Quality evaluation of NRTK correction transmission

Signal definition Data Analysis

Figure 3 illustrates a comparison between the benchmark solution and the solution via the GPRS combined with the long distance Internet link, in a 10-secondlong timeslot. The comparison is made in terms of the latency, the horizontal error and the vertical error respectively.

Because the receiver makes the solutions at 20 Hz, the latency will gradually increase by 0.05 second at each logging point over the time axis, until a new correction message is received, which brings the latency value back to zero. Therefore the height of the triangle shape in the latency figure can show the receiving time interval between two consecutive messages. In the benchmark solution it can be seen that,

due to the ideal transmission environment, the messages are received at a nearly constant one-second-interval; while in the combined link solution, the messages are received at varying intervals, which can show that the condition of the transmission route was unstable. It can be seen from the latter solution that there is a triangle with a two-second height. This shows a message loss in its scheduled time.

Because of the delay, the same message may arrive at different times in two links. Accordingly, in the latency chart of Figure 3, the channel delay results in a separation between the two different colour triangles.

Considering the zero transmission delay in the ideal benchmark solution, this separation can determine the message transmission delay in the combined link solution, as marked in the figure.

Statistics of both the message delay and loss in different solutions are shown in Table 2. It can be seen that the average delay of the GPRS and the long distance Internet link are at the same level, and the average delay of the combined link is close to a second (0.85 sec). Both the standard deviation of the message delay and the message loss percentage show the stability of the transmission, where the GPRS link is relatively more stable than the long distance Internet link, and the combined link is the worst. Comparing to the GPRS link, the long distance Internet link suffered much more message loss, and 20% of the messages were lost in the combining link.

Figure 4 shows the distribution of the message delay, for the three transmission scenarios. The long distance Internet link shows a greater spread in the delay than the GPRS link, due to its larger number of packet switchings during the transmission. There is no delay higher than two seconds in these two links, because if a message arrives later than the following message, it will simply be rejected by the receiver and will be treated as a message loss. The combined link does have a small percentage of message delays higher than two seconds, because

the following message may also not arrive on time due to its rigorous environment. With increasing message delay and loss, the receiver may not have the latest RTK corrections on time and only can ‘predict’

the present correction from the past RTK data [12]. The time-sensitive error will increase and the positioning solution will drift away from the true coordinates. In Figure 3, it can clearly be seen that, during the message-lost period, both the horizontal error and the vertical error are increased and a ‘degradation peak’ is formed. The statistics of the horizontal error and the vertical error are shown in Table 3 and Table 4 respectively. It can be seen that both the GPRS link and the long distance Internet link introduce some ‘degradation peaks’ and degrade the positioning solution precision eventually.

Again, the long distance Internet link shows a larger infl uence than the GPRS link. Comparing to the benchmark solution, the precision of the combined link solution is degraded by 60%.

Conclusion

This paper presents a study to investigate the quality of the NRTK data transmission methods. The two transmission variations, message delay and loss, were both observed whilst the GPRS link and the long distance Internet link were used. The combination of the two links was shown to have a 0.85 second average delay and 20% message loss.

It is demonstrated that these variations might introduce 60% degradation in the precision of the positioning solution in a static test. When designing a RTK network in a large-area, this transmission infl uence should be considered, and as a compensation, a more frequent message sending scheme could be considered.

References

[1] iSuppli report, 2007, Portable Navigation Devices Growth

[2] Meng X., Yang L., Aponte J., Hill C., Moore T. and Dodson A., 2008,

Development of Satellite Based Positioning and Navigation Facilities for Precise ITS Applications, the 11th International IEEE Conference on Intelligent Transportation Systems

[3] Namie H., NishiKawa K., Sasano K., Fan C. and Yasuda A., 2008, Development of Network-Based RTK-GPS Positioning System Using FKP via a TV Broadcast in Japan, IEEE Transactions on Broadcasting, 54(1), pp. 106-111

[4] Marel H., 1998, Virtual GPS Reference Stations in the Netherlands,

ION GPS’98 Proceedings, pp. 49-58

[5] Kumar M. D., Homer J., Kubik K. and Higgins M., 2005, Effi cient RTK Positioning by Integrating Virtual Reference Stations with WCDMA Network, Journal of Global Positioning Systems, 4(1-2), pp. 48-55

[6] Retscher G. , 2002, Accuracy Performance of Virtual Reference Station (VRS) networks, Journal of Global Positioning Systems, 1(1), pp. 40-47

[7] Fotopoulos G. and Cannon M. E., 2001, An Overview of Multi- Reference Station Methods for Cm-Level Positioning, GPS Solutions, 4(3), pp. 1-10

[8] Radio Technical Commission for Maritime Services, 2003, Networked Transport of RTCM via Internet Protocol Version 1.0, RTCM paper 167-203/SC104-315

[9] Cruddace P., Wilson I., Greaves M., Euler H-J., Keenan R. and Wuebbena G., 2002, The Long Road to Establishing a National Network RTK Solution, 22th FIG International Congress [10] Meng X., Dodson A., Moore T. and Roberts G., 2007, Ubiquitous Positioning, GPS World, June

[11] National Marine Electronics Association, 1998, NMEA 0183 Standard for Interfacing Marine Electronic Devices Version 2.30

[12] Leica Geosystems, 2007, GPS 1200 Technical Reference Manual Version 1.1