**5. Discussion**

GNSS-R is a promising and low-cost technique for ocean altimetry on different platforms. New GNSS signals, designed with better performance, bring opportunities for improving the accuracy and precision. We tested the performance of GNSS-R code-level altimetry based on new BDS-3 civil codes by conducting a coastal experiment for the first time. The solutions derived from B1C and B2a signals are achieved at one second intervals for a period of about eight hours. The final results show that the centimeter-level precision of GNSS-R altimetry based on B1C codes can be achieved; it is similar to that of the tide gauge.

Our results demonstrated that the precision of solutions from the two new civilcodesare higher than those from conventional GPS C/A and BDS B1I. Furthermore, the solutions from B1C are better than those from B2a. The poor performance of the single-frequency band B2a was attributed to its poor signal quality and narrower bandwidth. In addition, we also found that the precision of the solutions can be affected by signal power.

One of a main feature of BDS-3 is its hybrid constellation, in which the GEO satellites can provide stable geometries for GNSS-R observations. However, in this paper, no solution was retrieved from the signals of the BDS-3 GEO satellite. They provide fundamental PNT service on the legacy B1I and B3I signals, while new B1C and B2a signals are used for providing SABS service. Unfortunately, their SABS services are still in testing. As the GNSS-R code-level altimetry performance of BDS-3 B1C and B2a signals is studied, we could not test BDS-3 GEO signals during this experiment.

In this work, the performance of GNSS-R altimetry based on the B1C and B2a signals was only tested on a very low platform when the sea surface was in a good condition. We plan to conduct experiments on higher platforms, such as using an unmanned aerial vehicle and/or plane to investigate their characteristics. We should find a proper and safe place to conduct experiments for different sea states.

The monostatic radar altimeter can accurately measure the vertical distance from its phase center to the sea surface, which enabled us to obtain the precise reflector height. In this work, we solved a bias for each satellite using the precise reflector height values. This is because the biases are caused not only by the sea surface roughness and the electromagnetic characters, but are also affected by the signal bandwidth and instrumental reasons [22–24]. So, further explorations for the biases require much more data over a long time. However, as our SDR runs very slowly and the IF data are too large to be stored, we cannot process and analyse long-time data in this work.

**Author Contributions:** Conceptualization, F.G., T.X. and N.W.; methodology, F.G., X.M.; software, F.G., X.M. and Y.H.; validation, N.W., X.M. and Y.H. and B.N.; formal analysis, F.G.; investigation, F.G., X.M. and Y.H. and B.N.; resources, F.G., X.M. and Y.H. and B.N.; data curation, X.M. and Y.H. and B.N.; writing—original draft preparation, F.G.; writing—review and editing, F.G., T.X. and N.W.; visualization, F.G.; supervision, T.X.; project administration, T.X.; funding acquisition, F.G., T.X. and N.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was jointly funded by the National Key Research and Development Program of China (2016YFB0501701) and the Program of the National Natural Science Foundation of China (41604003, 41704017, 41704018).

**Data Availability Statement:** The datasets analyzed in this study are managed by Institute of Space Science, Shandong University and can be made available by the corresponding author on request.

**Acknowledgments:** The authors thank the staff of Weihai Golden Bay Hotel who kindly provided help during the experiment.

**Conflicts of Interest:** The authors declare no conflict of interest.
