Algorithm 2: Combined Feature Space SVM

Input: Preprocessed radar data

Output: Log-likelihood ratio (LLR) for each target

TRAINING:

for i:M aspects do

Calculate distribution function for each target: Fn(x)

Calculate K–S statistic: Dn,m

end for

Determine number of features to use

Concatenate features into feature vector: x

Randomly generate training set and validation set

Perform training of SVM: wi, bi

Sort aspects from highest value to lowest value

Determine LLR for each target based on validation set

TESTING:

Determine aspects required for data collection: N

for i:N aspects do

for j:P testPoints do

Make radar measurement

Preprocess data: x

end for

end for

Perform classification: h(x)

Lookup LLR for each target