The IoT sector keeps on developing with applications in critical fields such as health, energy,... Effort to produce low-cost low- power devices are considerable in order to fit in the IoT business model. However, in the aforementioned application fields,

tant as cost and power. The issue can be tackled in different ways. The most natural one is the encryption of data. A complementary alternative applies at the signal-level before its transmission. This PhD aims at investigating signal processing techniques to prevent the IoT communication to be detected by an eavesdropper even when located in the neighborhood of either the source or the destination: in these cases, the assumption that the legitimate user and the eavesdropper experience a different channel no longer holds.

One of the key performance indicators in the IoT ecosystem is the energy consumption of the transmitter, especially for standalone battery-operated devices. It is thus necessary to come up with constant envelope signals in order to efficiently amplify the transmitted signal. Other physical properties of the equipment will be ignored in a first time. They could be exploited in a second phase to further enhance the security.