[en] The extension of fuel burn-up beyond previously accepted levels is currently being applied in varying degrees throughout the nuclear industry, with the aim of improving fuel economics and reducing the spent fuel volume. So it is necessary that the current fuel knowledge base should be extended. Modifications of fuel rod/assembly concepts, together with fuel management schemes, should be gradually implemented so that the operation of power reactors becomes even more reliable and flexible than it is today. Extrapolation to extended burn-up levels does not cause concern but will have to be made in steps, in order to demonstrate expected performance trends. The fuel testing programmes at the OECD Halden Reactor Project have over the years significantly contributed to the understanding of LWR fuel behaviour in the high burn-up range. A broad range of versatile and integrated in-reactor test rigs and high pressure loops have been developed which allow simulations of LWR irradiation conditions, comparative testing of alternative fuel rod designs and use of test segments pre-irradiated in power reactors. A number of in-core instruments and experimental techniques have been developed for detailed investigations of various aspects related to the thermal behaviour, fission product release and mechanical response of high burn-up LWR fuel rods, under a variety of operating conditions. The paper reviews recent measurements in the area of burnup-dependent steady-state and transient thermal behaviour of fuel rods, intermixing of fission and helium filler gases in the pellet cladding gap, fission gas release kinetics under changing heat loads and power excursions (burst release) and dimensional changes of fuel rods subjected to cyclic load changes. (author). 14 refs, 12 figs