|
![]() |
Click for larger image |
Network Centric Warfare, also commonly referred as Network Enhanced Capability, is widely documented and described. The diagram above shows a typical scenario, illustrating the role of various components. Key features include:

The key benefit of HF Radio is that it provides communication over very long distances (worldwide with a suitable aerial). It achieves this with "sky wave" communication, where the HF signal is reflected off the ionosphere (illustrated above) for long distance communication, or using Near Vertical Incidence Skywave (NVIS) for shorter distances. This benefit comes with a number of drawbacks:
There are many alternatives to HF Radio, and it is useful to understand the pros and cons of each. A clear conclusion is that there is no perfect mechanism.
Transmissions at radio frequencies higher than HF (VHF, UHF, EHF, SHF) overcome all of the drawbacks of HF, and can provide high bandwidth data communications enabling key communication capabilities.
The main restriction of all of these frequencies is that they are limited to 'line of sight'. For some technologies, beyond line of sight (BLOS) can be achieved by use of a slightly curved communications path. This gives a little extra distance, but distance is still a key constraint.
Radio technologies provide communication over many miles, except in terrain where line of sight communication is more restricted. Higher frequency communication will be used in preference to HF, and so HF is primarily used for longer distance communication where other radio communication is not an option.
Satellite communication is becoming the preferred choice for long distance communication, offering relatively high bandwidth. However, there are a number of drawbacks:
To overcome the line of site constraint of most radio technology, an alternative to satellite is to put a platform in a 'high' and visible position that can act as a relay between radios that do not have direct line of sight communication. This can be thought of as a low level satellite. The two main options for providing this are:
A major issue for using either approach for long term communication (perhaps as an alternative to satellite) is the vulnerability to attack of the node.
In terms of planning an overall architecture the technology constraints lead to several possible views as to how HF Radio fits into the overall communications picture:
If options 2, 3 and/or 4 are chosen, it is critical that HF Radio is well integrated into the Network Centric Warfare architecture, and that effective application functionality is provided to support a full set of mission critical applications operating over HF Radio.
The core model of IP use is end to end communication with IP running over all of the networking components.
IP is a key element of the Network Centric Warfare architecture, as it allows a wide range of applications to be used over all of the varying network technologies that used. IP is the key standard that joins everything together.
The next sections look at how IP can be integrated with HF Radio. The first approach shown runs IP directly over the HF Radio, treating it like any other network component. While architecturally clean, the performance and management implications of this approach are very bad. A second approach is shown, which optimizes traffic over the HF link, and is the architecture recommended for HF Radio use.

With a direct application of this model to HF Radio, IP is operated over the HF link, leading to the sort of scenario shown above. This is an elegant architecture, and gives the key benefit that endpoints do not need to be aware of network topology, which can automatically adapt to changing situations and availability of communication links. This is a powerful and flexible model that works well for terrestrial networks, satellite and high speed radio links.
Unfortunately, this architecture does not work well for HF Radio. There are a number or reasons for this:
An alternate approach is now proposed as illustrated below:

At the endpoints of the HF connection, this approach uses an application relay or proxy (the exact nature of this component will depend on the application). The key point of this architecture is that the server level components sending data over the HF link are close to the link and can use special protocols that are optimized for sending data over HF. This architecture optimizes performance over the HF link, in a way that is simply not possible with end to end IP.
The protocols used over the link will most likely be based on STANAG 5066. This is discussed in the Isode White Paper STANAG 5066: The Standard for Data Applications over HF Radio. From the overall architecture, the key point is that special protocols are used to optimize link usage, including use of broadcast and dealing with systems in EMCON.
A key aspect of this architecture is that the optimized protocols are at the server level, and that the relay/proxy will support standard protocols on the non-HF side. This means that the HF specific protocols are hidden from the non-HF components. In particular, standard clients and applications can be run on the end systems, so that the HF optimization is transparent to the end user and does not constrain the choice of end application product used.
The application relays provide a number of additional management advantages:
The key feature of this architecture is that it optimizes use of the HF Radio link, which is going to be the limiting performance factor whenever HF is used.
A benefit of the IP end to end architecture is that change of connectivity is transparent to the end user. This can still be achieved with the proposed architecture. To support use of HF Radio, the end applications need to be configured to make use of the Application Relays. Changing this configuration to use different links does not make sense, as it would force the end application to be aware of network connectivity. To make use of faster links, the application relays need to be aware of connectivity, and make appropriate decisions.
There will be three types of behavior, dependent on the application:
This combination will allow changeover between HF and faster links, in a manner that is reasonably transparent to the end applications.
A goal of Network Centric Warfare is to maximize information sharing. While not all applications are suitable for HF Radio, a number of applications are. This section sets out a list of applications that may be mission critical, are suitable for HF Radio, and could reasonably co-exist and share an HF link:
This is not intended as an exhaustive list, but to give a sense of what could be sensibly achieved over an HF link.
The key standards for optimized data transfer over HF Radio are in place. STANAG 5066 is the key HF integration standard, that is supported by many other specifications. There are specifications in place for both STANAG 4406 formal messaging and Internet mail to operate over STANAG 5066.
In order to support deployment of other applications in situations where the application relays/proxies are not provided by a single vendor, it is desirable to develop standards for other applications to operate over STANAG 5066.
Isode is a messaging and directory server vendor. Isode's strategy is to provide a range of application/relay and proxy servers for applications to operate efficiently over HF radio. Isode provides the necessary pieces to make the application/relay and proxy model work.
Isode currently offers support for STANAG 4406 formal messaging, with STANAG 5066 integration.
This paper has shown the key roles that HF Radio can play in support of Network Centric Warfare. It has shown that use of application relays offers substantially improved utilization of HF links and better application characteristics than a simple end to end IP approach. It also shows how Isode’s products will support efficient application deployment over HF Radio.