# Rate-Distortion Classification for Self-Tuning IoT Networks

## TL;DR

This paper proposes a machine learning-based method for automatically estimating rate-distortion curves in IoT sensor networks, enabling dynamic protocol tuning for energy-efficient lossy compression.

## Contribution

It introduces an automatic profiling approach that estimates rate-distortion curves on-the-fly using simple statistical features and machine learning models.

## Key findings

- Reliable on-the-fly estimation of rate-distortion curves.
- Uses only 10-20 statistical features for assessment.
- Supports dynamic protocol adaptation in IoT networks.

## Abstract

Many future wireless sensor networks and the Internet of Things are expected to follow a software defined paradigm, where protocol parameters and behaviors will be dynamically tuned as a function of the signal statistics. New protocols will be then injected as a software as certain events occur. For instance, new data compressors could be (re)programmed on-the-fly as the monitored signal type or its statistical properties change. We consider a lossy compression scenario, where the application tolerates some distortion of the gathered signal in return for improved energy efficiency. To reap the full benefits of this paradigm, we discuss an automatic sensor profiling approach where the signal class, and in particular the corresponding rate-distortion curve, is automatically assessed using machine learning tools (namely, support vector machines and neural networks). We show that this curve can be reliably estimated on-the-fly through the computation of a small number (from ten to twenty) of statistical features on time windows of a few hundreds samples.