2019 Vol. 44, No. 3
Statistical model parameters of marine gravity field character are valuable to the approximation of the earth's gravitational field and the optimum programming of marine gravity surveying. In view of the distinction of satellite altimetry data with large coverage and even distribution, a study strategy is proposed to develop the analysis and calculation of the statistical models of marine gravity field character using the latest satellite altimetry data. The computation formulas of commission error and covariance function model parameters of gravity anomalies are given in detail. A generalized Bouguer anomaly is defined, and the change character of which is researched. Equal and unequal weight models are suggested to fit the experienced covariance function. A set of statistical model parameter of gravity field character for the sea area around China is calculated using 1'×1' grid of satellite altimetry data from more than 500 thousands of 5'×5' cell, and which show different change characteristics of marine gravity field from that of land area. The reliability of the calculated model parameters is validated using ship-borne gravity anomalies. Some corrections are presented to add to the above parameters, and a final set of statistical model parameter of marine gravity field character is obtained. A suggestion is given for use of the new model parameters.
Airborne LiDAR bathymetry (ALB) has the characteristics of high efficiency and cost-effectiveness to obtain submarine topography in the intertidal zone. However, the accuracy of ALB is seriously affected by oceanic hydrological factors and cannot meet the accuracy standard of International Hydrographic Organization (IHO). Traditional correction models are used to correct the ALB depth biases. However, because the factors considered are not comprehensive enough in the traditional models, the accuracy of the traditional correction model for depth bias is not high. A comprehensive bias correction model and its construction method are proposed, which take into account the parameters of depth of water, suspended sediment concentration, beam scan angle and flight height. The traditional and the comprehensive models are used to correct the ALB depth biases. The experimental results show that ALB depth biases corrected by the both models can meet the accuracy standard of IHO, and the comprehensive model is better than the traditional model.
This contribution can be mainly divided into 3 aspects:(1) Based on Bayesian theory, unknown parameters are treated as random varies and their non-informative prior distribution function is introduced. Mathematical analysis is carried out to drive the optimal Tikhonov regularization matrix in the sense of minimizing the mean square error (MSE) of the solutions. (2) Combining the efficient truncated singular value decomposition (eTSVD), a new regularization method is proposed. (3) Global Navigation Satellite System(GNSS) ambiguity resolution application of the new method is discussed. Least squares (LS) estimation, ridge estimation of L curve and the new algorithm are compared by a group of GNSS ambiguity resolution experiments. The results show that the MSE of the new algorithm is slightly smaller than ridge estimation of L curve and much smaller than LS, however, the computational cost of the new algorithm is slightly more than LS but much less than ridge estimation of L curve.
The performance of satellite gravimetry is determined not only by orbital parameters, sensitivity of payloads, resolution of data, et al, but also by inaccuracy of the Earth gravity recovered methods. In past years, the performance analysis results were unavoidably affected by the mathematical model error from recovering methods such as dynamic method, short-arc integrated method, et al. To solve this problem, space-wise least square method is present. The effects of each items which affected the performance of satellite gravimetry are evaluated by this method. The results indicate that the highest degree of the earth gravity model recovered is 240 for the satellite with 500 km orbital height. Then if the error of payload is 1×10-10 m·s-2·Hz-1/2, the degree of model recovered only approach to 136 with the accumulated gravity anomaly error 2.7 mGal and the accumulated geoid height error 14 cm. In order to achieve the best surveying performance, the orbital inclination should be greater than 89°. While, the max degree of the earth gravity model recovered should be greater than 240 so as to reduce the effect of high frequency gravity signal on low degree coefficients recovered. All these conclusions benefit to satellite designing and data processing.
On April 20th 2013, Ms 7.0 earthquake occurred on the southern section of the Longmen Shan fault, in Sichuan Province. In this research, the three-dimensional coseismic displacement field that GPS data from 33 continuous stations are used to invert for fault geometry parameters and the slip distribution. The MPSO (multiple peak particle swarm optimization) algorithm has been successfully applied to invert for the fault model parameters.Compared with the commonly used algorithm, such as genetic algorithm, simulated annealing and the overall minimum norm optimization algorithm, MPSO algorithm has the advantages of less control parameters, high global convergence efficiency, and is very suitable for nonlinear inversion of fault parameters. The result indicates that the seismogenic fault is a thrust with a strike of 206.47°, a dip of 44.11°, and length of 21.94 km. The minimum and maximum depth is 7.66 km and 17.84 km, respectively. The fracture surface was determined as listric surface by analysis. In order to obtain the detailed slip distribution, the listric surface model is set with the default dip angle varying from 50° on the top boundary of the surface to 21° on the bottom boundary. Our inversion result shows that the rupture surface has two slip peaks at different depths. The maximum slip is 0.68 m at the depth of 13 km. The inverted geodetic moment is 1.47×1019 N·m and the moment magnitude Mw is 6.74, which is consistent with the result of seismology.
Compared with very long baseline interferometry (VLBI) and superconducting gravimeter (SG) techniques, Global Positioning System (GPS) technique has the advantages of global high-resolution coverage, long-term continuous observations, and low environmental impact to determine the ocean tide loading (OTL) displacement, which can provide a technical means to establish the ocean tide models and theory significance and reference for studying the OTL effect. In order to establish the regional OTL displacement model, the GPS kinematic precise point positioning (PPP) was used to determine the OTL displacement parameters. Three-dimensional ocean tide loading displacements of eight constituents at 11 sites in Hong Kong with eight years of continuous GPS observations are estimated. Comparing the RMS misfit of the model values, we found that the accuracy of K2 and K1 constituents are worst and the others' RMS are smaller than 2 mm. Comparing our results with the kinematic PPP and static PPP results, we found that our improved kinematic PPP can significantly improve the accuracy of K1 constituent. And the accuracy of 8 constituents derived by our method can reach the same accuracy as the kinematic PPP results, for K1 constituent, the accuracy of our method results is slightly better than the kinematic PPP results. Finally, using the least square quasi method to establish the GPS regional OTL displacement model can provide the reference for OTL effect correction in the coastal area which are lack of high accuracy local tide model.
Global Navigation Satellite System (GNSS) ambiguity fixed solution can greatly improve the accuracy of GNSS/inertial navigation system (INS) integrated system. But it is difficult to achieve perfect real-time dynamic single-frequency cycle-slip detection merely by GNSS observations especially in complex environment. Inertial aided cycle-slip detection term (DT) based on station-satellite double-differences and satellite single-difference observations are derived. The error characteristic of the DT is analyzed comprehensively. DT error is influenced by the error of INS positioning increment and this error of each satellite is related to the angle between its station-satellite vector and that vector of the base satellite. Thus, it is of importance to select base satellite. It is proposed that two group of DT can be used together by selecting two different base satellites. The threshold of detection is estimated in a moving window, where the DTs, whose INS error is submerged in GNSS error, are eliminated to reflect the effects of INS error. The threshold has strong self-adaptability.
We present an improved gravity acceleration approximation function (IGAAF) suitable for space-borne GPS real-time onboard orbit determination. The test of analyzing IGAAF's performances demonstrates that IGAAF maintains the truncation error in the magnitude of 1×102-1×103 nm/s2 only with the computational burden less than a 45×45 spherical model and a 200-320 kB RAM requirement for fitting coefficients, and that its accuracy is slightly superior to the spherical models with optimal degree and order for autonomous orbit determination (105×105 model for GOCE, 85×85 for CHAMP, 65×65 for GRACE-A and 55×55 for ZY3 and TerraSAR-X). Compared to the spherical model, IGAAF does not decrease the orbit accuracies. The IGAAF method attains a good trade-off between computational efficiency and coefficient capacity without decreasing the orbit accuracy, so it is of strong engineering value for space-borne GPS autonomous orbit determination.
In the robust weighted total least squares(RWTLS) algorithm, its robustness of the robust model is highly related to the initial values. If the least squares or total least squares estimates is used as the initial value, it will be affected by gross error, and certainly impacted the robust characteristics of RWTLS estimates. Considering the correlation between the observed vector and the coefficient matrix, we first deduce the weighted least-squares solution of Partial-EIV model, and a new RWTLS algorithm of correlated observation is proposed to solve the initial values of robust iterations by using the median parameter method. Then the median parameter method is used to determine the initial value, and on this basis we propose a new robust estimated method, which is based on the standardized residual error and considered the influence of gross error both on observation and structure spaces. The experiment results show that the proposed estimated method has a good performance to resist gross error, and the presented solution is more accurate than the traditional method for line fitting, and with the increase of the number of gross errors, the stability of the algorithm is superior to the traditional method.
We develop a set of independent intellectual property rights software named MAGREAS (Mars gravity recovery and analysis software/system) to determine the Mars spacecraft orbit and solve Mars dynamic parameters. We introduce the design idea and basic structure of the software, and the three-way Doppler data of MEX (Mars express) from Chinese VLBI Network as well as the two-way Doppler data from ESA are analyzed. The results show that for the three-way Doppler tracking data, the post-fit RMS is 0.079 mm/s, and the maximum difference between post-fit orbit and precise orbit provided by Royal Observatory of Belgium is less than 100 m; for the two-way tracking data, the post-fit RMS is 0.067 mm/s and the maximum difference is less than 10 m. The results indicate MAGREAS is reliable and accurate. This software can provide reference for processing the orbital tracking data of Mars spacecraft of the upcoming Chinese Mars exploration mission.
For the issues of detecting insensitive cycle-slip in GNSS signals, this paper analyzes the generation and distribution of insensitive cycle-slip in three-dimensional coordinate system, researches the influence of the detection threshold on detecting insensitive cycle-slip, and then studies the effect of combined detection of combinations. A combination coefficient selection method is proposed for detecting insensitive cycle-slip. This method builds a function model based on the detection amount of combined coefficient, and screening the most sensitive combination coefficients. The method can detect insensitive cycle-slip with similar hop variables at three frequencies. As verified by the BeiDou triple-frequency measured data, multiple GF combinations effectively increase the detection probability of insensitive cycle-slip, and the appropriate number is two. The combination selected by proposed method effectively detect the manually added insensitive cycle-slip.
With external velocity reference, a fast damping method is proposed for fast suppression of the oscillation error of inertial navigation system (INS) by changing the structure of traditional dam-ping network. In traditional damping network structure, the internal damping method is easy to be influenced by the carrier acceleration without external velocity reference, while the external damping method has poor dynamic characteristic. So the imposed method increases a forward channel from external velocity to the damping network to compensate the damping error caused by the carrier acceleration. According to the dynamic process of system damping, the fast damping network parameters are designed and the adjustable parameters are input to the damping network considering the dynamic characteristic and error suppression in the damping process. The simulation and test results show that the settling time of the conventional damping algorithm is about 1 h, which is shortened to 10 min by using the fast damping algorithm with the external velocity reference.
Soil moisture has great significance in the researches of hydrology, meteorology and agriculture yield estimation. The quad-polarimetric SAR images can provide a lot of polarimetric features, the significance of the features in surface parameter retrieval have attracted attentions in previous researches with no final conclusions because of the complexity of terrain scattering. In this paper, random forest regression (RFR) is used for both soil moisture retrieval and the importance evaluation of polarimetric features of Radarsat-2 images in winter wheat fields. According to the score of importance, feature selection and combination are done for modelling. We evaluate the retrieval accuracy of models with different feature combinations. The results show that models of important features selected by RFR have RMSE(root mean square error) less than 6% which are better results compared to traditional models; when compared with support vector regression and artifical neural networks, the RFR also shows best retrieval accuracies, which proves that RFR is suitable for soil moisture retrieval and feature selection. The high retrieval accuracies of LBC-CPD(linear backscatter coefficients-Cloude-Pottier decomposition) and LBC-CPR(linear backscatter coefficients-circular polarimetric ratio) indicates these features can improve the retrieval accuracy of soil moisture.
In the era of absence of realistic downward-looking array synthetic aperture radar (SAR) system, the difficulties in data acquisition and further imaging algorithms research have been brought about. So the research of downward-looking array SAR three-dimensional (3D) simulation has great significance. In order to restore the aircraft's real flight path and to verify the topographical mapping ability of array SAR technology in high-rise buildings, the traditional simulation research based on the assumption of uniform linear motion is abandoned. And the non-ideal trajectory motion error model of MIMO(multiple input multiple output) downward-looking array SAR is analyzed and constructed with high-abrupt urban buildings. The 3D range-Doppler (RD) imaging algorithm of urban buildings MIMO downward-looking array SAR under non-ideal trajectory is proposed. Then the flight path and attitude modeling simulation technique of the aeronautical platform, and a fast and efficient echo simulation technique are used to carry out simulation experiments. The correctness and effectiveness of the imaging algorithm are verified by the imaging results.
An errors-in-variable (EIV) with arbitrary constraints is proposed for the purpose of building regularization from remote sensed data, in which the edge points are treated as measurements, the constrained slopes and intercepts of each edge are chosen as parameters. Assuming the measurement vector and the design matrix are mutually correlated, the scheme of calculating the dispersion matrix of augmented matrix is suggested. A generic constrained weighted total least squares(WTLS) algorithm is derived with an approximate accuracy assessment method, and the WTLS algorithm of a quadratic constrained EIV problem is given as a specific case. Theoretic analysis and data experiment demonstrate the advantages of an EIV model compared with a Gauss-Helmert model in building regularization problem, and the rapid convergence rate of proposed WTLS algorithm. It aims to promote WTLS adjustment methods, and to expand the applications of total least squares method in new surveying technology with a certain theoretical and practical significance.
This study present the results of measuring the motion characteristics of the South Inilchek glacier with the C-band Sentinel-1 images acquired from February 2015 to February. 2017. During the offsets estimation in offset-tracking method, we firstly adopt three-step method to achieve high precision overall registration, and then change the matched window size to estimate the local offset with the NCC(normalized cross correlation) algorithm, which was further separated to get glacier motion information. The observation results show that the significant high-speed flow area was the ice tongue, where the flow velocity reduced from the upper to the bottom and slowed obviously at the end, and increased from its both sides to the middle. Moreover, the glacier motion velocity had the same trend with the temperature variation seasonally. The glacier tongue moved with the maximum average velocity of 49 cm/d during the period of May to August, and with the slowest moving in the interval from November to February for about 25-30 cm/d. While in terms of annual changes the velocity in the summer of 2015 was higher than that in 2016 as a whole for about 1-3 cm/d, and there was no marked difference in the other seasons. Furthermore, the monitoring result was demonstrated by comparing the results from Sentinel-1 images in quantity with that from high resolution L-band PALSAR-2 images. The statistics of the sparse pixels in the tongue indicated that the average of the velocity difference was 3.48 cm/d and the standard deviation was ±3.78 cm/d, which confirmed the reliability of the results.
Trajectory data which contains mining valve is widely distributed and large-scale. How to organize trajectory data and retrieve trajectory data efficiently becomes very difficult to solve. We present a framework of adaptive index based on Geohash to organize the worldwide and large-scale trajectory data set. Different trajectory data sets will be covered by Geohash grid which is deepest, and then we take the grid as root node to generate adaptive Geohash-Trees. In order to quickly locate the corresponding index, we design trie on the basis of the feature of Geohash. Adaptive Geohash-Trees is a spatial index based on grid. It can divide the space according to the track density by adopting a variety of strategies which improves the efficiency of range query. Meanwhile, we design the algorithm of incremental insertion and update for the supporting of real-time update of trajectory data. Furthermore, this framwork has been migrated in Oracle. The experiment results verify that our approach in several aspects such as range query and occupied disk size performs much better than R-Trees.
The spatial pattern of landscape is fundamental to landscape's function and development. Landscape metric analysis is widely applied in fields such as land use/land cover patterns, ecosystem service, forest monitoring, urban sprawl, regional biodiversity, etc. There are high correlations between many presented landscape metrics, which not only result in redundancy but also influence the clarity and accuracy of interpretation. Meanwhile, most of previous researches on landscape metric analysis use raster data and few use vector data. In order to solve these problems, a case study of Guangzhou land use pattern was performed. Firstly, we calculate 44 common landscape metrics using vector data. Then, multivariate statistical analysis, a combination of Spearman correlation analysis and factor analysis, is applied to solve the problem of redundancy. A core set of six landscape metrics in vector format is extracted, which respectively describe landscape fragmentation, average patch size, average shape complexity, spatial isolation, shape complexity variation and land use similarity.
To overcome the shortcoming of the existing fractal interpolation methods that don't take into account the constraints of geographical bending characteristics and the fractal interpolation process is uncontrollable for coastline. A controlled fractal interpolation method for coastline is proposed by considering its geographical bending characteristic constraints. Firstly, the coastline is divided into several parts according to the bending characteristics and fractal characteristics of different coastal landform types, which will change the traditional integral fractal interpolation into a combination of several piecewise landform bending interpolation units. Secondly, the one-dimensional random midpoint displacement method is used as a fractal interpolation function for each divided unit of the coastline, and in order to maintain the bending characteristics of different landform units in the coastline, the fractal parameters of the interpolation function are restricted by the constraints of each unit fractal characteristics. Finally, the result curve is got by linking the interpolation units in order. The experiments show that this method can keep well the geographical bending characteristics and fractal characteristics of different landform units of coastline, and the process of fractal interpolation is controllable.
Recent years, orienting space platform to construct satellite-ground integrative spatial information network has become a hot topic of information network construction. Conventional information network construction tends to be flow driven to meet the communication requirement between nodes. The construction of spatial information network focuses more on taking tasks in the application business as guide, and pulls from communication resources to optimal recombination and cooperative service of computing and storage resources. How to form linkage mechanism between application tasks and limited network resources, which can effectively solve the capacity constrained problems of remote sensing data storage, processing and transmission, is the key technical problem to realize the efficient data processing in spatial information network. According to the mechanism of satellite-ground collaborative remote sensing data processing, based on the task driven spatial information network architecture, an approach of spatial information network oriented satellite-ground collaborative computing migration is presented. We build an optimization model of computing migration, propose the implementation technique of computing migration and verify the feasibility of the approach through results analysis of the simulation case.
In order to solve the problems such as strong limitations, low versatility and efficiency, and inaccurate topological difference identification of topological conflict detection and consistency maintenance method, et al, in the process of area entities incremental integration, a new topological conflict detection and consistency maintenance method based spatial relation model is proposed. The integrated spatial relationship expression model is designed with qualitative area entities topological relation types, and quantitative geometric features on the premise of area incremental entities. Area entities topological conflict detection and consistency maintenance algorithm is designed and integrated with a standard of incremental area entities topological relationships in the latest version of the database. Experimental results show that the experimental accuracy can reach 90%, and the method can do the work of topological conflict detection and consistency maintenance method in the process of area entities incremental integration very well with good pertinence and applicability.