One network, whether the signal comes from a tower or from orbit

Communication architectures connecting terrestrial, non-terrestrial, edge and intelligent systems toward AI-native Beyond-5G and 6G networks.

  • B5G / 6G
  • NTN
  • LEO / MEO / GEO
  • Direct-to-Device
  • ISAC
  • MEC & Edge
  • Private 4G/5G
  • RAN & Core
  • Network Slicing
  • AI-Native Networks

Terrestrial and non-terrestrial convergence: one network, whatever carries it

Terrestrial coverage stops at the edge of the build-out. Satellite coverage does not, but it has historically been a separate network with separate devices, separate contracts and separate operations.

Non-terrestrial networks bring satellite access into the same architecture as the ground network, including direct-to-device links that reach an ordinary handset. One orchestration layer places workloads, manages slices and hands sessions between the two without the device participating in the decision.

  • Coverage becomes continuous rather than a patchwork of separate systems.
  • Edge compute sits between the radio and the core, so latency-bound work never crosses the whole network.
Why it matters

Autonomous systems, vehicles and remote industrial sites need a connection that does not disappear at the edge of a coverage map.

AI-native orchestration slice placement · workload placement · autonomous management NTN satellite LEO · MEO · GEO direct-to-device Terrestrial RAN private & public 5G sensing return handset vehicle robot handover, decided above the device Edge / MEC low-latency compute Core cloud-native functions network slices

Integrated sensing and communications: the network as an instrument

A radio signal that reflects off a moving object carries information about that object. Integrated sensing and communications uses the same waveform to carry data and to sense the environment it passes through, rather than deploying separate radar infrastructure alongside the network.

That turns the network into a distributed sensor: presence, position, movement and environmental change, measured by infrastructure that is already there for another reason. AI-based prediction and localisation sit on top of the raw measurements.

Why it matters

Sensing you get as a by-product of coverage you were already paying for is a fundamentally different economic proposition to sensing you have to deploy.

one radio one waveform communication receiver data delivered a moving object sensing return range velocity

Components of the architecture

  • Beyond 5G / 6G

    The capability path from 5G-Advanced toward IMT-2030.

  • Non-terrestrial networks

    LEO, MEO and GEO satellite access integrated with the terrestrial network.

  • Direct-to-device

    Satellite links that reach an ordinary device without dedicated terminal hardware.

  • ISAC

    Sensing and communication sharing one waveform and one infrastructure.

  • MEC & edge computing

    Compute placed close enough to the radio to meet latency budgets.

  • Private 4G/5G networks

    Dedicated networks for industrial sites, campuses and operations.

  • RAN & core architectures

    Open, cloud-native and interoperable network functions.

  • Network slicing

    Isolated logical networks with their own guarantees on shared infrastructure.

  • AI-native networking

    Intelligence in the network itself, not bolted on to manage it.

  • Autonomous network management

    Closed-loop operation of the network as a system.

Applications: connectivity where the map runs out

  • 01 Remote and rural coverage
  • 02 Maritime, aviation and logistics
  • 03 Autonomous vehicles and drones
  • 04 Industrial private networks
  • 05 Emergency and resilient communications
  • 06 Distributed sensing and localisation

Standards and references

ITU and 3GPP for the 6G and NTN path, ETSI for sensing and edge, and O-RAN for open RAN — noting that O-RAN is an industry specification alliance rather than a formal standards body.

Research and resources

For further information, see the selected references, standards and research resources below.

Related