Detecting and Mitigating Coexistence Problems through RF Spectrum Analysis

With the rapid growth of the Internet of Things (IoT), an increasing number of wireless appliances is crowding the same unlicensed Industrial, Scientific, and Medical (ISM) frequency bands, causing severe cross-technology interference.

The latter leads to loss of packets, increased delays, and to a reduced performance, especially for wireless IoT devices operating at low power. It is hence crucial for a low- power wireless IoT device to get an understanding of the RF spectrum usage in its surrounding and dynamically adapt its protocol configuration accordingly, so to maximize the dependability of its communications. Such an understanding includes, for example, which channels are highly congested, as well as which (or how many) devices are operating on a given frequency and their traffic pattern.

Download as PDF

Student Target Groups:

  • Students of ICE/Telematics;
  • Students of Computer Science;
  • Students of Electrical Engineering.

Thesis Type:

  • Master Project / Master Thesis

Goal and Tasks:

  • Explore metrics that allow to characterize the RF spectrum usage (e.g., which channels are congested, how many devices are using these frequencies) in a compact and efficient way;
  • Explore the use of ML techniques to detect and classify wireless coexistence issues in real-time;
  • Explore how to make use of such a detailed RF spectrum usage characterization to increase coexistence across IoT devices;
  • Build a demonstrator where a more powerful device (e.g., Raspberry Pi) monitors the RF spectrum usage and communicates this information to a low-power IoT device (e.g., using CTC).

Required Prior Knowledge:

  • Knowledge of networked embedded systems;
  • Excellent C programming skills;
  • Experience with machine learning.


  • a.s.a.p.