The United States, which already has the use of two broad frequency bands for its GPS, is now seeking a third band in an attempt to meet the particularly heavy demands of civil aviation. The Europeans, who currently only have access to the few remaining bands, which are too narrow, are also seeking several broad bands capable of accommodating Galileo (which has taken over from the L-SATNAV, M-SATNAV and E-GNSS1 projects). The results obtained in early 1999 were hardly encouraging, with only a single band having been identified, the high frequency (5 GHz) of which entailed major technological constraints, and whose limits came far from achieving unanimous support.

However, on 25 January 1999, the Americans surprised everyone by deciding to establish themselves firmly within a band called L5 (1 164-1 188 MHz), which is densely occupied by navigation equipment using DME (distance measuring equipment) instruments, as well as by government equipment (see Figure 1). It gradually became clear that the new signal would only tolerate a limited number of DMEs and that, in several parts of the world, the excess DMEs had to be withdrawn. This seems paradoxical, given that the signal in question is supposed to assist civil aviation! In Europe, which has the highest number of DMEs, operating in L5 could in certain cases be impossible were it to prove unfeasible to withdraw enough DMEs.

Shortly afterwards, on 10 February 1999, the European Commission unveiled the broad outline of its Galileo project. Besides being a free public service comparable to GPS, it is intended to be even more efficient than GPS L5 in meeting air transport needs. Moreover, we have here the introduction of an innovative concept in the form of a made-to-measure commercial service for users with particular requirements.

Unfortunately, owing to a lack of available frequencies, the project was in danger of not being able to fulfil its declared aims. In particular, an L5-type solution seemed out of the question as long as the excess DMEs could not be removed.

However, a means of resolving the problem of compatibility between satellite navigation and DMEs was found in June 1999. It involves a two-bandwidth signal concept (referred to as "broadband/narrow-band"), which allows for signal reception to be adjusted according to the number of DMEs. This means that an adequate radionavigation service can be ensured while avoiding interference to DMEs. Preliminary studies have confirmed the promising potential of this concept.

Since then, two opposing schools of thought on the matter have been confronting each other in the different ITU forums:

In the course of the debates, which have continued for several months and have often been heated, the advocates of the second solution have gradually convinced increasing numbers of the aptness of their approach.

Furthermore, one of the two bands already open to satellite navigation, the so-called L2 band (1 215-1 260 MHz), can readily be extended towards the adjacent band, which is likewise mainly occupied by radars. Protection of the radars has been taken into account in the planned extension, but this raises the question of the protection of these same radars in the initial band; the issue remains open and promises to give rise to some interesting developments.

Finally, after a few tentative trials with current users, a satisfactory band was found in the vicinity of 5 GHz. New horizons have thus been opened up for the frequencies needed by Galileo. In October 1999, the European Commission confirmed the suitability of the frequencies envisaged for its project.

Since the end of 1999 the minds behind Galileo have been highly imaginative in their efforts to make optimum use of the various frequency options available to them. In particular, the broadband/narrow-band concept is surpassing itself by generating a number of unexpected developments. Creativity is in full swing.