Orthogonal Frequency Division Multiple Access (OFDMA) is the access scheme chosen by 3GPP for downlink in its LTE UTRAN. As OFDMA can provide intra-cell orthogonality, the main interference in downlink is inter-cell interference, which can severely limit the throughput of users near the cell edge. Under limited traffic conditions, fractional load as interference control scheme can be considered, transmitting only in a subset of the available physical resource blocks (PRBs). The inter-cell interference can be reduced avoiding the overlapping of the resources. However, defining the set of PRBs to transmit in each cell is not a trivial operation. In this thesis, autonomous resource selection algorithms are developed to allocate these resources utilizing only the information available internally within each cell. The algorithms proposed are implemented on a system level simulator and tested for a constant and variable number of users, under both low and high loads. The best-performing algorithm demonstrates a significant reduction in inter-cell interference by means of overlapping avoidance and, thereby, a considerable improvement in SINR. Furthermore, a new method proposed allows dynamic adjustments on the cell load based on the instantaneous cell requirements estimated through the feedback of the users.