Hearing Aid Hardware and Software: The Basics
Key Concept: Theoretical Versus Actual Values
The theoretical limit of the dynamic range in a digital system can be calculated based on the length of the digital word. In real systems, however, the actual dynamic range, in this case referring to the range from the noise floor to the maximum input level possible of being processed, will almost always be less than the theoretical limit. Similarly, amplifiers will have a theoretical efficiency limit and directional microphones will have a theoretical limit of directivity; in real systems, however, these limits will rarely be reached due to real design and implementation limitations.
The primary function of the electronic components and sound processing is to work together to increase the amplitude in a frequency-specific manner. The frequency-specific increase in sound level for a signal at the output of a hearing aid, in comparison with the sound level at the input, is referred to as gain.
The theoretical limit of the dynamic range in a digital system can be calculated based on the length of the digital word. In real systems, however, the actual dynamic range, in this case referring to the range from the noise floor to the maximum input level possible of being processed, will almost always be less than the theoretical limit. Similarly, amplifiers will have a theoretical efficiency limit and directional microphones will have a theoretical limit of directivity; in real systems, however, these limits will rarely be reached due to real design and implementation limitations.
The primary function of the electronic components and sound processing is to work together to increase the amplitude in a frequency-specific manner. The frequency-specific increase in sound level for a signal at the output of a hearing aid, in comparison with the sound level at the input, is referred to as gain.
The Basic Components of a Digital Hearing Aid
- Pre-Amplifier
- Analog to Digital converter
- Digital Signal Processor
- Digital to Analog Converter
- Memory to hold the desired amplification characteristics
Simple Journey Through a Hearing Aid
- Acoustic pressure changes in the environment are picked up and transduced into an alternating voltage that mimics the original pressure changes. Usually, these pressure changes are also provided a small amount of amplification through the preamplifier
- Current changes are sent to an analog to digital (A/D) converter which represents the current level measured at discrete time intervals using strings of binary numbers referred to as digital words give a defined sampling rate. The digital signal consists of tiny, transient pulses of voltage (0 = no voltage, 1 = tiny voltage)
- Changes are made to the signal to reflect filtering, amplification, compression and other processing using the digital sound processor (via the processor) and different types of memory which acts to hold the digital signal at various stages as well as hold the instruction and timing information
- The signal is either converted to an analog electrical signal using digital to analog conversion or digitally converted to a very high rate digital signal which represents the electrical amplitude of interest within very small adjacent blocks of time with amplitude represented by digital pulse intensity
- the digital to analog conversion leads to a low voltage analog signal and the digital to digital conversion leads to a low voltage digital signal
- After either digital to analog or digital to digital conversion, the low voltage analog or digital signal is sent to the output transistors (amplifier) which use the low voltage to control the flow of the much higher voltage from the battery. This provides a higher voltage amplified output.
- The much higher voltage is then routed to the receiver coil. The receiver then acts to transduce the electrical signal into an analog change in sound pressure for delivery to the patient's ear.
Components of Hearing Aids and Digital Signal Processing
Sound entry path is through the port, most have a front volume and a rear volume, the front is where the sound enters first. Pressure changes happen in front volume container and displace electret backplate membrane. The rear volume is where the back plate is and takes the voltage and sends it to the digital sound processor. This is a simple diagram, the sound is coming from the front and will hit the diaphragm and vibrate that creating an output signal that is fed to the amplifier and output to the patient's ear.
Schematic of an electret hearing aid microphone
- modular to simplify the manufacturing process
- microphone needs to be as durable as possible
- electret microphones, along with a miniature microphone pre-amplifier, are typically manufactured inside a metal case
- commonly referred to as a pre-amplifier or buffer amplifier to provide amplification to the very small voltage produced by the microphone, producing a signal that is better suited for amplification by the main hearing aid amplifier while keeping the main amplifier from loading down the microphone
- engineers often refer to the microphone buffer amplifiers as the FET because it is sometimes comprised of a single field effect transistor
- microphones also typically include, spacers, shielding materials and other components
- together the entire microphone and its housing are sometimes referred to as the microphone can
- the most fragile piece of the microphone, the diaphragm, can be oriented so that it is somewhat protected from objects directly entering the microphone port
- electret comes from electrostatic + magnet
MEMS - Examples of microelectromechanical systems microphones
- commonly used in cell phones and introduced into hearing aids
- silicon-based microphones are manufactured by depositing and removing semiconductor materials from a silicon wafer to form a capacitor
- the rest of the parts are then etched out and include a flexible diaphragm, a stiff backplate, and damping holes with an electrical charge on the backplate
- optimizes the interaction between sensitivity and size
- sensitivity of an electret microphone is based in part on the size of the diaphragm
- the smaller the diaphragm, the lower the sensitivity and the higher the noise
- silicon microphones can be made that are highly sensitive with extremely small diaphragms
- design offers the potential for lower noise floors and better resistance to vibration
Directional Microphones
- If have a dual microphone array can have two microphones that talk to each other
- can have different spectral arrays and can get into elegant signal processing
- a signal directional microphone, to get it to have direction can have two ports, a front and a rear
- sound comes into the front port, hits the diaphragm and vibrates it, the same sound will travel around to the back port, will have a filter that adds delay
- there is a measurable delay that happens inside the system and is how the processing decides directionality
- when sound comes through the rear of the report, it still goes through the delay but it is about as long as it would take to get from the front port to the rear port so they cancel out
- the problem is, what if you have an aid in the canal where essentially the microphone ports are facing out? this is why we have adaptability
Sensitivity Plots
- modern hearing aid microphones commonly produce resonant frequencies at around 4 to 5kHz or higher
- picking up signals of interest is often complicated because listeners may desire to hear only one or a few of the many available signals
- the fact that the hearing aid wearer can often hear sound, but not the specific sounds he or she wants, continues to be one of the most common complaints of hearing aid wearers
Directional Sensitivity Patterns
- tells you the sensitivity of the microphone
- if you had an absolutely perfect omnidirectional microphone you would have a plot like the first
- bi-directional can have less sensitivity
- cardiod's have some directionality
- shotgun microphone is not used in hearing aids
Sine wave and Sampling
- the limits of this process is that the digital representation of sound is in discrete steps rather than a smooth and continuous (perfectly accurate) representation of the original signal over time
- this misrepresentation of the continuous amplitude change in the original signal introduces roundoff error and quantization error
- the original smooth continuous signal is squared off in stair steps and the amplitude of the signal is not perfectly represented at many instants of time
- these errors introduce quantization distortion, which is functionally a noise floor that limits the dynamic range of the DSP (digital sound processor)
- the more bits available (the larger the digital word) the smaller the error and the larger the maximum dynamic range
- the dynamic range of DSP is defined as the difference between the largest signal that can be digitized and the noise floor resulting from quantization distortion
- theoretically, each additional bit will double the number of possible values and in doing so provides a 6dB increase in dynamic range
- the staircase reconstruction has two negative effects
- introduces a constant delay equal to one half of the sampling period
- introduces new harmonics above the nyquist frequency referred to as images
- the differences in sampling rate are the differences in smoothing of the shape of the frequency response due to averaging
- we will be more closer to 128 bit sampling than 32 which allows for a more in depth analysis of the signal
Filters
- low pass filters let the low frequencies pass and have a specific cutoff frequency for the highs
- high pass filters let the high frequencies pass and have a specific cutoff frequency for the lows
- band pass allow a specific frequency spectrum to pass
- remember, the cutoff frequency is defined as the 3dB down point
Input/Output functions
- input + gain = output
- for the input/output function, we use a chart that has input on the x-axis and output on the y-axis
- the input/output function of a given hearing aid is then displayed by the diagonal line
- at any point on this line, gain can be determined by subtracting the input values from the output
- the place on this line where it bends or changes angle (deviates from 45 degrees) is called the compression knee-point
- the knee-point is where the compression begins and is also referred to as the compression threshold
The difference between linear amplification and a few different compression hearing aid settings. Notice how the output grows in a linear manner as the input increases up to 90 dB SPL. In other words, the gain stays the same as the input increases. A 70 dB input equals 90 dB output (20 dB of gain) and likewise an 80 dB input equals 100 dB output (again 20 dB of gain) When an input signal of 90 dB is reached, notice that this hearing aid reaches its MPO (maximum processor output). This means that the output will not significantly increase for higher inputs, which then indirectly reduces gain.
|
The compression ratio describes the amount of compression, or in simple terms, "the squash effect." Compression ratios can be calculated by dividing the change in input by the change in output. If an input signal change of 10 dB results in an output change of 10 dB, the ratio is 1:1, or linear. If the change input of 10 dB results in an output change of 5 dB, the ratio would be 2:1. Input/output functions for various compression ratios are illustrated.
|
The effect of compression on Attack and Release time
- The attack time is essentially how quickly the compression comes on
- the release time is how quickly the system comes out of compression
- can have the ability to change the attack and release
- most hearing aid companies will ask for age when programming the hearing aid
- there is research that attack and release is judged differently for how old you are
- the amount of time that this takes for the system to say that this is a really loud sound (attack time)
- the recognition that the sounds are soft (release time)
Output is what the hearing aid is putting out, if doing REAL ear measurements it includes the hearing aid output as well as ear canal. If looking at it on the test box, just what the hearing aid is putting out, the gain is how much is input into the system, essentially output minus input. To look at the linear line segment, it is 1:1 to the kneepoint. As the inputs get higher, it is really adding the same amount of gain until the compression knee point.
Battery Life
- 10, 312, 13, 675; in order of battery life
- once we have what comes out of the amplifier, the battery boosts the signal, then it is routed to the receiver coil which then changes the signal back into analog pressure sound
T-Coils
- a second major way to introduce signals into a hearing aid is through a telecoil, commonly referred to as a t-coil
- hearing aids that have t-coils often have a user control that can be switched from microphone to t-coil and microphone at the same time
- some modern ones are automatic; activated and deactivated based on the presence or absence of a magnetic field
- t-coils only pick up electromagnetic signals
- noise in the environment will not be picked up and amplified if the hearing aid microphone is deactivated
- useful for telephones because no noise
- easily bringing signals directly and economically from an external microphone to the hearing aid through the use of an induction loop in the room, meanwhile avoiding picking up noise
used for a looped system via a magnetic field. the room has an induction loop that loops around where folks are sitting and the telecoil can pick up the activity without using the sound pressure level. Areas can be looped like your sofa or chair, places where you sit all of the time, they are wirelessly based induction systems.