From brideout at haystack.mit.edu  Mon Sep 20 17:15:16 2004
From: brideout at haystack.mit.edu (William Rideout)
Date: Mon Sep 20 17:15:19 2004
Subject: [gps-developers] Next GPS processing improvement - the mapping
	function
Message-ID: <414F4864.1000308@haystack.mit.edu>

Now that the fully automated GPS processing method that get biases from minimum 
scalloping and least squares fitting is up and running well, we're now in a 
position to attack systematically the last big issue in GPS processing - the 
mapping function.  The mapping function is the ratio between the measured tec at 
a given elevation, and the vertical tec.

Now that receiver bias is well in hand, the success of the mapping function can 
be determined by a simple figure-of-merit.  This figure of merit is the average 
difference between two coincident measurements (after the best fit receiver bias 
has been accounted for).  Since the receiverBias.py module already has code to 
find coincident measurements (that is, measurements from two receivers where the 
pierce points are within X kilometers), writing code to determine this figure of 
merit after all other processing is complete will be easy.

Anthea has suggested that we also define a latitude-binned figure of merit; that 
is, the figure of merit defined above but limited to coincidences within a given 
range of latitude.  The point would be that the best mapping function might be 
latitude dependent.

So the basic procedure would be to process the same day with a series of mapping 
functions, and determine the figures of merit for each.

Let me know if you have any suggestions.

Bill

-- 
Bill Rideout
MIT Haystack Observatory
Email: brideout@haystack.mit.edu
Phone: 781 981-5624