Access the full text.
Sign up today, get DeepDyve free for 14 days.
S. Chan, Yuexian Zou (2004)
A recursive least M-estimate algorithm for robust adaptive filtering in impulsive noise: fast algorithm and convergence performance analysisIEEE Transactions on Signal Processing, 52
(2003)
Noise and the prematureinfant: physilological effects and practice impications
Ming Zhang, H. Lan, W. Ser (2003)
A robust online secondary path modeling method with auxiliary noise power scheduling strategy and norm constraint manipulationIEEE Trans. Speech Audio Process., 11
S. Kuo, D. Morgan (1999)
Active noise control: a tutorial reviewProc. IEEE, 87
S. Kuo, D. Morgan (1996)
Active Noise Control Systems: Algorithms and DSP Implementations
R. Etzel, S. Balk, C. Bearer, Mark Miller, K. Shea, P. Simon, H. Falk, Riva Miller, W. Rogan, J. Hendrick, L. Schell (1997)
Noise: a hazard for the fetus and newborn. American Academy of Pediatrics. Committee on Environmental Health.Pediatrics, 100 4
Yuexian Zou, Chan Shing-Chow, Ng Tung-Sung (1999)
A robust M-estimate adaptive filter for impulse noise suppression1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258), 4
Yuexian Zou, S. Chan, T. Ng (2004)
Title A robust M-estimate adaptive filter for impulse noise suppression
W. Carvalho, M. Pedreira, Maria Aguiar (2005)
Nível de ruídos em uma unidade de cuidados intensivos pediátricosJornal De Pediatria, 81
J. Standley (2002)
A meta-analysis of the efficacy of music therapy for premature infants.Journal of pediatric nursing, 17 2
G. Whitwell (1999)
The Importance of Prenatal Sound and MusicJournal of Prenatal and Perinatal Psychology and Health, 13
Priya Thanigai, S. Kuo (2007)
Intrauterine Acoustics Embedded Active Noise Controller2007 IEEE International Conference on Control Applications
H. Murooka, Y. Koie, N. Suda (1976)
[Analysis of intrauterine sounds and their tranquillizing effects on the newborn infant].Journal de gynecologie, obstetrique et biologie de la reproduction, 5 3
B. Rosner, N. Doherty (1979)
The Response of Neonates to Intra‐uterine SoundsDevelopmental Medicine & Child Neurology, 21
J. Tulloch, B. Brown, H. Jacobs, D. Prugh, W. Greene (1964)
Normal Heartbeat Sound and the Behavior of Newborn Infants‐‐A Replication StudyPsychosomatic Medicine, 26
Yuexian Zou, S. Chan, T. Ng (2000)
Fast least mean M-estimate algorithms for robust adaptive filtering in impulse noise2000 10th European Signal Processing Conference
W. Carvalho, M. Pedreira, Maria Aguiar (2005)
Noise level in a pediatric intensive care unit.Jornal de pediatria, 81 6
P. Bremmer, J. Byers, E. Kiehl (2003)
Noise and the premature infant: physiological effects and practice implications.Journal of obstetric, gynecologic, and neonatal nursing : JOGNN, 32 4
W. Gan, S. Kuo (2002)
An integrated audio and active noise control headsetIEEE Trans. Consumer Electron., 48
S. Kuo, Bob Lee, W. Tian (2001)
Real-time digital signal processing
(2006)
Real-Time Digital Signal Processing Implementations and Applications
P. Rousseeuw, A. Leroy (2005)
Robust Regression and Outlier Detection
Priya Thanigai, S. Kuo, R. Yenduri (2007)
Nonlinear Active Noise Control for Infant Incubators in Neo-Natal Intensive Care Units2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07, 1
A. Johnson (2001)
Neonatal response to control of noise inside the incubator.Pediatric nursing, 27 6
Hindawi Publishing Corporation Advances in Acoustics and Vibration Volume 2008, Article ID 495317, 9 pages doi:10.1155/2008/495317 Research Article Still in Womb: Intrauterine Acoustic Embedded Active Noise Control for Infant Incubators Lichuan Liu, Shruthi Gujjula, Priya Thanigai, and Sen M. Kuo Department of Electrical Engineering, Northern Illinois University, DeKalb, IL 60115, USA Correspondence should be addressed to Lichuan Liu, ll24@njit.edu Received 1 December 2007; Revised 4 February 2008; Accepted 5 March 2008 Recommended by Marek Pawelczyk Excessive noise in neonatal care units and inside incubators can have a number of detrimental effects on an infant’s health. We proposed a novel, audio-integrated approach to achieve active noise control (ANC) for infant incubators. We also presented the implementation of the robust, nonlinear filtered-X least mean M-estimate algorithm, for reducing impulsive interference in incubators. The healthcare application is further enhanced by integrating the “womb effect”, that is, by using intrauterine and maternal heart sounds, proven to be beneficial to infant health, for soothing the infant and masking the residual noise. A computer model for audio-integrated noise cancellation utilizing experimentally measured transfer functions is developed for simulations using real medical equipment noise. The simulation of the audio integrated ANC system produced optimal results and the system was further validated by real-time experiments to be robust and efficient. Copyright © 2008 Lichuan Liu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 1. INTRODUCTION the incubator adaptively [5, 6]. Another approach to create a healthier ambience in NICUs is the introduction of Neonatal intensive care units (NICU) house and treat intrauterine audio into the incubator that allows the infant to premature infants until their organ systems are considered feel comforted. Intrauterine audio is a combination of low- fully developed. These infants are enclosed in incubators, frequency sounds from the womb and includes the sound of as shown in Figure 1, that monitor their vital statistics and the muffled heartbeat which can be heard distinctly in the ensure that environmental conditions are maintained at background. However, neither playing soothing audio nor applying optimum levels. The incubators create the precise and con- sistent environment [1], such as temperature and humidity, an ANC system is individually efficient creating the need controlled by microprocessor. However, according to the for an integrated system that can reduce harmful equipment American Academy of Pediatrics [2], high noise levels are noise while simultaneously playing beneficial intrauterine common in the NICU and in incubators, causing consid- audio. To achieve this end, this paper proposes an innovative erable auditory damage to preterm infants [3]. The noise application for neonatal healthcare—the intrauterine acous- is typically due to ventilation or breathing equipment and tics embedded active noise controller. The integrated system human activity. Figure 2 is an example of the real incubator aims at recreating prenatal ambience for premature infants noise in time domain with segments marked by impulse due who are required to spend extended periods enclosed inside to respiratory pumps and the background equipment hum. infant incubators. The consequences of exposing infants to incubator noise Section 2 of the paper discusses the positive effects of vary from short-term effects such as sleep disturbance to playing uterine audio to premature infants. These positive long-term effects such as delayed speech development. To effects are both medical and psychological, and reflect reduce medical equipment noise and external noise from results from studies carried out over the last three decades. the NICU, passive control systems such as absorbers [4]are Section 3 focuses on developing an ANC system utilizing not always efficient. This puts forth a need for an active the filtered-X least mean square (FXLMS) algorithm for noise control (ANC) system that can cancel noise inside cancellation of broadband noise using transfer functions 2 Advances in Acoustics and Vibration 2. A STUDY OF NEONATAL RESPONSE TO UTERINE SOUNDS This section briefly reviews that the numerous benefits intrauterine audio has on neonatal growth from [5]. It is widely accepted that the brain of the fetus develops while it is inside the womb. An infant’s ears begin to develop when it is around eight weeks old and can be considered fully developed by the twenty-fourth week. The development of the inner ears and the nerve endings from the brain is so advanced that the baby can hear the muffled sounds of the heartbeat and the blood flowing through the umbilical cord. The human cochlear system, which is considered fully developed by the twenty-fourth week, transforms acoustic vibrations into nervous influx allowing infants to have an understanding of rhythm at a very early stage [7]. These sounds form an imprint on the fetal brain and it has Figure 1: Mobile incubator unit: Giraffe Incubator by GE Health- been verified that post birth, the infant is comforted while care. listening to it. Playing soothing audio has always been known to relieve stress and has in recent years become an established form of 0.8 therapy. There have been a number of studies that indicate Background equipment hum that music has a positive impact on premature infants yet 0.6 Impulse due to the kind of audio to be played is contentious. The various breathing pumps 0.4 available options include playing nature sounds, live and recorded music. But “womb music” has consistently been 0.2 considered the most favorable choice. According to [7] the womb is not a silent place and is typically awash with sounds. Sounds that are heard inside the uterus include maternal −0.2 heartbeat, respiration, intestinal gurgling and sounds from −0.4 blood vessels. The maternal heartbeat heard by the infant is amuffled version of the original as it passes through layers −0.6 0 1 2 3 4 5678 9 of tissues before reaching the infant. A study conducted by ×10 Time index n Rosner and Doherty in [8] states that “playing prerecorded intrauterine sounds to newborns reportedly soothes the Figure 2: Example wave form of incubator noise, sampling babies.” The study concluded that 90% of infants who lis- frequency f = 4KHz. tened to intrauterine audio were calmed down significantly. In another study conducted by Murooka et al. [9], the authors used a piezoelectric microphone to record and measured from the real GE Healthcare Girraffe incubator. analyze intrauterine sounds. The sounds were found to be The laboratory setup was modeled using the same incubator mainly from blood vessels and were found to produce a shown in Figure 1. Section 4 introduces the novel filtered- calming effect on 86% of the infants, and 30% of the infants X least mean M-estimate (FXLMM) algorithm that is were found to have increased sleep cycles. The authors found to be statistically robust in the presence of impulsive asserted that playing such sounds externally recreates the interference in the input. Section 5 outlines the audio- “in-utero” ambience for infants [9]. A pioneering study integration algorithm that introduces intrauterine audio conductedbySalk[10] exposed neonates to prerecorded and allows it to be played simultaneously while the ANC maternal heartbeat and concluded that test infants showed system is in operation. This integration serves two important increased weight gain and food intake. Flowers, McCain, and purposes—it provides a potential health benefit for infants Hilker combined uterine sounds with soft ballads and tested by utilizing womb sounds as heard by the infant and also the impact of music on nine African-American premature masks the residual noise after noise cancellation has been infants. The infants displayed improvement in respiration performed. The algorithm is intended to prevent interference rate, oxygen saturation, and time spent in sleeping [11]. from the soothing audio on the performance of the ANC This paper therefore proposes the utilization of sound algorithm and ensures that the audio is not cancelled files from a commercially available product—the Baby Sleep by the ANC system. The audio interference cancellation System [12]. The soothing audio consists of intrauterine filter also performs online modeling of the secondary heartbeat recorded through a condenser microphone, which path to enhance the performance of the ANC system. is a very accurate representation of uterine sounds as heard Section 6 shows the simulation and real time experiment by the infant. The heartbeats were taken at 72 beats per results. minute, the rate of a relaxed adult heart. They were combined Amplitude Lichuan Liu et al. 3 Magnitude response of incubator noise x(n) d(n) e(n) Primary path P(z) y (n) −5 y(n) Adaptive filter −10 S(z) W (z) −15 S(z) −20 −25 LMS −30 Figure 4: Block diagram of ANC system with the FXLMS algo- −35 rithm. −40 0 500 1000 1500 2000 Frequency (Hz) In Figure 4, S(z) which is the secondary path between e(n)and y(n), includes the secondary loudspeakers, error Figure 3: Magnitude response of the incubator wideband noise. microphones, and acoustic path between the loudspeakers and the error microphones. The secondary path is modeled offline and retained during the online operation of ANC. The with the sound of blood and fluid movement to produce an estimate compensates for the secondary-path effects [15]. “in-utero” effect for the infant. This audio was incorporated The output of the adaptive filter can be represented along with the ANC system and serves two main purposes. as [15] The ANC system is optimized to cancel equipment and external NICU noise to the maximum possible extent. The y(n) = w (n)x(n), (1) audio integration allows for the soothing audio to be played continuously without interfering with the ANC system. Also, w (n) w (n) ··· w (n) where w(n) = [ 0 1 L−1 ] is the coef- the integrated system can be considered cost effective as the ficient vector of the adaptive filter W (z)and x(n) = power amplifiers and loudspeakers used by the ANC system [x(n) x(n − 1) ··· x(n − L +1)] is the L× 1 reference sig- can be used for playing the soothing audio, thus maximizing nal vector. The signal y(n) is filtered through the secondary the utility of resources. path S(z) and is subtracted from the primary noise d(n)to generate the residual error e(n). The equations for simulation 3. ACTIVE NOISE CONTROL FOR THE INCUBATORS are given by The noise in incubator can be classified as broadband noise d(n) = p(n)∗x(n), becauseitcoversawide rangeoffrequencies [13]. The y (n) = s(n)∗y(n), (2) noise sources are some medical equipments in the ICU, such as a blowers, nebulizers, humidifiers, and pumps. Figure 3 e(n) = d(n) − y (n) = d(n) − s(n)∗ w (n)x(n) , shows the magnitude spectrum of the recorded sample of broadband incubator noise. We can find that the power of where ∗ denotes the convolution operator, and p(n)and s(n) the noise is spread over a wide spectrum of the noise signal. are the primary and secondary path responses, respectively. The ANC systems can be used to cancel this high-power All these operations are carried out by the system internally wideband noise. and the signals picked up in real-time ANC are the reference ANC is based on the principle of untilizing destructive signal x(n) and the residual error e(n). For the adaptive filter, interference to cancel unwanted noise. The objective of an the weight update equation is ANC system is to generate an “antinoise” to cancel the primary noise. The amount of noise which can be cancelled w(n +1) = w(n)+ μe(n)x(n), (3) depends on the accuracy of the amplitude and phase of this antinoise [14]. μ is the step size; x(n) = [x (n) x (n−1)··· x (n−L+1)] The block diagram of a feedforward broadband ANC is the reference signal vector x(n) filtered by the secondary system using the FXLMS algorithm is illustrated in Figure 4, path model S(z), where P(z) is the transfer function of the primary path from the noise source to the error microphones, S(z) is the transfer x(n) = s(n)∗x(n), (4) function of secondary path and S(z) is its estimate. The primary noise d(n) inside the incubator is cancelled by the where s(n) is an accurate estimate of s(n). antinoise y(n) generated by the adaptive filter W (z). The The experimental setup is shown in Figure 5.One antinoise is produced by the secondary loudspeakers and microphone is placed on either side of the infant head. The e(n) is the residual noise picked up by the error microphone. outputs from both are analyzed by a spectrum analyzer. Spectrum (dB) 4 Advances in Acoustics and Vibration Primary speaker to left mic −10 −20 −30 −40 −50 −60 00.10.20.30.40.50.60.70.80.91 Normalized frequency (× rad/sample) Primary speaker to right mic −10 −20 −30 −40 Figure 5: Experiment setup by using the GE Healthcare Giraffe −50 −60 Incubator. 00.10.20.30.40.50.60.70.80.91 Normalized frequency (× rad/sample) Figure 6: Magnitude responses of the primary paths. The cancelling loudspeakers are placed in the incubator, and can be seen behind the infant head. Typically the offline modeling of the secondary paths from the cancelling loudspeakers to the error microphones is using adaptive which is well known for its computational simplicity. It filters with the least mean square algorithm. The magnitude defines as responses of the primary paths P(z) from the experimental setup are shown in Figure 6. e (n) ⎪ ,0 ≤ e(n) <ξ , Typically, white noise is used for adaptive system iden- ⎪ tification. But it is found to be annoying especially in ⎪ 2 sensitive environments like the NICU. The proposed method ⎪ ξ e(n) − , ξ ≤ e(n) < Δ , utilizes offline modeling approach. Nature’s sound, in this ⎪ case, the sound of a flowing stream is used. Nature’s ξ ξ ξ e(n) − Δ ρ e(n) = − + sounds are preferred owing to their flat spectrum and their ⎪ 2 Δ + Δ 2 2 Δ − Δ 1 2 1 2 pleasing effect on the listener. The secondary paths estimator Δ ≤ e(n) < Δ , ⎪ 1 2 converged for a filter length of 30. Satisfactory results of offline modeling are shown in Figure 7. ⎪ ξ ξ ⎩ − , Δ ≤ e(n) , 2 Δ + Δ 2 1 2 4. NONLINEAR ALGORITHM FOR IMPULSE (7) NOISE SUPPRESSION where ξ , Δ ,and Δ are the threshold parameters. 1 2 The performance of the linear adaptive filters degrade The objective function is minimized by dramatically in the presence of impulse noise, therefore nonlinear algorithms are capable of reducing the adverse ∂J ∂ρ e(n) ∂e(n) MP(n) effects [16]. The FXLMM algorithm is a simple and robust ∇J = = . (8) MP ∂w(n) ∂e(n) ∂w(n) method. It employs the mean M-estimation error objective function and is capable of performing effectively in impulsive Let ψ [e(n)] be the first-order partial derivative of ρ[e(n)], environment [17–19]. (8)becomes The objective of the adaptive filter W (z) is to minimize the least M-estimate function criterion ρ[e(i)] where ρ(·) ∂e(n) is the M-estimate function. The coefficient vector w(n)is ∇J = ψ e(n) = ψ e(n) − s(n)∗x(n) MP ∂w(n) updated in the negative direction of the gradient vector (9) =−q e(n) e(n) s(n)∗x(n) . w(n +1) = w(n) − μ∇J (5) MP Define q[e(n)] ≡ ψ [e(n)]/e(n) as the weight function. Since and the objective function is s(n) is the impulse response of the secondary path and not available directly, we use its estimation to calculate the J ≡ E ρ e(n) ρ e(n),(6) = gradient, MP where E[·] is the expectation operator, ρ(·) is chosen to be ∇J −q e(n) e(n) s(n)∗x(n) =−q e(n) e(n)x(n). MP the Hampel three-part redescending M-estimate function, (10) Magnitude (dB) Magnitude (dB) Lichuan Liu et al. 5 Right secondary speaker to right mic Primary path x(n) d(n) e a(n) P(z) −50 −100 00.10.20.30.40.50.60.70.80.91 y (n) y(n) Normalized frequency (× rad/sample) Adaptive filter S(z) W (z) a (n)+ Right secondary speaker to left mic S(z) 0 − S(z) −50 a(n) −100 LMS 00.10.20.30.40.50.60.70.80.91 x (n) LMS Normalized frequency (× rad/sample) e(n) Audio source Left secondary speaker to left mic Figure 8: Block diagram of the audio-integrated ANC system. −50 −100 00.10.20.30.40.50.60.70.80.91 Normalized frequency (× rad/sample) coefficients of the adaptive filter [20]. The block diagram of the audio integration algorithm is shown in Figure 8.The Left secondary speaker to right mic soothing audio a(n)isadded to y(n)and canbeheard by the infant inside the incubator. −50 −100 At the acoustic summing junction, the antinoise y (n) 00.10.20.30.40.50.60.70.80.91 and the primary noise d(n) are combined to produce the Normalized frequency (× rad/sample) residual error e a(n). It contains the true error (residual noise) e(n) and the component of audio. Therefore, by Figure 7: Magnitude responses of the secondary paths. subtracting the audio from the residual error e a(n), we can get the true error, then the true error is used to update the weight vector of the adaptive filter W (z). It should be noted Substituting (10) into (5), we can get the weight vector that the audio signal passed the secondary path, and filtered update equation as by S(z), then it is subtracted. The z transform of residual error e a(n) can be expressed as [21] w(n +1) = w(n)+ μq e(n) e(n)x(n), (11) E A(z) = D(z) − S(z) Y (z)+ A(z) . (12) where μ is the step size parameter. Equation (11) is known as the least M-estimate algorithm and can be viewed as a The adaptive filter S(z) is used to cancel the audio inter- generalization of the LMS algorithm. It becomes identical to ference on the performance of W (z). This filter generates the LMS algorithm when noise e(n) is less than a threshold ξ . When the signal error e(n) >ξ , q[e(n)] in (11)decreases E(z) = E A(z)+ S(z)A(z). (13) and reaches 0 when e(n) > Δ . Thus, the least M-estimate algorithm is capable of reducing the effect of large signal Then we can get the following equation by substituting error during the updating process [17]. (12) into (13): 5. INTRAUTERINE ACOUSTICS EMBEDDED E(z) = D(z) − S(z)Y (z). (14) ACTIVE NOISE CONTROLLER We assume that S(z) = S(z) and the audio is uncorrelated This section develops an algorithm that can integrate the with the primary noise. Then (13) can be expressed in time “comfort” audio with the existing ANC system, and provide domain as an environment that is capable of improving the health of the infant by masking the undesired residual noise. The e(n) = d(n) − y(n)∗s(n) (15) comfort audio used is a combination of maternal heartbeat and other intrauterine sounds [12]. Research has proven which is the true error used to update W (z) by using the that playing womb sound to infant in incubator showed FXLMS system. significant benefit in the respiration rate, sleep cycle, and The main advantage of this algorithm lies in its ability oxygen saturation [11]. Unfortunately, there are two main to model the secondary path online. This involves the issues with the integration of audio to the ANC system need estimation of the secondary path in parallel with the to be considered: first, the audio signal can act as interference operation of the ANC system. The S(z) filter is modeled to the ANC system and impede proper adaptation; and through a system identification scheme. It uses soothing second, the ANC system can cancel the intended soothing audio as the reference signal and treats the secondary path sound. Hence, a method must be devised to subtract the as the unknown system. This makes the algorithm sensitive audio from error signal before it is used to update the to time-varying secondary paths. Magnitude Magnitude Magnitude Magnitude (dB) (dB) (dB) (dB) 6 Advances in Acoustics and Vibration Simulated spectrum of left error microphone after Primary path1 P (z) intrauterine audio is added d (n) 20 y (n) S (z) W (z) − 1 e (n) 0 −10 S (z) e (n) 1 −20 FXLMS e (n) −30 x(n) S (z) −40 W (z) −50 e (n) S (z) 2 y (n) −60 0 50 100 150 200 250 300 350 400 450 500 Frequency (Hz) Primary path2 P (z) d (n) ANC OFF Figure 9: Block Diagram of the 1 × 2 × 2FXLMS algorithm. ANC ON Figure 10: Simulated spectra at left error microphone before (ANC OFF) and after (ANC ON) active noise control. The key advantages of the intrauterine acoustic embed- ded ANC system can be summarized as follows. (i) It re- establishes pre-natal ambience thus fostering infant health. where w (n)and w (n) are weight vectors of the adaptive 1 2 (ii) The secondary path is modeled online making the filters W (z)and W (z), respectively, μ and μ are the 1 2 1 2 system more receptive to changes in the environment. (iii) It step sizes, s (n), s (n), s (n), and s (n) are the impulse 11 12 21 22 is successful in masking residual error and in preventing the audio from interfering with the updation. (iv) The responses of S (z), S (z), S (z), and S (z), respectively. 11 12 21 22 audio integration does not require supplementary hardware, Similar to the multichannel ANC system (as shown in existing speakers and power amplifier of the ANC system can Figure 9), we extended the single-channel audio-integrated be used making it cost effective. ANC system (as shown in Figure 8) into a 1 × 2 × 2mul- tichanel system. In this multichannel intrauterine acoustics embedded ANC system, the two adaptive filters W (z)and 6. SIMULATION AND EXPERIMENT RESULTS W (z) are used to update the two antinoise y (n)and y (n). 2 1 2 6.1. Multichannel FXLMS algorithm 6.2. Simulation results In the previous sections, we described the single channel ANC system. In this section, an example of multichannel To evaluate the performance of the innovative intrauterine ANC system, 1 × 2 × 2 FXLMS algorithm is used for real acoustic embedded ANC, we investigate the noise cancella- experiment. Figure 9 shows the multichannel feedforward tion achievement through simulation and real time experi- ANC system using the 1 × 2 × 2 FXLMS algorithm. In this ment. system, two secondary speakers and two error microphones In the simulation, we apply the intrauterine acoustics are used independently. These two error microphones pick embedded ANC system described in Section 6.1. The input up the residual errors e (n)and e (n)atdifferent positions, 1 2 reference noise is taken from an incubator noise audio file thus able to form two individual quiet zones centered at the at first. The ANC system is simulated with measured P (z), error microphones. The ANC algorithm used two adaptive P (z), S (z), S (z), S (z), and S (z). A 60-tap filter with filters W (z)and W (z) to generate antinoise y (n)and y (n) 2 11 12 21 22 1 2 1 2 step size of 0.1 is used for the adaptive noise cancellation to drive the two independent secondary speakers. In Figure 9, filter W (z)and W (z). Theresidualnoise is foundtobe 1 2 d (n)and d (n) are the primary noises to be cancelled, S (z), 1 2 11 16 dB lower than the input on average. The plots illustrating S (z), S (z), and S (z) are the secondary path transfer 12 21 22 the spectra of noise before (ANC OFF) and after (ANC functions, and P (z)and P (z) are the primary path transfer 1 2 ON) cancellation at left error microphone and right error functions. microphone after assigning intrauterine audio are shown in The multichannel FXLMS algorithm is summarized as Figures 10 and 11. follows: To demonstrate the impulse noise suppress by non- linear algorithms, the noise signal is interspersed with y (n) = w (n)x(n), i = 1, 2, high-amplitude random impulses (30 dB higher than back- w (n +1) ground). The impulses are at time n = 40000, 52000, and = w (n)+μ e (n)x(n)∗s (n)+e (n)x(n)∗s (n) , i= 1, 2, 64000 and last for a length of 100 samples. The FXLMM i i i i1 i i2 algorithm was implemented for the audio-integrated ANC (16) Spectrum (dB) Lichuan Liu et al. 7 Comparision of MSE for LMS and LMM algorithms Simulated spectrum of right error microphone after intrauterine audio is added (right error microphone) 20 0.02 0.018 0.016 0.014 0.012 −10 0.01 −20 0.008 −30 0.006 0.004 −40 0.002 −50 0 50 100 150 200 250 300 350 400 450 500 0123 4 5 6 7 8 9 ×10 Frequency (Hz) Time index n ANC OFF FXLMS FXLMM ANC ON Figure 11: Simulated spectra at right error microphone before Figure 13: Learning curves at right error microphone for the (ANC OFF) and after (ANC ON) active noise control. FXLMS and FXLMM algorithms (μ = 0.0007, impulse occurred at n = 40000, 62000, and 64000). Comparision of MSE for LMS and LMM algorithms (left error microphone) Cancellation at left error microphone 0.025 0.02 −10 0.015 −20 −30 0.01 −40 0.005 −50 −60 0123 4 5 6 7 8 9 −70 0 50 100 150 200 250 300 350 400 ×10 Time index n Frequency (Hz) FXLMS ANC OFF FXLMM ANC ON Figure 12: Learning curves at left error microphone for the FXLMS Figure 14: Real-time noise cancellation at left error microphone in and FXLMM algorithms (μ = 0.0007, impulse occurred at n = the incubator. 40000, 62000, and 64000). system. The probabilities θ , θ ,and θ for determining 6.3. Real-time experiment results Δ Δ ξ 1 2 the threshold were taken to be 0.05, 0.025, and 0.005, respectively, for 95%, 97.5%, and 99.5% confidence that the A real-time experiment is set up as shown in Figure 5 with error vector was in the interval [0, ξ ], [0, Δ ], and [0, Δ ], the real GE Healthcare incubator, we use a 200 Hz sinusoidal 1 2 respectively [17]. A 60-tap adaptive filter with a step size signal generated by a loudspeaker as the primary noise (60 dB of 0.0007 was implemented. The results of incubator noise higher than the background), two antinoise loudspeakers cancellation are shown in Figures 12 and 13. are fixed in the incubator, two error microphones are The simulation results show that the FXLMM algorithm placed near baby’s ears to pick up the noise residue, the behaves in an identical manner to the FXLMS algorithm until primary microphone is set on the top of the incubator in before the impulses are encountered. The FXLMS algorithm, order to collect the primary noise signal. A TI TMS320C30 however, exhibits a degraded system performance with a very DSP is used for the ANC system. The assembly language high mean-squared error (MSE) in the presence of impulses. is used for software developing in order to achieve the The FXLMM algorithm is found to be more robust while real-time processing requirement [22]. For the real-time handling impulses. Comparing the MSE plots of the two experiment setup, the sampling frequency is 1.938 KHz, two algorithms shows that the FXLMM algorithm has superior 220-tap filters with the convergence factor of 0.0003 were performance in the presence of impulses and is more effective used for the adaptive noise cancellation filters W (z)and in suppressing the adverse influence of impulse noise. W (z). Error squared Spectrum (dB) Error squared Amplitude (dB) 8 Advances in Acoustics and Vibration Cancellation at right error microphone [4] A. N. Johnson, “Neonatal response to control of noise inside the incubator,” Pediatric Nursing, vol. 27, no. 6, pp. 600–605, 0 2001. [5] P. Thanigai, S. M. Kuo, and R. Yenduri, “Nonlinear active −10 noise control for infant incubators in neo-natal intensive care −20 units,” in Proceedings of the IEEE International Conference on −30 Acoustics, Speech and Signal Processing (ICASSP ’07), vol. 1, pp. 109–112, Honolulu, Hawaii, USA, April 2007. −40 [6] P. Thanigai and S. M. Kuo, “Intrauterine acoustic embedded −50 active noise controller,” in Proceedings of the IEEE International −60 Conference on Control Applications (CCA ’07), pp. 1359–1364, −70 Singapore, October 2007. 0 50 100 150 200 250 300 350 400 [7] G. E. Whitwell, “The importance of prenatal sound and Frequency (Hz) music,” http://www.birthpsychology.com/lifebefore/sound1 .html. ANC OFF ANC ON [8] B. S. Rosner and N. E. Doherty, “The response of neonates to intra uterine sounds,” Developmental Medicine and Child Figure 15: Real time noise cancellation at right error microphone Neurology, vol. 21, pp. 723–729, 1979. in the incubator. [9] H. Murooka, Y. Koie, and N. Suda, “Analysis of intrauterine sounds and their tranquilizing effects on the newborn infant,” Journal de Gynecologie, Obstetrique et Biologie de la Reproduc- The real-time noise cancellation results based on the tion, vol. 5, pp. 367–376, 1976. real time experiment are shown in Figures 14 and 15.By [10] J. D. Tulloch, H. L. Jacobs, D. G. Prugh, and W. A. Greene, comparing the spectra of the noise before and after cancel- “Normal heartbeat sound and the behavior of newborn lation, we can find that the residual noises are 30.231 dB infants—a replication study,” Psychosomatic Medicine, vol. 26, lower than the original noise at the left error microphone no. 6, pp. 661–670, 1964. and 23.685 dB lower at the right error microphone. After the [11] J. M. Standley, “A meta-analysis of the efficacy of music ANC system, the high power noise is dramatically reduced therapy for premature infants,” Journal of Pediatric Nursing, into an acceptable range and not harmful any more. vol. 17, no. 2, pp. 107–113, 2002. [12] The miracle baby sleep system, http://www.babysleepsystem 7. CONCLUSION .com/index.htm. [13] W. B. Carvalho,M.L.G.Pedreira, andM.A.L.deAguiar, In this paper, a novel neonatal healthcare application, the “Noise level in a pediatric intensive care unit,” Jornal de intrauterine acoustics embedded active noise controller, has Pediatria, vol. 81, no. 6, pp. 485–498, 2005. been presented. The integration algorithm created a bene- [14] S. M. Kuo and D. R. Morgan, “Active noise control: a tutorial ficial environment for the infant and allowed the residual review,” Proceedings of the IEEE, vol. 87, no. 6, pp. 943–973, noise from the ANC system to be masked. The ANC system involved an adaptive method of noise cancellation using the [15] S. M. Kuo and D. R. Morgan, Active Noise Control Systems statistically robust FXLMM algorithm. It allowed for stable Algorithms and DSP Implementations, John Wiley & Sons, New operation of the ANC system in the presence of impulsive York, NY, USA, 1996. interference in the input. Real transfer functions measured [16] P. J. Rousseeuw and A. M. Leroy, Robust Regression and Outlier from a laboratory setup were used to develop a computer Detection, John Wiley & Sons, New York, NY, USA, 1987. model for simulation of the ANC system. The integration [17] Y. Zou, S.-C. Chan, and T.-S. Ng, “Least mean M-estimate algorithm was proven to be highly advantageous as it allows algorithms for robust adaptive filtering in impulse noise,” IEEE the secondary path to be modeled online making the system Transactions on Circuits and Systems II, vol. 47, no. 12, pp. more sensitive to changes in the environment. The real time 1564–1569, 2000. controller was found to be cost effective and displayed stable [18] Y. Zou, S.-C. Chan, and T.-S. Ng, “A robust M-estimate performance in the real incubator. adaptive filter for impulse noise suppression,” in Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP ’99), vol. 4, pp. 1765–1768, Phoenix, REFERENCES Ariz, USA, May 1999. [1] G. Healthcare, Incubator the better care by design, http:// [19] S.-C. Chan and Y.-X. Zou, “A recursive least M-estimate www.gehealthcare.com/usen/perinatal/micro environments/ algorithm for robust adaptive filtering in impulsive noise: giraffe/products/giraffeincubator fnb.html. fast algorithm and convergence performance analysis,” IEEE Transactions on Signal Processing, vol. 52, no. 4, pp. 975–991, [2] R.A.Etzel,S.J.Balk, C. F. Bearer,M.D.Miller, andK.M. Shea, “Noise: a hazard for the fetus and newborn,” Pediatrics, vol. 100, no. 4, pp. 724–727, 1997. [20] M. Zhang, H. Lan, and W. Ser, “A robust online secondary [3] P. Bremmer, J. F. Byers, and E. Kiehl, “Noise and the pre- path modeling method with auxiliary noise power scheduling matureinfant: physilological effects and practice impications,” strategy and norm constraint manipulation,” IEEE Transac- Journal of Obstetric, Gynecologic, & Neonatal Nursing, vol. 32, tions on Speech and Audio Processing, vol. 11, no. 1, pp. 45–53, no. 4, pp. 447–454, 2003. 2003. Amplitude (dB) Lichuan Liu et al. 9 [21] W. S. Gan and S. M. Kuo, “An integrated audio and active noise control headsets,” IEEE Transactions on Consumer Electronics, vol. 48, no. 2, pp. 242–247, 2002. [22] S. M. Kuo, B. H. Lee, and W. Tian, Real-Time Digital Signal Processing Implementations and Applications, John Wiley & Sons, Hoboken, NJ, USA, 2006. International Journal of Rotating Machinery International Journal of Journal of The Scientific Journal of Distributed Engineering World Journal Sensors Sensor Networks Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Volume 2014 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 Submit your manuscripts at http://www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2010 Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com http://www.hindawi.com Volume 2014 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation Hindawi Publishing Corporation http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014
Advances in Acoustics and Vibration – Hindawi Publishing Corporation
Published: Apr 6, 2008
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.