The present lecture notes were written for the annual course on adaptive filters at Aalborg University. The content of the course is now a part of the annual course called Array and Sensor Signal Processing. The notes are written for the lecturer, but they may also be useful to the student as a supplement to his/her favourite textbook. Consequently, the notes are very concise and contain only what we believe to be the basics of adaptive filtering. Moreover, we have also made some important simplifications.
- We use real-valued numbers and not complex-valued numbers. Although the latter is more general, it is less confusing and leads to fewer errors when real-valued numbers are used.
- We only consider FIR adaptive filters.
- The signals have zero mean. This is a standard assumption used in most textbooks.
Each of the lectures contains an amount of material suited for a lecture lasting for approximately 90 minutes. The appendices contain a summary and some supplementary material.
These lecture notes are always work in progress. Therefore, if you have found an error, have a suggestion for a better statement and/or explanation, or just want to give us some feedback, then do not hesitate to contact us.