Modeling QoS in wireless monitoring systems as Markov chains

Wireless sensor networks (WSNs) are nowadays employed in several applications that require real-time event monitoring (telemedicine, remote environmental monitoring, industrial automation). This has been enabled by the availability of sensors that have become smaller, cheaper, and intelligent. Most of the development efforts in WSN design were focused on energy efficiency, dynamic self-organization and mobility of the nodes (sensors). Providing Quality of Services (QoS) support in WSNs is an emerging area of research due to resource constraints like processing capability, memory, bandwidth and power sources. In this paper, we provide a system configuration of end-to-end QoS in an hydrometric level sensor network comprising local sensors and wireless back-haul network technologies. In particular, we focus on adaptive bit-rate and sampling interval and information prioritization using a first-order Markov chain (FOMC) to model such aspects. We also report performances achieved through rated bandwidth allocation for the sensor data flow across the WSN.