Simulation Models
The standard protocol simulation library of OMNEST is the INET Framework. INET provides models for simulating a variety of communication networks, ranging from the Internet to Time-Sensitive Networking (TSN), mobile ad hoc networks, and more. INET has been specifically designed to facilitate experimentation with protocols and various architectures. Additionally, INET has support for cross-cutting features such as emulation, real-time (RT) and hardware-in-the-loop (HIL) simulation, visualization, advanced statistics collection, and more, and is accompanied by comprehensive documentation and showcase simulations.
A significant number of other simulation models and model libraries have been developed for OMNeT++ by various research groups and individual members of the user community, published under different open-source licenses. These models can also be utilized with OMNEST. An approximate catalog of these models is available on the Simulation Models and Tools page of the OMNeT++ website.
Below is a list of recommendations for simulating various types of networks.
Internet Simulations
The INET Framework should be used for simulating any of the protocols, technologies, and applications used on the Internet and other Wide Area Networks (WANs). INET includes the essential components such as protocols from the TCP/IP stack, common routing protocols, support for differentiated services, label switching (MPLS), static autoconfiguration, application models, and more.
Several packages, including xMIPv6 and VoIPTool, were once separate projects but have since been integrated into INET.
Wired LANs and WANs
The INET Framework is the recommended model library for simulating wired networks. INET provides models for PPP, and a wide range of Ethernet variants and features (including full duplex and half-duplex modes, cutthrough switching, and frame preemption), along with network devices built from them (switches, routers, access points, etc.). Support for switched networks is included, offering VLAN, spanning tree protocols (STP, RSTP), and TSN support.
Time-Sensitive Networking (TSN)
The INET Framework provides comprehensive support for Time-Sensitive Networking (TSN), which is a set of IEEE 802 standards aimed at making Ethernet networks deterministic with low latency and high reliability.
Key TSN capabilities supported by INET include:
- Time Synchronization: INET implements modules for the IEEE 802.1AS standard, which synchronizes clocks across network nodes to ensure coordinated communication. This is essential for ensuring accurate timing in time-critical applications such as industrial automation and automotive systems.
- Per-Stream Filtering and Policing: This feature, compliant with IEEE 802.1Qci, allows for the control and shaping of traffic streams. It can enforce bandwidth limits and prevent misbehaving streams from affecting the network.
- Scheduling and Traffic Shaping: INET supports both Time-Aware Shaping (IEEE 802.1Qbv) and Credit-Based Shaping (IEEE 802.1Qav). These methods help ensure low-latency, prioritized traffic transmission and reduce congestion.
- Frame Replication and Elimination for Reliability (FRER): This feature, defined in IEEE 802.1CB, enables the replication of critical frames across redundant paths to ensure reliability, which is particularly advantageous in systems where packet loss is unacceptable.
- Frame Preemption: With the implementation of the IEEE 802.1Qbu standard, high-priority traffic can interrupt lower-priority frames, improving latency for critical data.
- Cut-Through Switching: INET incorporates cut-through switching to reduce latency by initiating the forwarding process before the complete reception of a frame.
- Automatic Gate-Schedule Configuration: INET offers mechanisms for configuring gate control lists used in time-sensitive switches, which can be accomplished manually or using external solvers like SAT-based gate scheduling.
These features are modular and can be combined in various ways, enabling users to simulate complex TSN environments and explore different network configurations tailored to specific applications like automotive or industrial networks. The INET site contains several showcase examples demonstrating these features.
We recommend using the INET Framework for TSN simulations over other TSN model libraries.
Wireless Communication
The preferred choice for simulating wireless LANs with OMNEST is the INET Framework. The INET Framework features a sophisticated WiFi model with a modular architecture that closely follows the IEEE 802.11 standard. Other available protocol models include IEEE 802.15.4 and WSN protocols such as L-MAC and B-MAC. LoRa is available as a separate package (FLoRa). Abstract communication models for high-level modeling are also provided.
Key capabilities for wireless simulation in INET include:
- Optional detailed signal representation (time/frequency domain)
- Optional bit/symbol-level radio models
- Various modulation schemes
- Antenna models with optional directional selectivity
- Various path loss models
- Obstacle loss models
- Reception error models
- Coexistence/crosstalk modeling
- Background noise models, noise sources
- Abstract models such as unit disk for high-level modeling
Mobile Ad Hoc and Wireless Sensor Networks (MANETs, WSNs)
The recommended model for simulating mobile ad hoc networks (MANETs) and wireless sensor networks (WSNs) using OMNEST is the INET Framework. INET offers the required infrastructure for this domain, including mobility support, detailed radio models, low-energy MAC protocols, energy consumption, and battery models, among others.
An alternative choice is the INET fork named INETMANET, which is a superset of INET with many experimental (sometimes very experimental) MANET-related additions. Other options, including the MiXiM, Castalia, LSU SenSim, etc., are now considered obsolete.
Cellular (5G/LTE) Networks
Simu5G is an open-source 5G/LTE simulation framework that focuses on the data plane. As the successor to the widely-used SimuLTE 4G simulator, Simu5G introduces 5G New Radio access while maintaining its predecessor's strengths. Geared towards facilitating 5G research, Simu5G is fully customizable with a simple pluggable interface, allowing users to develop new modules for implementing custom algorithms and protocols.
Simu5G supports the simulation of scenarios where 4G and 5G coexist, including StandAlone (SA) and E-UTRA/NR Dual Connectivity (ENDC) deployments. Simu5G seamlessly integrates with the INET Framework, and can be used together with other model libraries, such as Veins for vehicular mobility simulations.
Vehicular Networks
A variety of open-source models for vehicular network simulations are available for OMNEST. Veins, which integrates OMNEST with the SUMO road traffic simulator using co-simulation, is a widely-used option for modeling Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communication, leveraging standards like IEEE 802.11p and DSRC for realistic simulations. Veins is typically used together with the INET Framework.
For more specialized needs, Artery extends Veins for V2X communication based on the ETSI ITS-G5 standard, making it ideal for cooperative driving scenarios. PLEXE enhances Veins with platooning capabilities, enabling detailed simulations of autonomous vehicle coordination. Veins may also be combined with Simu5G to incorporate cellular communication for long-range V2V and V2I studies. Veins VLC extends Veins with the ability to simulate Vehicular Visible Light Communication (V-VLC) alongside the default IEEE 802.11p communication, and includes realistic headlight and taillight path loss models.
In-Vehicle Networks
To simulate fieldbus communication protocols such as CAN and FlexRay, which enable reliable communication between electronic control units (ECUs) within vehicles, the FiCo4OMNET library can be utilized. FiCo4OMNeT allows engineers to simulate and analyze the performance and reliability of in-vehicle networks, assisting in the development of robust automotive communication architectures.
For modern car architectures that leverage Ethernet, Time-Sensitive Networking (TSN) is highly relevant (see the TSN section above). TSN facilitates deterministic real-time communication over Ethernet, which is vital for advanced driver-assistance systems (ADAS) and autonomous driving features.
Quantum Communication
The Quantum Internet Simulation Package (QuISP) offers a state-of-the-art, event-driven simulation of quantum repeater networks, envisioned to serve as the backbone of the future Quantum Internet. QuISP aims to model a full-scale Quantum Internet with up to 100 networks, each comprising up to 100 nodes. Its primary focus revolves around protocol design and the emergent behavior of complex, large-scale, heterogeneous networks, while maintaining a high level of realism in the physical layer.