Signal processing is a big feat of engineering, and is used everywhere. According to the IEEE Signal Processing Society, some of the main engineering challenges of signal processing are analog to digital conversion, filtering, echo/noise reduction, automatic gain, encoding and decoding.
A lot of solutions to these challenges are implemented on embedded systems – both hardware and software are important to solving these problems. Audio signal processing is an interesting example. Since systems in the future will be completely wireless if they aren’t already, the challenges are even more complex.
The widely known Bluetooth technology uses radio waves to transmit and receive data, and one of the most popular consumer uses of Bluetooth is for audio applications. In 2019, around 1.1 billion devices were created with bluetooth audio capabilities.
In a lot of cases though, Bluetooth audio quality is worse in comparison to physical forms of audio such as CDs or records. How can wireless audio signal processing be improved, and what does that mean?
One way would be to use different audio codecs – an endearing term given to the concept of a compression/decompression algorithm. Most Bluetooth uses SBC (low-complexity subband). The codec allows only specific data points to be transmitted without losing the overall audio waveform. An improved codec would be more of a software side embedded system solution. What defines an improvement is dependent on the desired outcomes, though. For some uses, lowest latency may be an improvement, for some, lossless codecs will be more applicable.
Another interesting idea is to use a different transmission frequency. Bluetooth uses 2.45 GHz – some leaders in audio technology are switching to the 5 GHz spectrum, which has a lot more radio frequency channels available than the 2.45 GHz frequency Bluetooth uses. With more channels available, a dynamic frequency selection can help eliminate issues associated with signal interference. At the first signs of trouble, the devices dynamically switch to a different channel. This requires new hardware designs in the embedded systems as well as software updates.
These are just two methods of improving wireless audio technology. There are a lot of possibilities to explore! Is there anything you have heard of lately that is being used or may be used in the future to improve audio signal processing?
-Andy Hua, Jason Wong, Nora Huynh
