Bands and channels
One of the first steps in the hearing aid’s processing sequence is to split the signal into a number of frequency regions, known as bands.
The division of the signal into frequency bands means that every frequency region can be handled individually in its own separate signal processing channel.
Precise and differentiated amplification
The separation of the signal into bands is done for two main reasons:
One is that completely flat hearing losses are rare. Most hearing losses require differentiated amplification in different frequency regions (compression).
The other reason is that features such as noise reduction and feedback handling can operate with more precision on the basis of finely divided frequency regions.
Nonlinear hearing aids
Dividing the signal into separated bands is especially important in connection with nonlinear hearing aids:
Imagine listening to a speaker through a 1-channel compressing hearing aid in a car. The loudest signal in the car will be the noise generated by the car. Car noise is primarily a low frequency signal. Most of the energy is centred in a lower frequency region than the energy of the speech signal.
The 1-channel hearing aid will only register one signal level, and as the dominating signal level is that of the car noise, it will be the level of the car noise rather than of the speech signal which will control the gain of the hearing aid.
As a result, the entire signal will be dampened because of the low-frequency noise.
More than one channel
If a hearing aid has more than one processing channel, however, low frequency noise will only affect the gain setting for the low frequencies.
This means that the dampening introduced as a result of the low-frequency car noise will only affect the low-frequency regions. Mid- and high-frequency regions will not be affected.
Important for noise reduction, speech enhancement and feedback handling
Noise reduction, speech enhancement and feedback handling all benefit from the existence of several processing channels because it means that gain can be regulated with more precision.
For example, in the Speech Enhancer noise reduction system, the aim is to optimise the gain settings of the hearing aid to achieve the best possible speech intelligibility. Multiple channels make it possible to monitor the frequency spectrum of the noise more precisely.
In the car noise example from before, the existence of multiple channels means that the Speech Enhancer is able to dampen only the narrow frequency region that is dominated by the car noise while actually adding a little extra amplification to the important speech-frequency regions in an attempt to provide better speech understanding. | 
