Kernel Adaptive Filters: A Panacea for Self-Interference Cancellation in Mobile Communication Transceivers?
Sprache des Vortragstitels:
Englisch
Original Tagungtitel:
International Conference on Computer Aided Systems Theory (EUROCAST 2019)
Sprache des Tagungstitel:
Englisch
Original Kurzfassung:
Kernel methods are powerful nonparametric modeling tools. The main idea is to transform the finite dimensional input data into a higher, possibly infinite dimensional space. In this so-called feature space the kernel trick can be applied: any inner product operation in the high-dimensional feature space is computed more efficiently by evaluating the kernel function. The kernel method can be
applied on adaptive filtering algorithms because they can be formulated such that the input vectors only occur as a part of an inner product.