Resource Optimization For Regenerative Relay Systems

OFDM-based multihop and two-way relay networks can not only benefit from traditional relay techniques but also from optimized resource allocation.Wireless resource allocation for relay networks is currently researched actively in academic field, although a lot of results emerged for multihop and two-way systems, most of them are for non-regenerative strategies. Even with few stud-ies put their focus on regenerative ones, they all based on the assumption that codes are applied on subcarrier basis. Resource optimization for symbol-level regenerative strategies is not well studied. Thus, our work will focus on the optimal resource allocation regime for symbol-level regenerative multihop and two-way relay networks. We will also study the resource allocation problem for code-level decode-and-forward relay where coded bits are mapped across multiple subcarriers. Proposed optimized resource allocation algorithms can significantly improve the system's capacity.The main contribution of this work is summarized as:1. Proposed a BER modified decoding algorithm for OFDM-based multihop and two-way relay networks. By utilizing the probability of error poten-tially introduced by regenerative strategy, a decoding algorithm is pro-posed to modify the LLR values output from the soft demapper. The al-gorithm can greatly increase the chance of successful decoding and boost the end to end performance. It is also discussed on how to calculate and obtain BERs for multihop, 3-step two-way, and 2-step two-way relay net-works.2. Study the resource optimization problem for OFDM-based multihop decode-and-forward relay networks. Based on the model of multi-staged BSC channels, the joint subcarrier pairing (SP) and power allocation (PA)problem is formed. We prove that jointly optimal solution of SP and PA can be reached separately. And in this work, we prove that optimal pair-ing scheme is to match incoming and outgoing subcarriers based on their sorted order of BERs. For either system or individual node power con-straints, we analyze the properties of optimal power allocations and pro-pose algorithms correspondingly. These algorihtms can help to effectively increase the system capacity.3. Investigate the resource allocation problem for OFDM-based 3-step code-level decode-and-forward two-way relay networks. We propose that for 3-step decode-and-forward two-way relays, subcarriers used by two com-municating node in the first two steps (time slots) could be shared and allocated optimally. Two algorithms are proposed: a fast algorithm and a balanced algorithm which can maintain the numbers of subcarriers al-located to two communication nodes are always the same. Next, based on the different power constraints we also study the optimal power allo-cation scheme and propose corresponding algorithms which can signifi-cantly boost the capacity .4. Study the resource optimization problem for symbol-level regenera-tive two-way relays, either for 3-step decode-and-forward or for 2-step denoise-and-forward. For 3-step decode-and-forward, relay will apply network coding (XOR) on the reconstructed symbols, which brings about another degree of freedom that can be exploited for optimization. We pro-pose a greedy subcarrier pairing algorithm as well as two optimal power allocation algorithms for different constraints. The proposed optimization algorithms can greatly improve the system throughput.