The invention discloses an on-
chip network
hardware Trojan horse detection platform based on
machine learning, and belongs to the technical field of calculation, reckoning or counting. A
machine learning-based network-on-
chip hardware Trojan horse detection platform is established by constructing a security detection module comprising a traffic
feature tracking extraction module, a
feature extraction register module, a change point detection module and a
random forest Trojan horse detection module. The flow characteristic tracking and extracting module is used for converting flow data in a network-on-
chip into a network-on-chip
flow time sequence model by analyzing network-on-chip flow characteristics. And the
feature extraction result register module stores the
feature data. The change point detection module extracts
change points and abnormal points in the on-chip network traffic
time sequence model through a Bayesian change
point method, and is used for separating normal data and change
point data. And the
random forest detection module accurately identifies network-on-chip abnormal behaviors by learning complex and mutually associated network-on-chip traffic characteristics.