Joint Optimization of Hybrid ARQ Performance in Wireless Packet Data Systems

We propose an analytical framework that optimizes the performance of various Hybrid Automatic Repeat reQuest (HARQ) schemes in wireless packet data systems. The optimization framework lends itself to maximizing the user throughput while the redundancy size per retransmission, the packet error rate, and the mapping of signal to interference plus noise ratios (SINR) into modulation and coding scheme (MCS) levels are jointly decided. After deriving a succinct throughput expression covering incremental redundancy (IR) scheme as well as Chase combining (CC) scheme, we show that there exists an optimal redundancy size for IR scheme when the number of retransmissions is limited, while on the other hand as small as possible redundancy size is optimal when the number of retransmissions is unlimited. Through extensive simulations of the IEEE 802.16e system which verify our analytical results, it is shown that IR scheme significantly outperforms CC scheme even with a practical packet error rate constraint.