5 Discussion and prospects

To secure the indoor communications, a certain level of confidentiality has to be guaranteed . This will force us to minimise at the maximum the transfer of the information outside the room where the communications take place. We showed starting from a simple example that the use of structured materials could guarantee the confidentiality. The artificial materials structured with one dimension already make it possible to have very good transmission of the visible region and a very high reflection at 1,55 ¼ m. The transmission is therefore lower than ˆ’ 70dB for angles of incidence lower than 20 °. Now, if the angle of incidence exceeds 30 ° the properties of the structure do not meet our aim any more.

To have a rejector filter at 1,55 ¼ m whatever the angle of incidence, it is possible to use more complex structures, initially a combination of multi-layer systems of different optical and geometrical characteristics. Another alternative consists in using structured materials according to two or three dimensions. Photonic crystals with three dimensions make it possible to carry out perfect reflectors whatever the incident angle of the light.

In parallel with the theoretical studies, it is necessary to consider the realisation criteria of such structures. Some of these criteria are the manufacturing cost of the definite structures which will be taken into account in the final choice of the solution. It is evident that one dimensional structures were cheaper and easier to fabricate than their counterpart in two dimensions. It is also necessary to take into account the integration of these structured materials in buildings that already exist or are to be built. Thus, we have to predict such structures which can be constructed during the manufacture of the glazing or to be added to existing glazing. Furthermore, for the choice of the materials used it will be necessary to take into account the reliability of the structure (reliability with respect to thermal variations, pollution, and cleaning product etc).

Until now, and in the preceding discussions we have not taken into account the nature of the connection (the source, the detector, the direct or diffuse transmission link etc). Thus, to completely define the confidentiality level of the system, we should integrate in our simulations the radiation diagram of the source. For a given structure, the effective confidentiality rate will depend on the nature of the link: direct, or diffuse.