Finally an update!

After a busy couple of weeks finally an update!

Two weeks ago our MTS310 sensor boards arrived, naturally we immediately connected one of them to a mote to see what would happen.

Unfortunately everything we were running at the time was in TinyOS 1, this meant that measuring sporadically did not work. After careful consideration and deliberation we decided that it was better for us to start working on an implementation in TinyOS-2.1.0 and risk delays in order to lay a proper foundation for the future. TinyOS-2 is a lot more reliable, but unfortunately implementations for the magnet sensor, the mesh network and even getting the data from our base station onto a PC are not yet available. Another issue is the lack of documentation and programming examples. This meant that we had to start all over with little to no outside help.

Since then we've been working hard on getting everything up and running.

One of the first orders of business was to create a very simple program that could help us measure the range our motes can send over. This program basically uses the LEDS as a binary counter and increases the count when a radio message is received. Using this program we tested the range on our MICAz motes, indoor we managed a distance of 95m but outside we got as far as 120m which exceeded all expectations we had. After the distance test, we also wanted to get an idea of how the signal was affected when two motes are on separate floors. After not seeing any problems with one mote on the third floor and the second mote on the second floor we even saw a strong signal go through two floors with one mote on the third and one mote on the first floor.

The next step was to configure the magnet sensors and create a program that would read out these sensors at a set interval and transmit this data over a wireless network. As there was little documentation available this took longer than expected, and we have since been behind on our original schedule.

Currently we are working on getting the data from the base station to the PC via a serial port forwarder that we are developing in C#. Getting the data on screen has since been realized.

The next step which we are currently working on is to format this raw data into neat packages so we can start really testing the magnet sensor and subsequently migrate this into a database.

Human Technology

Hello, my name is Erwin. I'm also part of the projectgroup Smart Movement but my role within the project is different. I'm a Human Technology student and I've got a different assignment. Oh, and the boy on the left would be me..

The project 'Smart Traffic Streams' contains two assignments: My colleagues are interested in making a proof of concept by testing the sensors (see previous post). I am interested in what you can do with surveyed traffic. In other words i'm investigating what applications can be develloped. The applications that I'll come up with are based on user needs and interests from third parties. I'm planning on interviewing these parties to see what they want. When that is done I can start generating idea's for applications. That's my specialty: Human Technology.

Concept Smart Traffic Monitoring

This is the first version of our plan for the proof of concept demonstration on the parking lot behind ZernikePlein 11 in Groningen. This plan is based on the worst case scenario, it will be updated when tests confirm our wireless motes have more range and the magnetic sensors are able to determine in which direction traffic is moving.

Wireless Sensor Nodes

Currently our goal is to use Crossbow's MICAz wireless module combined with a MTS310 sensor board which features a 2 axis magnetic field sensor to create sensor motes.
These sensor motes are going to detect traffic and relay information via a wireless mesh network to a central server

Crossbow MICAz 2.4 GHZ Wireless Module

Crossbow MTS310 SensorBoard

A rough start

Our project didn't have the best start. Because of this haven't made a lot of progress. However, now we can finally start working.

The biggest problem was that the sensorboard we want to use can't be delivered very fast. A possible solution for this was to use SunSPOT's instead of Crossbow nodes. However this would mean we had to start all over with making plans, and maybe lose more time.

After having discussed the problem with all the parties involved we have finally come to a decision. We will use Crossbow nodes with the MTS310 sensorboard. In the time it takes we have to wait for the sensorboard to arrive we can do a lot of preperation work.

After being a bit demotivated the last couple of day's, we now have a very good feeling about the project. Now we have a good supervisor, a clear goal and support from all the parties involved. We can finally get started.

This offcourse means you, the reader, can expect a lot more updates about the progress we make.

First blogpost from Smart Movement

Hello, and welcome to the blog of Smart Movement...

Smart Movement is a project group responsible for a very cool CENSI project. What CENSI is all about I will describe shortly, first the project. The project is called Slimme Verkeersstromen, which is Dutch for smart traffic streams. A bit cryptic (I know), I will elaborate.

The goal of the project is to measure traffic flow using a network of smart sensors. These measurements can be used to provide a valuable service. For example provide providing people with information about traffic congestion or warning drivers when it isn't safe to overtake.

The project is still in its infancy, we are still getting started. We will use this blog whenever we have something interesting to tell you. Things we will keep you informed about are (among others):

• Our progress
• Random cool stuff
• Random things we've found out

As said before, this is a CENSI project. CENSI stands for Center of Excellence for Intelligent Sensor Innovation. As the strange acronym implies, CENSI is about intelligent sensors. All kinds of diverse projects are carries out here. You can read more about it on the CENSI blog

So please check back often, or better yet, subscribe to the Atom feed.