Convex Combination of MIMO Filters for Multichannel Acoustic Echo Cancellation Danilo Comminiello, Simone Scardapane, Michele Scarpiniti, Raffaele Parisi, and Aurelio Uncini Department of Information Engineering, Electronics and Telecommunications (DIET) “Sapienza” University of Rome Via Eudossiana 18, 00184 Rome, Italy Email: danilo.comminiello@uniroma1.it Abstract—This paper introduces a new framework for mul- tichannel adaptive filtering, aiming at improving performance of an overall filtering system. The proposed architecture relies on the properties of the adaptive combination of filters which exploits the capabilities of different constituents, thus adaptively providing at least the behaviour of the best performing filter. Applying this concept to multichannel filtering systems, we define a scheme for the combination of multiple-input multiple- output (MIMO) filters. More precisely the proposed structure involves the combination of two different multiple-input single- output (MISO) systems for each MIMO output. We propose such framework with application to the multichannel acoustic echo cancellation (MAEC) with the goal of giving robustness to the system against impulsive background noise, and thus improving overall cancelling performance. Experimental results show the effectiveness of the proposed combined MAEC in the presence of adverse environmental conditions. I. I NTRODUCTION The growing availability of resources for data transmission and processing is encouraging the spread of hands-free speech communication systems which offer users an immersive expe- rience [1]–[3]. An immersive communication scenario is char- acterized by the use of multiple microphones and loudspeak- ers, which aim at capturing and reproducing desired signals while focusing on preserving the audio quality perceived by users. In such context, the presence of interfering sources and reverberation may cause a quality degradation of speech com- munications. In order to address this problems, microphones and loudspeakers are usually connected with signal processing systems, thus resulting in intelligent acoustic interfaces [3], [4]. One of the main problems of immersive scenarios is the multichannel acoustic echo cancellation (MAEC), which is caused by the multiple coupling between microphones and loudspeakers. MAEC may be seen as a straightforward gener- alization of the monophonic acoustic echo cancellation (AEC), but it entails more problems to tackle, such as nonuniqueness and slow convergence due to inter-channel correlation, and the poor ability to react to changes in the environmental conditions [5]. Such issues have roused remarkable interest over the years [6]–[9]. The effectiveness of an MAEC strictly relies on the design of a multiple-input multiple-output (MIMO) filtering system, whose main task is to estimate several acoustic impulse responses (AIRs), depending on the number of microphones and loudspeakers. It results in a large number of coefficients to adapt, therefore an appropriate choice of the adaptive algorithm becomes essential. Generally, in the time-domain, first-order adaptive algorithms, such as the least mean squares (LMS), are very attractive due to their simplicity and low computational cost. However, LMS-based algorithms do not take into account the cross-correlation statistics of the input signals, thus resulting in poor convergence performance [5]. Hessian-based algorithms, such as the recursive least squares (RLS), improve convergence abilities since they consider inter- channel correlations. Unfortunately, the RLS entails a larger computational cost and, moreover, it may perform worse than first-order algorithms in adverse environments [10]. This is the reason why here we consider the affine projection algorithm (APA) to adapt the MIMO filters, since it can be seen as a generalization of the normalized LMS (NLMS) algorithm, in which cross-correlations of the input signals are involved [11]. However, even a multichannel APA may suffer from ad- verse conditions, especially in the presence of impulsive noise or in changing environments, which may make the adaptation process unstable and reduce performance. In order to tackle this problem, we introduce a new adaptation framework which relies on the adaptive combination of MIMO filtering systems. Adaptive combination of filters is capable of automatically switching between constituents according to the best perform- ing filter, thus always providing the best possible performance [12]. Combined adaptive schemes are usually adopted with filters of the same family and complementary properties, but also with filters of different families or using different updating rules [13]–[15]. Recently, combination of filters was successfully applied to multiple-input single-output (MISO) systems [16]–[18] with application to adaptive beamforming for noise reduction. In this paper we exploit the properties of adaptive combi- nation of filters to perform a combination of MIMO filtering systems. In particular, we combine two MIMO systems, each one having different adaptation settings. We describe in detail a possible combining architecture for MIMO filters and we apply such technique to MAEC. As a consequence, an improvement of the overall performance of the canceller is expected, espe- cially in the presence of any impulsive noise that may cause a change in the environment. This paper is organized as follows: Section II describes the MAEC framework using the combined MIMO architecture updated by using the multichannel APA. The combination of adaptive MIMO filters is proposed in Section III. Section IV contains the evaluation of the effectiveness of the proposed framework with application to MAEC, and some conclusion are finally drawn in Section V. 8th International Symposium on Image and Signal Processing and Analysis (ISPA 2013) September 4-6, 2013, Trieste, Italy Special Sessions Multichannel Audio Algorithms: New Trends and Innovative Applications 778