Security of Geographical Information Systems (GISs) received only limited attention so far. At present, many GIS solutions do not offer any data protection, authenticity and role based data access. A GIS user can be easily mislead by the attacker providing altered data. A GIS user may use provided data against usage policy by falsely claiming ownership to the GIS data. Last but not least, such policy may require different resolutions to be available to different users. This work will explore feasibility of bringing security in GIS information distribution system by exploring the following:
1. Presently, data/image that is delivered to user agencies has no mechanism of non-repudiation in case of conflicting claims between data provider and user agency. It may happen that a critical mission carried out based on an inaccurate data provided by source agency fails. In such a scenario, it is necessary to have mechanism for non-repudiation to avoid conflicting claims by both the agencies. To solve this problem the student shall study and suggest suitable mechanism(s) based on existing cryptographic hash and signature.
2. It may happen that user agency may use data provided in a way that violates the usage policy of the provider. In that scenario provider agency would need to establish that the data used by user agency was actually provided by them. To do that, the provider agency would need to embed a watermark in its data and in case of conflict it can establish ownership of data by recovering the watermark from the image. To make the mechanism robust against malicious alterations, the watermark insertion and detection would be based on secret key known only to the provider. The student will propose a suitable solution in this area.
3. In military and security agencies, data provided is based on 'need to know' basis. Therefore, a resolution access control mechanism is proposed where each user can access different resolution for different user categories from the same image file. This is necessary as having multiple images for multiple user roles would increase the storage and communication costs. The solution of the student must enable users to access three levels of resolution based on a simple transformation, single key and transformation or two keys and transformation of the same image file. The mechanism will be robust in the sense that knowledge of the secret keys will be necessary for recovering the high-resolution details. The mechanism will also be correct in the sense that the higher resolution will be based on actual high-resolution data present in original image and not be based on enhancement techniques using image model priors.

Note: The usage of these mechanisms by actual GIS applications is outside the scope of the proposed thesis.