Abstract—In slow frequency hopping systems, it is critical

to detect, classify, and mitigate interference, as it degrades the

system performance. Generalized likelihood ratio test algorithm

was proposed for the detection of interference. However, it could

not classify the type of interference. In this paper, we propose

algorithms to classify between follower/stationary interference

and multitone interference/partial-band noise interference. Classification

of interference in various types of modulation schemes

is performed based on a ratio-testing approach, where the ratio

is compared with a predetermined threshold value. It is critical

to set the threshold at an optimal value in order to achieve a good

classification performance. The optimal value and classification

performance are obtained through simulation results.