Analysis of Multi-Constellation GNSS Signal Quality

The code pseudo-ranges and phase pseudo-ranges are affected by the systematic errors or biases and random noises. These errors can be classified into three groups, satellite related errors, propagationmedium related errors and receiver related errors (Hofmann-Wellenhof, Lichtenegger and Collins 2001; Leick 2004). Effective GNSS positioning depends on an understanding of the measurement error budget and eliminating or reducing those errors. Some systematic errors can be modelled and gives rise to additional terms in the observation equation, while other systematic effects can also be eliminated (or greatly reduced) by appropriate combinations of observables. However in the case of stand-alone positioning the use of relative positioning to reduce or eliminate biases is not a possible option. The GNSS signal observation quality and bias investigation involved a thorough review, analysis and quantification of the constellation error sources and mitigation processes, also, focusing on those effects which cannot be eliminated or modeled. Signal quality is usually represented as signal-tonoise ratio (SNR) or as carrier-to-noise ratio (C/No). Both of those parameters are essential to assess the performance of GNSS receiver and they are directly related to the precision of code-phase and carrierphase pseudo-range observations (Langley, 1997). This paper presents the work done, to evaluate the relationship between the SNR value generated by the receiver and the pseudo-range residual and, to develop an observation weight function, including the scale parameters that are specific to the observation types and linear combinations that were assessed, this ultimately may be used as a relative weighting scheme for combining data from multiple signals & constellations. The results obtained showed that the higher pseudorange residual values were associated with lower SNR values irrespective of GNSS constellation or observation signals. As the SNR values increased 4555 dBHz the pseudo-range residual values are generally reduced. The high residual values were observed mainly at the beginning and ends of the observation periods tested and maybe related to the ascension and descent of the satellite or multipath. The SNR weighing developed, produced comparably similar position solution results to that of elevation weighing, however this weighting method has the advantage of allowing the use of a combination of different observations and signals.