The application relates to the field of Internet of Vehicles
edge computing, and discloses a high-performance real-
time information processing system supporting Internet of Vehicles streaming computing, which comprises a vehicle-mounted terminal, a source node and a target node. The vehicle-mounted terminal generates a split mark containing a motion trajectory signature when entering an overlapping coverage area and appends the split mark to a service
data stream for transmission. The source node extracts an accumulated state according to the split mark, generates state algebra fragments and state gradient information, and encapsulates the state algebra fragments and the state gradient information into a
single frame control packet consistent with the inner five-tuple of the service
data stream. The target node card compresses the service data and the control packet into the same hardware receiving
queue through hashing. A computing engine performs
verification in combination with the motion trajectory signature, and performs
delay compensation according to the state gradient information. The five-tuple is modified to uniformly distribute the hardware
queue to avoid memory lock competition, the
replay attack is blocked in combination with the physical motion characteristics of the vehicle, and the
transmission time delay is corrected by using the gradient compensation, so that the safety and accuracy of the state
handover between nodes are guaranteed.