Archive for May, 2008

Ionospheric Error Analyst – a nav mashup Part 2

May 28th, 2008 | Category: Navigation Accuracy Library

Technorati Tags: ,,,,

I've started creating a nav mashup that takes several technologies and data from different sources to determine the near real time navigation error across CONUS. We left off just before I showed you how to create a custom error model to model the receiver noise. Let's get busy.

First, we need to derive a new noise model class from the GPSReceiverNoiseModel. Here's the code, I'll explain what it does below.

public class IonosphericNoiseModel : GpsReceiverNoiseModel
        public IonosphericNoiseModel(StreamReader IonoTecFile)
            m_tecFile = NoaaVerticalSlantTecFile.ReadFrom(IonoTecFile);
            m_GpsL1Frequency = 1575.5e6;
	NoaaVerticalSlantTecFile m_tecFile;

The IonosphericNoiseModel constructor takes a StreamReader that reads in a text file containing the Total Electron Count (TEC) data for a given time, and sets the GPS L1 frequency.

Once I have a TEC file, I need to override the GetTotalNoiseScalar method defined in the GPSReceiverNoiseModel class, to provide my own noise values. This is the meat of the functionality.

Read more

1 comment

AGI Components release 3 – Evaluators on Fire!

May 06th, 2008 | Category: Navigation Accuracy Library

So, it's the beginning of the month, that means a new AGI components release is here! You can get the release from the ADN here: I've profiled the performance of navigation accuracy calculations over a grid - usually the entire Earth - at different levels of granularity. I've started plotting the run times for these tests and wanted to show off the improvements in calculation speed. Kevin's DGL blog shows the how of evaluators and the changes that you need to be aware of, I'm going to cover the results from a nav point of view.

Here's the performance chart:

Release Performance R1-R3

The first two releases of AGI Components are show in red and blue. Release 3 is shown in the magenta line - the one with the really small slope. Release 3 shows about a 4400% improvement in calculation speed over R1 and R2.

Read more

No comments