Model updating method for protecting privacy and resisting abnormal data in mobile crowd sensing

The invention provides a privacy protection and abnormal data resistance model updating method in mobile crowd sensing. The method comprises a system initialization stage, a worker selection stage, a data encryption stage, a data aggregation stage and a data decryption stage. According to the method, abnormal model parameters can be filtered out and only normal model parameters are aggregated on the premise of protecting the privacy of local model parameters of workers, so that an accurate aggregation result is obtained, the training efficiency of a global model is improved, the training turns are reduced, and high-quality service in mobile crowd sensing is realized. According to the method, the abnormal data can be filtered in the data aggregation process without an additional process, so that the complexity of data processing is reduced. On the basis of the existing worker selection scheme, the introduction of the dimension conversion technology can greatly reduce the calculation overhead required for selec