12 Mar 2003
18 Mar 2003)
Markov Chain Model of RADAR Detector
% radar_detector models a Markov-Chain representation of
% a patrol car looking for speeding cars to ticket
% RADAR moves from state to state when bounce-back pulse is
% received. The state transistion is also dependent on the
% probability that the transition will occur.
% The states that are traveled through are an initialization
% stage (Init), two seeking substates where the car is detected
% or not (Acquired, not_Acquired), a waiting state (Wait) 5 seconds
% before the final state and a final state where the
% speed of the car has been determined (Locked).
% A parallel state is used to track the time taken to determine
% the car's speed. A ticket can be given if the speed was determined
% within 8 seconds - to allow the police officer to maintain visual
% contact with car.
% The 5 second is required to correctly determine the car's speed
% due to the number of RADAR pulses (2 to 3) recieved in that time.
% Transition probabilities were selected to mimic a moderately
% successful RADAR. If you wish to see a failure to ticket, try
% using the seed values of 20192, or 598345 in the uniform random
% number block.