基于嵌入式的助听器回波算法研究与实现

回声会导致助听器产生啸叫,损坏助听器设备,破坏患者的残余听力。为此,本文在助听器回声抵消模型的基础上,针对输入信号的能量变化,研究了基于归一化最小均方（Normalized Least Mean Squares, NLMS）的自适应助听器回声抵消算法。通过对比LMS和NLMS两种算法的MSE,ERLE等性能,研究发现NLMS算法在数字助听器模型中有更好的回声抵消性能。此外,将NLMS算法移植到嵌入式平台中,并通过实验对比算法的性能。因此,本研究对于助听器的回声抵消算法设计具有很强的实用价值。     

麦克风阵数字助听器实验平台研究与设计

随着现代数字信号处理技术的发展,数字助听器已经基本成为助听器的主流.在综合考虑助听器实时性及功耗,大小等诸多要求后,提出利用麦克风阵列改善助听器听觉效果,设计了基于麦克风阵列的数字助听器系统开发平台.平台以DSP芯片TMS320C6747为核心CPU,TLV320A1C32为音频AD采样芯片.在该平台上,对声源定位算法...     

基于麦克风阵列的数字助听器语音增强技术

针对噪声和混响环境下的助听器用户聆听上的困难,基于麦克风阵列的数字助听器设计能够很好的提高助听器在这种环境下的语音信噪比。本文研究了应用麦克风阵列进行数字助听器语音增强处理技术,提出了一种基于粒子群优化的改进粒子滤波算法,它将语音增强问题转换为从带噪语音中对纯净语音的估计过程,引入粒子群优化的方法来产生建议分布,使降噪结果更接近纯净语音,从而得到更好的语音增强效果。     

微系统技术助力数字助听器发展

随着人们对生活品质要求的提高,声障患者用耳内助听器的微型化、低功耗、无线化需求推动着微电子封装在该领域的技术进步。对现代助听器中涉及的微系统技术及这些技术给数字助听器带来的变革进行了调研。MEMS麦克风和扬声器技术,处理器、存储器芯片等IC堆叠采用的TSV技术,用于助听器连接智能手机的蓝牙天线Ai P技术,这些微系统技术有的已成熟应用于微型化数字助听器,有的还在进行不断的尝试以期早日应用于现代数字助听器中。数字助听器的发展要求微系统技术不断取得突破,微系统技术上的新材料、新方法、新工艺进步在微型数字助听器上得到了完美的展现。     

多通道数字助听器算法及低功耗VLSI设计

改进了多通道数字助听器中的听力补偿和噪声消除算法,并进行了低功耗VLSI设计.听力补偿方面,提出一种改进的多通道宽动态范围压缩(WDRC)算法;该方法降低了存储和计算开销,并抑制了对残余背景噪音的过度放大.噪声消除方面,利用语音谱和噪声谱的帧间相关性,改进了传统的多子带谱相减算法,使之在硬件实现时便于并行运算,同时不影响消噪性能.最后,综合采用多种低功耗设计方法,在SMIC的130nm工艺条件下,完成了基于上述算法的多通道数字助听器VLSI设计.后仿结果表明,该设计总功耗仅为228μW.     

一种基于声压级分段的数字助听器响度补偿方法研究

响度补偿技术是数字助听器的核心技术之一,目前大多数响度补偿算法均是笼统地通过低、中、高三段输入声压级来进行补偿增益值的计算的,这通常并不符合听力损失患者的实际需要。为此,本文针对传统三段补偿算法补偿效果的不足,提出了一种基于声压级分段的非等宽多通道响度补偿算法。实验仿真结果显示,该算法比传统的三段补偿算法更好的补偿了患者所需的听力损失,显著提升了患者的言语辨识率,且很好的消除了助听器内部噪声对患者正常使用造成的影响。     

基于加权次梯度投影算法的数字助听器自适应声反馈抑制

本文提出了一种利用加权自适应次梯度投影算法(Weighted Adaptive Projection Subgridient Method,WAPSM)进行声反馈抑制的方案.WAPSM算法来自于自适应次梯度投影算法(Adaptive Projection Subgridient Method,APSM),它以次梯度投影的超平面作为搜索区域来进行松弛投影.本文提出的算法将估计系统的先验知识以权重因子-在很多应用中为指数衰减-的方式加入APSM算法中提高算法性能.以WAPSM算法应用于助听器声反馈抑制的大量仿真实验表明,算法相比传统的NLMS算法和APSM算法在收敛速度、稳定性和精度方面取得了显著的进展.进厂步的实验表明,算法在以实际语音作为数字助听器输入信号时取得了优异的性能,并且在低信噪比条件下具有较强的鲁棒性.     

GA3280助听器开发平台分析

基于Gennum的助听器开发平台,完成了基于该平台的频域助听器系统设计,包括GA3280芯片介绍、开发板硬件资源分析、软件开发流程、频域助听器算法实现。通过简易的测试,设计的频域算法能实现助听效果。为开发基于GA3280芯片的助听器提供了软硬件设计参考。     

基于改进小波阈值法的语音去噪算法

针对数字助听器在接收和处理过程中,容易受到背景噪声干扰的问题,在传统小波阈值去噪的基础上提出了一种改进的阚值函数.该函数具有高阶连续可导,克服了传统小波阚值函数不可导的问题.利用该阈值函数对含噪语音的小波系数进行分帧实时阈值处理,以获得数字助听器中语音增强的效果.仿真结果表明:利用新阈值函数得到的语音去噪信号,其信噪比、均方误差和语音可懂度均优于其他非连续可导的阈值函数算法.     

数字移动电话与助听器兼容性测试

分析了移动电话对助听器的干扰成因。介绍了与助听器兼容性相关的标准和要求,包括美国ANSIC 63.19对助听器和数字移动电话之间电磁兼容性的要求、即将颁布的国家标准与现行通信行业标准YD/T1643—2007的差异,以及ANSI C 63.19：2007中HAC测试的内容和方法。     

Integrated frequency-domain digital hearing aid with the lapped transform

This paper presents a frequency domain digital hearing aid based upon the new lapped transform, which has the ability to eliminate the blocking effects inherent in traditional frequency domain filtering. Incorporated into the new digital hearing aid are the functions of frequency shaping, acoustic feedback cancellation and periodic noise reduction.<>     

基于人耳听觉特性的助听器响度补偿策略

设计面向老年性听力损伤患者的数字式助听器,对语音信号进行响度补偿时需要同时兼顾频率与声强级。针对人耳听觉的这一特性,归纳提出了动态压缩响度补偿策略。同时分别从对患耳敏感的低频域和高频域语音两个层面深入进行Matlab仿真验证。结果表明,此研究对于保证数字式助听器中响度补偿效果的合理性以及完整性具有重要意义。     

A PARCOR characterization of the ear for hearing aids

A possible characterization of the ear through the PARCOR algorithm is suggested. This leads to a digital filter heating aid that could be of assistance in compensating for partial hearing loss.     

Audiogram matching in hearing aid using approximate arithmetic

Multidimensional Systems and Signal Processing - Filter banks are the major signal processing blocks that dissipate large amount of power in a portable digital hearing aid device. The power...     

Reverberation canceling wireless aid for hearing impaired

The stringent requirements on size and power consumption constrain the conventional hearing aid devices. Besides providing an economic and user friendly aid, reverberation/echo cancellation is an important requirement. With the technological advancements, wireless hearing aids exploiting the usage of multi-microphones, mixed signals and RF signals processing circuits, digital signal processing units sound promising to overcome existing constraints. A new wireless hearing aid system is proposed in this paper. Unlike the previously reported wireless hearing aid concept, it needs only one way data transfer from body unit to the earpiece. It helps in minimizing power consumption in the earpiece RF-linked with a body unit with DSP based reverberation canceling scheme. For this purpose, a noise cancellation algorithm is developed based on beam-forming technique. The functioning of the whole system comprising an earpiece and a body unit has been ensured using the Advanced Design System^TM. The ADS compatible behavioral models were developed in order to enable the system level simulation. A comprehensive noise analysis is carried out and validated.     

Design of high-performance digital hearing aid processor

A digital hearing aid processor (DHAP) chip built around a general-purpose 16 bit DSP core is presented. The designed DHAP performs a nonlinear loudness correction of eight frequency bands based on acoustic measurements. The DHAP provides all the flexibility needed to implement audiological algorithms. In addition, the chip has a low power feature and 5,500×5,000 μm² dimensions that make it suitable for wearable hearing aids     

A true-1-V 300-/spl mu/W CMOS-subthreshold log-domain hearing-aid-on-chip

This paper presents a true very low-voltage low-power complete analog hearing-aid system-on-chip as a demonstrator of novel analog CMOS circuit techniques based on log companding processing and using MOS transistors operating in subthreshold. Low-voltage circuit implementations are given for all of the required functions including amplification and automatic gain control filtering, generation, and pulse-duration modulation. Based on these blocks, a single 1-V 300-/spl mu/A application specific integrated circuit integrating a complete hearing aid in a standard 1.2-/spl mu/m CMOS technology is presented along with exhaustive experimental data. To the authors' knowledge, the presented system is the only CMOS hearing aid with true internal operation at the battery supply voltage and with one of the lowest current consumptions reported in literature. The resulting low-voltage CMOS circuit techniques may also be applied to the design of A/D converters for digital hearing aids.     

Modeling the electromagnetic response of hearing aids to digitalwireless phones

Interference produced in hearing aids by the pulsed RF signal from digital wireless phones has become an increasingly important issue to wireless phone manufacturers and service providers, hearing aid manufacturers and users, and government regulatory agencies. Development and validation of a comprehensive model of the interaction would greatly benefit the efforts to achieve mutual electromagnetic compatibility (EMC). In order to develop reliable accurate methods to measure hearing aid immunity, an exact mathematical relationship must be demonstrated between the interference generated in hearing aids using a dipole with a standardized test signal [the input referenced interference level (IRIL)] and that produced by actual wireless phones with various signal formats [the overall input referenced interference level (OIRIL)]. A set of theoretical conversion factors has been developed and applied to predict OIRIL interference from the standard IRIL, measured immunity value. A square-law relationship exists within the linear response region of the hearing aid such that each 1 dB increase in RF power (or field strength in decibels V/m) results in a 2 dB sound pressure level (SPL) increase in acoustic power (or sound pressure level). Hence, the IRIL for any given field strength is obtained by doubling (in decibels) the change in field strength and adding the result to the reference IRIL (in decibels SPL). Subtracting 7.60 dB [for time-division multiple access (TDMA)-50 Hz] or 10.68 dB [for TDMA-217 Hz or code-division multiple access (CDMA)] from the IRIL predicts the corresponding OIRIL. The lower and upper limits of the predicted OIRIL are constrained by the measured ambient sound level and the amplifier saturation, respectively. The model predictions are valid when comparable field strength gradients and distributions, separation distances and orientations are maintained between the hearing aid and the RF emitter