An Adaptive Channel Reservation MAC Protocol Based on Forwarding Traffic of Key Nodes

Ze Liu, Bo Li, Mao Yang, Zhongjiang Yan

TL;DR

In multi-hop Ad Hoc networks, "key nodes" acting as traffic relays often become bottlenecks because standard protocols treat them like any other node. This paper introduces FTKN-CRM, an adaptive MAC protocol that identifies these critical nodes based on their forwarding load and grants them prioritized channel reservation opportunities, significantly boosting throughput and reducing latency compared to IEEE 802.11ax.

What Problem Does This Paper Address?

Wireless Ad Hoc networks (like those used in emergency relief or military operations) rely on nodes to relay data for one another. In these multi-hop topologies, certain "key nodes" inevitably end up at the intersection of multiple data paths.

The problem is that standard Media Access Control (MAC) protocols, such as CSMA/CA (used in Wi-Fi), are designed for fairness—they try to give every node an equal shot at the channel. However, a key node handling traffic for five neighbors needs much more "airtime" than a leaf node sending its own occasional data. When these key nodes can't get enough access, buffers overflow, packets drop, and the entire network's end-to-end performance collapses. Existing reservation schemes often fail to account for this inter-node variability in multi-hop settings.

Methods

The authors propose the Forwarding Traffic of Key Nodes - Channel Reservation MAC (FTKN-CRM). The protocol operates in three distinct phases:

  1. WRA Updating Phase: Nodes calculate a Weight of Reservation Ability (WRA). This isn't just a measure of how many packets are in the buffer; it weights packets based on how many hops they have already traveled. Packets that have traveled further are given higher priority to ensure they aren't dropped near the finish line.
  2. Reservation Phase: Nodes compare their WRA with their neighbors to calculate a "Correction Factor" ($\lambda$). This factor adaptively adjusts the Reservation Offset Time. Essentially, a node with a high relay load calculates a shorter wait time for its next reservation, allowing it to "book" the channel more frequently. To prevent the network from oscillating due to bursty traffic, they implement an N-step Inertia Factor to smooth out changes over time.
  3. Channel Access Phase: If a node fails to get a reservation and must compete via standard backoff, it adjusts its Contention Window (CW) based on its importance. Key nodes use a smaller window to win the "random draw" for channel access more often.

The protocol uses a special control frame called the RSI MPDU (Reservation Instruction) to broadcast these reservations to neighbors, who then remain silent during the reserved slots.

Datasets

As is common in MAC layer research, the authors used a custom simulation platform rather than a static dataset. They evaluated the protocol across four specific network topologies:

  • Short-hop topology: A simple three-region setup to test basic relaying.
  • Long-chain topology: A four-hop line to test cumulative delay.
  • Cross-topology: Two intersecting paths to simulate a high-interference "intersection."
  • Random topology: Nodes scattered in a 2000m x 2000m area to test real-world unpredictability.

The simulations were benchmarked against IEEE 802.11ax and SCRP (Soft Channel Reservation Protocol).

Key Insights and Ideas

  • Hop-Count Weighting: The most elegant idea here is weighting the WRA by hop count. It recognizes that a packet that has already consumed network resources (by being relayed three times) is "more valuable" to deliver than a brand-new packet.
  • Dimensionless Comparison: Rather than using a fixed threshold to define a "key node," the protocol compares a node's load to the average of its neighbors. This makes the protocol self-tuning regardless of whether the overall network load is light or heavy.
  • Energy Efficiency via Sleep: Because the protocol uses explicit reservations, neighbor nodes know exactly when they don't need to listen to the channel, allowing them to enter a low-power sleep mode, which significantly reduces energy consumption compared to 802.11ax.

Key Results

  • Throughput Stability: In random topologies with heavy traffic, FTKN-CRM maintained high throughput while 802.11ax and SCRP saw performance drops of nearly 30-50% as congestion took hold.
  • Latency Reduction: In the long-chain scenario, FTKN-CRM showed the lowest end-to-end delay. By prioritizing the intermediate relay nodes, packets moved through the "pipe" much faster.
  • Energy Savings: The protocol demonstrated significantly lower energy consumption (mJ/s) than 802.11ax because it reduced "disorderly competition" and allowed for more strategic sleep cycles.

Limitations and Future Work

  • Mobility: The paper focuses on static or semi-static topologies. In highly mobile environments (like vehicular networks), the neighbor WRA information might become stale too quickly for the "Inertia Factor" to be effective.
  • Overhead: While the authors mention "low signaling overhead," adding WRA fields to MAC headers and managing reservation tables does consume some bandwidth. A more detailed analysis of the "cost of the cure" vs. the "cost of the disease" in very low-bandwidth scenarios would be beneficial.
  • Hardware Implementation: The results are currently limited to simulations. Real-world radio effects (like non-binary capture effects) could impact how effectively the RSI MPDUs are received in crowded environments.

Summary updated

Abstract

Ad Hoc networks with multi-hop topology are widely used in military and civilian applications. One challenge for Ad Hoc networks is to design efficient Media Access Control (MAC) protocols to ensure the quality of service (QoS). In Ad Hoc networks, there is a kind of node called key node, which undertakes more forwarding traffic than other surrounding nodes. The number of neighbor nodes around key nodes is often large, and the surrounding channel environment and interference are often more complex. Thus, the key nodes can hardly get enough channel access opportunities, resulting in poor end-to-end performance. Therefore, we propose an adaptive channel reservation MAC protocol based on forwarding traffic of key nodes, which is aimed at alleviating the congestion for key nodes. Nodes initiate reservations for future transmission time according to the buffer status before sending packets and then calculate the Weight of Reservation Ability (WRA). The node adaptively adjusts its reservation opportunity by comparing the WRA with neighbor nodes, thus improving the channel access efficiency and ensuring the transmission opportunity of key nodes. Extensive simulation confirms that our proposed FTKN-CRM provides significant improvements in end-to-end performance over the IEEE 802.11ax protocol and other reservation access protocols.
BibTeX
@misc{liu2023adaptive,
  author = {Ze Liu and Bo Li and Mao Yang and Zhongjiang Yan},
  title = {{An Adaptive Channel Reservation MAC Protocol Based on Forwarding Traffic of Key Nodes}},
  year = {2023},
  eprint = {2304.14581},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2304.14581},
}
Figure