Physical Interference Modeling for Wireless Networks

39 concepts32 papers2 proposals

For decades, the design of wireless network protocols relied on a convenient fiction: the conflict graph. In this simplified view, wireless interference is treated as a binary, localized relationship. If node A transmits to node B, any node within a fixed "interference radius" is forbidden from transmitting simultaneously. This graph-based abstraction transforms complex radio frequency environments into clean, discrete combinatorial problems like vertex coloring or independent set selection.

However, real-world radio waves do not behave like tidy graph edges. In actual deployments, graph-based models frequently over-predict network capacity or fail to deliver expected throughput. To bridge this gap, a major research direction has shifted away from binary abstractions toward physical interference modeling, centered on the Signal-to-Interference-plus-Noise Ratio (SINR).

Understanding why this shift is necessary—and how it changes protocol design—requires looking at the physics of how signals collide and combine.

The Failures of the Binary Abstraction

The fundamental flaw of graph-based models is that they treat interference as a localized, non-additive phenomenon. In Effects of Interference on Wireless Mesh Networks: Pathologies and a Preliminary Solution, Ratul Mahajan demonstrates that binary conflict graphs oversimplify radio propagation by failing to account for environmental factors like multi-path fading and variable signal strengths, leading to severe capacity estimation errors.

Graph models fail in two distinct ways:

  1. The Problem of Additivity (Too Optimistic): Imagine a crowded room where multiple people are whispering far away from you. A graph model looks at each whisperer individually, determines they are too far away to disrupt your conversation, and labels them "non-interfering." In reality, the background noise from dozens of distant whispers adds up. In a wireless network, the cumulative background noise from many distant transmitters can easily corrupt a local transmission, even if no single transmitter is close enough to cause a conflict on its own.
  2. The Capture Effect (Too Pessimistic): Conversely, graph models often forbid concurrent transmissions that could actually succeed. If a nearby transmitter is incredibly loud, a receiver can often successfully decode its packet even if another node is transmitting nearby. This is known as the power capture effect.

By replacing binary ranges with a continuous physical representation of signal quality, physical models capture these dynamics accurately.

The Physics of Communication: The SINR Model

The physical interference model evaluates the success of a transmission using the SINR equation. For a receiver to successfully decode a message from a specific transmitter, the received signal power must outweigh the combined power of all other concurrent transmissions plus background noise by a specific threshold, $\beta$:

$$\text{SINR} = \frac{P_{\text{signal}}}{N + \sum P_{\text{interference}}} \ge \beta$$

This shift from an on/off switch to a continuous ratio fundamentally changes what is theoretically possible in a network. As Thomas Moscibroda et al. demonstrate in Protocol Design Beyond Graph-Based Models, designing protocols based on the actual physics of SINR allows for simultaneous transmissions through precise power management, unlocking throughput levels that are mathematically impossible under standard graph-based assumptions.

To ensure these mathematical models match reality, researchers have grounded them in hardware. In Joint Power Control and Scheduling Algorithm for Wi-Fi Ad-hoc Networks, Sang-Ho Oh et al. conducted hardware-level verification of the power capture effect on commodity Wi-Fi radios, measuring the exact SINR thresholds required for successful packet decoding to ensure their scheduling logic remained effective in practical deployments.

The Algorithmic Challenge: Global Coupling

If physical models are so much more accurate, why did early network protocols rely on graphs? The answer lies in algorithmic complexity.

In a graph model, interference is local. You only need to coordinate with your immediate neighbors. In an SINR model, interference is globally coupled. Every time a node anywhere in the network decides to transmit, it increases the denominator of the SINR equation for every other active receiver in the network. This makes scheduling transmissions an incredibly difficult optimization problem.

As noted by Shaojie Tang 0001 et al. in Maximum Weighted Independent Set of Links under Physical Interference Model, managing these intricate, cumulative interference dynamics requires specialized algorithms. To solve this without a omniscient central controller, researchers have designed localized approximation algorithms. For example, Yaqin Zhou et al. in Throughput Optimizing Localized Link Scheduling for Multihop Wireless Networks under Physical Interference Model introduced a localized scheduling algorithm that divides the network into shifting grids. By bounding the maximum possible cumulative interference from outside each grid cell, the authors proved that nodes can make independent, local scheduling decisions while still guaranteeing a constant fraction of the optimal global network throughput.

Expanding the Physical Model: Channels and Scaling

Realistic interference modeling extends beyond spatial distance and transmission power; it also encompasses how signals interact across different frequencies.

Traditional multi-channel protocols assume that channels must be completely non-overlapping to avoid interference. Challenging this assumption, Suman Banerjee et al. developed a quantitative model termed the 'I-factor' in Partially overlapped channels not considered harmful. This model moves beyond binary channel assessments to precisely measure the degree of interference between partially overlapped wireless channels, allowing protocols to utilize the spectrum much more granularly.

Furthermore, these physical constraints dictate how well a network can scale. In Leveraging Multiple Channels in Ad Hoc Networks, Magnús M. Halldórsson et al. explored how the additive nature of background noise and distant transmitters impacts parallel communication. They demonstrated that by using a hierarchical structure to manage this cumulative noise, distributed wireless algorithms can still achieve a linear speedup proportional to the number of available communication channels, even under strict, realistic SINR constraints.

By trading the clean simplicity of graphs for the continuous equations of physical propagation, this research direction provides the mathematical foundation necessary to optimize high-density, real-world wireless deployments.

Key Foundational Work

Protocol Design Beyond Graph-Based Models (2006) — contributed 3 concept elements to this direction.
Distributed Link Scheduling with Constant Overhead (2006) — contributed 1 concept elements to this direction.
Effects of Interference on Wireless Mesh Networks: Pathologies and a Preliminary Solution (2007) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_realistic_interference_modeling_99336b1e,
  title = {realistic interference modeling},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/realistic-interference-modeling/rdr_99336b1e1ea04f048b5bb796a2a6eee8},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_realistic_interference_modeling_99336b1e_v1,
  title = {realistic interference modeling},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/realistic-interference-modeling/rdr_99336b1e1ea04f048b5bb796a2a6eee8/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake