Indoor positioning with no added infrastructure than already deployed GNSS satellites or WLAN networks is a constant quest for many years. In the proposed paper, we are making a comparison of the main proposed techniques that are based on a local specific infrastructure in order to allow a standard or future GNSS receiver to work acceptably indoors. When dealing with these approaches, the main problems are no longer the received power level or the difficulty to acquire the navigation message, but rather the multipath and/or the near-far effect. As a matter of fact, one should consider that the multipath problem is also of prime importance for infrastructure-less solutions and that it is probably one aspect of the limited performance obtained.The proposed review starts with pseudolites and the identification of two difficulties: multipath and near-far, counterbalanced by the ability to achieve high accuracy by carrier phase enhanced measurements. Another trouble arose: the need for a rather precise synchronisation between the pseudolites. Of course, solutions have been given to these disadvantages, but leading to a more complex system as originally estimated. In order to solve the problem of synchronisation in a very simple way, repeaters appeared: the signal is picked up from the sky, leaving the synchronisation process back to the satellites. But in order to reduce the multipath effects, a sequential scheme was adopted, making it impossible, or at least very complex, to catty out carrier phase measurements. Thus, the accuracy, although quite enough in order to achieve continuity of GNSS service indoors, can no longer reach decimetre. The latest system is now based on so called repealites that can be seen as a simplified pseudolite system gathering the main advantages of both predecessors: simplicity, sub-metre accuracy and full compatibility with current receivers