Continuous processing and servicing of a new incoming session in LTE systems demands optimal tracking of the mobile user. In this letter, an optimal, information-theoretic framework is developed for tracking area update for next-generation LTE cellular systems. Shannon's entropy is used to characterize the location uncertainty of mobile user. The framework captures users' mobility patterns online and performs profile-based paging for optimizing the tracking area update cost. Simulation results demonstrate reductions in both update and paging costs in comparison to existing LTE systems.