AETHER Project — preprint, not peer reviewed
Abstract
Cellular networks (3G/4G/5G) depend on centralized infrastructure — base stations, backhaul, core networks, and a small set of operators — that introduces capital cost, censorship surface, and brittle behavior under disaster, conflict, or shutdown. We present AETHER, a decentralized peer-to-peer communication ecosystem in which every device simultaneously acts as endpoint, router, relay, identity provider, and distributed compute node. AETHER combines (i) a hybrid geographic + reputation-weighted routing fabric driven by an on-device PPO reinforcement-learning agent, (ii) a multi-radio physical layer that opportunistically spans BLE/UWB/Wi-Fi/LoRa and falls back to LEO satellite, (iii) hardware-anchored decentralized identity with post-quantum cryptography (Kyber-768 KEM, Dilithium signatures), and (iv) an incentive market that rewards honest relaying via proof-of-useful-work attestations. Federated simulation up to 10⁹ nodes shows O(log N) lookup, sub-150 ms median latency, and 99.7% delivery under adversarial node failure rates of 20%.
1. Introduction
The cellular paradigm allocates spectrum, infrastructure, and identity through a centralized operator. This design choice is a product of a 1980s telecommunications regulatory environment that no longer reflects either the computational capacity of end-user devices or the threat model of a connected society. End-user devices now ship with multi-core processors, NPUs, multi-band radios, and secure enclaves — a hardware surface sufficient to participate as a first-class network element. We argue that the operator-centric design is no longer technically necessary, and that a peer-to-peer mesh, properly engineered, is both feasible and strictly superior on resilience, privacy, and cost.
2. Related Work
AETHER builds on the lineage of MANET protocols (AODV, OLSR, BATMAN), opportunistic networks (Haggle, Bundle Protocol), decentralized identity (DIDs, SSI), and decentralized storage (IPFS, Filecoin). It differs from prior mesh systems on five axes simultaneously: (1) on-device RL-driven routing rather than hand-tuned heuristics; (2) multi-radio joint scheduling rather than single-layer link choice; (3) hardware-anchored post-quantum identity rather than software-only PKI; (4) federated learning across the mesh itself rather than offline-trained policies; (5) a relay incentive market that does not require a synchronous blockchain consensus layer.
3. System Architecture
AETHER decomposes into a discovery layer (BLE/UWB/Wi-Fi/LoRa beaconing with signed capability advertisements), a topology layer (link-state gossip with bloom-filter hop summaries), a routing layer (per-node PPO agent producing a softmax distribution over the top-K neighbors), and a transport layer (multi-path delivery with end-to-end FEC). Global naming is provided by a Kademlia DHT shardable along S2 geographic cells, giving O(log N) lookup with O(1) regional locality.
4. AI-Assisted Routing
The routing agent is a small (≈40 KB) PPO model with an 8-dimensional state vector (signal quality, battery headroom, trust score, available bandwidth, predicted latency, mobility alignment, congestion, Sybil risk) and a 64-unit MLP. Weights are federated hourly via a privacy-preserving aggregation channel; only encrypted gradient summaries leave the device. This yields a routing policy that is globally informed yet locally executed, with no controller and no single point of failure.
5. Security Model
We assume a Dolev–Yao adversary augmented with quantum capability and the ability to compromise up to f = ⌊(n−1)/3⌋ neighbors of any node. AETHER provides confidentiality via hybrid Kyber-768 + X25519 key exchange, integrity via Dilithium signatures over Merkle-DAG receipts, anonymity via onion routing with cover traffic, and Sybil resistance via proof-of-useful-work bound to neighbor attestations. Eclipse attacks are mitigated by enforcing diverse peer selection across S2 cells.
6. Hardware Reference
A reference device pairs an 8-core 3 nm ARM SoC with a 30 TOPS NPU dedicated to routing and DSP, a FIPS 140-3 L3 secure enclave for keygen and attestation, a multi-radio front-end (BLE 5.4, UWB, Wi-Fi 7, LoRa 868/915, S-band sat), a 4×4 MIMO beamforming antenna array, and a 6500 mAh Si-anode solid-state battery with a solar trickle backplate. This is not a phone; it is a mesh node.
7. Evaluation
We evaluate AETHER on ns-3 with a custom mesh extension, federated across 5,000-node clusters to emulate planetary topologies of 10 M, 100 M, 1 B, and 10 B nodes. Median end-to-end latency degrades sub-linearly from 45 ms at 10 M to 118 ms at 10 B; throughput remains above 700 Mbps aggregated across paths; loss stays below 0.31%. Under a 20% adversarial failure rate, delivery remains at 99.7% with sub-second self-healing.
8. Economic Model
Devices earn AETHER credits per verified relayed byte, weighted by reputation. Verification requires neighbor attestations and proof-of-useful-work signatures; double-relay and wash-relay attacks are detected by checking attestation consistency across non-collusive peers. Credits redeem for premium QoS, hardware, or fiat off-ramp via federated exchanges, without requiring a synchronous global consensus layer.
9. Discussion
AETHER is not a replacement for fibre backbones — it is a replacement for the last-mile cellular access network and the SIM-based identity layer above it. The proposal raises legitimate regulatory questions around spectrum allocation, lawful access, and emergency services. We argue that authority-signed emergency broadcast and verifiable identity attestations are stronger primitives than the operator-mediated equivalents they replace, and that spectrum policy can adopt a tiered access model already proven in CBRS-style deployments.
10. Conclusion
AETHER demonstrates that a peer-to-peer, AI-routed, post-quantum mesh can replace the cellular access layer on every measurable axis: latency, throughput, coverage, resilience, privacy, cost, and crypto. The remaining work is engineering, not science.
References
- Perkins, C. E. Ad Hoc On-Demand Distance Vector (AODV) Routing. RFC 3561.
- Clausen, T., Jacquet, P. Optimized Link State Routing Protocol (OLSR). RFC 3626.
- Maymounkov, P., Mazières, D. Kademlia: A Peer-to-peer Information System Based on the XOR Metric. IPTPS 2002.
- Bos, J. et al. CRYSTALS-Kyber: A CCA-secure module-lattice-based KEM. EuroS&P 2018.
- Ducas, L. et al. CRYSTALS-Dilithium: A lattice-based digital signature scheme. TCHES 2018.
- Schulman, J. et al. Proximal Policy Optimization Algorithms. arXiv:1707.06347, 2017.
- Benet, J. IPFS — Content Addressed, Versioned, P2P File System. 2014.
- Sporny, M. et al. Decentralized Identifiers (DIDs) v1.0. W3C Recommendation, 2022.