Coordinated Data Flow Control in IoT Networks.

An IoT cloud environment consists of connected physical devices communicating with the cloud, sending telemetry data and accepting actuation information. For sensors, the data flow is from the physical devices to the cloud. The IoT edge device is responsible for collecting this data and forwarding it to the cloud environment for processing. The time it takes for the data to be made available for processing in the cloud is critical, and the network connectivity, bandwidth and latency are the bottlenecks. In this work, we created a flow controller which adaptively controls the flow of the data from the edge device to the cloud. While rate limiting is a trivial technique to control data flow, it is crucial how the edge devices dynamically control the data rate by re-configuring the IoT devices to send data based on the current network condition and load on the Edge device. We tested this system with a simulated data flow from 10 sensors to a Raspberry Pi Device which performed the rate limiting.