Restoration scheme of mobility databases by mobility learning andprediction in PCS networks

摘要

This paper proposes a restoration scheme based on mobilitynlearning and prediction in the presence of the failure of mobilityndatabases in personal communication systems (PCSs). In PCSs, mobilityndatabases must maintain the current location information of users tonprovide a fast connection for them. However, the malfunction of mobilityndatabases may cause some location information to be lost. As a result,nwithout an explicit restoration procedure, incoming calls to users maynbe rejected. Therefore, an explicit restoration scheme against thenmalfunction of mobility databases is needed to guarantee continuousnservice availability to users. Introducing mobility learning andnprediction into the restoration process allows systems to locate usersnafter a failure of mobility databases. In failure-free operations, thenmovement patterns of users are learned by a neuro-fuzzy inference systemn(NFIS). After a failure, an inference process of the NFIS is initiatednand the users' future location is predicted. This is used to locate lostnusers after a failure. This proposal differs from previous approachesnusing a checkpoint because it does not need a backup process nornadditional storage space to store checkpoint information. In addition,nsimulations show that our proposal can reduce the cost needed to restorenthe location records of lost users after a failure when compared to thencheckpointing scheme