This thesis presents a novel wireless MAC-layer approach towards achieving channel access anonymity by preventing linkability and traffic analysis. Nodes autonomously select periodic TDMA-like time-slots for channel access by employing a channel sensing strategy, and they do so without explicitly sharing any identity information with other nodes in the network, thus preventing linkability. Without any message-based coordination, entire packet contents, including all addressing information, are hidden from other nodes and TDMA slot-allocation is achieved using blindfolded packet transmissions. The main idea behind this approach is to prevent identity exposure using encrypted packets, and thwart traffic analysis by taking advantage of the inherent periodic traffic pattern of TDMA protocols. The main contributions of this thesis are as follows. First, it presents ZEA-TDMA, a MAC protocol from the TDMA family, which is able to fulfill the aforementioned requirements. Additionally, an energy-aware approach is presented which can have applications in resource constrained sensor networks. Simulation-based experimental results have been presented to evaluate the functionality and performance of the proposed mechanisms. An analytical model has been developed to study the energy consumption of the proposed mechanism. In addition, a hardware module for wireless collision detection has also been developed. Finally, the thesis concludes with the system prototype implementation of the protocol and wireless network test-bed performance results, which demonstrate the functional feasibility of the developed concepts in this thesis.
Read
