ACoRN Members Only AreaSignal Processing
Signal Processing refers to various techniques to extract a signal of interest from imperfect measurements which may be marred by noise, distortions, interference, jamming, etc. Today it is synonymous with digital techniques as the processing is done via numerical algorithms running as a program on a computer or dedicated processor. Much of Signal Processing today is in the design of algorithms and in the analysis of their performance.
|
| Signal processing has wide application. From space communications... |
|
| ...to medical imaging. |
What characterizes a signal depends heavily on the application. In the context of communications, the nature of the desired signal is generally very well known as it corresponds to specifically designed modulations and coding. Despite the explicit information (data) being unknown, the waveform bearing the information is very well known and it is here, in the structure of the signal, that we can use to our advantage to reliably recover the data despite the presence of noise and interference.
Key Research Challenges
There are many application specific challenges in Signal Processing in the various domains of audio, video, speech, image, communication, geophysical, sonar, radar, medical, musical, etc. Here we emphasize the key generic challenges common to many of these domains and particularly communications.
Curse of Complexity
Often a very good model for the signal (structure) is known and, through theory, the best means to extract the signal is well-defined. However, often such an optimal solution has a complexity which cannot be implemented because the resources required are too great. The challenge then is to find a reduced complexity approximation that can be implemented and yet has a performance that approaches the performance of the optimal solution.
Closely related to this is determining theoretically that the reduced complexity scheme makes the best use of the limited resources (fixed complexity) and establishing theoretically that the performance is indeed close to optimal (rather than just appears to be close from running the system).
Systems and Subsystems
Complex engineering systems involve the cascades of subsystems and a hierarchy of processing. Classical design has involved optimizing the performance of subsystems which is a divide and conquer approach. The general challenge is to develop systematic approaches to joint subsystem optimization targeting superior performance and sharing of resources between the subsystems.
Australian Signal Processing Researchers
Note: You can search for ACoRN Members using the Member Search facility
