When Can Battlefield Command Support Systems Within Headquarters Go Wireless?

Introduction

The changing nature of warfare almost guarantees that more military forces will employ a networked command support system (CSS) on the battlefield. As an example the Australian Army employs its Battlefield Command Support System (BCSS) for enhanced capability and situational awareness in field deployments.

In the field, BCSS requires networked computing resources that are often interconnected with fibre-optic cabling. These deployments offer both high security and high data throughput but have disadvantages due to the inherent physical properties of the fibre.

For successful and secure field deployments it is paramount to ensure that any CSS HQ networks are easily and rapidly deployable and have sufficient security protection. Ease of deployment relies to some degree on minimal set up and dismantling times for equipment. Sufficient security protection relies on adequate encryption or protection from interception for the given classification of information being transmitted.

The advantages offered by wireless networks to BCSS HQ deployments such as reduced set-up and dismantling times have been recognised and funding has been allocated for the identification of suitable wireless networking technologies. As a tool, BCSS offers situational awareness, geographical information, intelligence processing, engineering support, combat service support, planning tools and battlefield messaging. This is achieved partially through the transmission of short-term tactical information from sources such as Global Positioning System (GPS) receivers and user contact reports.

As tactical information such as that used by BCSS could easily be put to use by opposing forces, it is imperative to safeguard this data as it is transmitted. While cabled systems are relatively immune to interception from the enemy, wireless networks are inherently more vulnerable since transmissions are broadcast via the radio frequency spectrum. The extent of this vulnerability is examined in this paper.

100BaseF fibre optic cabling offers a high data transfer rate of 100 Mb/s. The data transfer rate of wireless technologies depends on a number of factors including signal coding technique, output frequency and power level and channel attenuation characteristics. Different wireless technologies employ varying methods to overcome these factors and deliver the best possible transmission rate for a given transmission channel. However it is only recently that commercial-off-the-shelf (COTS) wireless equipment has become available that approaches or equals the data transfer rate of a standard 100BaseF fibre-optic system.

Despite the availability of wireless systems approaching the capacity of cabled networks, this paper shows that these may still be unsuitable for use in CSS deployments due to the insecurity of the encryption used. Future technologies are also revealed with some discussion of their potential feasibility for use in CSS HQ.

Current wireless technologies

There are several different COTS wireless technologies currently available that on the surface appear promising for use in a networked CSS. Most solutions are based on the IEEE802.11b standard. Equipment based on this standard has been deployed widely in commercial and domestic environments and is possibly approaching end of life due to encryption problems and low data transfer rates. The recent addition to the market of equipment based on the IEEE802.11a standard shows much more promise for use in the CSS’s as it offers data rates close to the speed of current optic-fibre deployments. Each of these technologies is discussed below.

Standard COTS ieee802.11b based equipment

The IEEE802.11b standard [1] specifies a higher speed extension to the original IEEE802.11 standard [2]. Two transmission methods, frequency hopping spread spectrum and direct sequence spread spectrum are described but comparisons of these schemes are not included this paper as they have little bearing on the equipment’s usefulness for Army field deployments. Attributes that directly affect the value of this equipment for Army field deployments include transfer rate, range and signal security.

IEEE802.11b-based equipment has a peak data transfer rate of 11 Mb/s. Compared to the 100 Mb/s fibre connection this is quite poor and is likely to be insufficient for some of the more resource-hungry CSS functions such as geographical information services (for example transfer of large map files between users). Further, this transfer rate is the peak possible rate with the equipment automatically reducing capacity as range is increased or as the signal is attenuated. This reduction in capacity occurs in discrete steps of 11 Mb/s, 5.5 Mb/s, 2 Mb/s and 1 Mb/s.

Range is dependent on the output power, channel characteristics and antennae employed. The Australian version of this equipment operates in the unlicensed industrial, scientific and medical (ISM) frequency range at 2.4–2.5 GHz. The Australian Communications Authority has limited the maximum output power in this frequency range at 100 mW. Amplifiers and high gain antennae can also be used to boost signal strength and range for a given receive area. Directional antennae are often used for point-to-point building links or where two fixed or semi-permanent points are to be bridged. High-gain omnidirectional antennae are used where the desired broadcast is point-to-multipoint.

The major concern with IEEE802.11b-based equipment is the Wired Equivalent Privacy (WEP) encryption used. Borisov et al have shown [3] that there are serious flaws with this encryption method and that WEP is subject to the four following types of attacks:

• statistical analyses of transmitted packets allowing decryption of traffic;
• the injection of new traffic from unauthorised stations;
• decryption of traffic for unauthorised users by tricking the access points; and
• real-time decryption of traffic through dictionary attacks.

Fluhrer et al [4] have also described a passive cipher-text-only attack against the RC4 encryption as used by WEP. Stubblefield et al have verified this attack method [5] and have proposed optimisations that reduce the time taken to decrypt the secret key to approximately 30 minutes.

There are now at least two open-source software tools publicly available that can crack WEP keys with no special knowledge of the mathematics needed by the user [6,7]. There are also websites such as Net Stumbler Dot Com [8] and Wardriving Dot Com [9] that are dedicated to mapping wireless deployments and exploiting weaknesses in these systems.

These vulnerabilities render WEP an ineffective encryption method and it would be irresponsible to deploy IEEE802.11b equipment with WEP encryption. There are software based methods such as the IEEE802.1x standard that attempt to resolve the WEP issue but these methods are only useful if implemented properly by all attached users. This can entail detailed architecture planning prior to deployment. IEEE802.1x is also subject to several vulnerabilities that a technically sophisticated and determined enemy could exploit [10].

Ieee802.11a-based equipment

The IEEE802.11a standard [11] offers a large improvement in data transfer rate compared to the IEEE802.11b standard. This is achieved through two major changes. The coding technique used at the higher rates is orthogonal frequency-division multiplexing as compared to complementary code keying specified in IEEE802.11b for the higher rates. The other major change is that this equipment operates in the 5-GHz ISM frequency range.

These two changes contribute to the peak data transfer rate being increased to 54 Mb/s. As with IEEE802.11b, there are standard discrete step-down rates specified in IEEE802.11a. These are 54 Mb/s, 48 Mb/s, 36 Mb/s, 24 MB/s, 18 Mb/s, 12 Mb/s, 9 Mb/s, and 6 Mb/s and even at this lowest rate data can be transferred at approximately half of the IEEE802.11b maximum. In addition to this, the company responsible for the first IEEE802.11a-based integrated circuits, Atheros [12], have included a proprietary ‘turbo mode’ in their circuitry that allows operation at up to 108 Mb/s.

Despite this vast improvement in capacity the encryption standard specified for IEEE802.11a is again WEP, but add-on confidentiality systems such as Virtual Private Network (VPN) connections may be applicable for the protection of information classified at higher than Restricted level. A VPN is an encryption system that may be implemented through software and / or hardware and there are several accredited information security products that would satisfy this application, at least in the Australian context [13].

Comparison of cabled and wireless field deployments

It is useful to draw some comparisons between cabled and wireless field deployments. Problems with field deployments of both cabled and wireless systems generally originate from the nature of the propagation medium of each. Understanding the nature of their associated propagation channel is necessary to allow identification of possible solutions to the issues encountered with each technology.

Another consideration when comparing wireless and cabled field deployments is the HQ type, layout and physical distances between and within HQ cells. It is expected that wireless connection would provide the most benefit for communications between HQ cells with separation approximating 100m. As personnel attached to a certain cell would be expected to carry out the majority of their duties within their physical cell precinct, wireless solutions within cells may only provide benefit to physically larger cell layouts.

Issues with cabled networks

Deploying fibre optic cables in the field introduces several issues mainly due to the physical nature of the glass propagation medium. These include:

• manual handling issues such as deployment time;
• connectors and cable are not field-serviceable;
• prone to mechanical damage by vehicles or livestock;
• cable length may not meet all deployment requirements; and
• entanglement and tripping hazards for personnel.

Another issue with cabled deployments is that architecture planning of the deployment is necessary before deployment. Failure to correctly design the network can lead to reduced capacity or to some users not being able to connect with others. Lack of appropriate network design can also lead to extraneous or insufficient equipment being deployed, which in turn affects logistics and support requirements.

Wireless solutions

The majority of the issues with cabled networks mentioned above could potentially be solved by wireless systems. Removing the glass propagation medium gives several advantages, some of which are:

• elimination of point-to-point cabling;
• reduction of set-up and dismantling times compared to wired systems;
• less chance of mechanical damage in the field; and
• greater range than wired systems if appropriate equipment is selected.

Potential issues with wireless equipment

Despite the overwhelming physical advantages offered by changing the propagation medium, wireless systems involve tradeoffs in security and transfer rate of data. The security issues arise because data is transferred at radio or microwave frequencies, is susceptible to interference, and is easily intercepted. Transfer rate is limited by environmental conditions such as vegetation, moisture content of the air and the topography of the area. As the wireless system relies on shared media rather than switching techniques there is also an associated reduction of throughput when compared with wired local area networks employing modern switching techniques.

The issues introduced by electromagnetic propagation of short-term tactical data can be placed in the electronic support (ES) and electronic attack (EA) categories. ES is defined as …actions undertaken to search for, intercept, identify and locate sources of intentional and unintentional radiated electromagnetic energy [14]. ES allows the production of operational intelligence, identification of EA targets and targeting for surveillance resources [15]. EA is defined as …the use of electromagnetic energy to attack … equipment, with the intent of degrading or destroying adversary combat capability [16].

ES issues raised by the use of wireless networking equipment include interception, location and analysis. EA issues include jamming, deception and neutralisation. Each of these issues is explained below in the context of a wireless CSS field deployment.

Interception

It is relatively easy to intercept the signals from current broadcast wireless equipment. This can be a passive attack with no way for the user to know that the signal has been intercepted. This type of attack may provide the opposing force with information about what communications protocols, frequencies and type of equipment is being used. Interception range is dependent on output power, channel characteristics and antenna type. Reducing the output power level for the purpose of reducing interception range presents a trade-off for signal-to-noise ratio (SNR). Similarly, directional antennae reduce interception range unless enemy are within the antennae beamwidth in which case interception range is increased when compared to omnidirectional antennae using the same power output. Balancing output power and antennae directivity is therefore an important step in reduction of interception range.

Location

The physical location of transmitters can be determined using direction finding (DF) techniques. This may provide an opposing force with physical locations of any wireless CSS users. In a CSS-equipped HQ it is probable that other radiators such as trunk-communications and combat-net-radio equipment provide better DF sources. The mechanisms described for interception range reduction also apply to location range although the feasibility of some DF techniques such as triangulation may be heavily reduced by the use of directional antennae.

Analysis

More serious than interception and location is the process of analysis, which involves the transformation of the transmitted data into a replication of the original information. For commercial WEP encrypted systems, packets can be collected from the transmission and the secret key decrypted using basic computing resources such as a notebook computer, appropriate software and a wireless network card. The original information and any information transmitted after decryption can be viewed and used by an opposing force. This type of passive attack cannot be detected and could provide an opposing force with access to all the CSS information.

Jamming

Jamming of wireless transmissions causes loss of packets and the user would find that they receive at a lower transfer rate. Increasing jamming signal strength could cause a complete loss of network connection for CSS users. In addition, to deliberate jamming, significant amounts of interference can be expected in ISM bands as there is no need for operational licenses and hence there are large quantities of equipment operating in these bands.

Deception

If the secret key has been breached through an analysis attack as described above it is possible for an opposing force to authenticate as a legitimate user and inject falsified traffic into the network. In a CSS context this may mean that false contacts are reported or that the messaging system is bombarded with spurious junk mail. Deception can also be practised through less difficult intrusions such as requests for authentication that impede the communications between legitimate users. These disruptions act as low-level denial-of-service attacks.

Neutralisation

If jamming power is high enough the opposing force may burn out wireless receivers. CSS users would then need to resort to more traditional tools and communication methods.

Future wireless technologies

Two future technologies show potential for use in the wireless network field deployments. One is based on improving older technology and the other is a radical departure from current wireless thinking.

The first of these is Harris Corporation’s SecNet11 Secure Wireless Local Area Network equipment [17]. This equipment uses a hardware cryptography module to boost the encryption level to Type 1 accreditation. The US National Security Agency (NSA) has accepted this project as a commercial communications security endorsement program. Although this system negates the effects of analysis and deception-type attacks, the maximum data transfer rate is still only 11 Mb/s. Harris has also proposed a technology roadmap for SecNet11 that includes frequency conversion to move signals out of the 2.4-GHz range and a SecNet54 product based on the IEEE802.11a standard. SecNet11 equipment is in production but will likely not be seen in Australia for several years due to US encumbrances.

The SecNet11 wireless solution offers CSS networks secure wireless transmission of tactical data. This is the primary advantage of SecNet11 over current wireless systems.

The second wireless communications technology that may arrive soon is ultra wide band (UWB) equipment. UWB technology is based on the transmission of very short bursts of radio frequency energy. The resultant waveforms are spread across an extremely wide frequency band (hence ultra wide band). UWB communications systems show promise for wireless field deployments as they offer low probability of interception, high tolerance to noise or tactical jamming and high data transfer rates. UWB technology has already entered service in applications such as UWB radar and through-wall soldier vision systems [18] and a field deployment of a mobile ad-hoc wireless network has also been demonstrated [19]. Spectrum management authorities were reluctant to allow UWB transmissions due to concerns that they may interfere with navigation beacons or GPS signals. The US Federal Communications Commission (FCC) recently approved the use of UWB equipment [20] but imposed tight restrictions on output power levels and operational frequencies to ensure that UWB systems do not interfere with other low power systems. It is expected that relevant authorities for other countries will follow the FCC’s lead and also allow UWB equipment with restricted output. These restrictions on output power may render UWB systems more vulnerable to non-intentional interference and tactical jamming, as SNR is directly proportional to output power in these systems.

It will likely be some time before a UWB solution for CSS networks becomes available, as this equipment is still undergoing testing and evaluation in the United States. In summary the advantages potentially offered by UWB to wireless deployments of a CSS are:

• due to the wide spread of signal and low power output, UWB systems offer low probability of intercept;
• UWB systems have high noise tolerance; and
• UWB systems may offer higher data rates than current wireless technologies.

Conclusions

The glass transmission medium of fibre-optic cabling has several disadvantages when used in Army field network deployments. These relate to manual handling and the fragility of the glass core. Fibre does offer high security and high capacity though, with typical 100BaseF field network deployments running at up to 100 Mb/s.

The majority of the disadvantages introduced by fibre optic cabling are overcome by the employment of wireless technologies. Wireless systems now offer similar capacities and are generally easier to deploy. Despite eliminating the majority of the physical problems caused by fibre-optic systems, wireless systems introduce a new set of issues related to the transmission method. Security through appropriate encryption becomes important as transmitted signals can be easily intercepted. Once intercepted they could be exploited or provide an opposing force with any transmitted tactical information. Attenuation due to topography, vegetation and environmental conditions may cause degradation in signal range and data transfer rates. Eventually this degradation may lead to loss of CSS capability for affected users. Loss of CSS capability in the field may force these users to resort to more traditional communication methods and less overall field capability.

It is believed that appropriate security necessary for wireless transmission of CSS information within headquarters can only be achieved through the use of additional encryption or by future technologies such as SecNet11 or UWB communications.

References

Author

Paul Arcus graduated from the University of South Australia in 2000 with the degree of Bachelor of Electronic and Microengineering with 1^st Class Honours. Paul worked for Saab Systems on the BCSS project before joining the Defence Science and Technology Organisation. Paul is currently studying for a PhD with the Systems Engineering and Evaluation Centre of the University of South Australia. He can be contacted through: paul.arcus@dsto.defence.gov.au