Experimental Study: Link-QualityEstimations in a City-DeployedWireless Sensor Network

Abstract: As the capabilities and scalability of the Internet grow largerand become more advanced, new opportunities arise that enableusers to make use of this enhanced manner in which to distributedata. One particular example of this is Wireless Sensor Networks -networks consisting of small sensors with wireless communicationcapabilities, which can monitor and record environmental phenomenawith the intention of forwarding the obtained information toindividual clients or centralized servers for data mining. Thesetypes of networks are often deployed in relatively hazardousenvironments, which may challenge the efficiency and performanceof the wireless connectivity between nodes. A joint effort between multiple partners, called the GreenIoTproject, aims to deploy a Wireless Sensor Network in Uppsala,Sweden. The partners of the project seek to learn more about thelevel of performance that can be expected by the network invarious scenarios, such as when the distance between thetransmitting nodes is increased and when the line of sight betweenthem is disrupted. The insights gained from this research are tobe used for optimal configuration and deployment of the nodes, inorder to ensure a reliable and high-performance network. Using theknowledge obtained from previous work in the area, methods andexperiments were developed for assessing the effects that thesedemanding situations have on the performance of the wirelesslinks. The results showed that the observed metrics of the links, such asthe strength of the received signal and the packet receptionratio, differed greatly depending on the environment in which theexperiments were conducted. Based on the outcomes of theexperiments, a conclusion was made that field tests on planneddeployment sites before installation in the testbed will likely benecessary to ensure sufficient reliability in the network. Wheninvestigating potential areas of improvements to the existingtestbed architecture, it was found in related work that techniquessuch as adaptive power and channel control provide higherperformance in lossy and high-density networks. Since the nodesused in the project are compatible with existing solutionsproviding this functionality, these kinds of enhancements shouldprove to be beneficial for the future operation of the GreenIoTtestbed.