A Privacy Scheme for Monitoring Devices in the Internet of Things

Zygmunt J. Haas, Ashkan Yousefpour

Introduction

In the Internet of Things (IoT), data privacy is often compromised not just by the content of communications, but by the metadata associated with them. Even when payloads are encrypted, the source and destination IP addresses in packet headers remain visible for routing purposes. This creates a "linking index" that allows adversaries to profile users. For instance, tracking packets from a smart meter can reveal when a home is unoccupied, while monitoring medical sensors can expose a patient's health history.

Traditional end-to-end encryption is often too computationally intensive for resource-constrained IoT devices and fails to solve the anonymity problem, as clear-text identifiers are still required for key management. This paper proposes a privacy-preserving scheme that utilizes "IP address hopping" to break this linkability without requiring complex encryption or infrastructure changes.

Proposed Methodology

The proposed scheme involves three primary entities: the IoT node, the Corresponding Node (CN) (the receiver), and a Central Node (a trusted coordinator). The system's security rests on a Tracking Function (TF), which serves as a pseudorandom number generator (PRNG) based on chaotic tent maps. These maps are highly sensitive to initial conditions, ensuring that outputs appear random to observers while remaining deterministic for authorized parties.

The Tracking Function and IP Generation

The IoT node and the CN share secret parameters and a timestamp. Every $\zeta$ seconds, both nodes calculate the next IP address in a sequence. As shown in the table below, the output of the PRNG is mapped to the network address space to generate a dynamic IP identity.

Figure 3: Example output of the Tracking Function showing how timestamps map to dynamic IP addresses.

Synchronization and Overlap

To maintain communication reliability despite network jitter and clock skew, the scheme introduces an Address Overlap ($\lambda$). When an IoT node hops to a new IP address, it retains its previous address for a short duration ($\lambda$). This ensures that packets already in flight are not discarded by the receiver during the transition.

Figure 2: Timing diagram of the IP address hopping mechanism, illustrating the hopping interval (ζ), overlap (λ), and network delay (d).

Performance and Security Analysis

The researchers evaluated the scheme using experimental testbeds across both Local-Area Networks (LAN) and Wide-Area Networks (WAN). A critical "threshold effect" was observed: if the hopping interval ($\zeta$) is too short, packet loss spikes due to synchronization issues. However, once $\zeta$ exceeds approximately 0.7s for LAN and 1s for WAN, packet loss becomes negligible (less than 1%).

Key Insights

  • IP as Identity: The pattern of changing IPs becomes the identity known only to the authorized receiver, flipping the traditional view of static IP identifiers.
  • Collision Management: Mathematical analysis shows that in a standard /24 subnet (256 addresses), the probability of two devices randomly hopping onto the same IP is statistically insignificant for small deployments (under 40 devices).
  • Efficiency: By implementing privacy at the network layer, the scheme avoids the per-packet overhead of heavy encryption, making it ideal for low-power hardware.

Conclusion and Future Work

The proposed IP hopping scheme successfully mitigates profiling and Denial of Service (DoS) attacks while remaining compatible with existing internet routing infrastructure. While the current model assumes static nodes and relies on a Central Node for initial parameter distribution, future work will explore extensions for mobile IoT nodes and methods to further enhance location privacy.

Summary updated

Abstract

Sufficiently strong security and privacy mechanisms are prerequisite to amass the promising benefits of the IoT technology and to incorporate this technology into our daily lives. This paper introduces a novel approach to privacy in networks, an approach which is especially well matched with the IoT characteristics. Our general approach is based on continually changing the identifying attributes of IoT nodes. In particular, the scheme proposed in this work is based on changing the IoT nodes' IP addresses, and because the changing patterns of the IP addresses appear random to a non-intended observer, an adversary is unable to identify the source or destination of a particular transmission. Thus, packets that carry information generated by a particular node cannot be linked together. The scheme offers additional security benefits, including DoS mitigation, is relatively easy to implement, and requires no changes to the existing networking infrastructure. We discuss the details of the implementation of the scheme and evaluate its performance.
BibTeX
@inproceedings{haas2015privacy,
  author = {Zygmunt J. Haas and Ashkan Yousefpour},
  title = {{A Privacy Scheme for Monitoring Devices in the Internet of Things}},
  booktitle = {IIoT},
  year = {2015},
  url = {https://openreview.net/forum?id=VhvzOt4ESl},
}
Figure