QoS Scheduling Algorithm for a Fog IoT Gateway

An Internet of Things (IoT) is characterized as distributed and heterogeneous, both in terms of technology and in the capacity of network processing. The IoT gateway is a component responsible for allowing communication between heterogeneous devices and middleware, regardless of network resources. In this work, we propose an approach to include a classifier, which analyzes the messages received by the gateway, separates them into queues, using criteria previously established by the user, and reserves the resources for sending data marked as critical to the Middleware. We used the Hierarchical Token Bucket (HTB) algorithm to perform the treatment of messages in the queues, aiming to provide fair and efficient treatment the requests serviced by the gateway. Experimental evaluation of our IoT gateway implementation shows that the prioritization of the requests that need more attention caused a reduction in the latency of the delivery of the messages in the highest priority queue obtaining average reduction of 24 percent without discards or losses of packages in this queue.