CAJ | 학술논문

针对无线Ad Hoc网络由于带宽限制和自组织的特点，容易导致网络负载过重产生拥塞问题，提出了基于贝叶斯网络的固定码率( CBR)发送速率调整的拥塞控制算法。该算法通过对影响网络拥塞状态的参数进行贝叶斯网络学习；然后，使用获得的贝叶斯网络( BN)对拥塞状态进行预测，根据预测结果对网络的发送速率进行调整来提高网络的服务质量；最后，利用NS2软件对上述算法进行仿真，结果表明改进算法能够使得网络服务质量得到提升。
침대무선Ad Hoc망락유우대관한제화자조직적특점，용역도치망락부재과중산생옹새문제，제출료기우패협사망락적고정마솔( CBR)발송속솔조정적옹새공제산법。해산법통과대영향망락옹새상태적삼수진행패협사망락학습；연후，사용획득적패협사망락( BN)대옹새상태진행예측，근거예측결과대망락적발송속솔진행조정래제고망락적복무질량；최후，이용NS2연건대상술산법진행방진，결과표명개진산법능구사득망락복무질량득도제승。
Aiming at the situation that it is easy for Ad Hoc Network to cause congestion for the overweight load with the bandwidth limitations and the characteristics of self-organization. This paper proposed a congestion control algorithm based on Constant Bit Rate ( CBR) of Bayesian Network to adjust sending rate. Firstly, the algorithm car-ried out Bayesian Network learning from the parameters influencing on its state. Secondly, it utilized the achieved Bayesian Network to predict the congestion state and then adjust the sending rate to improve the network Quality of Service ( QoS) . Finally, NS2 software was employed to implement the algorithm simulation. The simulation results demonstrate that the algorithm can improve the network Quality of Service.