M ag ag nn itit uu dd ee Relative Magnitude Relative Magnitude R R ee lat lat iv iv ee M M aa gg nn itit uu dd ee Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, x FOR PEER REVIEW 14 of 17 Figure 20 is the fluctuation intensity of the real target echo and the synthetic echo of the wideband echo envelope for different frequencies, where the pulse length is 3 ms. Figure 20a is the fluctuation intensity comparison of the real target wideband echo envelope for different frequencies; Figure 20b is the fluctuation intensity comparison of the synthetic wideband echo envelope for different frequencies. Appl. Sci. 2018, 8, x FOR PEER REVIEW 14 of 17 From Figures 18–20, we can see that the fluctuation intensity and its STD of the real target echo changes notably when the carrier frequency changes, but the fluctuation intensity and its STD of the Figure 20 is the fluctuation intensity of the real target echo and the synthetic echo of the synthetic echo remains almost the same when the carrier frequency changes. The experimental results wi Appl. deSci. band 2018 echo , 8, 1329 envelope for different frequencies, where the pulse length is 3 ms. Figure 20a 14 is of the 17 are consistent with the theoretical analysis and simulation results. fluctuation intensity comparison of the real target wideband echo envelope for different frequencies; Figure 20b is the fluctuation intensity comparison of the synthetic wideband echo envelope for -3 -3 x 10 x 10 different frequencies. From Figures 18–20, we can see that the fluctuation intensity and its STD of the real target echo 20-40kHz 40-80kHz 40-80kHz changes notably when the carrier frequency changes, but the fluctuation intensity and 2i0 ts ST -40kH D z of the synthetic echo remains almost the same when the carrier frequency changes. The experimental results are consistent with the theoretical analysis and simulation results. -3 -3 x 10 x 10 6 6 20-40kHz 12 40-80kHz 4 4 40-80kHz 20-40kHz 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) 6 6 (a) (b) Figure 18. (a) Fluctuation intensity comparison of the real target wideband echo envelope with different Figure 18. (a) Fluctuation intensity comparison of the real target wideband echo envelope with frequencies, the pulse length is 1 ms; (b) fluctuation intensity comparison of the synthetic wideband different frequencies, the pulse length is 1 ms; (b) fluctuation intensity comparison of the synthetic echo envelope with different frequencies, the pulse length is 1 ms. wideband echo envelope with different frequencies, the pulse length is 1 ms. 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 -3 -3 Azimuth(deg) x 10 x 10 Azimuth(deg) Figure 19 is the standard deviation of the fluctuation intensity of the real target echo and the (a) 16 (b) synthetic 16 echo of the wideband echo envelope for different frequencies, where the pulse length is 1 ms. 40-80kHz 20-40kHz 20-40kHz Figure 19a compares the standard deviation of the fluctuation intensity of the real tar4 get’s 0-80kwideband Hz Figure 18. (a) Fluctuation intensity comparison of the real target wideband echo envelope with echo envelope for different frequencies; Figure 19b compares the standard deviation of the fluctuation different frequencies, the pulse length is 1 ms; (b) fluctuation intensity comparison of the synthetic intensity of the synthetic wideband echo envelope for different frequencies. wideband echo envelope with different frequencies, the pulse length is 1 ms. -3 -3 x 10 x 10 40-80kHz 20-40kHz 20-40kHz 40-80kHz 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) Figure 19. (a) STD comparison of the real target wideband echo envelope with different frequencies, the pulse length is 1 ms; (b) STD comparison of the synthetic wideband echo envelope with different frequencies, the pulse length is 1 ms. 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) Figure 19. (a) STD comparison of the real target wideband echo envelope with different frequencies, Figure 19. (a) STD comparison of the real target wideband echo envelope with different frequencies, the pulse length is 1 ms; (b) STD comparison of the synthetic wideband echo envelope with different the pulse length is 1 ms; (b) STD comparison of the synthetic wideband echo envelope with different frequencies, the pulse length is 1 ms. frequencies, the pulse length is 1 ms. Figure 20 is the fluctuation intensity of the real target echo and the synthetic echo of the wideband echo envelope for different frequencies, where the pulse length is 3 ms. Figure 20a is the fluctuation intensity comparison of the real target wideband echo envelope for different frequencies; Figure 20b is the fluctuation intensity comparison of the synthetic wideband echo envelope for different frequencies. Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, 1329 15 of 17 Appl. Sci. 2018, 8, x FOR PEER REVIEW 15 of 17 -3 -3 x 10 x 10 12 20-40kHz 20-40kHz 40-80kHz 40-80kHz Appl. Sci. 2018, 8, x FOR PEER REVIEW 15 of 17 -3 -3 x 10 x 10 12 20-40kHz 20-40kHz 6 14 40-80kHz 40-80kHz 8 4 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) Figure 20. (a) Fluctuation intensity comparison of the real target wideband echo envelope with different Figure 20. (a) Fluctuation intensity comparison of the real target wideband echo envelope with frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the synthetic wideband different frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the synthetic 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 echo envelope with A dif zim fer uent th(de fr gequencies, ) the pulse length is 3 ms. wideband echo envelope with different frequencies, the pulse length is 3 ms Azimut . h(deg) (a) (b) For further comparison, Figure 21 is the difference in the fluctuation intensity between the two From Figures 18–20, we can see that the fluctuation intensity and its STD of the real target echo Figure 20. (a) Fluctuation intensity comparison of the real target wideband echo envelope with different carrier frequency echoes. Figure 20a compares the difference in the fluctuation intensity changes notably when the carrier frequency changes, but the fluctuation intensity and its STD of the different frequencies, the pulse length is 3 ms; (b) fluctuation intensity comparison of the synthetic from Figure 18, and Figure 21b compares the difference in the STD from Figure 19. In Figure 21, we synthetic echo remains almost the same when the carrier frequency changes. The experimental results wideband echo envelope with different frequencies, the pulse length is 3 ms. can see that the fluctuation intensity and its STD of the real target echo change notably when the are consistent with the theoretical analysis and simulation results. carrier frequency changes, but the fluctuation intensity and its STD do not present notable differences For further comparison, Figure 21 is the difference in the fluctuation intensity between the two For further comparison, Figure 21 is the difference in the fluctuation intensity between the two when the carrier frequency changes. Then, the real target echo and synthetic echo could be different carrier frequency echoes. Figure 20a compares the difference in the fluctuation intensity from different carrier frequency echoes. Figure 20a compares the difference in the fluctuation intensity discriminated by the difference in those two features. The experimental results demonstrate the Figure 18, and Figure 21b compares the difference in the STD from Figure 19. In Figure 21, we can from Figure 18, and Figure 21b compares the difference in the STD from Figure 19. In Figure 21, we feasibility of the method that is proposed. see that the fluctuation intensity and its STD of the real target echo change notably when the carrier can see that the fluctuation intensity and its STD of the real target echo change notably when the frequency changes, but the fluctuation intensity and its STD do not present notable differences when carrier frequency changes, but the fluctuation intensity and its STD do not present notable differences -3 -3 x 10 x 10 the carrier frequency changes. Then, the real target echo and synthetic echo could be discriminated when the carrier frequency changes. Then, the real target echo and synthetic echo could be by the difference in those two featur Real es. The experimental results demonstrate the feasibility of the discriminated by the difference in those two features. The experimental results demonstrate the Real Synthesized method that is proposed. feasibility of the method that is proposed. synthesized -3 -3 x 10 x 10 8 Real Real Synthesized synthesized 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) (a) (b) Figure 21. (a) Fluctuation intensity difference comparison of Figure 18; (b) STD difference comparison 0 20 40 60 80 100 120 140 160 180 of Figure 19. 0 20 40 60 80 100 120 140 160 180 Azimuth(deg) Azimuth(deg) 6. Discussion and Conclusions (a) (b) In this study, the characteristics of echo envelope fluctuation intensity were studied, two features Figure 21. (a) Fluctuation intensity difference comparison of Figure 18; (b) STD difference comparison Figure 21. (a) Fluctuation intensity difference comparison of Figure 18; (b) STD difference comparison of wh of ich Figur were e 19 cha . racterized. The data processing results of the simulation and scaled benchmark of Figure 19. tests are consistent. The echo envelope fluctuation intensity and its STD features strongly depend on 6. Discussion and Conclusions the carrier frequency: when the carrier frequency increases, the echo envelope fluctuation intensity increases correspondingly. The magnitude of the echo fluctuation intensity and its STD vary with In this study, the characteristics of echo envelope fluctuation intensity were studied, two features azimuth, which is minimum on the beam, and maximum on the keel-line. This is because the time of which were characterized. The data processing results of the simulation and scaled benchmark tests are consistent. The echo envelope fluctuation intensity and its STD features strongly depend on the carrier frequency: when the carrier frequency increases, the echo envelope fluctuation intensity increases correspondingly. The magnitude of the echo fluctuation intensity and its STD vary with azimuth, which is minimum on the beam, and maximum on the keel-line. This is because the time Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Relative Magnitude Appl. Sci. 2018, 8, 1329 16 of 17 6. Discussion and Conclusions In this study, the characteristics of echo envelope fluctuation intensity were studied, two features of which were characterized. The data processing results of the simulation and scaled benchmark tests are consistent. The echo envelope fluctuation intensity and its STD features strongly depend on the carrier frequency: when the carrier frequency increases, the echo envelope fluctuation intensity increases correspondingly. The magnitude of the echo fluctuation intensity and its STD vary with azimuth, which is minimum on the beam, and maximum on the keel-line. This is because the time difference between the echo highlights in the keel-line is greater than that in the beam direction, causing a larger magnitude fluctuation of the echo envelope. The magnitude of the fluctuation intensity changes as the pulse length changes—when the pulse length increases, the magnitude of the echo fluctuation intensity decreases. The fluctuation intensity and its STD could be used to characterize the fluctuation in the echo envelope magnitude, but the sensitivity of the standard deviation of the fluctuation intensity is relatively low, compared to that of the fluctuation intensity, especially for a narrowband signal. Based on these echo envelope fluctuation characteristics, we propose a novel method for discriminating between a real underwater target echo and synthetic echo. The feasibility of the method was verified through a sea experiment. It should be noted that the waveform structure of the synthetic echo becomes more like the real target echo when the number of highlights simulated by the echo decoy increases. The performance of this proposed method for a more complex synthetic echo will be studied in further work. Author Contributions: All authors contributed significantly to the work presented in this manuscript. Y.C. proposed the novel method and wrote the paper; B.J. and Z.W. conceived and designed the experiments; G.L. and S.L. contributed with valuable discussions and scientific advice. Funding: This research was funded in part by the Foundation of Science and Technology on Underwater Test and Control Laboratory under grant no. 9140C260201130C26096, and the APC was funded by that too. Acknowledgments: The authors are grateful to Jun Fan from Shanghai Jiao Tong University for the acoustic scattering program code based on the Planar Element Method that originally developed by Jun Fan. Conflicts of Interest: The authors declare no conflict of interest. References 1. He, X.Y.; Jiang, X.Z.; Liu, J.Y. Submarine echo simulation based on echo highlight model. J. Unmanned Undersea Syst. 2001, 9, 15–18. 2. Lin, W.; Liu, L.J.; Xu, Y. Simulation of Underwater target echo based on highlight model. Appl. Mech. Mater. 2014, 536–537, 39–42. [CrossRef] 3. He, C.; Zhao, A.B.; Zhou, B.; Song, X.; Niu, F. Parameter measurement research of sonar echo highlights. J. Acoust. Soc. Am. 2014, 135, 2303. [CrossRef] 4. Cerqueira, R.; Trocoli, T.; Neves, G.; Joyeux, S.; Albiez, J.; Oliveira, L. A novel GPU-Based sonar simulator for real-time applications. Comput. Graphics 2017, 68, 66–76. [CrossRef] 5. Gokhan, I.; Ozgur, B.A. Three-dimensional underwater target tracking with acoustic sensor networks. IEEE Trans. Veh. Technol. 2011, 60, 3897–3906. 6. Luo, J.H.; Han, Y.; Fan, L.Y. Underwater acoustic target tracking: A review. Sensors 2018, 13, 112. [CrossRef] [PubMed] 7. Seo, Y.; On, B.; Im, S.; Shim, T.; Seo, I. Underwater Cylindrical Object Detection Using the Spectral Features of Active Sonar Signals with Logistic Regression Models. Appl. Sci. 2018, 8, 116. [CrossRef] 8. Meng, Q.X.; Yang, S.E. A wave structure based method for recognition of marine acoustic target signals. J. Acoust. Soc. Am. 2014, 137, 2242. [CrossRef] 9. Fan, J. Study on Echo Characteristics of Underwater Complex Targets. Ph.D. Thesis, Shanghai Jiaotong University, Shanghai, China, 2001; p. 16. 10. Jia, H.J.; Li, X.K.; Meng, X.X. Rigid and elastic acoustic scattering signal separation for underwater target. J. Acoust. Soc. Am. 2017, 142, 653–665. [CrossRef] [PubMed] Appl. Sci. 2018, 8, 1329 17 of 17 11. Yang, M.; Li, X.; Yang, Y.; Meng, X. Characteristic analysis of underwater acoustic scattering echoes in the wavelet transform domain. J. Mar. Sci. Appl. 2017, 16, 93–101. [CrossRef] 12. Wu, Y.S.; Li, X.K.; Wang, Y. Extraction and classification of acoustic scattering from underwater target based on Wigner-Ville distribution. Appl. Acoust. 2018, 138, 52–59. [CrossRef] 13. Xu, G.F.; Li, C.X.; Ding, S.Q. Study on extracting the information of elastic scattering wave. Ocean Technol. 2006, 25, 94–98. 14. Tang, W.L. Highlight model of echoes from sonar targets. ACTA Acoust. 1994, 19, 92–100. 15. Lou, S.C.; Chao, R.M.; Ko, S.; Lin, K.M.; Zhong, J.X. A simplified signal analysis algorithm for the development of a low cost underwater echo-sounder. Measurement 2011, 44, 1572–1581. [CrossRef] 16. Wang, M.Z.; Huang, X.W.; Hao, C.Y. Model of an underwater target based on target echo highlight structure. J. Syst. Simul. 2003, 15, 21–25. 17. Muller, M.W.; Au, W.W.; Nachtigall, P.E.; Allen, J.S., III; Breese, M. Phantom echo highlight amplitude and temporal difference resolutions of an echolocating dolphin, Tursiops truncates. J. Acoust. Soc. Am. 2007, 122, 2255–2262. [CrossRef] [PubMed] 18. Dankiewicz, L.A.; Helweg, D.A.; Moore, P.W.; Zafran, J.M. Discrimination of amplitude-modulated synthetic echo trains by an echolocating bottlenose dolphin. J. Acoust. Soc. Am. 2002, 112, 1702–1708. [CrossRef] [PubMed] 19. Chen, Y.F.; Li, G.J.; Wang, Z.S.; Zhang, M.W.; Jia, B. Statistical feature of underwater target echo highlight. Acta Phys. Sin. 2013, 62, 084302. 20. Chen, Y.F.; Li, S.; Wang, Z.S.; Li, G.J.; Gao, F. Echo magnitude fluctuation feature of underwater target. J. Ship Mech. 2017, 21, 218–227. 21. Chen, Y.F.; Jia, B.; Li, S.; Wang, Z.S.; Li, G.J. Discrimination of real underwater target echo and synthetic echo based on envelope modulation rate. J. Harbin Eng. Univ. 2016, 37, 1467–1472. 22. Nell, C.W.; Gilroy, L.E. An Improved BASIS Model for the BeTSSi Submarine; DRDC Atlantic TR 2003-199; Defence R&D Canada: Ottawa, ON, Canada, November 2003. 23. Fan, J.; Zhu, B.L.; Tang, W.L. Modified geometrical highlight model of echoes from non-rigid surface sonar target. Acta Acust. 2001, 26, 545–550. 24. Fan, J.; Li, J.L.; Liu, T. Transition characteristics of echoes from complex shape targets in water. J. Shanghai Jiaotong Univ. 2002, 36, 161–164. 25. Fan, J.; Tang, W.L.; Zhuo, L.K. Planar elements method for forecasting the echo characteristics from sonar targets. J. Ship Mech. 2012, 16, 171–180. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png
Applied Sciences
Multidisciplinary Digital Publishing Institute
http://www.deepdyve.com/lp/multidisciplinary-digital-publishing-institute/feature-of-echo-envelope-fluctuation-and-its-application-in-the-qTRj0xt1iX