checking the validity of a signal

[shamma]

I have a two data set with x representing the azimuth position and the y representing the signal strength. I would like to use a formula to make sure that the graph I am getting is similar to a signal where the value is low then it peaks and finally it drops again to the same level. what is the equation to accomplish that mathematically. Also, I want to remove outliers first, what is the most recommend formula to reach that ?

[SIZE=-1]

[/SIZE][SIZE=-1]

[/SIZE]

 

[Ishuda]

For a normal distribution, outliers are generally considered points outside 2-4 standard deviations. A discussion of outliers is presented at
http://www.itl.nist.gov/div898/handbook/prc/section1/prc16.htm
and, as it indicates, outliers may be real data rather than errors is data and they may contain valuable information.

Several techniques exist for 'digging signal out of noise' and techniques differ depending on the type of noise. If there is no noise, then it is a just a matter of normalizing the data. For example the 'signal'
{10, 20, 40, 20, 10} is the same as {1, 2, 4, 2, 1} if you divided the first by 10. One number used for normalization is the division by the square root of the sum of the squares so that the sum of the squares of the resulting data always sums to one (if this is data with a magnitude and phase, for example, normalization is done on the separate components or possibly only on the magnitude (signal strength in your data)). Another normalization used is the division by the sum of the magnitudes so that the sum of the resulting magnitudes sum to one.