A Cross Layer Autonomic Enabled Architecture for VANETs

Similar to other ad hoc networks, VANETs need to change and adapt to their surrounding environment. Having various internal and external facilities to change network structure and parameters, VANETS are a good target for self-adaptive and self-managing systems. In order to achieve an autonomic system, selfmanagement should be implemented in the system architecture. To address the autonomic management issues there is a need for a holistic architecture that works without considering predefined constraints and problem conditions. In this paper we present an autonomic enabled architecture using new component which enables cross layer connections. In this architecture we have customized elements of MAPE-K loop, namely Touchpoint and Knowledge, to enhance its applicability in vehicular ad hoc networks. By customizations in the concept of Touchpoint, our presented architecture supports the extraction of data from the cross layer connections between non-neighbors layers, In addition to the layers and their parameters. Furthermore we have classified the knowledge elements of MAPE-K loop in VANETs and described the role and usage of these elements. The applicability of this architecture is verified through simulation of two scenarios by in-team developed simulator. Simulation results show the applicability of the proposed architecture for VANETs in different situations.