[5599] | 1 |
|
---|
| 2 | \documentclass[10pt]{beamer}
|
---|
| 3 | \usetheme{umbc2}
|
---|
| 4 | \useinnertheme{umbcboxes}
|
---|
| 5 | \setbeamercolor{umbcboxes}{bg=violet!12,fg=black}
|
---|
| 6 |
|
---|
| 7 | \usepackage{longtable}
|
---|
| 8 | \usepackage{tabu}
|
---|
[5607] | 9 | \usepackage{subeqnar}
|
---|
[5599] | 10 |
|
---|
| 11 | \newcommand{\ul}{\underline}
|
---|
| 12 | \newcommand{\be}{\begin{equation}}
|
---|
| 13 | \newcommand{\ee}{\end{equation}}
|
---|
| 14 | \newcommand{\bdm}{\begin{displaymath}}
|
---|
| 15 | \newcommand{\edm}{\end{displaymath}}
|
---|
| 16 | \newcommand{\bea}{\begin{eqnarray}}
|
---|
| 17 | \newcommand{\eea}{\end{eqnarray}}
|
---|
[5607] | 18 | \newcommand{\bsea}{\begin{subeqnarray*}}
|
---|
| 19 | \newcommand{\esea}{\end{subeqnarray*}}
|
---|
[5599] | 20 | \newcommand{\mb}[1]{\mbox{#1}}
|
---|
| 21 | \newcommand{\mc}[3]{\multicolumn{#1}{#2}{#3}}
|
---|
| 22 | \newcommand{\bm}[1]{\mbox{\bf #1}}
|
---|
| 23 | \newcommand{\bmm}[1]{\mbox{\boldmath$#1$\unboldmath}}
|
---|
| 24 | \newcommand{\bmell}{\bmm\ell}
|
---|
| 25 | \newcommand{\hateps}{\widehat{\bmm\varepsilon}}
|
---|
| 26 | \newcommand{\graybox}[1]{\psboxit{box .9 setgray fill}{\fbox{#1}}}
|
---|
| 27 | \newcommand{\mdeg}[1]{\mbox{$#1^{\mbox{\scriptsize o}}$}}
|
---|
| 28 | \newcommand{\dd}{\mbox{\footnotesize{$\nabla \! \Delta$}}}
|
---|
| 29 | \newcommand{\p}{\partial\,}
|
---|
| 30 | \renewcommand{\d}{\mbox{d}}
|
---|
| 31 | \newcommand{\dspfrac}{\displaystyle\frac}
|
---|
| 32 | \newcommand{\nl}{\\[4mm]}
|
---|
| 33 |
|
---|
[5601] | 34 | \title{Processing GNSS Data in Real-Time}
|
---|
[5599] | 35 |
|
---|
| 36 | \author{Leo\v{s} Mervart}
|
---|
| 37 |
|
---|
[5601] | 38 | \institute{TU Prague}
|
---|
[5599] | 39 |
|
---|
[5601] | 40 | \date{Frankfurt, January 2014}
|
---|
[5599] | 41 |
|
---|
| 42 | % \AtBeginSection[]
|
---|
| 43 | % {
|
---|
| 44 | % \begin{frame}
|
---|
| 45 | % \frametitle{Table of Contents}
|
---|
| 46 | % \tableofcontents[currentsection]
|
---|
| 47 | % \end{frame}
|
---|
| 48 | % }
|
---|
| 49 |
|
---|
| 50 | \begin{document}
|
---|
| 51 |
|
---|
| 52 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 53 |
|
---|
| 54 | \begin{frame}
|
---|
| 55 | \titlepage
|
---|
| 56 | \end{frame}
|
---|
| 57 |
|
---|
| 58 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 59 |
|
---|
| 60 | \begin{frame}
|
---|
[5601] | 61 | \frametitle{Medieval Times of GNSS (personal memories)}
|
---|
| 62 |
|
---|
| 63 | \begin{description}
|
---|
| 64 | \item[1991] Prof. Gerhard Beutler became the director of the Astronomical Institute, University of
|
---|
| 65 | Berne. The so-called Bernese GPS Software started to be used for (post-processing) analyzes of
|
---|
| 66 | GNSS data.
|
---|
| 67 | \item[1992] LM started his PhD study at AIUB.
|
---|
| 68 | \item[1992] Center for Orbit Determination in Europe (consortium of AIUB, Swisstopo, BKG, IGN, and
|
---|
| 69 | IAPG/TUM) established. Roughly at that time LM met Dr. Georg Weber for the first time.
|
---|
| 70 | \item[1993] International GPS Service formally recognized by the IAG.
|
---|
| 71 | \item[1994] IGS began providing GPS orbits and other products routinely (January, 1).
|
---|
| 72 | \item[1995] GPS declared fully operational.
|
---|
| 73 | \end{description}
|
---|
| 74 |
|
---|
[5599] | 75 | \end{frame}
|
---|
| 76 |
|
---|
| 77 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 78 |
|
---|
[5601] | 79 | \begin{frame}
|
---|
| 80 | \frametitle{CODE-Related Works in 1990's}
|
---|
[5599] | 81 |
|
---|
[5601] | 82 | \begin{itemize}
|
---|
| 83 | \item The Bernese GPS Software was the primary tool for CODE analyzes (Fortran~77).
|
---|
| 84 | \item IGS reference network was sparse.
|
---|
| 85 | \item Real-time data transmission limited (Internet was still young, TCP/IP widely accepted 1989).
|
---|
| 86 | \item CPU power of then computers was limited (VAX/VMS OS used at AIUB).
|
---|
| 87 | \end{itemize}
|
---|
| 88 |
|
---|
| 89 | In 1990's high precision GPS analyzes were almost exclusively performed in post-processing mode.
|
---|
| 90 | The typical precise application of GPS at that time was the processing of a network of static
|
---|
| 91 | GPS-only receivers for the estimation of station coordinates.
|
---|
| 92 |
|
---|
| 93 | \end{frame}
|
---|
| 94 |
|
---|
| 95 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 96 |
|
---|
[5599] | 97 | \begin{frame}
|
---|
[5601] | 98 | \frametitle{Tempora mutantur (and maybe ``nos mutamur in illis'')}
|
---|
[5599] | 99 |
|
---|
[5603] | 100 | \includegraphics[width=0.7\textwidth,angle=0]{pp_vs_rt.png}
|
---|
[5602] | 101 |
|
---|
[5603] | 102 | \vspace*{-2cm}
|
---|
| 103 | \hspace*{6cm}
|
---|
| 104 | \includegraphics[width=0.4\textwidth,angle=0]{ea_ztd_21h.png}
|
---|
| 105 |
|
---|
| 106 |
|
---|
[5599] | 107 | \end{frame}
|
---|
| 108 |
|
---|
[5601] | 109 |
|
---|
[5599] | 110 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 111 |
|
---|
[5601] | 112 | \begin{frame}
|
---|
[5604] | 113 | \frametitle{O tempora! O mores!}
|
---|
[5601] | 114 |
|
---|
[5604] | 115 | \begin{itemize}
|
---|
| 116 | \item people want more and more \ldots
|
---|
| 117 | \item everybody wants everything immediately \ldots
|
---|
| 118 | \item \hspace*{2cm} and, of course, free of charge \ldots
|
---|
| 119 | \end{itemize}
|
---|
| 120 | \vspace*{5mm}
|
---|
| 121 | In GNSS-world it means:
|
---|
| 122 | \begin{itemize}
|
---|
| 123 | \item There are many new kinds of GNSS applications - positioning is becoming just one of many
|
---|
| 124 | purposes of GNSS usage.
|
---|
| 125 | \item Many results of GNSS processing are required in real-time (or, at least, with very small
|
---|
| 126 | delay).
|
---|
| 127 | \item GPS is not the only positioning system. Other GNSS are being established (for practical but
|
---|
| 128 | also for political reasons).
|
---|
| 129 | \item People are used that many GNSS services are available free of charge (but the development and
|
---|
| 130 | maintenance has to be funded).
|
---|
| 131 | \end{itemize}
|
---|
| 132 |
|
---|
| 133 | \begin{block}{But \ldots}
|
---|
| 134 | \end{block}
|
---|
| 135 |
|
---|
[5601] | 136 | \end{frame}
|
---|
| 137 |
|
---|
| 138 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 139 |
|
---|
| 140 | \begin{frame}
|
---|
[5604] | 141 | \frametitle{Nihil novi sub sole}
|
---|
[5601] | 142 |
|
---|
[5606] | 143 | Each GNSS-application is based on processing code and/or phase observations
|
---|
| 144 | \vspace*{-3mm}
|
---|
[5604] | 145 | \begin{eqnarray*}
|
---|
| 146 | P^i & = & \varrho^i + c\;\delta - c\;\delta^i + T^i + I^i + b_P \\
|
---|
| 147 | L^i & = & \varrho^i + c\;\delta - c\;\delta^i + T^i - I^i + b^i
|
---|
| 148 | \end{eqnarray*}
|
---|
| 149 | where
|
---|
| 150 | \begin{tabbing}
|
---|
| 151 | $P^i$, $L^i$ ~~~~~~~ \= are the code and phase measurements, \\
|
---|
| 152 | $\varrho^i$ \> is the travel distance between the satellite
|
---|
| 153 | and the receiver, \\
|
---|
| 154 | $\delta$, $\delta^i$ \> are the receiver and satellite clock errors, \\
|
---|
| 155 | $I^i$ \> is the ionospheric delay, \\
|
---|
| 156 | $T^i$ \> is the tropospheric delay, \\
|
---|
| 157 | $b_P$ \> is the code bias, and \\
|
---|
| 158 | $b^i$ \> is the phase bias (including initial
|
---|
| 159 | phase ambiguity).
|
---|
| 160 | \end{tabbing}
|
---|
| 161 | Observation equations reveal what information can be gained from processing GNSS data:
|
---|
| 162 | \begin{itemize}
|
---|
| 163 | \item geometry (receiver positions, satellite orbits), and
|
---|
| 164 | \item state of atmosphere (both dispersive and non-dispersive part)
|
---|
| 165 | \end{itemize}
|
---|
| 166 | The observation equations also show that, in principle, GNSS is an
|
---|
| 167 | \textcolor{blue!90}{interferometric} technique -- precise results are actually always relative.
|
---|
| 168 |
|
---|
[5605] | 169 | \end{frame}
|
---|
[5604] | 170 |
|
---|
[5605] | 171 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 172 |
|
---|
| 173 | \begin{frame}
|
---|
| 174 | \frametitle{Challenges of Real-Time GNSS Application}
|
---|
[5606] | 175 | \begin{itemize}
|
---|
[5608] | 176 | \item Suitable algorithms for the parameter adjustment have to be used (filter techniques instead
|
---|
| 177 | of classical least-squares).
|
---|
[5606] | 178 | \item Reliable data links have to been established (between rover station and a reference station,
|
---|
| 179 | between receivers and processing center, or between processing center and DGPS correction
|
---|
| 180 | provider).
|
---|
| 181 | \item Software tools for handling real-time data (Fortran is not the best language for that).
|
---|
| 182 | \item Fast CPUs.
|
---|
| 183 | \end{itemize}
|
---|
[5605] | 184 |
|
---|
[5606] | 185 | As said above -- GNSS is an interferometric technique. Processing of a single station cannot give
|
---|
| 186 | precise results. However, data of reference station(s) can be replaced by the so-called corrections
|
---|
| 187 | (DGPS corrections, precise-point positioning etc.) These techniques are particularly suited for
|
---|
| 188 | real-time applications because the amount of data being transferred can be considerably reduced.
|
---|
| 189 |
|
---|
[5601] | 190 | \end{frame}
|
---|
| 191 |
|
---|
[5607] | 192 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 193 |
|
---|
| 194 | \begin{frame}
|
---|
[5609] | 195 | \frametitle{Algorithms -- Kalman Filter}
|
---|
[5607] | 196 |
|
---|
| 197 | \begin{small}
|
---|
| 198 |
|
---|
| 199 | State vectors $\bmm{x}$ at two subsequent epochs are
|
---|
| 200 | related to each other by the following linear equation:
|
---|
| 201 | \bdm
|
---|
| 202 | \bmm{x}(n) = \bmm{\Phi}\; \bmm{x}(n-1) + \bmm{\Gamma}\;\bmm{w}(n)~,
|
---|
| 203 | \edm
|
---|
| 204 | where $\Phi$ and $\Gamma$ are known matrices and {\em white noise} $\bmm{w}(n)$ is a random
|
---|
| 205 | vector with the following statistical properties:
|
---|
| 206 | \bsea
|
---|
| 207 | E(\bmm{w}) & = & \bmm{0} \\
|
---|
| 208 | E(\bmm{w}(n)\;\bmm{w}^T(m)) & = & \bmm{0} ~~ \mbox{for $m \neq n$} \\
|
---|
| 209 | E(\bmm{w}(n)\;\bmm{w^T}(n)) & = & \bm{Q}_s(n) ~.
|
---|
| 210 | \esea
|
---|
| 211 |
|
---|
| 212 | Observations $\bmm{l}(n)$ and the state vector $\bmm{x}(n)$ are related to
|
---|
| 213 | each other by the linearized {\em observation equations} of form
|
---|
| 214 | \bdm \label{eq:KF:obseqn}
|
---|
| 215 | \bmm{l}(n) = \bm{A}\;\bmm{x}(n) + \bmm{v}(n) ~ ,
|
---|
| 216 | \edm
|
---|
| 217 | where $\bm{A}$ is a known matrix (the so-called {\em first-design matrix}) and
|
---|
| 218 | $\bmm{v}(n)$ is a vector of random errors with the following properties:
|
---|
| 219 | \bsea\label{eq:KF:resid}
|
---|
| 220 | E(\bmm{v}) & = & \bmm{0} \\
|
---|
| 221 | E(\bmm{v}(n)\;\bmm{v}^T(m)) & = & \bmm{0} ~~ \mbox{for $m \neq n$} \\
|
---|
| 222 | E(\bmm{v}(n)\;\bmm{v^T}(n)) & = & \bm{Q}_l(n) ~.
|
---|
| 223 | \esea
|
---|
| 224 |
|
---|
| 225 | \end{small}
|
---|
| 226 |
|
---|
| 227 | \end{frame}
|
---|
| 228 |
|
---|
| 229 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 230 |
|
---|
| 231 | \begin{frame}
|
---|
| 232 | \frametitle{Classical KF Form}
|
---|
| 233 |
|
---|
| 234 | Minimum Mean Square Error (MMSE) estimate $\widehat{\bmm{x}}(n)$ of vector
|
---|
| 235 | $\bmm{x}(n)$ meets the condition
|
---|
| 236 | $E\left((\bmm{x} - \widehat{\bmm{x}})(\bmm{x} - \widehat{\bmm{x}})^T\right) =
|
---|
| 237 | \mbox{min}$ and is given by
|
---|
| 238 | \begin{subeqnarray}\label{eq:KF:prediction}
|
---|
| 239 | \widehat{\bmm{x}}^-(n) & = & \bmm{\Phi} \widehat{\bmm{x}}(n-1) \\
|
---|
| 240 | \bm{Q}^-(n) & = & \bmm{\Phi} \bm{Q}(n-1) \bmm{\Phi}^T +
|
---|
| 241 | \bmm{\Gamma} \bm{Q}_s(n) \bmm{\Gamma}^T
|
---|
| 242 | \end{subeqnarray}
|
---|
| 243 | \begin{subeqnarray}\label{eq:KF:update}
|
---|
| 244 | \widehat{\bmm{x}}(n) & = & \widehat{\bmm{x}}^-(n) +
|
---|
| 245 | \bm{K}\left(\bmm{l} -
|
---|
| 246 | \bm{A}\widehat{\bmm{x}}(n-1)\right) \\
|
---|
| 247 | \bm{Q}(n) & = & \bm{Q}^-(n) - \bm{K}\bm{A}\bm{Q}^-(n) ~,
|
---|
| 248 | \end{subeqnarray}
|
---|
| 249 | where
|
---|
| 250 | \bdm \label{eq:KF:KandH}
|
---|
| 251 | \bm{K} = \bm{Q}^-(n)\bm{A}^T\bm{H}^{-1}, \quad
|
---|
| 252 | \bm{H} = \bm{Q}_l(n) + \bm{A}\bm{Q}^-(n)\bm{A}^T ~.
|
---|
| 253 | \edm
|
---|
| 254 | Equations (\ref{eq:KF:prediction}) are called {\em prediction},
|
---|
| 255 | equations (\ref{eq:KF:update}) are called {\em update} step of Kalman filter.
|
---|
| 256 |
|
---|
| 257 | \end{frame}
|
---|
| 258 |
|
---|
| 259 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 260 |
|
---|
| 261 | \begin{frame}
|
---|
| 262 | \frametitle{Square-Root Filter} \label{sec:SRF}
|
---|
| 263 | \begin{small}
|
---|
| 264 | Algorithms based on equations (\ref{eq:KF:prediction}) and
|
---|
| 265 | (\ref{eq:KF:update}) may suffer from numerical instabilities that are primarily
|
---|
| 266 | caused by the subtraction in (\ref{eq:KF:update}b). This deficiency may be
|
---|
| 267 | overcome by the so-called {\em square-root} formulation of the Kalman filter
|
---|
| 268 | that is based on the so-called {\em QR-Decomposition}. Assuming the
|
---|
| 269 | Cholesky decompositions
|
---|
| 270 | \be \label{eq:SRF:defsym}
|
---|
| 271 | \bm{Q}(n) = \bm{S}^{T} \bm{S} , \quad
|
---|
| 272 | \bm{Q}_l(n) = \bm{S}^T_l \bm{S}_l, \quad
|
---|
| 273 | \bm{Q}^-(n) = \bm{S}^{-T}\bm{S}^-
|
---|
| 274 | \ee
|
---|
| 275 | we can create the following block matrix and its QR-Decomposition:
|
---|
| 276 | \be \label{eq:SRF:main}
|
---|
| 277 | \left(\begin{array}{ll}
|
---|
| 278 | \bm{S}_l & \bm{0} \\
|
---|
| 279 | \bm{S}^-\bm{A}^T & \bm{S}^-
|
---|
| 280 | \end{array}\right)
|
---|
| 281 | =
|
---|
| 282 | N \left(\begin{array}{cc}
|
---|
| 283 | \bm{X} & \bm{Y} \\
|
---|
| 284 | \bm{0} & \bm{Z}
|
---|
| 285 | \end{array}\right) ~ .
|
---|
| 286 | \ee
|
---|
| 287 | It can be easily verified that
|
---|
| 288 | \bsea\label{eq:SRF:HK}
|
---|
| 289 | \bm{H} & = & \bm{X}^T\bm{X} \\
|
---|
| 290 | \bm{K}^T & = & \bm{X}^{-1}\bm{Y}\\
|
---|
| 291 | \bm{S} & = & \bm{Z} \\
|
---|
| 292 | \bm{Q}(n) & = & \bm{Z}^T\bm{Z} ~ .
|
---|
| 293 | \esea
|
---|
| 294 | State vector $\widehat{\bmm{x}}(n)$ is computed in a usual way using the
|
---|
| 295 | equation (\ref{eq:KF:update}a).
|
---|
| 296 | \end{small}
|
---|
| 297 | \end{frame}
|
---|
| 298 |
|
---|
[5609] | 299 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 300 |
|
---|
| 301 | \begin{frame}
|
---|
| 302 | \frametitle{Data Transfer -- NTRIP}
|
---|
| 303 |
|
---|
| 304 | In order to be useful data have to be provided in a well-defined \textcolor{blue}{format}.
|
---|
| 305 | RTCM (Radio Technical Commission for Maritime Services) messages are widely used for GNSS data in
|
---|
| 306 | real-time.
|
---|
| 307 |
|
---|
| 308 | \vspace*{5mm}
|
---|
| 309 |
|
---|
| 310 | In addition to a format the so-called \textcolor{blue}{protocol} has to be defined. Using a given
|
---|
| 311 | protocol the data user communicates with the data provider.
|
---|
| 312 |
|
---|
| 313 | For GNSS data, the so-called \textcolor{blue}{NTRIP} streaming protocol is used.
|
---|
| 314 | \begin{itemize}
|
---|
| 315 | \item NTRIP stands for Networked Transport of RTCM via Internet Protocol.
|
---|
| 316 | \item NTRIP is in principle a layer on top of TCP/IP.
|
---|
| 317 | \item NTRIP has been developed at BKG (together with TU Dortmund).
|
---|
| 318 | \item NTRIP is capable of handling hundreds of data streams simultaneously delivering the data
|
---|
| 319 | to thousands of users.
|
---|
[5627] | 320 | \item NTRIP is world-wide accepted (great success of BKG).
|
---|
[5609] | 321 | \end{itemize}
|
---|
| 322 |
|
---|
| 323 | \end{frame}
|
---|
| 324 |
|
---|
| 325 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 326 |
|
---|
| 327 | \begin{frame}
|
---|
| 328 | \frametitle{NTRIP}
|
---|
| 329 |
|
---|
| 330 | Efficiency of data transfer using NTRIP is achieved thanks to the GNSS Internet Radio /
|
---|
| 331 | IP-Streaming architecture:
|
---|
| 332 |
|
---|
[5610] | 333 | \begin{center}
|
---|
[5609] | 334 | \includegraphics[width=0.7\textwidth,angle=0]{ntrip.png}
|
---|
[5610] | 335 | \end{center}
|
---|
[5609] | 336 |
|
---|
| 337 | \end{frame}
|
---|
| 338 |
|
---|
[5610] | 339 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 340 |
|
---|
| 341 | \begin{frame}
|
---|
[5619] | 342 | \frametitle{NTRIP Users}
|
---|
| 343 |
|
---|
| 344 | \includegraphics[width=0.5\textwidth,angle=0]{numberRegisteredUsers_1.png}
|
---|
| 345 | \includegraphics[width=0.5\textwidth,angle=0]{activeClients_month_1.png}
|
---|
| 346 | \begin{center}
|
---|
| 347 | \includegraphics[width=0.5\textwidth,angle=0]{casterTransfer_1.png}
|
---|
| 348 | \end{center}
|
---|
| 349 |
|
---|
| 350 | \end{frame}
|
---|
| 351 |
|
---|
| 352 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 353 |
|
---|
| 354 | \begin{frame}
|
---|
[5610] | 355 | \frametitle{BKG Ntrip Client (BNC)}
|
---|
| 356 |
|
---|
| 357 | An important reason why NTRIP has been widely accepted is that BKG provided high-quality public
|
---|
| 358 | license software tools for its usage. One of these tools is the so-called \textcolor{blue}{BKG
|
---|
| 359 | Ntrip Client}.
|
---|
| 360 |
|
---|
[5611] | 361 | \begin{itemize}
|
---|
| 362 | \item BNC source consists currently of approximately 50.000 lines of code
|
---|
[5628] | 363 | \item development started 2005 (LM and Georg Weber)
|
---|
[5627] | 364 | \item BNC uses a few third-party pieces of software (e.g. RTCM decoders/encoders)
|
---|
| 365 | \item BNC has a good documentation (thanks Georg Weber)
|
---|
[5611] | 366 | \end{itemize}
|
---|
[5610] | 367 |
|
---|
[5611] | 368 | \begin{block}{BNC is intended to be}
|
---|
| 369 | \begin{itemize}
|
---|
| 370 | \item user-friendly
|
---|
| 371 | \item cross-platform
|
---|
| 372 | \item easily modifiable (by students, GNSS beginners)
|
---|
| 373 | \item useful (at least a little bit ...)
|
---|
| 374 | \end{itemize}
|
---|
| 375 | \end{block}
|
---|
[5610] | 376 |
|
---|
[5612] | 377 | \begin{block}{BNC is not only an NTRIP client \ldots}
|
---|
| 378 | \end{block}
|
---|
[5611] | 379 |
|
---|
| 380 | \end{frame}
|
---|
| 381 |
|
---|
| 382 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 383 |
|
---|
| 384 | \begin{frame}
|
---|
[5629] | 385 | \frametitle{BNC Basic Usage}
|
---|
| 386 | \includegraphics[width=0.6\textwidth,angle=0]{screenshot12.png}
|
---|
| 387 |
|
---|
| 388 | \vspace*{-4cm}
|
---|
| 389 | \hspace*{4cm}
|
---|
| 390 | \includegraphics[width=0.5\textwidth,angle=0]{screenshot24.png}
|
---|
| 391 | \end {frame}
|
---|
| 392 |
|
---|
| 393 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 394 |
|
---|
| 395 | \begin{frame}
|
---|
[5613] | 396 | \frametitle{Data QC in BNC}
|
---|
| 397 | \begin{center}
|
---|
[5614] | 398 | \includegraphics[width=0.9\textwidth,angle=0]{bnc_qc2.png}
|
---|
| 399 | \end{center}
|
---|
| 400 | \end {frame}
|
---|
| 401 |
|
---|
| 402 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 403 |
|
---|
| 404 | \begin{frame}
|
---|
| 405 | \frametitle{Data QC in BNC}
|
---|
| 406 | \begin{center}
|
---|
[5613] | 407 | \includegraphics[width=0.9\textwidth,angle=0]{bnc_qc1.png}
|
---|
| 408 | \end{center}
|
---|
| 409 | \end {frame}
|
---|
| 410 |
|
---|
| 411 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 412 |
|
---|
| 413 | \begin{frame}
|
---|
[5611] | 414 | \frametitle{Precise Point Positioning with PPP}
|
---|
| 415 | \begin{center}
|
---|
| 416 | \includegraphics[width=0.9\textwidth,angle=0]{ppp1.png}
|
---|
| 417 | \end{center}
|
---|
| 418 | \end {frame}
|
---|
| 419 |
|
---|
| 420 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 421 |
|
---|
| 422 | \begin{frame}
|
---|
[5615] | 423 | \frametitle{Principles of Precise Point Positioning}
|
---|
| 424 | \framesubtitle{Observation Equations}
|
---|
| 425 |
|
---|
| 426 | The PPP is based on the processing of the ionosphere-free linear combination of phase
|
---|
| 427 | observations
|
---|
| 428 | \be
|
---|
| 429 | L^{ij}_3 = \varrho^{ij} - c\delta^{ij} + T^{ij} + \bar{N}^{ij}_3 ~,
|
---|
| 430 | \ee
|
---|
| 431 | where the ambiguity term is given by
|
---|
| 432 | \be
|
---|
| 433 | \bar{N}^{ij}_3 = N^{ij}_3 - l^{ij}_3
|
---|
| 434 | = \frac{c\;f_2}{f^2_1-f^2_2}\;(n^{ij}_1-n^{ij}_2) + \lambda_3\;n^{ij}_1 - l^{ij}_3
|
---|
| 435 | \ee
|
---|
| 436 | and (optionally) the ionosphere-free linear combination of code observations
|
---|
| 437 | \be
|
---|
| 438 | P^{ij}_3 = \varrho^{ij} - c\delta^{ij} + T^{ij} + p^{ij}_3 ~,
|
---|
| 439 | \ee
|
---|
| 440 | where the code bias $p^{ij}_3$ is the linear combination of biases
|
---|
| 441 | $p^{ij}_1,p^{ij}_2$
|
---|
| 442 | \end{frame}
|
---|
| 443 |
|
---|
| 444 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 445 |
|
---|
| 446 | \begin{frame}
|
---|
| 447 | \frametitle{Principles of PPP Service}
|
---|
| 448 |
|
---|
[5630] | 449 | The server has to provide the orbit corrections and the satellite clock corrections
|
---|
| 450 | $c\delta^{ij}$. That is sufficient for a client processing phase observations only.
|
---|
[5615] | 451 |
|
---|
| 452 | Using the code observations on the client-side is not mandatory. After an initial convergence
|
---|
| 453 | period (tens of minutes) there is almost no difference between a phase-only client and the client
|
---|
| 454 | that uses also the code observations. However, correct utilization of accurate code observations
|
---|
| 455 | improves the positioning results during the convergence period.
|
---|
| 456 |
|
---|
| 457 | Client which processes code observations either
|
---|
| 458 | \begin{enumerate}
|
---|
| 459 | \item has to know the value $p^{ij}_3$ (the value must be provided by the server -- the most
|
---|
| 460 | correct approach), or
|
---|
| 461 | \item has to estimate terms $p^{ij}_3$, or
|
---|
| 462 | \item neglect the bias (de-weight the code observations -- not fully correct).
|
---|
| 463 | \end{enumerate}
|
---|
| 464 | Options (2) and (3) mean that the benefit of using the code observations on the client-side (in
|
---|
| 465 | addition to phase observations) is minor only.
|
---|
| 466 |
|
---|
| 467 | \end{frame}
|
---|
| 468 |
|
---|
| 469 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 470 |
|
---|
| 471 | \begin{frame}
|
---|
[5630] | 472 | \frametitle{PPP Options in BNC}
|
---|
[5626] | 473 | \begin{itemize}
|
---|
| 474 | \item single station, SPP or PPP
|
---|
| 475 | \item real-time or post-processing
|
---|
| 476 | \item processing of code and phase ionosphere-free combinations, GPS,
|
---|
| 477 | Glonass, and Galileo
|
---|
| 478 | \end{itemize}
|
---|
| 479 | \begin{center}
|
---|
| 480 | \includegraphics[width=0.9\textwidth,angle=0]{ppp_opt1.png} \\[2mm]
|
---|
| 481 | \includegraphics[width=0.9\textwidth,angle=0]{ppp_opt2.png}
|
---|
| 482 | \end{center}
|
---|
| 483 | \end {frame}
|
---|
| 484 |
|
---|
| 485 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 486 |
|
---|
| 487 | \begin{frame}
|
---|
[5616] | 488 | \frametitle{PPP of Moving Receiver by BNC}
|
---|
| 489 | \begin{center}
|
---|
| 490 | \includegraphics[width=0.6\textwidth,angle=0]{screenshot32.png}
|
---|
| 491 | \end{center}
|
---|
| 492 | \end{frame}
|
---|
[5615] | 493 |
|
---|
[5616] | 494 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
[5615] | 495 |
|
---|
[5616] | 496 | \begin{frame}
|
---|
[5618] | 497 | \frametitle{PPP -- Server-Side}
|
---|
[5615] | 498 |
|
---|
[5616] | 499 | \includegraphics[width=0.8\textwidth,angle=0]{igs_map.png}
|
---|
| 500 |
|
---|
| 501 | \vspace*{-2cm}
|
---|
| 502 |
|
---|
| 503 | \hspace*{2cm}
|
---|
| 504 | \includegraphics[width=0.8\textwidth,angle=0]{bnc_rtnet_flow.png}
|
---|
| 505 |
|
---|
[5618] | 506 | \end{frame}
|
---|
[5616] | 507 |
|
---|
[5618] | 508 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 509 |
|
---|
| 510 | \begin{frame}
|
---|
| 511 | \frametitle{PPP -- Server-Side}
|
---|
| 512 | \begin{center}
|
---|
| 513 | \includegraphics[width=0.9\textwidth,angle=0]{bnc_feed.png}
|
---|
| 514 | \end{center}
|
---|
[5615] | 515 | \end{frame}
|
---|
| 516 |
|
---|
| 517 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 518 |
|
---|
[5620] | 519 | \begin{frame}
|
---|
[5631] | 520 | \frametitle{Server-Side -- RTNet (www.gps-solutions.com)}
|
---|
| 521 | \includegraphics[width=0.9\textwidth,angle=0]{GPSS_home.png}
|
---|
| 522 | \end{frame}
|
---|
| 523 |
|
---|
| 524 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 525 |
|
---|
| 526 | \begin{frame}
|
---|
[5632] | 527 | \frametitle{Server-Side -- RTNet (www.gps-solutions.com)}
|
---|
[5633] | 528 | \includegraphics[width=0.9\textwidth,angle=0]{gpss_team.png}
|
---|
| 529 | \end{frame}
|
---|
| 530 |
|
---|
| 531 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 532 |
|
---|
| 533 | \begin{frame}
|
---|
| 534 | \frametitle{Server-Side -- RTNet (www.gps-solutions.com)}
|
---|
[5634] | 535 | \includegraphics[width=0.7\textwidth,angle=0]{rtnet_menu.png}
|
---|
| 536 |
|
---|
| 537 | \vspace*{-3cm}
|
---|
| 538 | \hspace*{4cm}
|
---|
| 539 | \includegraphics[width=0.7\textwidth,angle=0]{rtnet_schema.png}
|
---|
[5632] | 540 | \end{frame}
|
---|
| 541 |
|
---|
| 542 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 543 |
|
---|
| 544 | \begin{frame}
|
---|
[5633] | 545 | \frametitle{Server-Side -- RTNet (www.gps-solutions.com)}
|
---|
| 546 | \includegraphics[width=0.5\textwidth,angle=0]{eq_monitoring.png}
|
---|
| 547 | \includegraphics[width=0.5\textwidth,angle=0]{tsunami.png}
|
---|
| 548 | \begin{center}
|
---|
| 549 | \includegraphics[width=0.4\textwidth,angle=0]{veripos.png}
|
---|
| 550 | \end{center}
|
---|
| 551 | \end{frame}
|
---|
| 552 |
|
---|
| 553 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 554 |
|
---|
| 555 | \begin{frame}
|
---|
[5620] | 556 | \frametitle{PPP -- Server-Side}
|
---|
| 557 | \begin{center}
|
---|
| 558 | \includegraphics[width=0.9\textwidth,angle=0]{ac_results.png}
|
---|
| 559 | \end{center}
|
---|
| 560 | \end{frame}
|
---|
| 561 |
|
---|
| 562 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 563 |
|
---|
| 564 | \begin{frame}
|
---|
| 565 | \frametitle{PPP -- Server-Side}
|
---|
| 566 | \begin{center}
|
---|
| 567 | \includegraphics[width=0.9\textwidth,angle=0]{ac_results2.png}
|
---|
| 568 | \end{center}
|
---|
| 569 | \end{frame}
|
---|
| 570 |
|
---|
| 571 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 572 |
|
---|
[5621] | 573 | \begin{frame}
|
---|
[5623] | 574 | \frametitle{Combination using Kalman filtering}
|
---|
| 575 | The combination is performed in two steps
|
---|
| 576 | \begin{itemize}
|
---|
| 577 | \item[1.] The satellite clock corrections that refer to different broadcast
|
---|
| 578 | messages (different IODs) are modified in such a way that they all refer
|
---|
| 579 | to common broadcast clock value (common IOD is that of the selected
|
---|
| 580 | ``master'' analysis center).
|
---|
| 581 | \item[2.] The corrections are used as pseudo-observations for Kalman filter
|
---|
| 582 | using the following model (observation equation):
|
---|
| 583 | \begin{displaymath}
|
---|
| 584 | c_a^s = c^s + o_a + o_a^s
|
---|
| 585 | \end{displaymath}
|
---|
| 586 | where
|
---|
| 587 | \begin{tabbing}
|
---|
| 588 | $c_a^s$ ~~ \= is the clock correction for satellite s estimated by \\
|
---|
| 589 | \> the analysis center a, \\
|
---|
| 590 | $c^s$ \> is the resulting (combined) clock correction for
|
---|
| 591 | satellite s, \\
|
---|
| 592 | $o_a$ \> is the AC-specific offset
|
---|
| 593 | (common for all satellites), and \\
|
---|
| 594 | $o_a^s$ \> is the satellite and AC-specific offset.
|
---|
| 595 | \end{tabbing}
|
---|
| 596 | \end{itemize}
|
---|
| 597 | The three types of unknown parameters $c^s$, $o_a$, $o_a^s$ differ in their
|
---|
| 598 | stochastic properties: the parameters $c^s$ and $o_a$ are considered to be
|
---|
| 599 | epoch-specific while the satellite and AC-specific offset $o_a^s$ is assumed
|
---|
| 600 | to be a static parameter.
|
---|
| 601 | \end{frame}
|
---|
| 602 |
|
---|
| 603 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 604 |
|
---|
| 605 | \begin{frame}
|
---|
[5634] | 606 | \frametitle{PPP -- Combination of Corrections}
|
---|
[5621] | 607 | \begin{center}
|
---|
| 608 | \includegraphics[width=0.9\textwidth,angle=0]{combination_1.png}
|
---|
| 609 | \end{center}
|
---|
| 610 | \end{frame}
|
---|
| 611 |
|
---|
| 612 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 613 |
|
---|
| 614 | \begin{frame}
|
---|
[5634] | 615 | \frametitle{PPP -- Combination of Corrections}
|
---|
[5621] | 616 | \begin{center}
|
---|
| 617 | \includegraphics[width=0.9\textwidth,angle=0]{combination_2.png}
|
---|
[5635] | 618 |
|
---|
| 619 | \includegraphics[width=0.6\textwidth,angle=0]{dailyRMS_GLONASS.png}
|
---|
[5621] | 620 | \end{center}
|
---|
| 621 | \end{frame}
|
---|
| 622 |
|
---|
| 623 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 624 |
|
---|
| 625 | \begin{frame}
|
---|
[5634] | 626 | \frametitle{PPP -- Combination of Corrections}
|
---|
[5621] | 627 | \begin{center}
|
---|
[5635] | 628 | \includegraphics[width=0.8\textwidth,angle=0]{combination_3.png}
|
---|
[5621] | 629 | \end{center}
|
---|
| 630 | \end{frame}
|
---|
| 631 |
|
---|
| 632 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 633 |
|
---|
[5622] | 634 | \begin{frame}
|
---|
[5634] | 635 | \frametitle{PPP -- Estimated Troposphere}
|
---|
| 636 | \includegraphics[width=0.5\textwidth,angle=0]{tropo1.png}
|
---|
| 637 | \includegraphics[width=0.5\textwidth,angle=0]{tropo2.png}
|
---|
[5622] | 638 |
|
---|
[5634] | 639 | \includegraphics[width=0.5\textwidth,angle=0]{tropo3.png}
|
---|
[5622] | 640 | \end{frame}
|
---|
| 641 |
|
---|
| 642 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 643 |
|
---|
| 644 | \begin{frame}
|
---|
[5623] | 645 | \frametitle{Principle of our PPP-RTK Algorithm}
|
---|
| 646 | For a dual-band GPS receiver, the observation equations may read as
|
---|
| 647 | \begin{eqnarray*}
|
---|
| 648 | P^i & = & \varrho^i + c\;\delta - c\;\delta^i + T^i + b_P \\
|
---|
| 649 | L^i & = & \varrho^i + c\;\delta - c\;\delta^i + T^i + b^i
|
---|
| 650 | \end{eqnarray*}
|
---|
| 651 | where
|
---|
| 652 | \begin{tabbing}
|
---|
| 653 | $P^i$, $L^i$ ~~~~~~~ \= are the ionosphere-free code and phase measurements, \\
|
---|
| 654 | $\varrho^i$ \> is the travel distance between the satellite
|
---|
| 655 | and the receiver, \\
|
---|
| 656 | $\delta$, $\delta^i$ \> are the receiver and satellite clock errors, \\
|
---|
| 657 | $T^i$ \> is the tropospheric delay, \\
|
---|
| 658 | $b_P$ \> is the code bias, and \\
|
---|
| 659 | $b^i$ \> is the phase bias (including initial
|
---|
| 660 | phase ambiguity).
|
---|
| 661 | \end{tabbing}
|
---|
| 662 | The single-difference bias $b^{ij} = b^i - b^j$ is given by
|
---|
| 663 | \begin{displaymath}
|
---|
| 664 | b^{ij} = \displaystyle\frac{\lambda_5-\lambda_3}{2}\;(n_5^{ij} + b_5^{ij})
|
---|
| 665 | + \lambda_3\;(n_1^{ij} + b_1^{ij})
|
---|
| 666 | \end{displaymath}
|
---|
| 667 | where
|
---|
| 668 | \begin{tabbing}
|
---|
| 669 | $n_1^{ij}$, $n_5^{ij}$ ~~~~ \= are the narrow-lane and wide-lane integer ambiguities \\
|
---|
| 670 | $b_1^{ij}$ \> is the narrow-lane (receiver-independent) SD bias \\
|
---|
| 671 | $b_5^{ij}$ \> is the wide-lane (receiver-independent) SD bias
|
---|
| 672 | \end{tabbing}
|
---|
| 673 | \end{frame}
|
---|
[5615] | 674 |
|
---|
[5623] | 675 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 676 |
|
---|
[5615] | 677 | \begin{frame}
|
---|
[5623] | 678 | \frametitle{Principle of our PPP-RTK Algorithm (cont.)}
|
---|
| 679 | Receiver-independent single-difference biases $b_1^{ij}$ and $b_5^{ij}$ have
|
---|
| 680 | to be estimated on the server-side.
|
---|
| 681 | \begin{itemize}
|
---|
| 682 | \item Narrow-lane bias $b_1^{ij}$ may be combined with satellite clock
|
---|
| 683 | corrections $\Longrightarrow$ \textbf{modified satellite clock corrections.}
|
---|
| 684 | \item Wide-lane bias have to be transmitted from the server to the client
|
---|
| 685 | (this bias is stable in time and can thus be transmitted in lower rate).
|
---|
| 686 | \end{itemize}
|
---|
[5615] | 687 |
|
---|
[5623] | 688 | On the client-side the biases $b_1^{ij}$ and $b_5^{ij}$ are used as known
|
---|
| 689 | quantities. It allows fixing the integer ambiguities $n_5^{ij}$ and
|
---|
| 690 | $n_1^{ij}$. The technique is called Precise Point Positioning with Ambiguity
|
---|
| 691 | Resolution (PPP~AR) or PPP~RTK, or zero-difference ambiguity
|
---|
| 692 | fixing (the latter term not fully correct because the ambiguities are
|
---|
| 693 | actually being fixed on single-difference level).
|
---|
[5615] | 694 | \end{frame}
|
---|
| 695 |
|
---|
| 696 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 697 |
|
---|
[5623] | 698 | \begin{frame}
|
---|
| 699 | \frametitle{Performance}
|
---|
| 700 | \begin{center}
|
---|
| 701 | \includegraphics[width=0.75\textwidth]{kir0.png}
|
---|
| 702 | \end{center}
|
---|
| 703 | \vspace*{-5mm}
|
---|
| 704 | \begin{block}{Standard deviations (N,E,U)}
|
---|
| 705 | \vspace*{3mm}
|
---|
| 706 | \begin{small}
|
---|
| 707 | \hspace*{2cm}
|
---|
| 708 | \begin{tabular}{l|ccc|ccc}
|
---|
| 709 | \mbox{} & \multicolumn{3}{c|}{10-60 min} & \multicolumn{3}{c}{30-60 min} \\
|
---|
| 710 | float & 0.034 & 0.026 & 0.026 & 0.010 & 0.009 & 0.011 \\
|
---|
| 711 | fix & 0.007 & 0.003 & 0.016 & 0.007 & 0.003 & 0.012
|
---|
| 712 | \end{tabular}
|
---|
| 713 | \end{small}
|
---|
| 714 | \end{block}
|
---|
| 715 | \end{frame}
|
---|
[5615] | 716 |
|
---|
[5623] | 717 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 718 |
|
---|
[5615] | 719 | \begin{frame}
|
---|
[5623] | 720 | \frametitle{Challenges}
|
---|
| 721 | There are still both principal and technical problems and challenges:
|
---|
| 722 | \begin{itemize}
|
---|
| 723 | \item Principal problems:
|
---|
| 724 | \begin{itemize}
|
---|
| 725 | \item Convergence time: PPP~RTK in the form outlined above provides
|
---|
| 726 | accuracy similar (or even slightly better) to RTK but the convergence
|
---|
| 727 | time is longer.
|
---|
| 728 | \item There is a degradation in accuracy with the age of corrections.
|
---|
| 729 | \item Glonass ambiguity resolution: is it possible to resolve Glonass
|
---|
| 730 | ambiguities? (yes, it is possible but it implicates introducing new
|
---|
| 731 | parameters - does it really improve the results?)
|
---|
| 732 | \item ...
|
---|
| 733 | \end{itemize}
|
---|
| 734 | \item Technical problems:
|
---|
| 735 | \begin{itemize}
|
---|
| 736 | \item Availability of data in real time (reference network, high-precision
|
---|
| 737 | satellite orbits).
|
---|
| 738 | \item Very high CPU requirements on the server-side.
|
---|
| 739 | \item Solution robustness on the server-side
|
---|
| 740 | (problems with reliable DD ambiguity resolution).
|
---|
| 741 | \item ...
|
---|
| 742 | \end{itemize}
|
---|
| 743 | \end{itemize}
|
---|
[5615] | 744 | \end{frame}
|
---|
| 745 |
|
---|
| 746 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 747 |
|
---|
| 748 | \begin{frame}
|
---|
[5623] | 749 | \frametitle{Challenges (cont.)}
|
---|
| 750 | \begin{block}{Longer convergence time}
|
---|
| 751 | In case of a standard RTK the very short convergence time is being achieved
|
---|
| 752 | thanks to the combined DD ambiguity resolution on both $L_1$ and $L_2$ when
|
---|
| 753 | the differential ionospheric bias can either be neglected (short baselines)
|
---|
| 754 | or its influence is mitigated (stochastic ionosphere estimation with
|
---|
| 755 | constraints).
|
---|
[5615] | 756 |
|
---|
[5623] | 757 | On the contrary, the outlined PPP~RTK algorithm is in principle based on
|
---|
| 758 | processing single (ionosphere-free) linear combination and resolving only
|
---|
| 759 | one set of (narrow-lane) initial phase ambiguities.
|
---|
| 760 | \end{block}
|
---|
| 761 | \begin{block}{Possible solutions}
|
---|
| 762 | \begin{itemize}
|
---|
| 763 | \item third carrier
|
---|
| 764 | \item multiple GNSS (Glonass ambiguity resolution?)
|
---|
| 765 | \item processing original carriers (instead of ionosphere-free linear
|
---|
| 766 | combination) and modeling the ionosphere?
|
---|
| 767 | \item ?
|
---|
| 768 | \end{itemize}
|
---|
| 769 | \end{block}
|
---|
| 770 | \end{frame}
|
---|
[5615] | 771 |
|
---|
[5623] | 772 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
[5615] | 773 |
|
---|
[5623] | 774 | \begin{frame}
|
---|
| 775 | \frametitle{Challenges (cont.)}
|
---|
| 776 | \begin{block}{Age of corrections 0 s}
|
---|
| 777 | \begin{center}
|
---|
| 778 | \includegraphics[width=0.6\textwidth]{age1.png}
|
---|
| 779 | \end{center}
|
---|
| 780 | \end{block}
|
---|
| 781 | \end{frame}
|
---|
[5615] | 782 |
|
---|
[5623] | 783 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
[5615] | 784 |
|
---|
[5623] | 785 | \begin{frame}
|
---|
| 786 | \frametitle{Challenges (cont.)}
|
---|
| 787 | \begin{block}{Age of corrections up to 35 s}
|
---|
| 788 | \begin{center}
|
---|
| 789 | \includegraphics[width=0.6\textwidth]{age2.png}
|
---|
| 790 | \end{center}
|
---|
| 791 | \end{block}
|
---|
[5615] | 792 | \end{frame}
|
---|
| 793 |
|
---|
| 794 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 795 |
|
---|
| 796 | \begin{frame}
|
---|
[5623] | 797 | \frametitle{Real-Time Data Availability}
|
---|
| 798 | \framesubtitle{IGS network: very good global coverage:}
|
---|
| 799 | \vspace*{-5.5cm}
|
---|
| 800 | \begin{center}
|
---|
| 801 | \includegraphics[width=0.9\textwidth]{map.pdf}
|
---|
| 802 | \end{center}
|
---|
| 803 | \end{frame}
|
---|
[5615] | 804 |
|
---|
[5623] | 805 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
[5615] | 806 |
|
---|
[5623] | 807 | \begin{frame}
|
---|
| 808 | \frametitle{Real-Time Data Availability (cont.)}
|
---|
| 809 | \begin{tabular}{cc}
|
---|
| 810 | \includegraphics[width=0.4\textwidth]{100A_lat.png} &
|
---|
| 811 | \includegraphics[width=0.4\textwidth]{101A_lat.png} \\
|
---|
| 812 | \includegraphics[width=0.4\textwidth]{102A_lat.png} &
|
---|
| 813 | \includegraphics[width=0.4\textwidth]{104A_lat.png}
|
---|
| 814 | \end{tabular}
|
---|
[5615] | 815 |
|
---|
[5623] | 816 | Gaps in reference network data may degrade the PPP~RTK server performance
|
---|
| 817 | considerably!
|
---|
[5615] | 818 | \end{frame}
|
---|
| 819 |
|
---|
| 820 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 821 |
|
---|
| 822 | \begin{frame}
|
---|
[5623] | 823 | \frametitle{Technical issues}
|
---|
| 824 | \begin{block}{CPU-requirements on the server-side}
|
---|
| 825 | Processing a global reference network is a very CPU-intensive
|
---|
| 826 | task. Numerically stable forms of the Kalman filter (square-root, UDU
|
---|
| 827 | factorization etc.) require very fast hardware.
|
---|
[5615] | 828 |
|
---|
[5623] | 829 | Possible solutions:
|
---|
| 830 | \begin{itemize}
|
---|
| 831 | \item Processing optimization (estimating various kinds of parameters in
|
---|
| 832 | different rates)
|
---|
| 833 | \item Parallel processing
|
---|
| 834 | \item Advanced hardware (GPS Solutions uses GPU-accelerated library)
|
---|
| 835 | \end{itemize}
|
---|
| 836 | \end{block}
|
---|
| 837 | \begin{block}{Reliable DD ambiguity resolution on the server-side}
|
---|
| 838 | Reliable double-difference ambiguity resolution on the server-side remains
|
---|
| 839 | the crucial issue of the PPP~RTK technique.
|
---|
| 840 | \end{block}
|
---|
| 841 | \begin{block}{Dissemination of PPP~RTK corrections}
|
---|
| 842 | \begin{itemize}
|
---|
| 843 | \item data links
|
---|
| 844 | \item formats (standardization?)
|
---|
| 845 | \item optimization of correction rates (bandwidth)
|
---|
| 846 | \end{itemize}
|
---|
| 847 | \end{block}
|
---|
[5615] | 848 | \end{frame}
|
---|
| 849 |
|
---|
| 850 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 851 |
|
---|
| 852 | \begin{frame}
|
---|
[5623] | 853 | \frametitle{Satellite orbits}
|
---|
[5611] | 854 |
|
---|
[5623] | 855 | Predicted part of the IGS ultra-rapid orbits (available in real-time) is
|
---|
| 856 | sometimes not sufficient for the processing of a global reference network
|
---|
| 857 | (with narrow-lane ambiguity resolution). We have been forced to implement
|
---|
| 858 | the real-time orbit determination capability in our main processing tool
|
---|
| 859 | RTNet (Real-Time Network software).
|
---|
| 860 | \begin{center}
|
---|
| 861 | \includegraphics[width=0.75\textwidth]{rtnet_pod.png}
|
---|
| 862 | \end{center}
|
---|
| 863 | \end{frame}
|
---|
[5611] | 864 |
|
---|
[5623] | 865 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 866 |
|
---|
| 867 | \begin{frame}
|
---|
| 868 | \frametitle{Regional versus global PPP~RTK services}
|
---|
| 869 | Currently we are routinely running both regional and global PPP~RTK service
|
---|
| 870 | demonstrators in real-time (some of the results will be shown below).
|
---|
[5611] | 871 | \begin{itemize}
|
---|
[5623] | 872 | \item in principal there is no difference between a global and regional
|
---|
| 873 | service as far as the data processing, algorithms etc. is concerned
|
---|
| 874 | \item global PPP~RTK service has at least the following two advantages
|
---|
| 875 | \begin{itemize}
|
---|
| 876 | \item[1.] a single correction stream can serve all users
|
---|
| 877 | \item[2.] all satellites are tracked permanently (helps ambiguity
|
---|
| 878 | resolution)
|
---|
| 879 | \end{itemize}
|
---|
| 880 | \item global PPP~RTK service is much more challenging (data availability,
|
---|
| 881 | CPU-requirements on the server-side, DD ambiguity resolution on long
|
---|
| 882 | baselines, the highest requirements for the accuracy of the satellite
|
---|
| 883 | orbits)
|
---|
[5611] | 884 | \end{itemize}
|
---|
| 885 |
|
---|
[5623] | 886 | \end{frame}
|
---|
| 887 |
|
---|
[5611] | 888 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 889 |
|
---|
| 890 | \begin{frame}
|
---|
[5623] | 891 | \frametitle{Services monitoring}
|
---|
| 892 | Reliable, production-quality PPP~RTK service requires sophisticated
|
---|
| 893 | monitoring tools.
|
---|
| 894 | \begin{tabular}{cc}
|
---|
| 895 | \includegraphics[width=0.6\textwidth]{monitor1.png} & \\[-1.5cm]
|
---|
| 896 | & \hspace*{-3cm} \includegraphics[width=0.6\textwidth]{monitor2.png}
|
---|
| 897 | \end{tabular}
|
---|
| 898 |
|
---|
| 899 | \end{frame}
|
---|
| 900 |
|
---|
| 901 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 902 |
|
---|
| 903 | \begin{frame}
|
---|
| 904 | \frametitle{Results}
|
---|
[5636] | 905 | \includegraphics[width=0.6\textwidth]{tsunami.pdf}
|
---|
| 906 |
|
---|
| 907 | \vspace*{-5mm}
|
---|
| 908 | \hspace*{4cm}
|
---|
| 909 | \includegraphics[width=0.6\textwidth]{301C_RAR_POS_2014-01-22.png}
|
---|
[5623] | 910 | \end{frame}
|
---|
| 911 |
|
---|
| 912 | %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
---|
| 913 |
|
---|
| 914 | \begin{frame}
|
---|
| 915 | \frametitle{Results (cont.)}
|
---|
| 916 | \begin{center}
|
---|
| 917 | \includegraphics[width=0.9\textwidth]{nrcan.png}
|
---|
| 918 | \end{center}
|
---|
| 919 | \end{frame}
|
---|
| 920 |
|
---|
[5599] | 921 | \end{document}
|
---|