MISEN: A Mobile Indoor White Space Exploration Method

With the exponential growth of the demand for wireless communications, people are faced with the shortage of spectrum resources. To tackle this problem, researchers consider about the further usage of TV channels with low occupancy, referred as white spaces. In this paper, we propose a mobile platform based indoor white space exploration method, namely MISEN, to profile the white space distribution inside a building. In this method, we design a tensor completion based algorithm based on the framework of Alternating Direction Method of Multipliers (ADMM) to recover a complete spectrum map, where we consider the linear dependency of white space information. Moreover, we build a prototype of MISEN and evaluate the performance in real scene. It is shown that MISEN senses on average 18:7% more white spaces with 20:0% less false alarm rate than the state of art methods.