在人类已知生物声呐中，蝙蝠声呐是最高效的，这得益于一套精妙的定位系统——回声定位，因而具有极高的仿生学研究价值。蝙蝠通过喉部发声，主动调节发声器官来产生动态变化的回声定位信号，该环节在蝙蝠声呐实现精准定位的过程中非常关键。虽然蝙蝠声呐仿生研究已取得相当进展，但蝙蝠声呐系统的工作机制尚未完全揭示，主要体现在蝙蝠回声定位信号建模不精确和蝙蝠发声系统参数化模型不完善，因此，建模回声定位信号和发声系统对蝙蝠声呐仿生研究具有十分重要的意义。本文针对蝙蝠回声定位信号和发声系统开展参数建模研究，通过大量的蝙蝠回声定位信号实测数据，结合蝙蝠生物特征，运用信号处理手段，建立了基于蝙蝠信号包络和相位的信号参数模型和蝙蝠发声机理的等效系统模型。本文的主要研究内容分为以下三个方面：   

第一，对大棕蝙蝠回声定位信号实测数据进行初步的信号处理工作，获取信号的特征信息。使用短时傅里叶变换（Short Time Fourier Transform，STFT）、希尔伯特-黄变换（Hilbert-Huang Transform，HHT）和维格纳-威勒变换（Wigner Ville Distribution，WVD）对信号数据进行时频分析，结果显示该蝙蝠信号具有两个下调频谐波结构，这与所用蝙蝠信号的认知一致。

第二，针对蝙蝠回声定位信号的建模问题，提出基于包络项和相位项建模的信号模型。信号的包络项建模为高斯函数的加权和函数，相位项依据信号瞬时频率构建，瞬时频率建模为三类拟合函数，均通过线性和非线性最小二乘法解得模型参数。实测蝙蝠声信号的结果表明，相较于现有的回声定位信号模型，基于反比例函数构建相位项的回声定位信号模型不仅能够实现对实测蝙蝠声信号的良好近似，而且具有较高的适用性。

第三，为构建具有生物特征信息的蝙蝠发声系统模型并应用于蝙蝠声呐仿生研究，开展基于蝙蝠喉部发声机理的时变系统建模工作。
结合蝙蝠生物研究，将蝙蝠发声系统描述为时变自回归（Time-Varying Autoregressive，TV-AR）模型，模型参数变化轨迹建模为分段常数和连续变化的两种形式，依据两种参数变化形式，进行相应系统模型研究，使用正则化最小二乘法和基函数法进行参数求解。实测蝙蝠信号数据的建模结果表明，所提基于TV-AR的系统模型实现了高斯白噪声为输入，蝙蝠回声定位信号为输出的蝙蝠发声系统仿真。相较于现有的蝙蝠发声系统模型，系数连续变化的发声系统模型不仅更加符合蝙蝠发声器官变化特征，而且实现了蝙蝠发声系统的良好模拟。对将来进一步揭示蝙蝠声呐系统的工作机制和促进蝙蝠声呐仿生研究具有重要参考价值。

The bat sonar is the most efficient of all known human biosonars, which benefits from an elaborate navigation system, echolocation, and is therefore of great bionomic research value. The bat vocalizes through its larynx and actively adjusts its vocal organs to generate dynamically changing echolocation signals, this link is critical in the process of achieving accurate positioning of the bat sonar. Although considerable progress has been made in bat sonar bionic research, the working mechanism of the bat sonar system is still not fully revealed, which is reflected in the inaccurate modeling of bat echolocation signal and the lack of parametric model of bat vocal system, therefore, modeling echolocation signal and vocal system is of great importance for bat sonar bionic research. In this paper, a parametric modeling study is conducted for bat echolocation signals and vocal systems. Through a large number of bat echolocation signals, a parametric model of signals based on bat signal amplitude and phase modeling, and an equivalent system model of bat vocal mechanisms are established by combining signal processing and bat biological characteristics.
The main research of this paper is divided into the following three areas:

First, preliminary signal processing work is performed on the measured bat echolocation signal data to obtain the signal characteristic information. Short Time Fourier Transform (STFT), Hilbert-Huang Transform (HHT) and Wigner Ville Distribution (WVD) are used to perform time-frequency analysis of the signal data. The results demonstrate that the bat signal has two down-modulated harmonic structures, which is consistent with the perception of the bat signal used.

Second, for the modeling problem of bat echolocation signal, a signal model based on envelope term and phase term modeling is proposed. The envelope term of the signal is modeled as a weighted sum function of Gaussian functions, the phase term is constructed based on the instantaneous frequency of the signal, and the instantaneous frequency is modeled as three types of fitted functions, all solved by linear and nonlinear least squares methods to obtain the model parameters. The results of the measured bat acoustic signals show that compared with the existing echolocation signal models, the echolocation signal model based on the inverse proportional function to construct the phase term can not only achieve a good approximation of the measured bat acoustic signal, but also has a high applicability.

Third, in order to construct a bat vocal system model with biometric information and apply it to bat sonar bionic research, work on a system model based on the bat laryngeal vocal mechanism is carried out. The bat vocal system is described as a Time-Varying Autoregressive (TV-AR) model based on bat biology studies. The model parameter variation trajectory is modeled as both segmented constant and continuously varying forms.
Based on the two forms of parameter variations, the corresponding system models are studied and the parameters are solved using the regularized least squares method and the basis function method. The modeling results of the measured bat signal data show that the proposed TV-AR-based system model achieves the simulation of bat vocal system with Gaussian white noise as the input and bat echolocation signal as the output.
Compared with the existing bat vocal system model, the vocal system model with continuously varying coefficients is not only more consistent with the characteristics of bat vocal organ changes, but also achieves a good simulation of the bat vocal system.It is an important reference value to further reveal the working mechanism of bat sonar system and promote bat sonar bionic research in the future.

[1] PROTHERO D R. The princeton field guide to prehistoric mammals[C] // . 2016.
[2] TSAGKOGEORGA G, PARKER J, STUPKA E, et al. Phylogenomic analyses elucidate the evolutionary relationships of bats[J]. Current Biology, 2013, 23 : 2262 – 2267.
[3] ALTRINGHAM J D. Bats: Biology and behaviour[C] // . 1996.
[4] THOMAS J A, MOSS C F, VATER M. Echolocation in bats and dolphins[C] // . 2003.
[5] WOTTON J M, SIMMONS J A. Spectral cues and perception of the vertical position of targets by the big brown bat, Eptesicus fuscus[J]. The Journal of the Acoustical Society of America, 2000, 1072 : 1034 – 41.
[6] SUMIYA M, ASHIHARA K, YOSHINO K, et al. Bat-inspired signal design for target discrimination in human echolocation[J]. The Journal of the Acoustical Society of America, 2019, 145 4 : 2221.
[7] MING C, HARO S, SIMMONS A M, et al. A comprehensive computational model of animal biosonar signal processing[J]. PLoS Computational Biology, 2021, 17.
[8] JAKOBSEN L, RATCLIFFE J M, SURLYKKE A. Convergent acoustic field of view in echolocating bats[J]. Nature, 2012, 493 : 93 – 96.
[9] TAUB M, YOVEL Y. Segregating signal from noise through movement in echolocating bats[J]. Scientific Reports, 2020, 10.
[10] JAKOBSEN L, OLSEN M N, SURLYKKE A. Dynamics of the echolocation beam during prey pursuit in aerial hawking bats[J]. Proceedings of the National Academy of Sciences, 2015, 112 : 8118 – 8123.
[11] LINNENSCHMIDT M, WIEGREBE L. Sonar beam dynamics in leaf-nosed bats[J]. Scientific Reports, 2016, 6.
[12] KOTHARI N B, WOHLGEMUTH M J, MOSS C F. Dynamic representation of 3D auditory space in the midbrain of the free-flying echolocating bat[J]. eLife, 2018, 7.
[13] LUO J, GOERLITZ H R, BRUMM H, et al. Linking the sender to the receiver: vocal adjustment by bats to maintain signal detection in noise[J]. Scientific Reports, 2015, 5.
[14] SCHOUTEN G, STECKEL J. Principles of biological echolocation applied to radar sensing: applying biomimetic sensors to achieve autonomous navigation[J]. IEEE Signal Processing Magazine, 2019, 36 : 98 – 111.
[15] KUC R. Generating cognitive maps using echo features from a biomimetic audible sonar[J]. The Journal of the Acoustical Society of America, 2019, 145 4 : 2084.
[16] YOST W A, PASTORE M T, ZHOU Y. Discrimination of changes in spatial configuration for multiple, simultaneously presented sounds[J]. The Journal of the Acoustical Society of America, 2019, 145 4 : EL310.
[17] ELIAKIM I, COHEN Z, KÓSA G, et al. A fully autonomous terrestrial bat-like acoustic robot[J]. PLoS Computational Biology, 2019, 14.
[18] YAMADA Y, WATABE Y, HIRYU S, et al. 3D acoustic localization based on echolocation strategy of bats[J]. 2020 IEEE/SICE International Symposium on System Integration (SII), 2020 : 1133 – 1138.
[19] BAIER A L, WIEGREBE L. Flutter sensitivity in FM bats. Part I: delay modulation[J]. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 2018, 204 : 929 – 939.
[20] BAIER A L, STELZER K-J, WIEGREBE L. Flutter sensitivity in FM bats. Part II: amplitude modulation[J]. Journal of Comparative Physiology. A, Neuroethology, Sensory, Neural, and Behavioral Physiology, 2018, 204 : 941 – 951.
[21] GREITER W, FIRZLAFF U. Representation of three-dimensional space in the auditory cortex of the echolocating bat P. discolor[J]. PLoS ONE, 2018, 12.
[22] JONES G J F, TEELING E C. The evolution of echolocation in bats[J]. Trends in ecology & evolution, 2006, 21 3 : 149 – 56.
[23] SIMMONS J A. Perception of echo phase information in bat sonar.[J]. Science, 1979, 204 4399 : 1336 – 8.
[24] SIMMONS J A, FERRAGAMO M J, MOSS C F. Echo-delay resolution in sonar images of the big brown bat, Eptesicus fuscus[J]. Proceedings of the National Academy of Sciences of the United States of America, 1998, 95 21 : 12647 – 52.
[25] SIMMONS J A, HOUSER D S, KLOEPPER L N. Localization and classification of targets by echolocating bats and dolphins[C] // . 2014.
[26] JONES G J F, SIEMERS B M. The communicative potential of bat echolocation pulses[J]. Journal of Comparative Physiology A, 2011, 197 : 447 – 457.
[27] SCHUCHMANN M, PUECHMAILLE S J, SIEMERS B M. Horseshoe bats recognise the sex of conspecifics from their echolocation calls[C] // . 2012.
[28] YOH N, SYME P, ROCHA R, et al. Echolocation of central amazonian‘whispering’ phyllostomid bats: call design and interspecific variation[J]. Mammal Research, 2020, 65 : 583 – 597.
[29] LEISER‐MILLER L B, SANTANA S E. Functional differences in echolocation call design in an adaptive radiation of bats[J]. Ecology and Evolution, 2021, 11 : 16153 – 16164.
[30] TAN X, LIN A, SUN K, et al. Greater horseshoe bats recognize the sex and individual identity of conspecifics from their echolocation calls[J]. Animals : an Open Access Journal from MDPI, 2022, 12.
[31] LUO B, LEISER‐MILLER L B, SANTANA S E, et al. Echolocation call divergence in bats: a comparative analysis[J]. Behavioral Ecology and Sociobiology, 2019, 73.
[32] SUN C, ZHANG C, LUCAS J R, et al. Territorial calls of the bat hipposideros armiger may encode multiple types of information: body mass, dominance rank and individual identity[J]. Animal Cognition, 2021, 24 : 689 – 702.
[33] BRUALLA N L M, WILSON L A B, DOUBE M, et al. The vocal apparatus: an understudied tool to reconstruct the evolutionary history of echolocation in bats?[J]. Journal of Mammalian Evolution, 2023, 30 : 79 – 94.
[34] METZNER W, MÜLLER R. Ultrasound production, emission, and reception[C] // . 2016.
[35] CARTER R T, ADAMS R A. Ontogeny of the larynx and flight ability in jamaican fruit bats (Phyllostomidae) with considerations for the evolution of echolocation[J]. The Anatomical Record, 2014, 297.
[36] METZNER W, SCHULLER G. Vocal control in echolocating bats[J]. Handbook of Behavioral Neuroscience, 2010, 19 : 403 – 415.
[37] HÅKANSSON J, MIKKELSEN C, JAKOBSEN L, et al. Bats expand their vocal range by recruiting different laryngeal structures for echolocation and social communication[J]. PLOS Biology, 2022, 20.
[38] AUSTIN K B, THOMAS H H, JR. P R M, et al. Capturing the auditory experience of behaving bats: a preliminary study[J]. 2009 International Conference on Computing, Engineering and Information, 2009 : 123 – 126.
[39] ZHUANG Q, MÜLLER R. Noseleaf furrows in a horseshoe bat act as resonance cavities shaping the biosonar beam[J]. Physical review letters, 2006, 97 21 : 218701.
[40] AUSTIN K B, MOOSMAN P R, THOMAS H H. Eavesdropping on echolocation: recording the bat’s auditory experience[J]. 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2011 : 7682 – 7686.
[41] JAKOBSEN L, BRINKLØV S, SURLYKKE A. Intensity and directionality of bat echolocation signals[J]. Frontiers in Physiology, 2013, 4.
[42] TESHIMA Y, NOMURA T, KATO M, et al. Effect of bat pinna on sensing using acoustic finite difference time domain simulation[J]. The Journal of the Acoustical Society of America, 2022, 1516 : 4039.
[43] TESHIMA Y, YAMADA Y, TSUCHIYA T, et al. Analysis of echolocation behavior of bats in“echospace” using acoustic simulation[J]. BMC Biology, 2022, 20.
[44] HIRYU S, BATES M E, SIMMONS J A, et al. FM echolocating bats shift frequencies to avoid broadcast– echo ambiguity in clutter[J]. Proceedings of the National Academy of Sciences, 2010, 107 : 7048 – 7053.
[45] KOUNITSKY P, RYDELL J, AMICHAI E, et al. Bats adjust their mouth gape to zoom their biosonar field of view[J]. Proceedings of the National Academy of Sciences, 2015, 112 : 6724 – 6729.
[46] MÜLLER R. Numerical analysis of biosonar beamforming mechanisms and strategies in bats[J]. The Journal of the Acoustical Society of America, 2010, 128 3 : 1414 – 25.
[47] MÜLLER R. A numerical study of the role of the tragus in the big brown bat[J]. The Journal of the Acoustical Society of America, 2004, 116 6 : 3701 – 12.
[48] MÜLLER R, HALLAM J. Knowledge mining for biomimetic smart antenna shapes[J]. Robotics Auton. Syst., 2005, 50 : 131 – 145.
[49] 王福勋. 基于蝙蝠耳的仿生麦克风轮廓模型的数值分析 [D]. [S.l.] : 山东大学, 2010.
[50] 王福勋, 张智伟, 武鹏飞. 大菊头蝠对耳屏下沟的声学功能 [C] // 2009 年度全国物理声学会议论文集. 2009.
[51] 张智伟, 王福勋, 高丽. 高鞍菊头蝠鼻叶鞍状结构摆动对超声定位影响的数值研究 [C] // 2009年度全国物理声学会议论文集. 2009.
[52] 庄桥, 陈晓乾, 张美生. 埃及裂颜蝠超声辐射波束形成结构的数值研究 [J]. 山东建筑大学学报,2017, 32(2) : 7.
[53] 张智伟, 王福勋, 赵昕. 鞍状叶长度对菊头蝠超声定位的影响 [C] // 中国声学学会第十一届青年学术会议会议论文集. 2015.
[54] ZHANG S, MA X, DONG Z, et al. Ultrasonic spatial target localization using artificial pinnae of brown long-eared bat[J]. bioRxiv, 2020.
[55] MASTERS W M, JACOBS S C, SIMMONS J A. The structure of echolocation sounds used by the big brown bat Eptesicus fuscus : some consequences for echo processing[J]. Journal of the Acoustical Society of America, 1991, 89 : 1402 – 1413.
[56] BALLERI A, GRIFFITHS H D, HOLDERIED M W, et al. Bat-inspired multi-harmonic waveforms[J]. 2010 International Waveform Diversity and Design Conference, 2010 : 000086 – 000089.
[57] DONNELLY D. The fast Fourier transform for experimentalists. Part VI. Chirp of a bat[J]. Computing in Science & Engineering, 2006, 8 : 72 – 78.
[58] DICECCO J, GAUDETTE J E, SIMMONS J A. Multi-component separation and analysis of bat echolocation calls[J]. The Journal of the Acoustical Society of America, 2013, 133 1 : 538 – 46.
[59] NAGAPPA S, HOPGOOD J R. Frequency tracking using Monte Carlo methods: application to bat echolocation signals[J]. 2008 16th European Signal Processing Conference, 2008 : 1 – 5.
[60] HERMAN K, GUDRA T. New approach in bats’ sonar signals parametrization and modelling[J]. Physics Procedia, 2010, 3 : 217 – 224.
[61] LI T, MA X. Numerical research on the acoustic deatures of the vocal tract in a horseshoe bat, Rhinolophus hipposideros[J]. 2013 International Conference on Computer Sciences and Applications,
2013 : 549 – 553.
[62] MA X, LI T, LU H. The acoustical role of vocal tract in the horseshoe bat, Rhinolophus pusillus[J]. The Journal of the Acoustical Society of America, 2016, 139 3 : 1264 – 71.
[63] LAURE DESUTTER T R, SUEUR J. The sound library of the Muséum national d’Histoire naturelle[EB/OL]. . https://sonotheque.mnhn.fr/ q=chiropt%C3%A8res/.
[64] 丁玉美, 高西全. 数字信号处理 (第二版)[M]. [S.l.] : 数字信号处理（第二版） , 2001.
[65] ANGELOSANTE D, GIANNAKIS G B. Group lassoing change-points in piecewise-constant AR processes[J]. EURASIP Journal on Advances in Signal Processing, 2012, 2012 : 1 – 16.
[66] TARAGNA M. Identification of time-varying processes[J]. Automatica, 2000, 38 : 1434 – 1436.
[67] FAN J, LI R. Variable selection via nonconcave penalized likelihood and its oracle properties[J]. Journal of the American Statistical Association, 2001, 96 : 1348 – 1360.
[68] YUAN M, LIN Y. Model selection and estimation in regression with grouped variables[J]. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2006, 68.
[69] DOBIGEON N, TOURNERET J-Y, DAVY M. Joint segmentation of piecewise constant autoregressive processes by using a hierarchical model and a Bayesian sampling approach[J]. IEEE Transactions on Signal Processing, 2006, 55 : 1251 – 1263.
[70] REICH C, NICOLAOU C, MANSOUR A, et al. Bayesian estimation of recurrent changepoints for signal segmentation and anomaly detection[J]. 2019 27th European Signal Processing Conference(EUSIPCO), 2019 : 1 – 5.
[71] FEARNHEAD P. Exact Bayesian curve fitting and signal segmentation[J]. IEEE Transactions on Signal Processing, 2005, 53 : 2160 – 2166.
[72] ZOU R, WANG H, CHON K H. A robust time-varying identification algorithm using basis functions[J]. Annals of Biomedical Engineering, 2003, 31 : 840 – 853.