No abstract

No abstract

The increasing number of music available for download and subscriptions motivates the need for new solutions in organizing music for consumers. In this paper, several approaches for automatic genre classification of music using polyphonic timbre models are evaluated.Specifically, we compare the performance of the Gaussian mixture model (GMM), the Support Vector Machine (SVM), and the k-nearest neighbor (k-NN). Features are extracted to model the major attributes of timbre such as spectral envelope, range between tonal and noiselike character, and spectrotemporal evolution of sound. To address the scalability problem, a modified filter-andrefine method is integrated with the k-NN classifier. Results show that the 1-NN classifier with filter-and refine method achieved the highest classification accuracy on the GTZAN and ISMIR2004 datasets.

MUSIC is a popular subspace method because of its superior performance in estimating direction-of-arrival (DOA). However, the performance of the algorithm in a multipath environment has not been fully addressed. This study evaluates the performance of MUSIC algorithm in estimating DOA in such situation. Experimental verification was performed in an enclosed room with varied parameters including the number of sensors in the antenna array and inter-element spacing. Results show that the algorithm localizes the signal with a deviation of 30 for a four-element, Vd2 spaced array.

⎯ ⎯ ⎯ ⎯ In this paper, we present a model-based Register Transfer Level (RTL) design of a simple joint parameter method for Carrier Frequency Offset and I/Q imbalance estimation. The estimation scheme is based on a proposed algorithm that uses only two short training symbols for faster estimation. The simple yet robust algorithm is then optimized for Field Programmable gate Array (FPGA) implementation. Simulation results show that the performance of the 16-bit fixed-point system approximates that of the ideal performance of the estimation algorithm.