Using middleware as a certifying authority in LBS applications

TLDR: The authors proposed an alternative solution which helps in avoiding a bottleneck in the existing system in terms of performance and availability as the entire client's service transactions are routed through the middleware to the actual Location Based Service Providers (LSP).

Abstract: The trusted middleware is the most commonly used solution to address the location privacy in location based services as generally such service providers are un-trusted entities that can be adversary attack sensitive points. The authors proposed an alternative solution which helps in avoiding a bottleneck in the existing system in terms of performance and availability as the entire client's service transactions are routed through the middleware to the actual Location Based Service Providers (LSP). In the proposed solution, the client and the LSPs can directly communicate with the same level of location security, privacy and anonymity. The trusted middleware is used as certifying authority that generates authentication certificates which contains the Proxy Identity (also called Pseudonyms), and the services subscribed with validity period. The encrypted certificate fulfills the authentication requirements at the LSP servers. In this paper we are reporting the implementation of the proposed system as a proof of concept using Struts Technology of Java. While evaluating the system features such as response time, delay, drop rate etc., the Google Map's location services and the internet browser have been considered as a service provider and client respectively. Performance analysis of our solution and that of prevalent architecture is done using Packmime model for http traffic generation of NS2 (Network Simulator 2) tool. The comparative graphs of the simulation results show that the proposed solution is better in terms of throughput, response time, drop rate and scalability in comparison to the existing middleware architectures in which the request response is every time routed through middleware, thus increasing the overheads.