SECA: Snapshot-based Event Detection for Checking Asynchronous Context Consistency in Ubiquitous Computing (1305.3105v1)

Abstract: Context-consistency checking is challenging in the dynamic and uncertain ubiquitous computing environments. This is because contexts are often noisy owing to unreliable sensing data streams, inaccurate data measurement, fragile connectivity and resource constraints. One of the state-of-the-art efforts is CEDA, which concurrently detects context consistency by exploring the \emph{happened-before} relation among events. However, CEDA is seriously limited by several side effects --- centralized detection manner that easily gets down the checker process, heavy computing complexity and false negative. In this paper, we propose SECA: Snapshot-based Event Detection for Checking Asynchronous Context Consistency in ubiquitous computing. SECA introduces snapshot-based timestamp to check event relations, which can detect scenarios where CEDA fails. Moreover, it simplifies the logical clock instead of adopting the vector clock, and thus significantly reduces both time and space complexity. Empirical studies show that SECA outperforms CEDA in terms of detection accuracy, scalability, and computing complexity.