midlet + database + webtrac

Midlet - finished.

I was having a problem - where the GPRMC strings would super impose themselves ontop of each other every 10 or so NMEA sentences, so after cleaning that up - we have a clear output via gprs from the phone.

looks something like this:

UUID RMC addedDate
329e18d0-2125-11db-a98b-0800200c9a66 $GPRMC,070014.000,A,3357.7844,S,15103.0108,E,27.94,190.35,020806,,,A*49 8/02/2006 4:59:47 PM

obviously the formatting will be off. basically its a table with the three columns.


this then gets sent to the server via a HTTP.get.

for consistency - the phone as already mentioned - uses the same string as the WebTrac [about which, or lack of - I will talk about in a second]

In order to clean up the string - i had to bastardise the data stream, and ignore a large portion of the stream. Hopefully I'll get some time to re-write the parser.

-

At first I had all this being written to a flat file, which was fine, but then would require some work in order to be appended into a database. Since our database is 4 weeks late, I had to create one with just a table in it, and add all the data to it. this is mainly a data generation/aggregation exercise.

Andrew has had some trouble in delivering the database on time, and I hope that at this Friday's meeting - we will see some fruits of his promised labour.

-

It has also been decided that we will separate the portal, the db, and the plotting/graphing engine so to speak - apart.

this will allow all of us to work on the individual parts and to then combine them bit by bit - in order to achieve the final product.

the graphing engine will be built by Shawn using PHP.

midlet design

i have completed the first instalment of the midlet, not very pretty - but works.

i have the following working now:

- device discovery [used to find possible GPS devices around {this is done by looking for devices that support a serial connection}]
- reading all NMEA sentences from the GPS unit. was quite nice to see them all for the first time.
- parse to get the GPRMC sentence out, as this is the same sentence used by WebTrac-4, and in order to keep things consistent - thought we'd use the same one.

main part for this week after all the other assessments was to utilize GPRS to send the data back to the server.

whilst we have no database that i can put the data in. i built a small listener/parser on the server, to which the data is to be submitted via HTTP.GET
this then gets stored into a flat file [for now]

so - GPRS connection is now being made by the applet and GPS sentences successfully sent to the server for storage.

unfortunately each time i tested the code - i incurred a $1 fee, as part of vodafones wonderful pricing structure.

so i had to make sure that the code worked prior to running it. adds a little edge to debugging - as everytime you run your program - you're a dollar short :)

this weekend, the plan is to sign up to vodafones data bundle - which is to cost 29.95AUD per month, and it will include 100MB of data.

by my calculations - this should allow us to send:
104857600 characters via gprs in a month.

if we send 80 every second, this equates to:
13107seconds
21845minutes
910hours
37.9 days ----> which is GREAT

[hope my maths was right]

Another Google Map Test Page

Spent about 2 hrs working through the Google Map API's building up the "standard" controls that we will need on the Map.

Will need to add the following

1. Top Right Nav/Zoom control, map.addControl(new GSmallMapControl());
   
2. Map Type, Map/Satellite/Hybrid, map.addControl(new GMapTypeControl());
   
3. Overview Map. map.addControl(new GOverviewMapControl ());

I added a point and a popup for it, next I want to update this points location no a time basis.

Currently the page looks like.



Also been watching a series of video's on using Visual Web Developer 2005 Express Edition

Shawn

A Google Maps .NET Control

Roman pointed me to a article on a Google Maps .NET Control, it's located at

http://www.codeproject.com/aspnet/LatLaysFlat-Part1.asp

and is part of a 4 part series.

I have spent about 3 hrs going through this and have downloaded the code and started working with it.

Going to spend some more time working through Google Map API's so I can see how the .NET control ties into it.

Shawn

Friday, 21st July - Meeting

Today is the first meeting we've had as a group.

After a short discussion with Shawn, it became evident that both of us are a little scared of the remaining time and the amount of work that needs to be completed.

So in order to alleviate any stress, we've decided to set milestones for each person for every week.

By the end of next week, here's what is to be completed.

Andrew:
- Remaining documentation from last semester.
- Understand how to setup a DB within SQL Server

Shawn:
- Utilize the .NET toolkit for GoogleMaps

Roman:
- Complete the [GPS + Phone] MIDlet

Phone + Java progress

I have returned with a phone which is the Nokia 6230.

I’ve played around with Ben Hui’s Midlet framework, and have commenced to design something with will relay GPS co-ordinates back to the server.


So far I see it having the following steps:

1. Allow the input and storage of server settings and account information
2. Provide choice of near-by GPS devices
3. Verify Serial Data as being NMEA
4. Connect to server
5. Commence data relay
6. Provide options for allowing the midlet to be active or passive