Abstract
The quality of human speech through a communication system is referred to as mean opinion score and forms the basis of the project. It has been utilized as a key benchmark of sound quality over the years, and the need to have an optimal standard has been continuous. It is pertinent to note that GSM networks mainly offer voice services even as data demand has been on the rise. Many operators across around the world operate networks that yield MOS value that is below the standard in some areas. This study will comprehensively elaborate the causes of MOS degradations in some networks and suggest solutions to the problem. The solutions will be based on two main aspects, which include optimization of the radio environment and radio parameters (Hioki, 78).
Introductions
The fundamental service offered by GSM networks across the world is voice service. Multiple aspects influence the quality of services in these networks. However, the broad expectation from communication networks is to provide high performance, reliable and quality services. The complexity of the aspect of QoS in mobile networks is brought about by the radio environments involved in the communication process. Mobility in such networks is also a key contributor to the varying quality of services across mobile networks. However, it is pertinent to note that the user’s perception of the received voice is the most inherent determiner of their satisfaction. The mean opinion score is a pertinent measure used to determine the quality of sound received or perceived by telephone user. The MOS of a network is provided an s the arithmetic mean of multiple scores obtained across different points of the area covered by a network.
This study seeks to investigate the key aspects influencing the field MOS values as well as come up with solutions t enhance the DT MOS. This will ensure compliance with the standard MOS value required to satisfy QOS standards.
- Problem statement
The MOS value significantly influences the QOS of a network operator and several operators are running networks that do not satisfy the standards requirement for MOS. This denies users a high quality services for which they are paying for.
- Justification
Across different network operators, there is a quest to locate base station in an optimal location to ensure that the area of coverage has a high level of QOS. Customer satisfaction is key in every business and voice services providers should make certain that their services are of the utmost value considering the resources they have for network deployment. This area of improving MOS has not been comprehensively researched creating a gap in the field.
- Objectives
The study seeks to achieve the following key elements:
Chapter 2:
Literature review
2.1 Fundamentals of mean opinion score
An evaluation criterion for sound quality is pertinent to the determination of the compliance level as well as the quality of network operators. Evaluation criteria’s can be broadly categorized into subjective evaluations and objective evaluations. The subjective evaluation criterion has been inexistence for several decades, and a listener is expected to give a score of the sound received. The score is dependent on the perception of the listener. It has been argued that subjective evaluation is prone to bias if necessary precautions are not taken. The choice of the listeners is also a pertinent factored in the determination of the MOS value in subjective evaluations (Rocker, 17). In a test caution has to be taken to guarantee that multiplicity in terms of age, gender, as well as a native language influence, is put into consideration. The value selected must range between zero and five. On the other hand, objective evaluation provides an empirical approach that involves calculations to determine the MOS value of a network.
The complexity involved in administering a subjective evaluation triggered the quest by the ITU to develop standardized evaluation criterion. The introduction of objective quantitative algorithms to be used in the determination of MOS value was timely. Some key objective evaluation approaches include the ITU-T P.563 and ITU-T P.862 (PESQ). The P.563 approach was developed back in May 2004. It is pertinent to note that it was developed as a single-end objective measurement algorithm. This implies that P.563 can process only the received audio streams. Over the years, this approach to obtain MOS according to P.563 is commonly used than P.862. To enhance reliability of the data obtained several measurements should be performed and the mean of the scores obtained. The approach is not applicable to individual calls but is preferred for multiple calls (Bullinger, 31).
On the other hand, the ITU-T P.862 (PESQ) was developed through the joint efforts of British Telecom and KPN. However, the efforts were started way earlier it was accepted by the regulator as ITU-T Recommendation P.862 back in 2001. The approach involves comparison of original signals with another degraded signal and deduces through calculations the MOS value. It is pertinent to note that the MOS is similar to one obtained through subjective listening test. This intrusive test algorithm is highly enhanced to handle different radio environments. It is a powerful tool to test the performance of network equipment such as decoder, as well as the end-to-end speech quality (Marco, 93).
There several differences between the two algorithms which are of significance to this study. The differences include but not limited to requirements for the algorithms. For example, P.563 recommendations standard requires the audio file for scoring. This implies that there is no need for a reference file. On the other hand, the P.862 approach requires the original reference file as well as the file processed by the system for scoring. Moreover, the MOS value deduced using to P.563 is mainly based on MOSLQO and MOSLQS. On the other hand, the MOS value, obtained using P.862 is mainly the PESQ MOS.
Chapter 3: Methodology
The methodology to be used will put into consideration several assumptions. These assumptions are necessary to enhance the reliability of the outcome of the study. First, optional parameters involved in cell handover are assumed to have insignificant influence on the degradation of MOS. The enhancement of the MOS is dependent on factors an are interrelated. It is also assumed that the test devices to be used do not have a zero error and none will malfunction during the study.
3.1 Rx Qual versus C/I
It is relatable to note that the ground tests will be carried out using NEMO Handy for multiple C/I ratios. Moreover, the corresponding changes of MOS will be observed measured and conclusions deduced. This is instrumental in ensuring that the analysis will put into consideration the relationship between MOS and C/I. Other things are also put into perspective such as the existing radio environment and its contribution to the decisions of an engineer.
3.4 Transmission quality
The transmission quality is a pertinent parameter that must be put into consideration during the study to ensure that there is uniformity of results. Moreover, the reliability of results will be enhanced by ensuring that transmission quality variance in the networks is not skewed. Field tests are necessary to ascertain the impact of transmission quality on the MOS value (Yang, 59). In instances where there is a good transmission site, a comparative analysis with poor transmission sites will be carried out. The comparative analysis will be pertinent in identification of the parameters in need of alteration and the ultimate solutions to the challenges of the network.
3.5 Test methodology
The test methodology will involve simple field and simulated performance tests that are instrument in identifying the variation of MOS. The variations will be determined relative to changes in the radio environment parameters (Carayannis, 142). Moreover, the factors resulting to the degradation of MOS will be deduced using the simulated performances. The key areas of observation include co-channel interference, handover interferences. Implications of handover cell handover in relation to channel modes will also be investigated using different parameters.
Works Cited
Bullinger, H.. Technology guide: principles, applications, trends. Berlin: Springer, 2009. Print.
Carayannis, Elias G.. Knowledge creation, diffusion, and use in innovation networks and knowledge clusters a comparative systems approach across the United States, Europe, and Asia. Westport, CT: Praeger Publishers, 2006. Print.
Hioki, Warren. Telecommunications. 4th ed. Upper Saddle River, N.J.: Prentice Hall, 2001. Print.
Marco, Marco. Information systems crossroads for organization, management, accounting and engineering ; ItAIS: The Italian Association for Information Systems. Heidelberg: Physica-Verlag, 2012. Print.
Rocker, Carsten, and Martina Ziefle. Smart healthcare applications and services developments and practices. Hershey, Pa.: IGI Global (701 E. Chocolate Avenue, Hershey, Pennsylvania, 17033, USA), 2011. Print.
Yang, Harrison Hao, and Steve Chi Yuen. Collective intelligence and e-learning 2.0 implications of web-based communities and networking. Hershey, PA: Information Science Reference, 2010. Print.
