Addressing for Intelligent Routing in Mobile Military Networks
Military / Coalition Issue
In military networks, which are operating in increasingly congested and contested settings, it is vital to prioritise and control traffic both at the network and application layers. Further, the current use of IP addresses ties the source and destination of a packet to locations within the network, and thus highly mobile military networks suffer from needing to re-route and re-address network elements.
Core idea and key achievements
DAIS ITA research has shown how vector based approaches can be used to reconfigure battlespace communication plans, and to enable service discovery and orchestration in tactical coalition information systems.
The core idea is that a vector-based approach can also be used to represent both source and target identities (replacing IP addresses). This allows us to specify the source and target of a packet to include attributes such as Role, Type, Status (by no means an exhaustive list).
Thus, a battlefield order could be tagged with source “Mission Commander | Orders | Manual | One-off” and target “Soldier | Platoon 1 | Chat” to be issued to all the soldiers in platoon 1 in their chat application. The message would then be routed using intelligent routers with knowledge of the state of the network. It would replace both unicast and multicast traffic. To improve the performance on the wire, hardware architectures, such as phase-change-memory, in the router could be employed to act on the packets at “wire-speed”.
Implications for Defence
Applying these techniques would create a new level of resilience and flexibility in tactical networks to allow them to adapt to battlefield conditions, including electronic attacks. It would allow the battlespace network to prioritise traffic such as individual messages and order, and if the network is too congested to adopt delay-tolerant approaches for lower priority traffic (e.g. use store-and-forward messaging).
Readiness & alternative Defence uses
The core idea is currently technology readiness level (TRL) 1, but most of the related technology is at TRL2+, so enabling rapid development and demonstration of the concept. An issue that would need to be addressed is the confidentiality of addresses. They could be represented using random binary vectors, but these could conceivably be learned over time. One option is to investigate the impact of use either encryption or dynamic vector space mapping to thwart such an attack and to determine the performance implications of these approaches.
Resources and references
- Simpkin, Chris, et al. “A scalable vector symbolic architecture approach for decentralized workflows.” COLLA (2018).
- Simpkin, Chris, et al. “Constructing distributed time-critical applications using cognitive enabled services.” Future Generation Computer Systems 100 (2019): 70-85.
- Simpkin, Chris, et al. “Efficient orchestration of node-red iot workflows using a vector symbolic architecture.” Future Generation Computer Systems 111 (2020): 117-131.
The underlying vector symbolic architecture code has been published as Open Source.
Organisations
IBM UK, Cardiff University