I'm looking for a better way of doing origin-destination surveys to gather data for traffic modelling. We have in the past used video and numberplate recognition software and also surveys where we stop a random sample of vehicles and give them a card to hand to the people manning the next cordon. None of these methods are very reliable and generally involve huge manpower and delay to motorists.

I ideally would like to post out say 10,000 passive tags that people can choose to stick to their windscreen or put on their dash and then we would set up a whole lot of readers in a few cordons around a city centre for example. The readers would either need to have a range of 10m or so and be able to tell which way the vehicles were travelling or else a range of 5m-6m and not need to tell as we would put one on each side of the road.

I assume the cheapest way would be to have the data dropped straight into a smartphone and the reader be of a type that doesn't actually store any data but just identifies the tag.

I'm doing plenty of reading up on the subject but finding it pretty had going so if anyone can even point me to a relevant website (yes I have done a google search and am dredging through websites at the moment) that would be most appreciated. I'm not an electronics geek unfortunately but a mere civil engineer.

As an aside, I'm guessing that there will be plenty of people concerned about privacy but the way I see it there is no difference between this and the numberplate method. In fact in every project I have been involved in we standardise the data by making all o's into 0's, B's into 8's and the like straight away to eliminate one potential source of errors so immediately lose a fair chunk the identification potential. The data is never used to identify people as such but merely to provide an identifier to tell us where the car goes.

Cheers,
Chris