GNSS scintillation: detection, forecasting and mitigation

Lead Research Organisation: Newcastle University
Department Name: Electrical, Electronic & Computer Eng

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
 
Description The grant was aimed at modelling and mitigating the effect of scintillation on GNSS positioning and other applications. This scintillation can result in cycle slips or even phase lock loss in GNSS receivers which degrade the positioning accuracy. This is one of 3 linked grants; the other 2 at Nottingham and Bath universities. The text here will report the contribution from the work on the grant at Newcastle.

What were the most significant achievements from the grant?

(1) Modelling scintillation. Nowadays, both GPS and the new European Galileo satellite system broadcast three frequencies enabling more advanced three frequency correction schemes so that knowledge of correlations of different frequency pairs for scintillation conditions is desirable. To treat this question for the case of strong scintillation, a previously constructed simulator program, based on our hybrid method, was further modified to simulate the fields for both frequencies on the ground, taking account of their cross correlation and then the errors in the dual frequency range finding method caused by scintillation were estimated for particular ionospheric conditions and a realistic fully 3D model of the ionospheric turbulence. This work has also been extended to determine the variation of various parameters, including the spectral indices and other statistical moments of the field, as a function of the severity of the signal fluctuations and to estimate the rate of cycle slips for different scintillation levels (Gherm et al, 2011; Zernov et al., 2012). Generally weak scintillation models are used to determine scintillation levels as the theory is considerably more complex for strong scintillation conditions where saturation of the field can occur; an effect which is not reproduced by the weak scatter models. It has also been shown that our hybrid method can enable scintillation results for strong scatter conditions to be achieved by a much simpler method than standard strong scintillation theory and can show the correct field saturation effect which the weak scintillation methods fail to do (Strangeways et al., 2014). The problem with employing the generally used weak scintillation theory to model the scintillation is that it is for the strong scintillation conditions that the significant scintillation problems occur in GNSS receivers and positioning.
The simulator has also been used to generate time series of the amplitude and phase for particular scintillation conditions as a "test bed" for the effectiveness of different hardware/firmware scintillation mitigation methods for GNSS receivers as real scintillation data is not repeatable and also cannot easily be directly linked in practice with the contemporaneous ionosphere and irregularity conditions from which it arose which are rather difficult to measure. We have also supplied such outputs to other parties.

(2) Mitigating scintillation. A mitigation method has been developed (Strangeways et al, 2011) by which it has been shown that the 3D positioning error in dual frequency GPS positioning can been be reduced by more than 70% in strong scintillation conditions. This method utilises a positioning calculation which weights each satellite path in the positioning determination inversely according to the amount of scintillation on each and determines the respective scintillations levels on the paths. The method (Strangeways, 2009) is employed to enable the key parameters of the power spectral densities of amplitude and phase of the received signals to be determined just using time domain data so that time-consuming Fourier transforms are not required.

(3) Scintillation Mapping Function. A scintillation mapping function has been determined. In the Bath and Nottingham Universities linked grants a large amount of data has been collected of amplitude and phase scintillation levels from both high latitude and low latitude receiving locations where scintillation is most prevalent. In order to be able to use this data to estimate the scintillation at a particular time on a particular path a mapping function has been developed which incorporates the effect of changing elevation and azimuth of the received GPS satellite signal so that the scintillation observed for a vertical path can be mapped for a path at any elevation or azimuth (or vice versa) for any model of the irregularities and any electron density profile. This enables collected scintillation data to be used to make scintillation predictions for other transmitter to receiver paths for a variety of different conditions, ionosphere profiles and irregularity structures.

(4) Real time scintillation monitoring and mitigation using TEC measurements. Determination of scintillation levels using high pass (>1Hz) filtered TEC (Total Electron Content) measurements and construction of a regional scintillation alarm index (Tiwari et al; 2013, Tiwari and Strangeways, 2014). The scintillation indices in generic GPS receivers are usually estimated every minute and a special firmware in the GPS receiver is required to compute them which is costly in time and resources. It has been shown that regional mapping of scintillation is possible with the index derived in the above papers and in Europe this can be done by incorporating TEC measurements from the existing IGS GNSS network. The results in Tiwari and Strangeways (2014) shows that the effect of strong scintillation can mitigated in real-time using any software defined GNSS receiver or PLL adaptive receiver. For a generic receiver, scintillation mitigation in the position solution is also possible using the method described in Strangeways et al. (2011) and the indices as computed in Tiwari and Strangeways (2014).

(5) The development of a software receiver to determine optimum hardware architectures to minimise the effect of the scintillation on the receiver, in particular the signal tracking using the phase lock loop (PLL). In a generic GNSS receiver, the loop parameters in the PPL are fixed and so cannot be adapted to the ionospheric scintillation situation. In Tiwari et al (2011), a GNSS software receiver programmed in Matlab is described which can take information on current conditions from an ionospheric scintillation prediction programme (such as WBMOD) and update its tracking parameters in real-time to take account of any predicted scintillation level. A similar approach developed and validated using USRP (Universal Software Radio Peripheral) devices was tested was at Newcastle using a NovAtel GISTM receiver with a common clock reference as described in Tiwari et al. (2013b).

(6) Acquisition of weak GPS signals. In the GPS receiver, the acquisition of a satellite signal is necessary prior to the tracking. Considering, the weak signals that can arise in scintillation as well as some other propagation conditions, a PDBA (Parallel Double Block Processing) method has been developed which helps in acquiring the weak signals which may arise due to environmental noise, multipath or amplitude scintillation [Ahmed et al. 2014a; 2014b; 2014c].
Exploitation Route The method described by Strangeways et al (2011) could be generally employed to reduce the effect of scintillation in GNSS receivers in moderate and strong scintillation conditions. If scintillation indices are not available, the analogous scintillation index based on TEC (Tiwari et al, 2013) could be used instead as described in Strangeways and Tiwari (2013)
The regional alarming index derived in Tiwari and Strangeways (2014) could be used by the surveying industry to get alerts in real-time about ionospheric scintillation occurrence and then precautions could be taken to reduce their effect (such as time rescheduling of the surveying measurements) or the scintillation effect could be mitigated. This could make use of the regional alarming index to identify the conditions for which the parameters of the tracking loop of the GPS receiver should be updated. To do this the GNSS receivers could have an additional capability to accept the indices from a regional alarming index as discussed in Tiwari and Strangeways (2014). In a planned KTS project with the Nottingham Company NSL, this would be applied in a USRP device. Discussions are currently taking place for Dr Rajesh Tiwari to take up, through an EPSRC IAA KTS, a 9 month secondment (January to September 2015) at NSL which is one of the leading companies in Europe involved in satellite navigation. This secondment took place from January to September 2015; further details are given in the secondment entry which include use of our findings by the company NSL (Nottingham Scientific).
Sectors Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software)