COMPLEXITY REDUCED MULTIPATH MITIGATION IN GNSS WITH THE GRANADA BIT-TRUE SOFTWARE RECEIVER

Abstract

The estimation of several unknown channel amplitudes and taps in a multipath environment is an important problem when designing positioning receivers. A popular approach for these estimation problems are methods solving the maximum likelihood (ML) problem for each dimension separately like the expectation maximization (EM) approach and the more sophisticated space alternating generalized expectation maximization (SAGE) algorithm. However, both methods require a high computational complexity when used in spread spectrum systems due to long spreading sequences. Therefore, we apply methods for decreasing the computational complexity before executing the iterative estimation algorithms. The respective performance is assessed by means of computer simulations using the GRANADA Bit-True simulator which show a negligible degradation in estimation accuracy despite a substantial complexity reduction. The GRANADA (Galileo Receiver ANalysis And Design Application) Bit-True simulator is a powerful simulation tool for the design of navigation receivers respecting the special features of the future GALILEO system, e. g. the binary offset carrier (BOC) transmit pulse structure. In this paper, two important adaptations of low complexity channel estimation are presented appearing when considering the integration of multipath mitigation algorithms into GRANADA. The first one is the special way of complexity reduction designed for ML channel estimation which is adapted to the superposition of data and pilot signals in GRANADA. As a second important implementation issue, constellations and problems with non-integer oversampling factors will be analyzed in detail. Finally, we present solutions for the integration of the multipath estimation algorithms directly into code tracking loops. The proposed way of complexity reduced multipath mitigation is directly suited for the implementation into GRANADA data and pilot code tracking loops.