تجزیه و تحلیل عملکرد از برآوردگر DOA / TOA مشترک

In many applications such as radar and mobile communication, the multipath propagation effects are described as a sum of contributions of a large number of wavefronts that arrives at the sensor array in clusters of rays, distributed around a nominal direction of the signal sources. Based on this observation and on the work of Bengtsson and Ottersten (Proceeding of Norsig-98, IEEE Nordic Signal Processing Symposium, April 1998), this paper jointly estimate the directions of arrival and the times of arrival of scattered sources. A theoretical performance analysis is given in terms of asymptotic error variance and illustrated by a simulation study.

For several years, many algorithms have been proposed to resolve the problem of signal parameter estimation in sensor array processing. In fact, the signal can be separated at the array based on the knowledge of their spatial and/or temporal “signatures”. Traditionally, the studies assume that the received signals originate from far-field point sources and give rise to planar wavefronts which impinge on the array from a fixed direction of arrival (DOA). However, in some radio propagation scenarios, especially in urban areas, this assumption is clearly inappropriate. The array receives many rays from the vicinity of the source and the assumption on number of paths becomes restrictive [1] and [10]. This scattering situation, is modelised by a spatial distribution function of the DOA and times of arrival (TOA). In [3], the authors treated each scattered source as the superposition of two point sources under the assumption that the delay between the different rays is small and can be treated as shift in the gain factors. However, a joint estimation of the DOA and TOA can be used to enhance the power of the received signal and a good separation of rays. In this contribution, we confirm the approximation in [3] where the time delay was introduced in the model. A spatio-temporal processing technique, MUSIC, is used for channel parameters estimation. This paper is organised as follows: In Section 2, a modelling framework is presented for the space-time processing. The algorithm MUSIC, to estimate the parameters channel, is investigated in Section 3 followed by a Monté Carlo simulations in Section 4. The paper is concluded in Section 5.

Herein, we investigate the problem of jointly estimating the direction of arrival and time of arrival of local scatterers, approximated by two points sources. We presented an asymptotic accuracy analysis of the 2D-MUSIC estimator and an explicit expression, for its theoretical error variance have been provided.