【作者】 贺媛；

【导师】 郑侃；

【作者基本信息】 北京邮电大学 ， 通信与信息系统， 2013， 硕士

【摘要】 随着无线通信的发展和互联网技术的不断创新,移动互联网业务发展迅速,业务类型层出不穷,业务量不断增长,给无线通信系统提出了巨大的挑战。用户体验质量(QoE)是无线业务追求的最终目标。而QoE不仅受到网络质量的影响,它还和用户主观因素以及业务类型等因素息息相关。现有的用户体验质量评价体系主要基于客观的质量评价,且关注的因素主要是网络因素,无法有效的反映用户的真实体验。随着用户对业务体验的质量要求越来越高,现有的QoE研究已经无法满足用户和运营商的需求。基于无线业务QoE的研究现状,本论文提出了针对无线业务的统一的质量评价体系。该体系主要由三个步骤组成：样本采集、数据分析以及模型验证。样本采集包括选取合适的测试样本、主观测试、测试结果的搜集,以及建立样本数据库；基于样本数据库,数据分析主要完成在影响业务质量的各种因素与用户QoE之间建立函数映射模型的工作,并从用户的角度出发,设计不同的建模思路,最终建立适用于不同场景的映射模型；模型验证主要是为了确保模型的精确性,使用测试样本对模型进程测试,确定模型精确性能够满足建模需求。然后,针对统一质量评价体系的关键技术点,进行了科学而详细的论证和阐述。从实时业务和非实时业务两类不同业务的角度出发,分别选取视频业务和网页浏览业务作为各自的实施例,为这两种业务建立了统一的质量评价体系。除了详细描述三个步骤的具体实现细节之外,还描述了两种业务在建立统一质量评价体系时的异同点。通过对实现步骤的描述和对比,既说明了从两种不同业务类型的角度进行建模分析的科学性,也证明了统一建模思想的可实施性。最后,数据分析与模型验证确保了整个建模体系的精确性,也证明了建立合理和科学的函数映射模型的重要性。对比传统的质量评价体系,本论文建立的质量评价体系在各方面性能上都有明显的改善和提升。更多还原

【Abstract】 Recently, with the rapid development of wireless communication and internet technology, wireless services have significantly increased, which poses the great challeges to wireless communication systems. As the pursuit of wireless services, quality of experience (QoE) is not only influenced by network quality but also related to user subjective perception and service type, etc. The existing QoE evaluation systems are usually based on objective quality evaluation, and mainly pay attention to network factors, which cannot effectively reflect the user’s real perception. As users expectation to service quality is growing higher and higher, the existing QoE evaluation systems are unable to meet the needs of users and operators.A unified QoE evaluation system for wireless services is proposed. The proposed system mainly consists of three steps:sample acquisition, data analysis and model validation. Sample acquisition includes the selection of test samples, subjective test, collection of the results, and eventually building a sample database. Based on the sample database, data analysis mainly consists of abstraction of QoE-influencing factors and establishment of the mapping functions between the factors and QoE. This system can be conducted from different perspectives according to users’perception, such as real time services and non-real time services. Model validation uses test samples to validate the model. For real-time and non real-time services, video and web-browsing services are choosen to establish quality evaluation system as examples, respectively. The similarities and differences are described between the two kinds of services in establishing the unified QoE evaluation system. Based on the description and analysis, it is proved that the implementation of the unified modeling on different services is able to obtain the service QoE accurately and feasibily through building rich sample database, applying reasonable mathematical analysis tool, etc. Analysis and model evaluation results demonstrate that the proposed QoE evaluation system attains significantly improvement in all aspects compared with the traditional ones.更多还原