A binaural hearing aid set-up where the left and right ear devices are connected by a rate-constrained wireless link is considered, and the performance gain resulting from beamforming is quantified. Each device is assumed to have two or more microphones. The transmitting device sends a signal at a rate R to the receiving device where it is combined with the locally available signals to obtain a minimum mean-squared error estimate of the desired signal. Different practically realizable choices for the signal to be transmitted are considered: an estimate of the desired signal, an estimate of the interfering signal (relevant for multi-microphone interference cancellation), and the unprocessed signal. It is not obvious which scheme provides the best rate-gain trade-off. In fact, it is shown that this varies depending on the configuration of the desired and interfering sources. This paper provides a framework to quantify and thus compare the performance of the above-mentioned schemes in terms of the rate R and the resulting beamforming gain.
[1]
Henning Puder,et al.
Signal Processing in High-End Hearing Aids: State of the Art, Challenges, and Future Trends
,
2005,
EURASIP J. Adv. Signal Process..
[2]
Toby Berger,et al.
Rate distortion theory : a mathematical basis for data compression
,
1971
.
[3]
Martin Vetterli,et al.
Rate-Constrained Beamforming for Collaborating Hearing Aids
,
2006,
2006 IEEE International Symposium on Information Theory.
[4]
Hirosuke Yamamoto,et al.
Source Coding Theory for Multiterminal Communication Systems with a Remote Source
,
1980
.
[5]
Olivier Roy,et al.
Collaborating Hearing Aids
,
2006
.
[6]
R. Duda,et al.
Range dependence of the response of a spherical head model
,
1998
.