Novel data pollution attack defense method oriented to Internet of Things system

The invention provides a novel data pollution attack defense method oriented to an Internet of Things system, and the method comprises the following steps: after a server collects updates of all terminals trained in the current round, randomly grouping the updates to form a secondary result; for each secondary result, randomly selecting a terminal from a set of all terminals according to a preset rule and authorizing the terminal, and detecting and evaluating the received secondary result by each terminal; enabling the server to collect an evaluation report of the terminal on the secondary result, and counting the number of times that a certain update is judged to have potential maliciousness; enabling the server to adjust the weight, including adding a penalty coefficient to each update based on the number of times that each update is judged to be malicious; taking a penalty coefficient as a weight to calculate an average value, and aggregating secondary results. In order to resist data pollution attacks pos