InterimReportRomanMuntean.rtf

{\rtf1\ansi\ansicpg1252\deff0\nouicompat\deflang2057{\fonttbl{\f0\fswiss\fprq2\fcharset0 Calibri;}{\f1\fnil\fcharset0 Calibri;}}
{\colortbl ;\red0\green0\blue0;\red255\green255\blue255;}
{\*\generator Riched20 10.0.17134}\viewkind4\uc1 
\pard\widctlpar\qc\b\f0\fs24\lang1033\par
Interim Report\par
CS440 Final Year Project\par
2018 \endash  2019\par
\par
\par
\b0 Roman Muntean\par
15566617\par
\par
Supervisor: Dr Stephen Brown\par
\par
\ul\f1\fs22\lang9 Location Tracking  with Dead-Reckoning\ulnone\f0\fs24\lang1033\par
\par
\par

\pard\widctlpar\qj\ul\b Goals of this project\par
\par
\ulnone\b0 The primary goal of this project is to set up a fully functional GPS tracking environment and use it to test methods of improving the tracking precision using dead-reckoning calculations.\b\par
\par

\pard\widctlpar\cf1\highlight2\ul\lang6153 Problems encountered and progress so far\par
\par
\cf0\highlight0\ulnone\b0\f1 The initial challenge of the project was to program a functioning GPS node and establish communication with the Sigfox backend. The tracking node in use is a SiPy by Pycom on a Pytrack GPS board which is programmable using MicroPython. Having no experience with Python first steps were to install and practice with the Pymakr plugin for Atom to run code on the SiPy. Using the online Pycom documentation and git repository examples, the Sipy was programmed to receive and print the GPS coordinates of the unit to a connected computer. The only real issues at this stage were: Seperating failures due to code and issues with signal form each other \par
and an initial difficulty in understanding the correct folder structure for python import commands.\par
\par
The next step in the project was to establish communication between the unit and the middleman: The Sigfox backend. The SiPy is a multi-network (Sigfox, WiFi and BLE) development platform and the available documentation covered sending messages from the unit to the Sigfox back-end over the Sigfox network. For initial tests the message data consisted of arbitary short strings.\par
\par
Upon receiving access to an established server from the unviersity, PHP and sigfox callbacks were the predetermined method of passing on messages from the Sigfox backend to the server. This was a difficult stage due to the number of technologies that would have to be implemented from some theoretical familiarity into practice.\par
The sigfox callback was implemented as a GET request to the server with supervisor help and the API documentation. PHP scripting took some initial practice due to a lack of previous experience and the files were then placed onto the server using filezilla. PuTTY was installed for command line interaction with the server from Windows. The PHP script at this stage simply writes the message data to a server text file. Due to there being no external write permission on the server, the messages were not recorded initially which caused confusion in testing, but this was shortly fixed.\par
\b\f0\lang1033\par

\pard\widctlpar\qj\ul Next steps\par
\ulnone\par
\b0 The current project stage is to send full GPS coordinates as the Sigfox message data. Due the a 12 Byte limitation on the message data length, this poses the technical challenge of compressing the message length on the unit and unpacking the bits on the server end to re-create the original message.\par
\par
Next instead of writing the messages to a text file a database infrastructure will need to be impelmented to handle the data. The currently planned tool for this is MongoDB.\par
\par
For the final part of the establishing the main technical infrastructure of the project, a user front-end for dispalying GPS location of the unit is planned. Once this is complete testing to improve accuracy using dead-reckoning will take place.\ul\b\par
\par
General comments\par
\ulnone\b0\par

\pard\sa200\sl276\slmult1 At this midway stage of the project a lot of progress has already been made in realizing the ideas and goals established at the outset. A lot of experience has been gained with using a suite of tools and languages and out of necessity, topics where none or only a shallow understanding existed have been fleshed out and improved through practice, research and discussion. By the time the project is complete, goal is to not only have improved my personal understanding but to add something new to the field.\par
}