Type1 in-band relaying is supported by Long Term Evolution Advanced (LTE-Advanced). Based on the frame structure defined in 3GPP LTE, a down link radio frame is partitioned into access and backhaul sub-frames. In the access sub-frames, the direct link and access link transmissions can take place simultaneously with full frequency reuse. In the backhaul sub-frames, the DeNB (Donor evolved Node B) allocates orthogonal resources for different DeNB->;RN transmission as well as co-scheduled DeNB->;UE transmissions. Macro UE scheduled during the backhaul sub-frames will avoid the interference from RN in downlink, enjoying improved SINR. However, this will decrease the efficiency of radio resource reuse, and potentially reduce cell throughput. In this paper, the performance of relay enhanced networks with partial macro UEs scheduled during the backhaul sub-frames is investigated through system level simulation. A simplified resource allocation algorithm (SRA) which aims to maximize the throughput of worst UEs in relay enhanced networks is proposed to calculate the resource allocation among DeNB->;RN transmission, RN-UE transmissions, as well as DeNB->;UE transmissions in backhaul sub-frames and access sub-frames. Average cell throughput gain and 5%-tile user throughput gain of proposed resource allocation algorithm (SRA) is compared with an exhaustive search algorithm.