Basis of acoustic discrimination of Chinook salmon from other salmons by echolocating Orcinus orca

The "resident" ecotype of killer whales (Orcinus orca) in the waters of British Columbia and Washington State have a strong preference for Chinook salmon even in months when Chinook comprise less than about 10% of the salmon population. The foraging behavior of killer whales suggests that they depend on echolocation to detect and recognize their prey. In order to determine possible cues in echoes from salmon species, a series of backscatter measurements were made at the Applied Physics Laboratory (Univ. of Wash.) Facility on Lake Union, on three different salmon species using simulated killer whale echolocation signals. The fish were attached to a monofilament net panel and rotated while echoes were collected, digitized and stored on a laptop computer. Three transducer depths were used; same depth, 22° and 45° above the horizontal plane of the fish. Echoes were collected from five Chinook, three coho and one sockeye salmon. Radiograph images of all specimens were obtained to examine the swimbladder shape and orientation. The results show that echo structure from similar length but different species of salmon were different and probably recognizable by foraging killer whales.     

Compressive sensing: a strategy for fluttering target discrimination employed by bats emitting broadband calls

When foraging, so-called FM-bats emit sequences of frequency modulated (FM) calls in order to detect, identify, and localize edible prey. Once a potential target has been detected, various call and call sequence parameters, such as frequency sweep, pulse duration, and inter pulse interval (IPI) vary. In this paper, the possible functions of the variation of the IPI are studied. In particular, it is conjectured that the IPI patterns are an adaptive behavior that optimizes the signal design parameters in order to improve information retrieval. Such an irregular sampling strategy would be useful whenever bats need to characterize signal modulation (e.g., the wing beat of an insect) using a call emission rate lower than the signal modulation of interest. This problem can be recast as extracting features, in this case the joint acoustic and modulation frequency representation, from signals sampled at frequencies well below the Nyquist cut-off frequency. To study the possibility of such target classification using a sub-Nyquist sampling scheme, results derived in the context of compressive sensing are used. Processing echoes collected from both rotating computer fans and fluttering locusts, it is shown that such a strategy would allow FM-bats to discriminate between targets based on their different fluttering rates.     

Evidence for spatial representation of object shape by echolocating bats (Eptesicus fuscus)

Big brown bats were trained in a two-choice task to locate a two-cylinder dipole object with a constant 5 cm spacing in the presence of either a one-cylinder monopole or another two-cylinder dipole with a shorter spacing. For the dipole versus monopole task, the objects were either stationary or in motion during each trial. The dipole and monopole objects varied from trial to trial in the left-right position while also roving in range (10-40 cm), cross range separation (15-40 cm), and dipole aspect angle (0 degrees -90 degrees ). These manipulations prevented any single feature of the acoustic stimuli from being a stable indicator of which object was the correct choice. After accounting for effects of masking between echoes from pairs of cylinders at similar distances, the bats discriminated the 5 cm dipole from both the monopole and dipole alternatives with performance independent of aspect angle, implying a distal, spatial object representation rather than a proximal, acoustic object representation.     

Changes in signal parameters over time for an echolocating Atlantic bottlenose dolphin performing the same target discrimination task

This study documents the changes in peak frequency, source level, and spectrum shape of echolocation clicks made by the same dolphin performing the same discrimination task in 1998 and in 2003/2004 with spherical solid stainless steel and brass targets. The total average peak frequency used in 1998 was 138 kHz but in 2003/2004 it had shifted down nearly 3.5 octaves to 40 kHz. The total average source level also shifted down from 206 dB in 1998 to 187 kHz in 2003/2004. The standard deviation of these parameter values within time periods was small indicating a consistent difference between time periods. The average parameter values for clicks used when exposed to brass versus steel targets were very similar indicating that target type did not greatly influence the dolphin's average echolocation behavior. The spectrum shapes of the average clicks used in 1998 and in 2003/2004 were nearly mirror images of each other with the peak energy in 2003/2004 being concentrated where the 1998 clicks had the lowest energy content and vice versa. Despite the dramatic differences in click frequency content the dolphin was able to perform the same discrimination task at nearly the same level of success.     

Developmental changes in ultrasonic vocalizations by infant Japanese echolocating bats, Pipistrellus abramus

Developmental changes in vocalizations by Pipistrellus abramus were investigated during the first post-natal month. Vocalizations by pups on the day of birth were frequency-modulated ultrasounds from 30.0 ± 4.0 kHz to 19.3 ± 1.9 kHz with multiple harmonics. The terminal frequency of the second harmonic (TF(2)) of pup vocalizations corresponded to that of the fundamental (TF(1)) in adult bats (41.4 ± 2.6 kHz), suggesting that pup vocalizations can be easily detected by the mother. In addition, there are two types of infant vocalization: short duration echolocation precursor and long duration isolation calls, which showed separate developmental patterns over time.     

Vocal control of acoustic information for sonar discriminations by the echolocating bat, Eptesicus fuscus

This study aimed to determine whether bats using frequency modulated (FM) echolocation signals adapt the features of their vocalizations to the perceptual demands of a particular sonar task. Quantitative measures were obtained from the vocal signals produced by echolocating bats (Eptesicus fuscus) that were trained to perform in two distinct perceptual tasks, echo delay and Doppler-shift discriminations. In both perceptual tasks, the bats learned to discriminate electronically manipulated playback signals of their own echolocation sounds, which simulated echoes from sonar targets. Both tasks utilized a single-channel electronic target simulator and tested the bat's in a two-alternative forced choice procedure. The results of this study demonstrate changes in the features of the FM bats' sonar sounds with echolocation task demands, lending support to the notion that this animal actively controls the echo information that guides its behavior.     

Echo-intensity compensation in echolocating bats (Pipistrellus abramus) during flight measured by a telemetry microphone

An onboard microphone (Telemike) was developed to examine changes in the basic characteristics of echolocation sounds of small frequency-modulated echolocating bats, Pipistrellus abramus. Using a dual high-speed video camera system, spatiotemporal observations of echolocation characteristics were conducted on bats during a landing flight task in the laboratory. The Telemike allowed us to observe emitted pulses and returning echoes to which the flying bats listened during flight, and the acoustic parameters could be precisely measured without traditional problems such as the directional properties of the recording microphone and the emitted pulse, or traveling loss of the sound in the air. Pulse intensity in bats intending to land exhibited a marked decrease by 30 dB within 2 m of the target wall, and the reduction rate was approximately 6.5 dB per halving of distance. The intensity of echoes returning from the target wall indicated a nearly constant intensity (-42.6 +/- 5.5 dB weaker than the pulse emitted in search phase) within a target distance of 2 m. These findings provide direct evidence that bats adjust pulse intensity to compensate for changes in echo intensity to maintain a constant intensity of the echo returned from the approaching target at an optimal range.     

Stabilization of perceived echo amplitudes in echolocating bats. II. The acoustic behavior of the big brown bat, Eptesicus fuscus, when tracking moving prey.

Big brown bats, Eptesicus fuscus, can be trained to use echolocation to track a small microphone with a food reward attached when it is moved rapidly toward them. This situation mimics prey interception in the wild while allowing very precise recording of the sonar pulses emitted during tracking behavior. The results show that E. fuscus intensity compensates, reducing emitted intensity by 6 dB per halving of target range so that the intensity incident upon the target is constant and echo intensity increases by 6 dB per halving of range. This increase in echo intensity is effectively canceled by the reduction in auditory sensitivity due to automatic gain control (AGC) of 6 to 7 dB per halving of range. Intensity compensation behavior and AGC therefore form a dual-component, symmetrical system that stabilizes perceived echo amplitudes during target approach. The same system is present in the fishing bat, Noctilio leporinus, suggesting that it may be widespread in echolocating bats. Correlation analysis shows that, despite large changes in the duration of the pulses emitted by E. fuscus during an approach, the pulse frequency structure is such that the spatial image of the target perceived along the range axis is highly stable. Pulse duration is not reduced in the manner theoretically necessary to eliminate potential echo distortion effects due to AGC, but is reduced in such a way that this distortion is insignificant. During the terminal buzz, a high degree of temporal overlap (relative to pulse duration) occurs between emitted pulse and returning echo.     

双线阵目标左右舷分辨概率分析

基于二元假设检验理论,研究了双线列阵对目标方位左右舷分辨的正确和错误概率。根据阵列接收窄带信号的概率密度函数,定义了左右舷判决统计量,当信号快拍数足够大时,左右舷判决统计量接近正态分布,从而可得到窄带信号的左右舷分辨概率。利用子带信号左右舷分辨概率的频率加权或波束加权,得到频率或波束加权的宽带信号左右舷分辨概率。理论和数值分析表明,窄带信号的左右舷分辨概率取决于双阵间距,而宽带信号的左右舷分辨概率对阵间距的宽容性较大。     

Integral array of acoustic sensors for micro-liter liquid discrimination

It is shown experimentally that the sensing properties of normal acoustic waves propagating in a piezoelectric plate are sensitive to the mass load and the viscosity and electric conductivity of a liquid deposited on the plate. The sensing properties depend on (a) the direction of propagation, (b) the mode number for each direction, and (c) the plate thickness normalized to the acoustic wavelength for each mode and each direction. Based on these properties, a prototype of an acoustic liquid sensor array with different patterns of selectivity is developed and tested. The sensors are located on a single plane of a piezoelectric plate and do not use any sorbent film for sensing. The array is capable of identifying types and sorts of liquids according to their physical effects on the absorption of normal waves with different numbers. The total number of acoustic responses is presented in the form of histograms that are acoustic images of the analyzed substances’ flavors. Having no sorbent film improves the reliability and long-term consistency of sensing.     

运动目标声信号的时频特征研究

提高声被动目标定位精度需要研究运动目标在有风条件下目标声信号的特征。本文从声学波动方程出发，对无风和有常速风情况下的波动方程进行了求解，得到了运动目标声信号的延迟关系；分析了解的主要影响项，并得到了其解析信号；用解析信号对运动目标声信号的延迟特性和时频特征进行了仿真计算。     

水声目标辐射噪声的局部线性预测分析

从动力学的观点,对水声目标辐射噪声进行相空间的局部线性预测。计算了辐射噪声序列的平均预测误差与邻域大小的关系曲线,从曲线的类型可判断辐射噪声序列的线性性,并得到合适的预测函数类型及参数。文中对线谱和连续谱成分分别进行了局部预测。此外,还讨论了使预测误差最小的模型参数的选取。     

Range discrimination by big brown bats (Eptesicus fuscus) using altered model echoes: implications for signal processing

The sonar emissions of two big brown bats (Eptesicus fuscus) were modeled to create a "normal" echolocation signal for each bat which was then used as an artificial echo to synthesize a phantom target. The bat's task was to indicate which of two phantom targets (presented singly) was the "near" target and which the "far" target. Threshold range discrimination at a nominal target distance of 80 cm was about 0.6 cm for both bats. The normal signal was then modified to change the relative energy in each harmonic, the signal duration, the curvature of the frequency sweep, the absolute frequency, the phase of the second and third harmonics relative to the first, or the Doppler shift of the signal. To determine which modifications affected ranging performance, the altered models were used in tests of range discrimination that were interleaved on a day-to-day basis with tests using the normal model. Of the 12 modifications tested, only those changing the curvature of the frequency sweep affected performance. This result appears not to be predicted by current models of echo processing in FM bats. Eptesicus may be able to compensate for certain types of distortions of a returning echo, an ability possibly related to Doppler tolerance or to the characteristics of the natural variation in a bat's emissions.     

Correlational analysis of acoustic cues for the discrimination of auditory motion.

The sound of a source moving in a straight path and passing directly in front of the listener on the azimuthal plane was synthesized over headphones to include three dynamic cues for motion: Doppler effect, overall intensity, and interaural time difference. Discriminability of a change in displacement, velocity, and acceleration of this source was measured using a standard two-interval, forced-choice procedure. In each case, the relative reliance or weight given to the three acoustic cues was estimated from correlations of the listener's response with small independent pertubations imposed on cues from trial to trial. Group estimates of threshold agreed well with results from past studies, while the obtained pattern of weights depended on the individual, starting velocity, and discrimination task. For the discrimination of displacement at moderate velocity (10 m/s), responses were most highly correlated with intensity or interaural time difference. For the discrimination of velocity and, to a lesser extent, acceleration, responses were most highly correlated with Doppler effect. At higher velocity (50 m/s) responses in all discrimination tasks were most strongly correlated with Doppler effect with few exceptions. Randomizing source spectrum or roving distance of the source from trial to trial did not significantly affect the pattern of results. The results suggest that motion perception is mutable, and not in all cases based on a single invariant acoustic cue.     

A physical basis for designing integrated acoustic network systems for underwater observations

A concept for designing network systems for underwater observations based on the joint development of hydroacoustic radiating and receiving elements integrated with different channels of communication is discussed. The main requirements for such systems, principles, and algorithms of their operation under nonstationary conditions of actual marine waters are formulated. The results from several numerical and field experiments demonstrating the performance of certain elements of such systems in typical shallow water basins are presented.     

一种快速水声目标波达方向估计方法

以基追踪为重构算法的水声目标波达方向估计方法在实际的应用中往往运算速度较慢,针对这一问题,在分析水声目标空间稀疏特性的基础上,结合压缩感知理论框架下波达方向估计的特点,提出一种快速水声目标波达方向估计方法。该方法通过逐步减小逼近参数的方式来得到l₀范数最优解,实现水声目标信号的波达方向估计。通过计算机仿真从成功率、运算时间、分辨角度等多个方面与基追踪算法进行比较分析,实验表明:在成功率和分辨角度上所提出算法与基追踪算法性能相当,但运算时间却仅仅是基追踪算法的1/11。     

声场干涉结构相似度匹配的目标运动参数估计

针对水中运动声源定位问题,建立了一种基于声场干涉结构谱向量相似度匹配的目标运动参数估计方法。结合声场干涉结构水平纵向相关特性,利用目标辐射噪声LOFAR谱图中不同频率的谱值序列,基于两方位-两距离目标运动要素计算方法获取的目标距离解向量,构造声场干涉结构谱向量相似度匹配代价因子,实现目标运动参数估计。仿真数据与试验数据验证结果表明:在绘制的目标运动要素解算结果可视化分布图中可清晰呈现出目标初距、航向和速度结果。该算法实现过程简单、解算效率高,对目标方位误差有较好的鲁棒性。     

移动平台超短基线阵实现水下目标高精度定位

针对自主式水下航行器如AUV (Autonomous Underwater Vehicle)等水下移动平台,提出了一种超短基线多元阵的快速收敛的高精度声定位方法。通过对比超短基线多元阵定位误差的克拉美罗下界,对阵型结构和阵元数目进行了优化,确定优化的阵列为9元立方阵;融合移动平台行进过程中的多次定位结果,进一步降低定位误差;并利用梯度下降法实现平台轨迹优化,使得系统最快收敛到需要的定位精度。仿真结果表明,在1000 m距离内,利用优化轨迹可以将定位均方根误差快速收敛至预设的0.2%以内.此外,对于海试实验采集数据的目标定位结果显示,对距基阵约700 m及400 m的目标,5元阵定位标准差相较4元阵缩小了约2 m及1 m,验证了多元阵定位方法在实际海洋环境中的有效性.