The use of 8.33kHz spacing to solve the European VHF frequency congestion problems was proposed at the EURRAN meeting in Vienna in 1994 and this plan was then endorsed at the COM/OPS meeting in 1995 and subsequently adopted by ICAO. It present 8.33kHz spacing will only be implemented in Europe as a need for this is not foreseen in the rest of the world.

The implementation plan

A plan for the implementation of 8.33kHz in Europe has been developed by the ICAO Frequency Management Group (FMG). The plan has been approved by EATCHIP and EUROCONTROL and steps are now in hand to introduce the plan.

The key points of the plan are:

Introduction will be in the core area of Europe from 1st Jan 1999. There is a transition period during which existing 25kHz frequencies will be converted to 8.33kHz from this date until 31st Dec 1999. During this transition period, the plan states that, if there is a problem with an 8.33kHz frequency (e.g. interference), it would be possible for that frequency to go back to 25kHz. It is not clear how realistic that will be in practice. It will at least as a minimum require both sets of ground equipment to be maintained in working order during the transition.

The States involved are: Austria, Belgium, France, Germany, Luxembourg, Netherlands, Switzerland, UK.

8.33kHz spacing will be introduced above FL245 in all states except France where it will be required above FL195. This is a cause for concern for controllers in countries surrounding France as they will have to determine if a non-equipped aircraft can continue at FL240 or if it will enter France at any point in which case it must descend to FL190. This appears to make an already complicated situation even more difficult.

The carriage of 8.33kHz airborne equipment in these areas above these flight levels will be mandatory for all aircraft except for aircraft granted State exemptions.

8.33kHz spacing will take place in a dedicated sub-band from 132.000mHz to 134.800 mHz

Aircraft should indicate 8.33kHz equipped on the flight plan starting 3 months before the implementation date. There is the intention to try to measure the amount of non-equipped aircraft in this period to judge the amount of additional traffic that will be expected to fly below the minimum 8.33kHz level.

CFMU/IFPS will process the flight plan and check 8.33kHz status. If the aircraft files above FL245 and the airport of departure is within the IFPS zone then the flight plan will be rejected. If the departure airport is outside the IFPS zone then a warning message will be sent. It is not clear if the system will cope with the need to recognise FL195 if going over France.

The frequency numbering arrangements are shown in Appendix A. This system results in controllers working with 8.33kHz frequencies having to say longer numbers (normally only one more number) then at present. It is hoped that this will only be a transitional problem as it was getting used to the new numbers when the spacing was reduced from 50kHz to 25kHz.

The use of the longer numbers is supposed to indicate to the pilot that he is being transferred to an 8.33kHz frequency. If the aircraft is not equipped, it will not be possible to select the longer frequency number given. There is clearly a problem where the pilot is told to contact e.g. 132.055 and selects 132.05 either because he did not hear the last number, plain old finger trouble, or the pilot of a non-equipped aircraft does not understand and just selects the closest number he can find. In this case it is possible for an aircraft to cause interference (see below).

Channeling/frequency pairing :

This should properly be referred to as frequency blocking. If an aircraft in the 8.33kHz area selects a 25kHz frequency in the dedicated 8.33kHz band and transmits on it he will block the 8.33kHz frequency he selects and one or two 8.33kHz frequencies on each side. The controller and other aircraft will hear a blocking tone and probably will not be able to hear who is causing the problem.

There are a number of ways this could happen. It could be a selection of the wrong frequency by mistake as I outlined above. This will undoubtedly happen. Another scenario is an aircraft flying below the 8.33kHz area (possibly with a 25kHz only radio, but it could just as easily be with an 8.33kHz set) who is instructed to contact e.g. 122.02 and mishears or selects 132.02 by mistake.

It would appear that there are no protection measures against this. The impact of this problem is not clear but it seems certain that there will be more frequency blocking problems. Firstly, there are more possible mistakes to make and secondly because one transmission will block several frequencies.

It is very difficult to establish what quality to expect from 8.33kHz spacing. It is known that the bandwidth is less than 25kHz, particularly at the top (high frequency) end. It is generally excepted that a woman’s voice comes over more clearly on RTF because it is higher pitched than a mans. With age, the high frequencies in the hearing are lost first. This could be a problem for older controllers and pilots but it would appear that nothing has been done to investigate this.

A trial has been done by the American National Standards Institute. This takes the reduction in bandwidth and converts it into the number of syllables in speech that could potentially be misunderstood. (Speech comprehensibility). It compares 25kHz and 8.33kHz and the results for 8.33kHz are 70% and 100% worse. This does not necessarily translate directly into errors, but if it did then the number of errors on 8.33kHz would be twice as much as with 25kHz. As a result of this it is expected that there will be a significant increase in “say again” type events as less words are understood.

The UK NATS and German DFS reports referred to in the transition plan state that the quality is acceptable or no worse than 25kHz. This is hardly a glowing recommendation! The nature of the UK trials is not known. The German report acknowledges that the specification is at the lower end of what could reasonably be expected due to the need to balance acceptable quality with reasonable costs. It also acknowledges the reduction is bandwidth and speech comprehensibility. The report concludes that these reductions will be accepted by pilots and controllers as they become used to the system and that there will be no reduction in flight safety as a result.

Information on the German trial appears to indicate that the test was from one ground transmitter site to one ground receiver site. If the test was ground to ground this cannot be considered a realistic operational trial. There is also concern from those who took part in the trials that a tape recording was used, not live transmissions. The test was also done in laboratory conditions with no background noise or distractions.

A report on the introduction of 8.33kHz by the German Pilots Association expresses many similar concerns. This is an extremely expensive ‘stop gap’ system and other options for resolving the problems may not have been fully explored. Even if the change to 8.33kHz spacing is essential then realistic trials should take place to include factors such as flight deck noise and atmospheric interference. It considers that any reduction in speech comprehensibility is unacceptable. It is also concerned that the system will be introduced with little testing resulting in pilots and controllers having to do this testing in a live traffic environment. The report considers that the implementation of 8.33kHz spacing should be delayed until these problems have been resolved.

States need to develop data processing systems to display the aircraft equipage to the controller. How reliable will this information be in real life? If an aircraft goes u/s at the last moment and is replaced with a non-equipped aircraft it is doubtful that a change message will be sent. There will be extra workload for surrounding countries in descending the non-equipped aircraft to avoid the 8.33kHz area and they will also need the flight planning and data display equipment.

What will be the proportion of non-equipped aircraft that we will have to deal with? It seems likely that there will always be some non-equipped aircraft. Operators with limited financial resources and occasional visitors may not consider it worth equipping for a few flights. There may also be others such as possibly US carriers inbound to London, Paris or Amsterdam who will accept having to descend a bit early rather than have the expense of re-equipping.

The introduction of 8.33kHz will have a major impact on the European ATC system. If the introduction is not carefully managed then the results will be serious. There is the possibility of non-equipped aircraft entering 8.33kHz, the increased possibility of frequency blocking, and the additional workload placed on controllers both around and below the 8.33kHz airspace. All these factors may potentially have an adverse impact on the safety of the ATC system.

The identification and control of non-equipped aircraft relies on the flight planning arrangements working reliably and these systems need to be in place and properly tested before 8.33kHz spacing is introduced.

The quality of 8.33kHz spacing appears to be unproven. It would appear that no realistic operational trials have been done. It is essential that these take place and the quality of 8.33kHz spacing established beyond any reasonable doubt before the system is introduced.

The implementation of 8.33kHz channel spacing should not take place until the speech quality has been tested in a realistic operational environment to ensure that flight safety will not be jeopardised.

To limit controller workload, the procedures and equipment to identify non-equipped aircraft and to deal with the mixed operational environment must be in place before 8.33kHz spacing is introduced. This applies in both the core area of Europe and also in those states that must identify and re-route non-equipped aircraft.