Part one of our two-part series on modeling and simulation for 5G networks begins with a look at 4G and 5G networks and how the necessity for standardization has evolved and been implemented to date. The blog also delves into how this increasing pressure on standardization has brought to light the need for modeling and simulation platforms in 5G networks to assist with cyber situational awareness, proficiency training, and network analysis.
Commercial Network Standards
Commercial mobile networks are complex. These networks are typically architected, designed, and deployed across large physical regions, and can span multiple countries. As global cellular networks evolve from current 3G and 4G systems to future 5G networks, they are enabling and proliferating a more ubiquitous all-things-connected world. Nations around the world desperately need an efficient, high-fidelity approach for understanding, evaluating, and ultimately protecting critical infrastructure from cyber threats. Innovative new techniques for Modeling and Simulation (M&S) of 5G mobile networks will enable effective proficiency training, cyber situational awareness, network analysis, and mission rehearsal exercise support for cyber-physical activities. They will also accurately represent and be interoperable with the new frequencies available with 5G (e.g.,mmWave and cmWave) Access Points (AP), smart buildings and homes, mobile-to-mobile, telematics, and sensor networks.
All commercial networks operating today adhere to some standard for communications using well-defined interfaces and protocols. This ensures interoperability and enables seamless communications between connected systems. The mobile network is no exception. Since the early-90s, mobile network interfaces and protocols have been defined and standardized by the Third Generation Partnership Project (3GPP) (www.3gpp.org). This standardization group has provided the basis for interoperability among hardware and software vendors.
Modern networks in use today are on the fourth major iteration of mobile network standards. This is referred to as 4G Long Term Evolution (LTE). These networks are comprised of several different components enabling high-quality voice and data services to subscribers. The Network Elements (NEs) that compose a 4G LTE network is shown in the following diagram. This network depicts various mobile devices connected to an Evolved Packet Core (EPC) via a Radio Access Network(RAN).
Figure 1: Logical view of a 4G EPC
Depending on the size and scope of the network being deployed, each one of the logical NE’s shown in Figure 1 can be anywhere from the size of a pizza box to the size of several refrigerators and be load balanced across several physical instantiations. As networks have evolved today, several of these NEs are also deployed as virtualized instances that exist as software deployed within cloud computing centers.
From the late 2000s through today, 4G LTE and smartphones brought high-speed networking to people, enterprises, and governments on the move. However, increased demand for bandwidth, low-latency, and ubiquitous connectivity requires an evolution towards the fifth generation of mobile networks. As shown in the diagram below, significant complexity has been added to support multi-function 5G networks. Many believe that 5G will be the first network architected to support massive amounts of Internet of Things (IoT) communicating on a machine to machine basis. 4G was all about connecting humans to data. 5G is going to be about ubiquitous communications globally and massive machine to machine communications. From the warfighter’s perspective, this increases both the challenges in performing modeling and simulation as well as being able to understand how the network will perform when perturbated by kinetic or cyber effects.
Figure 2: Logical view of a 5G Core
High Fidelity Modeling and Simulation of Mobile Networks
Modeling and simulation platforms for mobile networks are useful for many reasons: wired and wireless network planning, network analysis, and cyber threat analysis. For example, from a network planning perspective, understanding where to place cell towers and optimizing that placement has a significant impact on cost and operational effectiveness of the network. Operators and researchers continue to analyze networks to create new layouts that optimize cost and efficiency. Being able to first model these networks in a completely simulated domain to understand how radio frequency(RF) may theoretically perform based on terrain and environmental conditions controllable within the network environment is important. The fidelity of these platforms can then be increased with the addition of hardware in the loop capabilities. This allows the user to understand down to the bit level how a system will react and perform under specific conditions. The following diagram provides a notional view of the split between Live, Virtual, and Constructive (LVC) for modeling and simulation of mobile network environments.
Figure 3: Notional View of a Modeling and Simulation Platform providing LVC capabilities
Modeling and simulation of 4G/5G systems are ideal for proficiency training, network analysis, cyber situational awareness, and mission rehearsal for the warfighter. Utilizing a M&S platform that can provide high fidelity hardware in the loop capability for focused areas of a network allows the warfighter to understand to a high-fidelity, bit accurate traffic activity, and system responses when the network is perturbated. In addition, the terrain map and configuration of each live, virtual or simulated NE can be quickly changed to meet scenario requirements. This approach provides a global scope view of the entire network, while also providing high-fidelity understanding of system operation for localized focus areas within the network.
Stay tuned for Part Two of this blog, where we will delve further into techniques for modeling and simulation in 5G environments. This blog is based on a joint whitepaper featuring the integrated CACI LiveRAN and SCALABLE Network Technologies EXata® Modeling and Simulation platform (referred to as the Platform).