Store Carry and Forward for Delay Tolerant Networks

88 concepts76 papers2 proposals

Traditional internet routing relies on a fundamental assumption: a continuous, end-to-end physical path must exist between the sender and the receiver. If any link along the path breaks, the connection fails and packets are dropped. However, in highly dynamic, sparse, or extreme environments—such as low-Earth orbit satellite constellations, underwater sensor arrays, or remote rural communities—this assumption is routinely violated.

As surveyed by Yongping Xiong et al. in Opportunistic Networks, these challenged environments suffer from intermittent connectivity, where network partitions are the norm rather than the exception. To enable communication when stable paths do not exist, researchers developed the store-carry-and-forward paradigm. This approach shifts routing from a purely spatial problem to a spatio-temporal one, trading local storage and latency for eventual data delivery.

The Mechanics of Spatio-Temporal Flow

Why does store-carry-and-forward work? The core intuition is that physical mobility can bridge temporal gaps in network topology. When a node has data to send but no active link to the destination, it does not discard the packets. Instead, it buffers the data in local memory, physically "carries" it as it moves through space, and "forwards" it when an opportunistic encounter with another node occurs.

Traditional Routing (Spatial):
[Source] ------> [Relay] ------> [Destination]  (Requires simultaneous links)

Store-Carry-and-Forward (Spatio-Temporal):
[Source] ------> [Relay]
                   |
                   v (Relay physically moves over time)
                   |
                   +--------------> [Destination] (Link forms later)

In this framework, local buffers are not merely temporary queues; they are a foundational requirement for network flow. As formalized by Eleni C. Akrida et al. in Temporal Flows in Temporal Networks, buffers allow data to traverse edges that are never active simultaneously. By holding data until the next valid outgoing link appears, nodes can successfully propagate information across highly fragmented networks.

Managing Replication: From Epidemics to Optimization

The simplest way to ensure a message reaches its destination in an unpredictable network is to replicate it infinitely—a strategy known as epidemic flooding. In Epcast: Controlled Dissemination in Human-based Wireless Networks by means of Epidemic Spmeans of Epidemic Spreading Models, Cecilia Mascolo et al. modeled this behavior using mathematical epidemic frameworks (like the Susceptible-Infected-Recovered model). They demonstrated that treating data dissemination like a viral outbreak can guarantee delivery within a target timeframe.

However, unconstrained replication quickly exhausts the limited storage and bandwidth of mobile nodes. To make store-carry-and-forward practical, protocols must intelligently constrain how many copies of a message are made. For instance, Longbo Zhang et al. introduced the Dynamic Spray and Wait Routing Protocol for Delay Tolerant Networks, which uses a probability-based stopping mechanism. By calculating the cumulative delivery success probability of current message holders, the protocol slashes redundant transmissions while maintaining low latency.

When contact windows are exceptionally brief—such as high-speed vehicles passing roadside access points—nodes cannot afford to transmit entire large files at once. To maximize throughput during these fleeting encounters, Yong Li et al. developed a heuristic algorithm in Collaborative data dissemination in opportunistic vehicular networks. By framing data distribution as a submodular optimization problem, the authors showed that fragmenting data across multiple passing vehicle relays significantly increases the probability of successful reassembly at the destination within strict deadlines.

Exploiting Predictable and Social Mobility

Rather than replicating data blindly, advanced store-carry-and-forward protocols exploit the fact that physical mobility is rarely completely random. By understanding the underlying patterns of node movement, networks can select the most effective relays.

In human-centric networks, mobility is driven by social routines. Feng Xia leverages this insight in PIS: A Multi-dimensional Routing Protocol for Socially-aware Networking, demonstrating that integrating physical proximity with shared interests and social relationships creates a highly predictive utility function for relay selection. Nodes pass data to individuals who are socially closer to the destination, drastically reducing network overhead.

[Source] ---> [Casual Acquaintance] ---> [Close Friend of Target] ---> [Destination Target]

Conversely, in aerospace and satellite networks, mobility is governed by deterministic orbital mechanics. As highlighted by Radhika Radhakrishnan et al. in Survey of Inter-satellite Communication for Small Satellite Systems: Physical Layer to Network Layer View, small satellites in Low Earth Orbit (LEO) experience predictable link disruptions due to line-of-sight obstructions. Because these contact patterns are scheduled and known in advance, systems can utilize contact graph routing. As extended by Ethan Fettes et al. in Next-Generation Satellite IoT Networks: A HAPS-Enabled Solution to Enhance Optical Data Transfer, intermediate nodes (such as High Altitude Platform Stations) can be scheduled to act as stratospheric buffers, holding data until cloud cover clears or a satellite passes directly overhead.

Real-World Implementations: From Wildlife to Data Mules

The store-carry-and-forward paradigm has enabled connectivity in regions where deploying traditional infrastructure is economically or physically impossible.

  • Long-Term Animal Tracking: In Delay-Tolerant Networking for Long-Term Animal Tracking, Philipp Sommer et al. designed a three-tier architecture to track nomadic animals across continental scales. Energy-harvesting tags on animals buffer location data for weeks at a time, forwarding it only when they randomly pass within range of a strategic ground gateway.
  • Bridging the Digital Divide: In rural or developing regions, physical transportation systems can be co-opted as network infrastructure. Salah Abdeljabar proposes using school buses and delivery trucks as "data mules" to physically transport digital educational content between urban internet gateways and remote schools, bypassing the need for costly cellular or fiber deployments.

By treating isolation and mobility as routing assets rather than failures, store-carry-and-forward mechanisms ensure that data can reliably navigate the temporal gaps of an unconnected world.

Key Foundational Work

Adapting asynchronous messaging middleware to ad hoc networking (2004) — contributed 3 concept elements to this direction.
Epcast: Controlled Dissemination in Human-based Wireless Networks by means of Epidemic Spreading Models (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_intermittent_connectivity_2c5ee256,
  title = {intermittent connectivity},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/intermittent-connectivity/rdr_2c5ee256247e4d49a75e17eaefdc4da2},
  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_intermittent_connectivity_2c5ee256_v1,
  title = {intermittent connectivity},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/intermittent-connectivity/rdr_2c5ee256247e4d49a75e17eaefdc4da2/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