More specifically, there is a need for a fundamental and comprehensive data quality assessment of mobile phone data, which is intrinsically linked to the definition of the key indicators and to the system of variables used. It also includes data description, validation, weighting, sample selection, aggregation, assumptions, and restrictions. At the beginning of the project, it was decided that the statistical data, including mobile phone data, are produced and distributed by the National Statistical Office. In this respect, a number of solutions were identified, the most important of which are as follows:
In case of mobile phone data, there is a need to standardize measurement, in order to assure that the aggregated variables reflect what is really intended by the analysts, the majority of which are not professionals in telecommunications. In addition, the survey construction methods need to be agreed. Thus, the following standard parameters were considered:
In addition, data quality requirements include the assurance of the following aspects. Firstly, certain data are not yet available, either because they are not collected, or because of wrong or corrupted data entry in case of the existing data. In the case of mobile phone data, this is mainly related to the existence of small cells, which have not been deployed as of yet. This data are not necessarily wrong, but they have not been collected in statistical surveys. On the other hand, we should also distinguish between passive and active mobile phone data, which is a further dilemma. Secondly, there is a basic asymmetry between the cost of the data collection and of the data analysis. Since new data are generally collected at a low cost, data analysis is much more expensive, which is why the limitations in data collection impact data analysis even less. This asymmetry is perceived more as a cost factor than as a limitation.
Dr. Chong Liu received his B.Eng. degree from the State Key Laboratory of Radio Frequency Engineering of the Chinese Academy of Sciences, Beijing, in 1988, his M.Eng. degree from McGill University, Montreal, QC, in 1990, and his Ph.D. degree from the University of Toronto, Toronto, ON, in 1995. He was a postdoctoral fellow in the Department of Electrical and Computer Engineering, University of Waterloo, Waterloo, ON, from 1995 to 1997. He is currently a Professor with the University of Waterloo, Canada. His current research interests include mathematical modeling and analysis of wireless communication systems, particularly in the areas of MIMO, multi-user detection, and energy efficiency. He serves as an Associate Editor for the IEEE Transactions on Communications, the IEEE Transactions on Vehicular Technology, and the IEEE Journal on Selected Areas in Communications. He has been a member of the technical program committee of many international conferences, and has been the chair of the technical program committee of a number of conferences, including the IEEE International Symposium on Personal, Indoor and Mobile Radio Communications, 2006, and the IEEE International Conference on Communications, 2009 and 2010. 827ec27edc