Wireless technologies are penetrating all human activities deeper and deeper as these technologies progress, become accepted and are even greedily demanded by societies and consumers. This generates a large variety of devices and use cases, in order to cope with vastly different requirements. Extremely simple, low-cost and low consumption connected objects of the internet of things (IoT) will be deployed, as well as further sophisticated multi-antenna / multi-band ultra high throughput smartphones or access points operating at hugely different frequencies . The radio link quality is, in all cases, dominated by the characteristics of the propagation channel and antennas. This can be translated into attenuations varying over many orders of magnitude, or other more refined performance related parameters. While the impact of the propagation has been studied in detail over several decades since the early GSM developments, much less has been done regarding terminals. Statistical channel models have been developed with increasing complexity, in order e.g. to account for wide bands or multiple antennas (MIMO). However, although the impact of antennas in their use context may degrade the link budget by tens of dB, the modeling of that part of the overall channel is in general very crude, surprisingly. In this presentation, we will address that problem and show how statistical models of antenna performance jointly with the propagation can be devised, intended to represent the variable “effective” performance of a radio link, whether it was for a very simple or a sophisticated device. Among the benefits of such models is the performance evaluation of competing physical layer schemes or device designs. Among the open questions is the trade-off between the representativeness/accuracy of the models and their complexity, which may harness the practical applicability of such models. Further perspectives of the approach towards future 5G networks will be discussed, either regarding the large variety of use cases, or the anticipated technology roadmap (such as mm wave spectral bands).