Autonomous Networking and Discovery in DaaS-IoT Overlay Architecture

Introduction

The rapid growth of distributed Internet of Things (IoT) infrastructures is increasingly demanding networking architectures that can operate with minimal human intervention. Traditional networking models often rely on pre-configured identifiers and centralized services to enable node discovery and topology formation. While effective in static or controlled environments, these approaches struggle to adapt in dynamic, heterogeneous, and large-scale scenarios.

In this context, we present an autonomous networking and discovery layer developed within the DaaS-IoT (Device-as-a-Service for the Internet of Things) framework. The objective of this work is to enable automatic network formation, dynamic node mapping, and time-synchronized topology building without requiring manual configuration.

Our approach introduces a self-organizing overlay network in which nodes autonomously announce their presence, establish logical connections, and elect a timing master in a fully distributed manner. This discovery mechanism significantly reduces deployment complexity, improves scalability, and strengthens network adaptability in real-world scenarios such as edge and embedded systems.

Background and Motivation

Traditional IoT networking relies heavily on static configurations, centralized registries, and pre-assigned identifiers to enable node discovery and communication. These solutions are suitable in controlled or small-scale deployments, but they introduce critical limitations when applied to distributed and dynamic environments, where nodes may frequently join, leave, or move across the network.

Manual configuration of node identifiers and routing information becomes a scalability bottleneck, leading to increased deployment time, reduced flexibility, and potential inconsistencies in large or mobile networks. Moreover, centralized service discovery solutions often create single points of failure, making the entire system less resilient to disruptions.

In this context, self-organizing discovery mechanisms offer a promising alternative. By enabling nodes to autonomously announce their presence, detect neighbors, and establish logical mappings without external coordination, the network can dynamically adapt to topological changes. This paradigm aligns closely with the goals of the DaaS-IoT framework, which aims to provide a distributed, resilient, and lightweight infrastructure for IoT ecosystems.

The motivation behind this work is therefore to design a discovery and networking layer capable of operating without manual configuration, scaling seamlessly, and remaining robust in the presence of dynamic network conditions. This approach supports a wide range of scenarios, including industrial IoT, sensor networks, and embedded systems with intermittent connectivity.

Networking and Discovery

The Networking and Discovery layer of the DaaS-IoT framework provides the foundational mechanisms for building autonomous and adaptive overlay networks. Its primary goal is to enable nodes to join and organize into a functional topology without requiring predefined identifiers or centralized coordination.

When discovery is enabled, each node broadcasts a beacon to announce its presence in the environment. Other nodes within communication range can detect this signal, establish mutual awareness, and automatically generate the logical identifiers and mappings needed to participate in the overlay. This process eliminates the need for manual configuration of node identifiers, reducing deployment complexity and supporting large-scale, heterogeneous networks.

The resulting network is fully distributed: no central authority is required to assign addresses, maintain routing tables, or orchestrate connections. Instead, connectivity emerges dynamically through local interactions between nodes. This self-organizing mechanism allows the topology to evolve over time, supporting scenarios in which nodes may join, leave, or relocate without disrupting overall connectivity.

Security and trust boundaries are also embedded at the discovery level. Nodes can be configured to accept or reject incoming discovery requests, enabling the formation of controlled sub-networks within a larger deployment. In addition, the discovery process is tightly coupled with the system’s time synchronization layer, which ensures that once nodes discover each other, they also align to a common temporal reference. This integration improves the stability of distributed operations, particularly in latency-sensitive IoT applications.

Key Features

• Autonomous Beaconing: Nodes periodically broadcast presence signals
• Dynamic Identifier Assignment: Logical IDs generated automatically during discovery
• Distributed Coordination: No central authority required for network formation
• Security Integration: Trust boundaries enforced at discovery level
• Time Synchronization: Temporal alignment integrated with discovery process

Autonomous network formation in DaaS-IoT overlay architecture

Implementation Overview

The networking and discovery layer is implemented as an integral component of the DaaS-IoT software stack. Its design prioritizes modularity, cross-platform compatibility, and low configuration overhead. The current implementation enables nodes to participate in a shared overlay network through a lightweight discovery protocol that operates independently of specific hardware or communication drivers.

When a node is initialized, the discovery service is activated by default. Each node periodically emits a beacon to signal its availability, while simultaneously listening for beacons from other nodes. This mechanism supports dynamic peer detection, enabling the creation of logical links between participants without requiring a pre-existing network configuration. Once mutual discovery occurs, nodes can exchange control information to establish their position within the overlay.

A key aspect of this mechanism is its ability to assign and manage node identifiers dynamically. Unlike previous versions of DaaS, nodes can now join the network without pre-defined SID or DIN values; these identifiers are generated or negotiated automatically during the discovery phase. This significantly simplifies the deployment of large-scale or ad hoc networks, allowing nodes to self-organize without external provisioning steps.

Security considerations are integrated into the implementation: nodes can enforce policies that filter or reject discovery requests based on trust boundaries or deployment scenarios. Additionally, the discovery layer interfaces with the time synchronization subsystem to ensure a stable and consistent temporal reference once connectivity is established.

The implementation is designed to be portable across operating systems and network technologies. This flexibility allows the same discovery mechanism to operate in heterogeneous environments, including wired, wireless, and hybrid deployments. The resulting architecture can be seamlessly integrated into larger distributed IoT infrastructures with minimal setup and operational cost.