Tactical Consequences of Radio Spectrum Out-of-Band Properties

Tactical communications . for ground-based operations requires many co-located communication systems on combat vehicles. Typical frequency bands for such communications are the 30–88 MHz band for army combat radio and the harmonized NATO band 225–400 MHz. As a result of increasing demands of different communication services and larger bandwidth, the amount of co-located communication systems in these bands is continuously increasing. A consequence of this is that the used frequencies will be less separated, meaning that out-of-band properties will be of severe importance for the performance of the individual systems.

Furthermore, another practical consequence is that the possible co-location distance between different combat vehicles is directly dependent on the out-of-band properties of in-going systems. In this paper, we present examples of how these co-location distances can be affected of out-of-band properties and how this in turn will affect the possible separation distances between combat vehicles, the possible communication range and the service available on the radio links. We show the importance of making tactical considerations already in the specification of requirements for out-of-band properties of tactical communication systems.

 

Keywords: Intersystem interference, tactical wireless communications, electromagnetic interference, frequency hopping, co-location, out-ofband emissions.

Introduction

Tactical communications for ground-based operations requires lots of co-located communication systems on the combat vehicles. Typical frequency bands for such communications are the 30 – 88 MHz band for army combat radio and the harmonized NATO band 225 – 400 MHz. In these frequency bands different types of communication systems must co-exist e.g. frequency-hopping (FH) and fixed-frequency systems. Systems for tactical communications designed to be used for larger communication distances are mixed with short-range systems for internal communication in- and between combat vehicles.

 

Furthermore, in Europe the lower and upper part of the 225-400 MHz band is planned to be used for civilian services e.g. 225-240 MHz for DAB and DVB-T2 and 380-400 MHz for TETRA. This means a reduction of the available frequency range for tactical communications. As a result of increasing demands of different communication services and larger bandwidth, the amount of co-located communication systems in these bands is continuously increasing. Moreover, unintentional electromagnetic interference from electric equipment in the vehicles can also affect communication performance. All electronic devices produce radiated electromagnetic interference that can cause severe interference problems for co-located wireless systems. Several interference accidents in military applications have been reported the last decades which highlight the need of controlling this interference in every military operation and platform.

 

Civilian electronic systems are in general allowed to produce considerably higher levels of radiated electromagnetic interference than military specified equipment. This means that an increased amount of so called Commercial off the Shelf (COTS) products in military applications therefore automatically increase the intersystem-interference levels.

 

A consequence of the increasing amount of wireless systems on the platforms is that the communication frequencies used will be less separated, meaning that out-of-band properties will be of high importance for the performance of the individual systems and for the overall wireless network. The increasing amount of radio systems leads to a larger number of antennas on each vehicle. Together with all other practical requirements on a combat platform, optimal antenna locations are not always possible to achieve, which in turn degrades communication performance. Furthermore, another practical consequence is that the possible co-location distance between different combat vehicles will be directly dependent on the out-of-band properties of in-going systems.

 

Altogether, considering the different requirements above, the integration of tactical communications on combat vehicles is a really challenging work that needs careful attention in the early design phases of a military platform.

One fundamental way of reducing intersystem-interference on a platform is to reduce the interference from out-of-band emissions from wireless transmitters. Out-of-band properties for radio systems can be either specified for a certain application or by referring to a standard requirement. In order to determine what requirement that is necessary, dedicated analyses must be done and often a trade-off between the desired properties, possible technical solutions and the economic cost must be done.

 

Typically such analyses are done in the integration work for a certain platform. However, including possible co-location distance between combat vehicles in such analyses is also necessary since there is a direct connection between spectrum properties and co-location consequences. The latter is not so often done to the knowledge of the authors and no open publications have been found on this matter.

 

Intersystem interference can affect wireless communication systems in different ways:

  • communication disruptions,
  • reduced communication range,
  • increased time delay of data,
  • reduced data rate; reduction of possible wireless services on the link,
  • increased range for hostile jammers.

The most difficult interference problems to handle are those that do not give obvious communication problems such as disrupted links. If a communication link is disrupted, the operator recognizes this immediately and can prepare counteractions. However, if the interference gives a more gradual degradation, it is considerably more difficult for the operator to be aware of an interference problem. It is also important to note that every intersystem-interference problem gives a hostile jammer an advantage since the jammer can obtain the same impact on a larger distance than in the case with no interference present at the victim receiver.

 

Thus, out-of-band interference can degrade the communication performance without a clear warning to the user. In this paper we present examples of how possible co-location distances between combat vehicles can be affected by out-of-band properties and how this in turn will affect the possible separation distances in a vehicle convoy, the possible communication range and the service available on the radio links. We show the importance of making tactical considerations already in the setting of requirements for out-of-band properties of tactical communication systems.

 

The paper is organized as follows. In the next section, requirements on out-of-band properties are discussed. Furthermore, the concept of orthogonal frequency hopping sequences is introduced as a technique to avoid frequency collisions when several FH systems are present. Also, the properties of the two frequency bands are summarized. This section is followed by examples of the impact of different out-of-band properties on communication performance for a group of combat vehicles. We show the tactical consequences of out-of-band emissions in terms of necessary separation distances between vehicles, available services or a reduced communication range. Finally, the paper is concluded.

Frequency measures for co-location

Requirements on out-of-band properties

In general, the out-of-band domain is defined to start at a frequency offset of 0.5 times the necessary bandwidth and extends up to 2.5 times the necessary bandwidth. However, for very narrowband and wideband emissions, there may be exceptions for the upper boundary. For these cases the upper boundary is defined in ITU-R SM.1539 [5]. Here, we use an extended definition since performance will be affected by emissions further above/beyond the carrier frequency. This will be shown later in the paper.

 

Out-of-band properties for radio systems can be either directly customized and specified for a certain application or by referring to a standard requirement. For military radio systems, one typical standard requirement of out-of-band properties is the RE103 (Radiated Emission) in MILSTD-461F [1]. The requirement says: “Harmonics, except the second and third, and all other spurious emissions shall be at least 80 dB down from the level at the fundamental. The second and third harmonics shall be suppressed to a level of -20 dBm or 80 dB below the fundamental, whichever requires less suppression.” The requirement is not applicable within the bandwidth of the transmitted signal or within ±5 percent of the fundamental frequency, whichever is larger.

 

For civilian systems, ITU-R SM.1541 [2] covers definitions of various terms used in the context of unwanted emissions, general out-of-band limits for many services and principally describes ways to measure conformance with the limits. However, Out-of-band limits for some modern digital systems are not always given in general, e. g. digital cellular systems, short range devices (SRD), personal communication systems (PCS). ITU-R SM.329 [3] covers unwanted emissions in the spurious domain. It describes measurement methods and contains general limits. Examples of measures to reduce out of-band emissions are

  • spectral efficient modulation schemes,
  • sharper transmitter filters,
  • linear amplifiers.

Furthermore, measures for improved interference performance can also be done in the receiving system e.g.

  • sharper receiver filters,
  • co-location filters,
  • active interference cancellation [6],
  • hybrid spread-spectrum methods [7].

For example, to prevent that the out-of-band transmission of JTIDS/ MIDS terminals interfere with air navigation and flight safety, the system out-of-band emission is strictly specified and the terminals are equipped with an interference protection feature to monitor their terminal transmissions.

 

Reduction of out-of-band emissions is costly so there is always a trade-off between requirements, the choice of measures and economy. The typical way of creating specifications for the out-of-band emissions is by analyzing the impact on co-located systems within a platform or at a certain separation distance. For digital communication systems, the bit error probability (BEP) required for a certain quality of service (QoS) determines the maximum tolerable amount of out-of-band emissions. For instance, if the BEP is larger than, say 10-3, voice may still be unaffected but data transfer is impossible.

 

Furthermore, the relation between the signal-to-interference ratio (SIR) in a receiver and the availability of a certain service is not linear but has a threshold behaviour and typical maximum limits of BEP for some services may be e.g.

  • < 10-2 for voice,
  • < 10-5 for data with moderate requirements,
  • < 10-7 for data with high requirements.

If, for example, frequency-hopping systems with overlapping frequency bands are co-located, the resulting BEP will show a threshold behaviour with respect to SIR, see Fig. 1. In Fig. 1 the resulting BEP caused by co-location of two FH-systems within the 30-88 MHz band is shown, and is calculated according to [9]. The solid line in Fig. 1 represents the BEP when the all out-of-band products are considered, for out-of-band emissions shown in Fig. 2. The dashed line (fundamental only) in Fig. 1 represents the case when only the main spectrum lobe is considered. Each step in the BEP is caused by one of the spectrum lobes so that the lowest step (around SIR~0 dB) is caused by direct collisions of the carriers.

Figure 1: The resulting BEP of two co-located FH -systems, using the same frequency band, will result in a threshold behaviour. Dashed line represents the BEP when only the fundamental frequency in the transmitted spectrum is considered.
Figure 1: The resulting BEP of two co-located FH -systems, using the same frequency band, will result in a threshold behaviour. Dashed line represents the BEP when only the fundamental frequency in the transmitted spectrum is considered.

The second lowest step (around SIR ~ -5 dB) is caused by the largest out of-band components up to 100 kHz (approx.) colliding with the centre frequency of the receiver. The step related to the highest BEP value (around SIR~-40 dB) is caused by the out-of-band components above 100 kHz in the transmitter spectrum. Hence, an improvement of the out-of-band characteristics may not always improve the QoS for a system, since the BEP is affected in steps and the maximum permitted BEP for a certain service must be fulfilled. Thus, an improvement of the out-of band characteristics must be large enough to pass the threshold in BEP for a specific service.

Figure 2: Example of out-of-band emission spectra for a transmitter used in the analysis.
Figure 2: Example of out-of-band emission spectra for a transmitter used in the analysis.

Finally, it is also important to notice that even if a certain limit of the bit error probability is achieved, the maximum communication range will also be affected. We will therefore include such analyses in our scenario examples in the next section and show that a large reduction in communication range can appear.

Frequency-hopping sequences

Co-located frequency-hopping systems often use the same frequency band and are separated by different FH sequences, i.e. stacked nets. The benefits of that all systems use the whole band is that it is harder for a hostile jammer to disrupt the communications if a large band is used. A drawback of using the same band is the interference from other FH transmitters. Hence, for a frequency hopping system, it is beneficial to avoid that several users use the same frequency at the same time.

 

To handle this, the frequency hopping sequences may be designed in a certain way. For sequences of a specific length and number of frequency hopping channels a bound exist of the autocorrelation of a sequence [10] and sequences fulfilling the criterion are called optimal or orthogonal frequency-hopping sequences. Examples of optimal frequency hopping sequences can be found in [11], [12]. However, in reality, the system’s out-of-band properties may aggravate the possibility to obtain negligible interference from other users. Since the transmission spectrum often has a decaying emission power from the carrier frequency and outwards, a possible approach is to put restrictions on how close the used simultaneous carrier frequencies may be located. That is, the carrier frequency used by a user must be at least of from all other used carrier frequencies at a certain moment, in the network.

 

This approach may handle carrier frequency collision and avoid considerable interference from other user’s out-of-band emission in your own network. Other radio networks, however, cannot be coordinated in this way. Such situations stress the need for error correction that is implemented over several frequency hops. The problem with stacked nets is acknowledged in [8], where it is

highlighted that the number of concurrent hopping patterns is limited to 20 before error rates produce significant impact in JTIDS/MIDS. If additional interference appears, the number of nets needs to be reduced further.

Military frequency bands

The military band 30-88 MHz is typically used for frequency-hopping, military army combat, radios. Such systems usually have 25 kHz channels and a relatively large number of frequency channels to hop between. The equipment is often mounted on military vehicles or in masts and the co-location issues may arise when several antennas are grouped together. In co-operation involving collaboration between nations or when several nations are sharing the same camp, interference issues is likely to appear. Another demanding situation is several vehicles moving in convoy.

 

The harmonized military frequency band at 225-400 MHz is used by a large number of different military radio systems. The bandwidth of the systems are usually wider than in the 30-88 MHz band and for FH systems, bandwidths in the order of 1 MHz is not unusual, with the consequence that the number of frequency channels becomes less. This band accommodates FH systems (Saturn, HQI/II with future versions, national versions of army combat radios), as well as fixed frequency systems (Marlin, national versions of hand held radios). In addition, in some countries also other systems are allocated in parts of this band, such as Tetra systems. There is also a great pressure from completely civilian services to be able to use parts of this band and, for example, in the lower part of the band radio and/or television transmissions are already in use in some countries. In general, the 225-400 MHz band involves more different actors in the same frequency band than in the 30-88 MHz band, with actors from the army, the air force and the navy, as well as civilian actors.

 

In these bands, emission from unintentional interferers is very common. It can for example be interference from engines, electronic devices, in particular civilian computers and other transmitting devices, such as radar systems and electronic warfare equipment. The frequency utilization of the radio systems is often preplanned and coordinated. Nevertheless, due to out-of-band properties of other radio systems, unintentional interference and unexpected co-location situations, interference from other radio systems cannot be avoided and interference issues appear. Altogether, radio systems in these bands will be exposed to sever interference problems, which may even increase in the future.

Tactical consequenses

Analysis tool genesis

For the analyses of the impact from spectrum properties on co-location distances, communication range and service availability, the research tool GENESIS is used. GENESIS is a software research tool developed for intersystem interference analyses of military camps and mobile platforms. The main features of GENESIS are very briefly summarized below.

  • Analysis methods dedicated for modern digital telecommunication systems have been developed to prevent intersystem-interference problems in military and homeland-security applications.
  • Parameter-reduced calculation models for fast analysis. Such models are used both for electromagnetic modeling and modeling of interference impact on digital radio receivers.
  • Interaction with the network analysis tool OPNET Modeler in order to analyze the intersystem-interference impact on large communication networks.

In GENESIS, the operator works in an advanced 3D graphic environment, see Figure 3. All objects are stored in a data base with all relevant parameters (object attributes) connected to every object. If one wish to alter any parameters that can be done directly in the object. The objects consist of radio systems, antennas, vehicles, containers etc and with all relevant parameters specified. GENESIS has been tested and evaluated by military officers and military engineers and is now used for education and intersystem interference analyses. In GENESIS, the quality of the communication is visualized by different colors, such as good, poor or failing (green, yellow or red) communications.

Figure 3: A snapshot of the operators view in GENESIS . In [4], GENESIS is used to determine necessary co-location between medical electrical equipment and tactical radio systems at military camps, not to cause interference.
Figure 3: A snapshot of the operators view in GENESIS . In [4], GENESIS is used to determine necessary co-location between medical electrical equipment and tactical radio systems at military camps, not to cause interference.
The acceptable BEP for a system depends on the supported service; e.g. voice communications can often tolerate a higher BEP than data communications. For each system and service, two BEP levels are set in GENESIS and used to show the link quality by different colors. In GENESIS, calculations of BEP can be performed over an area. The results can show resulting BEP for different placements of a receiver or of an interfering equipment. The former can be used for example when evaluating suitable places for a receiver at a camp with many interfering equipment. The latter is often used when analyzing which distances between a receiver and certain equipment that are needed to fulfill a certain requirement.

Co-location distances for combat vehicles

Two fundamental scenarios are used:

  1. Three combat vehicles containing frequency-hopping systems for communication in three different networks in the 30-88 MHz band.
  1. Two combat vehicles containing frequency-hopping systems for communication in two different networks in the 300 MHz band.

For each scenario, the necessary co-location distance to achieve the required BEP is determined as well as the reduction of communication range. The transmitter spectrum in Fig. 2 is used in the evaluation of the convoy radio performance when the 30-88 MHz band is used. In this scenario, the instantaneous bandwidth is 25 kHz. The out-of-band emission is evident in several channels outside the intended frequency band; 400 kHz consists of 16 separate frequency hopping channels. In Fig. 4, a convoy of three vehicles is shown and the analyzed receiver is in the middle vehicle. The communication system is a FH system with 2320 hopping frequencies in the band 30-88 MHz.

 

Without any co-located vehicles, the BEP is 6∙10-20 and the quality of the communication is excellent. However, when two radio systems are co-located with the receiver the BEP increases and the communication is disrupted.

 

Both the transmitters have transmission spectra shown in Fig 2 that fulfills RE103 in MIL-STD-461F. All systems are using the same frequency band and can accidentally use the same hopping frequency at the same time. This is expected, since the co-location is close and the out-of-band emission levels are high. Moreover, since the number of FH-channels is limited, the probability is relatively high for collisions. This results in a large number of disturbed frequency channels. Orthogonal frequency hopping patterns can improve the performance by separating the hopping patterns by a number of channels.

 

Here, the patterns are separated by 8 channels, with the result that the BEP decreases to 8∙10-3, and communications are possible but with degraded performance. However, the communication range for the system with good communication quality is severely degraded; only 6% of the range for the original link is obtained. A way to further improve the performance is to attenuate the transmitter spectra. If the spectrum is modelled for the same frequencies but with a level of -80 dB, the BEP is reduced slightly to 6∙10-3 in the receiver if orthogonal hopping patterns are used.

 

However, a larger improvement is obtained on the range of the communication link, which now is 25 % of the maximum range. The reduction in range is large since a large improvement in SIR is needed to achieve the BEP; this is also illustrated by the BEP performance for a FH system shown in Fig. 1. The area in which the communication link quality is degraded is shown in Figure 4. The necessary distance between the interfering transmitters and the receiver is about 40 meter. In the UHF band, the number of available frequencies is more limited.

Figure 4: Co-location of three VHF radio systems analyzed in GENESIS .
Figure 4: Co-location of three VHF radio systems analyzed in GENESIS .

Here, it is assumed that the frequency hopping systems use a frequency band of 100 MHz, i.e. 80 different channels for a bandwidth of 1.25 MHz. The transmitter spectrum is modelled as an attenuation of 80 dB at the 2 nearest channels at both sides of the fundamental. The co-location results in a BEP of 6∙10-3 and negligible communication range for the communication system. When orthogonal frequency hopping is used and the direct collisions are avoided the BEP is 9∙10-5 and the range is 40% of the maximal range for a system without interference, see Fig. 5. The necessary separation between the systems is shown in Fig. 5, showing that a separation of about 40 meter is needed.

Figure 5: Co-location of two UHF radio systems analyzed in GENESIS .
Figure 5: Co-location of two UHF radio systems analyzed in GENESIS .

The impact of intersystem interference is more severe for the UHF system than for the VHF system, since the UHF system have fewer FH channels. This illustrates the need of a fair amount of frequencies to perform frequency hopping on if the performance should not be severely degraded.

Conclusions

We have shown that possible co-location distances between combat vehicles can be heavily affected by out-of-band properties of the wireless communication systems on board. We have also shown how the co-location will affect the possible communication range and the service available on the radio links. The results show the importance of making tactical considerations already in the specification of requirements for out-of-band properties for tactical communication systems.

 

The paper shows that there can be a significant impact of the radio transmitter’s out-of-band emission and that, without coordination of the used frequency patterns for frequency hopping systems, the intersystem interference between the radio networks can be crucial. This is a consequence of the fact that the number of frequency channels is in practice limited. This raises the need for:

  • coordination of the frequency hopping sequences, including optimal sequences, coordination between systems and frequency separation of used frequency channels of frequency hopping systems.
  • strict limits on the out-of-band emission of the transmitter, which can be controlled by proper radio design and filters. Here, relevant design and spectrum requirements are important, taking tactical scenarios and the possible platform configuration with other systems into account.

These challenges need to be considered in the design phase of the development of the radio system. This will be increasingly important since the number of co-located systems is expected to increase at the same time as the available frequency bands are strictly limited.

 

Peter Stenumgaard, Karina Fors, Kia Wiklundh, Sara Linder
Swedish Defence Research Agency, FOIDept. of Robust Telecommunications, Sweden

 

References

  • [1] MIL-STD-461F, ”Requirements for the control of electromagnetic interference characteristics of subsystems an equipment”, 10 December 2007.
  • [2] ITU-R SM.1541, Unwanted emissions in the out-of-band domain, 2001
  • [3] ITU-R SM.329, Unwanted emissions in the spurious domain, 2012
  • [4] Peter Stenumgaard, Karina Fors, Kia Wiklundh, Sara Linder “Interference on Tactical Radio Systems from Collocated Medical Equipment on Military Camps” IEEE Communications Magazine, Vol 50, no 10, pp, 63-69, October 2012.
  • [5] ITU-R SM.1539, Variation of the boundary between the out-ofband and spurious domains required for the application of Recommendations ITU-R SM.1541 and ITU-R SM.329, 2002
  • [6] Zahedi-Ghasabeh, A.; Tarighat, A.; Daneshrad, B., “Active Interference Cancelation for user coexistence in the presence of I/Q imbalance,” Proceedings of IEEE MILITARY COMMUNICATIONS CONFERENCE, 2010 – MILCOM 2010, pp.261,265, Oct. 31 2010-Nov. 3 2010.
  • [7] Olama, M.M.; Killough, S.M.; Kuruganti, T.; Carroll, T.E., “Design, Implementation, and Evaluation of a Hybrid DS/FFH Spread-Spectrum Radio Transceiver,” Proceedings of IEEE Military Communications Conference (MILCOM), 2014, pp.1368-1373, 6-8 Oct. 2014
  • [8] C. Kopp, “NCW101: An introduction of network centric warfare, part 3”, Strike Publication, Air Power Australia, 2008.
  • [9] Sara Linder, Karina Fors, Kia Wiklundh, Peter Stenumgaard,” Intersystem interference model for frequency hopping systems”, Proceedings of EMC Europe 2012 International Symposium on Electromagnetic Compatibility, Rome, Italy 17-21 September 2012, pp. 1-4.
  • [10] A. Lempel and H. Greenberger, “Families of sequences with op timal hamming-correlation properties,” Information Theory, IEEE Transactions on, vol. 20, no. 1, pp. 90–94, Jan 1974.
  • [11] R. Fuji-Hara, Y. Miao, and M. Mishima, “Optimal frequency hopping sequences: a combinatorial approach,” Information Theory, IEEE Transactions on, vol. 50, no. 10, pp. 2408–2420, Oct 2004.
  • [12] W. Chu and C. Colbourn, “Optimal frequency-hopping sequences via cyclotomy,” Information Theory, IEEE Transactions on, vol. 51, no. 3, pp. 1139–1141, March 2005.