2012 Vol. 37, No. 9
The satellite networks have being developed to today,and its techniques have being developed maturely.International standards have being gradually developed,and markets also have being boomed.But the limitations in recent satellite networks,which are high-latency pipe structure,non-standard satellite IP function,lack of compatibility and expansibility,et al.,block the rapid development of them comparing with the ground Internet.We focus on the IP based next generation satellite network.Firstly,the status of recent satellite networks and the characteristics of the next generation satellite network have been introduced.Then,several key technologies in IP based next generation satellite network and its seamless integration with ground Internet have been discussed in detail from the routing scheme,transfer protocol,mobility management and QoS(Quality of Service) models.At last,the researching hot-topics and futures works of the next generation satellite networks have been concluded and prospected.
In order to improve service performance of navigation constellation,many experts consider the mixed constellation configuration of GEO satellite,IGSO satellite and MEO satellite.In the design of the new constellation configuration or the research of the existing constellation performance,the inter-satellite link(ISL) is needed to established,and the connectivity and robustness of the ISL are needed to analyze.The characteristics of the mixed constellation configuration are analyzed,and the characteristics of ISL establishment and the service performance of GEO satellite,IGSO satellite and Walker-δ constellation are discussed firstly.Then,the method to determine ISL connectivity in the mixed constellation by the adjacency matrix is put forward,and the standards of the ISL robustness are given using the number of ways between any two satellites,cutting point,cutting edge,k-connectivity degree and k-edge connectivity degree of graph theory.The minimum required number of ISLs of a k-connectivity′s constellation and the construction method of k-connectivity′s constellation are given for the first time.Finally,the 3GEO+3IGSO+24MEO constellation configuration is simulated.The criteria for ISL establishment,the connectivity and robustness of this mixed constellation configuration are studied,and many important conclusions are obtained.
The analysis of ΔLOD is of profound significance on climate forecasting,disaster prevention,deep space exploration,geodesy and other fields.With extremely high precision and angular resolution,VLBI plays the irreplaceable role on the analysis of the changes of the length of day.Based on this,the ΔLOD sequence and its precision are estimated from VLBI observations during 1985-2010 with comparison with the results published by IVS.The half-month,month,half-year,year periods and longer periods of the changes of ΔLOD is extracted with FFT and wavelet analysis.At the same time,the integration curve map of the ΔLOD is made to analyze the seasonal and long-term changes of earth rotation.At last,the causes of the changes of LOD are analyzed.
Based on the analysis of common regional troposphere surface models such as four-parameter surface model and a curve model,we propose a new kind of regional troposphere delay model based on multi-faceted functions.We use our zero-difference troposphere estimate software to process the CORS data of one province to analyze results of various types of model.The experimental results show that multi-faceted function model fitting residual RMS of the troposphere is about ±6 mm,and the convergence in the initial stage is much smaller than four-parameter surface model,greater stability between the epoch,there is no significant ups and downs,to better describe the regional troposphere.At the same time,we use different troposphere solutions to get PPP results,the two ways are random walk and model fitting.The experimental results indicate that model fit method can reduce the number of unknown parameters,and enhance strength and stability,which will improve the accuracy of the initial epochs.The accuracies of positioning after the convergence are with considerable accuracy.This new method can help to get dm-level PPP results quickly,but it doesn′t have obvious effect for improving the speed of convergence to cm-level.
The precision of dynamic precise point positioning using Kalman filtering will be degraded,even be divergent when the outliers exist.Particle filtering is applied to control the influences of the observational outliers,and improve the accuracy of positioning.Particle filtering is a kind of nonlinear filter with non-Gaussian distribution,and it can obtain accurate parameters by random sample.The weight of each particle is defined based on the probability densities of the observational errors,predicted state errors as well as the important distribution in order to control the influences of contaminated particles to the positioning results.Kalman filtering is employed to get the important sampling to slow down the degeneracy of the particle.The free-ionosphere ambiguities are fixed before data processing to reduce the number of parameters in the state vector.An actual dynamic GPS data set is employed to test the particle filter procedure.The procedure of the particle filtering can efficiently control the influences of the observational outliers,and improve the accuracy of the dynamic precise point positioning.
We put forward a method of using LSSVM to fit the satellites' clock error data,and give the concrete step of the method.For the both ends dissipate phenomenon,we make use of the method of adding window to improve accuracy.The calculate example using the clock error data from IGS shows that the add window LSSVM method has higher accuracy than the existing method,and the polynomial kernel function is the better kernel function than others.At last,we use the adding window LSSVM to fit the clock error data of different satellite clocks.
Kalman filter is widely used in the area of kinematic positioning and navigation.However,it doesn′t have the ability to resist the influence of measurement outliers,hence its performance is easy impacted by the observation outliers or kinematic state disturbing.In order to guarantee the reliability of the navigation with precise dynamic model,a model set,which contains many different observation models,is established.An improved Kalman filtering,in which the design matrix of the observational model is substituted by its expectation is proposed to control the influences of the measurement outliers.An integrated GPS/INS navigation example is given to show that the modified Kalman filtering algorithm works well.
Based on the GPS velocity field data from 1999-2007,we estimate velocity residuals and unbiassedness of rigid model,uniform strain model and linear strain model about block movement.Results show that block movement characteristics can be well described by linear strain model.According to the complexity of the fault belt deformation,we also give slip velocity of each fault in east boundary of Sichuan-Yunnan block using above three models to solve average sliding rate and near-field rate of faults in the research area.On the basis of judging possible deformation width of each fault,we obtain the abnormal region are Anninghe fault,which are relatively closed and the high rate of shear strain accumulation in the Xianshuihe-Anninghe-Zemuhe-Xiaojiang faults.
Based on the analysis of fitting residual errors of GPS velocity,the errors of deformation parameters,and the applicability of block model,we use GPS velocity fields,strain rate fields,slipping rate of faults and GPS profile results from 1999-2007 and from 2007-2009 to identify the deformation difference of principle faults on the north section of North-South Seismic Zone.The results show that the laevorotation deformation of Zhuanglanghe Fault has increased,which the dextrorotation deformation rate calculated from Eulerian model is about 1.5±0.3 mm·a-1 in the period 1999-2007,and the laevorotation deformation rate is about 5.3±0.4 mm·a-1 in the period 2007-2009.On the other hand,the laevorotation slipping rate of north boundary of Qilian block has increased from 2.5±0.2 mm·a-1 to 7.0±0.2 mm·a-1 in these two periods.Otherwise,the extrusion deformation of Liupanshan and Haiyuan Faults has strengthened too,but the deformation magnitude is not big.These above deformation characteristics illustrate that the region of east side of Zhuanglanghe Fault has accumulated high strain energy.
An attitude determination method for a deep-space mission spacecraft is derived and its performance is analyzed.The attitude determination system is composed of inertial measurement units(IMU),an X-ray Pulsar scanner,onboard computers and electronic devices.IMU data is processed to maintain real-time knowledge of spacecraft attitude relative to an inertial reference frame.X-ray pulsar scanner data is processed using unscented Kalman filter(UKF) to estimate and correct the attitude determination errors and the gyro drift compensation errors.The results of two X-ray pulsars availability analysis for deep-space attitude determination mission are presented.Linear covariance analysis techniques are used to evaluate nominal attitude determination performance,the effects of sensor measurement accuracy variations,and the effects of gyros misalignment errors.A semi-physical simulation scheme is devised to verify the integrated attitude determination algorithm based on the data from NASA HEASARC.Results of a nonlinear simulation analysis of attitude determination performance are also shown.Compared to only using pulsars or IMU,the attitude determination accuracy of integrated algorithm is higher,and the performance of processing nonlinear state functions for UKF is superior to EKF.It can satisfy the system requirements of high precision and stabilization.
By analyzing the reason of masking and combining the method of posterior probability of observation error and the influence of Kullback-Leiber divergence,a Bayes unmasking method for gross errors detection is proposed based on the idea of one-by-one searching.The experimental results show that this method is successful for gross errors detection.The method not only detects the gross errors of masking effectively and has a bstter result,but also has a simple computational process and higher efficiency.
Aiming at the complexity of image line feature matching,we propose a matching strategy turning to the establishment of object topology relation table of describing object structure based on the object structure information.The method takes advantage of the graphical relations of image point and line features obtained by image feature extracting to build object topology relation table of describing object structure.Candidacy polygon features are acquired through segmentation geometry graphics of left and right image.The arc-polygon file of association graph of left and right image is obtained using vector data topology relation automatism algorithm.Then,the transmission match is carried out by the method of depth-first traversal based on base match and topology relation table.As the relationship tables consistency shows the composition of objects,features elements of consistency,the matching search scope of whole image is made smaller,thereby the image matching accuracy and speed is enhanced.
A new hierarchical semantic model method is proposed to overcome the disadvantages existing in most of the relative methods.Firstly,a multi-scale segmentation is employed to obtain the candidate object regions.Secondly,semantic model is utilized to get objects semantic distribution information.Lastly,in the extraction processing stage,a correlative function based on the semantic model and the candidate object regions is built up and maximized to extract the objects.Experimental results demonstrate the precision,robustness,and effectiveness of the proposed method.
Several polarimetric characteristics are extracted from ALOS PALSAR data,containing coherent and non-coherental characteristics.Then,through qualitative and quantitative indicators,their abilities to distinguish five typical urban land covers including artificial features,bare land,farmland,woodland and water are analyzed and compared.The experimental results show that circular polarization correlation coefficient,linear polarization correlation coefficient,total power and XPI are the best set of polarization characteristics to classify these land covers.Classification using such a group of characteristics can achieve an overall accuracy of 75.5% with Kappa being 0.651 1.In addition,we find that artificial features are easier to be distinguished from other land cover types,while bare land and farmland are prone to be mixed up.
The self-calibration block adjustment based on the additional parameters is applied to ADS40 imagery geometric positioning.Upon the detailed investigation on the imaging error properties of the three-line-scanner CCD sensor,the suitable self-calibration parameter model is built up.Then three types of exterior orientation parameter models and the self-calibration block adjustment model are set up.Finally the normal block adjustment test and the self-calibration block adjustment are carried out using the ADS40 data.Experimental results prove that the self-calibration block adjustment technique can efficiently compensate the system errors of ADS40 image and significantly enhance the positioning accuracy.
The CCD sensors on board HJ-1A/B satellites launched by China can capture images with high spatial and temporal resolution,which makes them have great potential in quantitatively monitoring the water quality of inland lakes by remote sensing technologies.The atmospheric correction problem is the first important issue to be resolved.With the assistance of quasi-synchronous Terra/MODIS aerosol data,the HJ-1A/B CCD image on October 24,2009 was done the atmospheric correction processing using FLAASH model.The atmospheric effects have been effectively removed,and the atmospheric correction method worked well at the red and green bands with relatively high remote sending reflectance,while it worked poorly at the near-infrared and blue bands with relatively low remote sensing reflectance.It may be related with signal to noise ratio and land adjacent pixels effect at different bands.This study can provide reference for the atmospheric correction of inland lake HJ-1A/B satellites CCD images.
For the highly turbid waters along the China coastal region,MODIS atmospheric correction products derived using shortwave infrared(SWIR) bands always have signal saturation and stripe noise.We demonstrate an improvement method with neural network for MODIS/Terra atmospheric correction products using quasi-synchronization HJ-1A/B satellites CCD images and in-situ data.The average relative error of the improved Rrs and MODIS Rrs is 13.3% and the average relative error of the improved Rrs in signal saturation area and in-situ data is 28.2%.The results show that this method is an effective way to significantly repair the blank area in the MODIS/Terra products and remove the stripe noise with acceptable error.
Taking Poyang Lake as an example,numerical simulation and remote sensing are fully taken advantage of to monitor the water extend change.Water extend extracted from MODIS images is used to initialize numerical simulation boundaries of water body and calibrate simulation parameters,and also used to be compared with the simulated water extend.With this method,the water extend dynamic change of Poyang Lake from July 8,2001 to November 30,2001 is simulated.Results show that the average absolute error of simulated water level is less than 15 cm.Water extend of numerical simulated and extracted from the MODIS cloud-free images have a good agreement,and the average relative error of water area is 5.7%.This study shows that by making use of satellite remote sensing data,precision of water extend hydrodynamic numerical simulation of the Poyang Lake can be increased.This method can make up for the deficiency of water dynamic monitoring by optical remote sensing in bad weather conditions,thus providing a powerful technique support for water environment monitoring of lakes.
The serious subsidence in reclamation area may cause inaccurate topographic survey based on bench points.Time series DInSAR is an effect method to detect subsidence velocity.We analyze the interferogram phase model,and use amplitude dispersion values and temporal coherence to select coherent points.According to the different space-time characteristics,we separate and extract the deformation phase,and detect subsidence velocity.24 Envisat ASAR images acquired from Oct.2007 to Feb.2010 were used to detect subsidence velocity.The results show that the age of Hydraulic-Fill Soil will affect the subsidence.
An empirical model of ocean surface wind speed and significant wave height relation is proposed by analyzing data obtained by buoy in-situ observations at the coverage of radar beam.The model′s parameters are estimated by minimum standard Euclidean norm method from six months′ buoy in-situ measuring data and the estimation results show the model′s stability.The wind and wave model is applied to HFSWR OSMAR071′s wind speed inversion.Compared with four months′ buoy in-situ observation wind speed,the correlation coefficient is 0.6 and the RMSE is 2.7 m·s-1 which confirmes the model′s availability,which verifies the validation of the empirical wind speed inversion.
To denoise the severely polluted documental image produced by the process of scanning or duplicating,characteristics of noise and text content are investigated.Gray morphology with constructed four direction structuring elements is applied to filter the severely background noise and retain directional text content information.Local structural tensor is brought in to describe the local gradient difference between noise and text content,making the scale element dynamic,which will be larger when process probable noise area and smaller when it is closed to the text.By this way,denoising is satisfying and easily lost stroke and punctuation are well saved.Finally gray morphologic reconstruction is applied to recover original text content information,picking up image only containing text information.Comparing with normal image processing methods,this proposed approach has got efficient effect,which is suitable to deal with this kind of severely polluted documental image.
Adaptive green noise halftoning algorithm with the optimal hysteresis constant was proposed.A random number was used to select one of the pre-designed different error diffusion filters and alter the hysteresis constant based on the image gray level.Regular strips and nonuniform visual effect which are created because of point aggregation in a general green noise algorithm are effectively eliminated.Experiment results show that the halftone image produced by this algorithm can maintain the amplitude characteristic and guarantee visual uniformity.
To establish large-scale multi-level database of land use,we propose a new method to generalize polygon roads to polygon-polyline mixed roads.When integrating road polygon datasets,the network road polygons are divided into three types,retention,deletion and reduced maintenance parts.Deletion parts are removed,and the reduced maintenance parts are changed into linear roads,and the retention parts and the linear roads form the new mixed road dataset of the new target scale.Moreover,the deletion and reduced maintenance road polygons will be divided into small pieces and then incorporate into polygons which matches these pieces in order to ensure the land use data having no gaps.
The evaluation and guarantee of the quality of automated map generalization is recently a hot point at home and abroad.We mainly study on the quality assessment of scattered polygons.In order to get the final assessment of the whole cartographic area′s generalization,we propose a method containing both qualitative and quantitative ways.According to the criterion of some critical indices of features,this method begins with the presentation of these indices.By comparing with the criterion,every feature will have a specific assessment result for its generalization.After integrating all the results,we can get the final assessment of generalization.At last,taking the generalization of rural settlements as an example,we introduce the process of quality assessment in detail,which has been proved to be feassible.
We propose an improved algorithm of combination(aggregation and amalgamation) of parcels on the point of computational geometry.Aggregation is done with the concept of buffer,by which the so-called bridge can be merged efficiently,while amalgamation is done using skeleton.The process of amalgamation is influenced by both the curves shared by parcels and the types of the parcels.After being splitted by skeleton,the chosen parcel derives a set of polygons to be combined by the neighbor parcels one by one.This algorithm keeps the natural shape of parcels,and has taken the amount of the father level of shown parcels into account.
A new method of morphing for two linear features is proposed by considering their BLG-tree(binary line generalization tree) structures.First,the structures of the linear features are identified using Douglas-Peucker algorithm,and represented by the BLG-tree.Second,corresponding nodes are determined by matching the nodes of the BLG-trees from higher level to lower level.In this case,the two linear features can be partitioned into a set of corresponding line segments.Third,the linear interpolation algorithm is employed for morphing every pair of the corresponding line segments.Finally,the simulated and real experiments are implemented and the results show that the proposed method can improve the accuracy of morphing,and is effective in keeping the structural characteristics of linear features.
A routing protocol based on position service information is proposed which adopts the cooperation of micro-scale routing within a road limits and macro-scale routing between roads for VANETS.The optimized Dijkstra algorithm is employed to calculate an oriented optimum route for source and destination.The simulation results show that the proposed protocol enhances the throughputs and reduce the end-to-end delay.
By describing the spatial-temporal characteristics of statistical data and extending the dynamic visual parameters of dynamic spatial visualization,three types of dynamic statistical charts are designed.Those charts are based on dynamic spatial visualization of electronic map and are used in representing the multi-temporal statistical information.According to time factor technology using in map animation and the interpolation technology using in the key-frame,the idea and the flow of dynamic statistical charts are designed and built.Meanwhile,the research results was adopted in a program,which realized the visualization expression of the statistical data which is changing as time.Finally,we achieved the expected effect of dynamic spatial visualization expression of statistical data,which indicating that the research has certain promotion value.