For multi-relay orthogonal frequency division multiplexing (OFDM) systems, efficient subcarrier pairing and resource allocation are significant to boost the system performance. We study the joint optimization problem of subcarrier assignment, subcarrier pairing and power allocation, to maximize the system transmission rate under a total power constraint. Many previous works formulate such a problem as a mixed integer programming problem, then transform it into a convex optimization problem by continuous relaxation and pair all the subcarriers using Hungarian method, which requires high computation complexity. In this paper, we assign each subcarrier in both hops to the relay that has the highest channel gain on that subcarrier. Since the numbers of subcarriers assigned to each relay in both hops, denoted as n1(k) and n2(k), are probably different, we first sort the subcarriers in descending order and pair the front min {n1(k),n2(k)} subcarriers assigned to each relay in sorted way, then we assign and pair the remaining subcarriers using Hungarian method. Due to the decrease of matrix dimensions when using Hungarian method, the computation complexity of the suboptimal algorithm is reduced. Simulation results show that the suboptimal algorithm provides the similar performance to that of the optimal algorithm with much lower computation complexity.