中文摘要：

为了最大化正交频分复用（orthogonal frequency division multiplexing,OFDM）无线通信系统中非实时业务的有效吞吐量,提出了一种子载波与数据包长度的联合分配（joint sub-carrier and packet size allocation,JSPSA）策略。该策略较以往的子载波分配或数据包长度优化策略,有更大的优化自由度。基于JSPSA策略,详细分析平坦衰落信道下联合分配算法有效吞吐量函数的单调特征,证明该函数为单峰函数。利用频率选择性信道与平坦衰落信道下有效吞吐量函数单调性特征的相似性,通过引入一种新的等效信道增益,提出了一种频率选择性信道下的JSPSA算法。仿真结果表明,该算法的有效吞吐量接近于JSPSA策略下最优的枚举法,与已有的贪婪比特分配、等误比特率功率分配（greedy algorithm bit allocation/equal bit error ratio power allocation,GABA_EBPA）算法相当,而复杂度却大幅度降低,在复杂度和性能之间取得了较好的折中。

英文摘要：

A joint sub-carrier and packet size allocation（JSPSA） strategy is proposed to enhance the goodput in orthogonal frequency division multiplexing（OFDM） systems.It produces more degrees of optimization freedom than the traditional optimization strategies.Based on the JSPSA strategy,the monotonicity of the goodput function of the joint allocation is studied in the flat fading channel.It is proved that the goodput function is a unimodal function.Furthermore,it is shown that the monotonicity of the goodput function in the selective fading channel is similar to that in the flat fading channel.Therefore,a JSPSA algorithm for the selective fading channel is presented by introducing a new equivalent channel gain.Analytical and simulation results illustrate that the performance of the proposed JSPSA algorithm is almost the same as the enumeration method and is comparable to the existing greedy algorithm bit allocation/equal bit error rate（BER） power allocation（GABA_EBPA） algorithm.The JSPSA algorithm makes a better trade off between complexity and performance by reducing the complexity significantly.