XonGrid is a XLL Excel Add-in written in C and C++. It provides 5 Excel functions to perform 1D, 2D, 3D, 4D and any dimension (ND) interpolation from scattered data.
Update: Available for 32 and 64 bits Excel versions

## Screenshots!

1D Interpolation Input

Interpolation Example

3D Interpolation Input

1D Example

2D Example

{Array Formula} Example

What is an {Array Formula}?

# Compile XonGrid from Source with XLW and Code::Blocks

1. Download and install MinGW according these XLW instructions (see MinGW Section)
2. Install the Code::Blocks version which NOT includes MinGW
5. Open the XonGrid.cbp project file

Note: Respect the order of steps 1, 2 and 3 or read this

# Use the XonGrid.xll Excel Add-in

The Add-in provides natives Excel function for multi-dimension interpolation. To use the interpolation functions, click on the fx button on the formula bar, select the XonGrid Add-in v0.4 category, select a function then enter your data. Examples of usage are included with the XonGrid archive.

# XonGrid Interpolation Methods

Available interpolation methods are presented on the table below. The Kriging interpolation is highly recommended (with Method = 1.5).

 Function Name Number of Parameters Weighting of Parameters Linear Interpolation Nearest Point Interpolation Kriging Interpolation Interp1d(…) 1 Not Applicable Method = -1 Method = 0 1 ≤ Method ≤ 1.99 Interp2d(…) 2 Weighting = 0 or 1 Not Available Interp3d(…) 3 Interp4d(…) 4 InterpNd(…) Any

## 1D Linear Interpolation (Method = -1)

XonGrid considers a linear variation of the function between known values.

## Weighting of Parameters

In an Euclidean space, the distance between 2 points A and B is given by:

This technique is used if Weighting = 0

If the coordinates are non-homogeneous physical parameters (e.g. Volts and kg), calculate the distance without weighting has no meaning. To avoid this, XonGrid can calculate the distance with normalised parameters.

Normalised space

Then the distance in the normalised space is:

This technique is used if Weighting = 1

## Nearest Point Interpolation (Method = 0)

XonGrid returns the function value of the closest point.

## Kriging interpolation (1 ≤ Method ≤ 1.99)

XonGrid uses Ordinary Kriging interpolation with the power variogram presented below.

with 1 ≤ B ≤ 1.99 (B = 1.5 is a good choice)

The resulting linear system is solved with the Crout LU decomposition with pivoting method. The number of point in the data set is limited to 1024 due to Crout LU solver limitation.

# FAQ

• How much XonGrid cost ?
• XonGrid is a free software. You are free to use, share and modify it (see license for details)
• Interpolation is slow !
• If you are performing multiple interpolation on the same data set, you should try array formula as shown on examples files.