IIR Filters for Blind Deconvolution using Information Maximization Kari Torkkola Motorola, Phoenix Corporate Research Laboratories 2100 E. Elliot Rd, MD EL508 Tempe, AZ 85284, USA tel: (602)413-4129, fax: (602)413-5934, email: A540AA@email.mot.com Abstract Bell and Sejnowski have showed that a suitable criterion to blind deconvolution using an FIR-filter is to maximize the information at the filter output passed through a nonlinear function. In this paper we will simplify their derivation of the adaptation equations. We will show that the simplified approach is readily applicable to more complex filter structures, such as recursive filters, We will also present examples of blind echo cancellation applied to speech signals together with a method that avoids whitening the signals in the procedure.