Towards a Versatile Problem Diagnosis Infrastructure for LargeWireless Sensor Networks

Abstract

Abstract. In this position paper, we address the issue of durable maintenance of a wireless sensor network, which will be crucial if the vision of large, long-lived sensornets is to become reality. Durable maintenance requires tools for diagnosing and fixing occurring problems, which can range from internode connectivity losses, to time synchronization problems, to software bugs. While there are solutions for fixing problems, an appropriate diagnostic infrastructure is essentially still lacking. We argue that diagnosing a sensornet application requires the ability to dynamically and temporarily extend the application on a selected group of nodes with virtually any functionality. We motivate this claim based on deployment experiences to date and propose a highly nonintrusive solution to dynamically extending a running application on a resource-constrained sensor node.

Debugging Low-Power and Lossy Wireless Networks: A Survey

“…Diagnosis and management of failures by the network itself can be realised via distributed network protocols and application support [53]. There is a trade-off between prolonging the network's lifetime and providing high quality fault management schemes [26].…”

“…Iwanicki et al [53] argue that diagnosis failures requires extra application support on a selected group of nodes. Furthermore, they claim that networks will be run by specialised technicians, and therefore include them in the debugging process.…”

Multi-hop Cluster Hierarchy Maintenance in Wireless Sensor Networks: A Case for Gossip-Based Protocols

“…Moreover, the overlay can be synthesized and maintained without human intervention, thereby minimizing deployment efforts and facilitating unattended operation. For these reasons, a number of WSN application proposals can be founded on a multi-hop cluster hierarchy, for example, object tracking [1], reactive tasking [6], energy-efficient centralized data collection [3], scalable network monitoring [7], and multi-dimensional range querying [5], to name a few. By and large, multi-hop cluster hierarchy is a compelling paradigm for organizing nodes in large WSNs.…”