Sound pollution is becoming an increasingly pressing issue in today’s world. To effectively address it, it must be measured. To this end, Serval was developed, an edge-ai powered sound recognition solution. Its lack of accuracy, however, makes it difficult to deploy. This thesis examines the potential for improving this solution while staying within its technical limitations in order to raise the accuracy to satisfactory levels. Multiple aspects of Serval were evaluated and compared to the current stateof-the-art: its data augmentation, the embedding it uses, and the hardware it runs on. Alternatives for each of these components were evaluated and each aspect was optimized.
The results show that after these improvements, the single-label F1-score increased from 0.60 to 0.76, and the single- and multi-label combined F1-score increased from 0.64 to 0.67. Finally, power consumption has been reduced by 14%, partially thanks to the usage of specialized hardware. One issue that has yet to be adequately addressed is the size of the dataset. By increasing the number of samples, the accuracy could be further improved.

ProRail uses multiple custom built software applications to control the train infrastructure in the Netherlands. At this moment these applications experience downtime when they are updated. This interrupts processes that are related to controlling the train services. ProRail asked us to redesign an existing application in such a way that updates can be applied with zero downtime.

To solve this problem we use redundancy with automatic load balancing. We compared multiple solutions and we chose to use Docker’s swarm functionality. We then redesigned the non-critical application Brugkijker, an application to monitor trains approaching a bridge. In the end we could successfully run multiple instances of Brugkijker in Docker and show how the application can be incrementally updated, inducing no downtime for the service as a whole and without losing data coming from upstream. This proposed approach can be used by ProRail to make other noncritical and critical applications highly available.