Super-Resolution Time of Arrival for Indoor Localization

We present the results of a large number of measurements of the indoor radio channel at 5.8 GHz taken in a typical office environment. The time of arrival (TOA) for the measurements are analyzed using two recently proposed super- resolution algorithms, and compared with traditional approaches utilizing the inverse FFT. We find that the performance of the FFT is equal to or better than the super-resolution techniques. Based on a statistical analysis of our measurements, we show that the information content of the signal relating to the TOA is concentrated in just a few coefficients on the leading edge of the impulse response. Drawing on this result, we propose a pattern matching approach for the determination of the TOA which performs significantly better than existing techniques.