Anti-censorship techniques are important as the ability to freely access information is key to a democratic society. This issue has become even more critical in the information age. In this paper, we present an anti-censorship scheme based on the use of moving target defense implemented with Mobile IPv6. Our approach leverages the participation of content providers (the servers) and relies on unmodified standard Mobile IPv6 protocol. We utilize multiple care-of-addresses to assign each user to access groups and rotate them in a random shuffing that occur at some parametized time interval. We present the various aspects of this approach (such as blocking probability) and reason about its efficacy using an analytical (probabilistic) model. We also present the results of our initial prototype implementation, which was used to look at practical parameters and to understand the implementation overhead. Our analysis shows that it would require a tremendous amount of effort and resources for the adversary to overcome our scheme and censor the end-users.