No abstract

In this paper, a survey of the mobile data traffic growth over the last decade is performed, showing historically how data consumption patterns have evolved. Based on the results from the survey and the most important factors for unprecedented traffic growth from last years, a set of user-centric services and scenarios are identified as the most prevalent by 2020. Based on service characteristics and subscribers' behaviour and consumption regarding traffic generation, four user segments are derived. All these factors allow introducing an impact model that characterises user segment's behaviour over post-4G cellular networks, namely, its consistency and stability towards traffic generation, data consumption and service usage, providing mobile network operators with relevant information to predict user segment dispersion, stability and risk. Real and estimated market data are used to test the impact model, and most relevant results are shown. Overall, the proposed impact model aims to bring together technological, sociological and also economical perspectives into a single analytical framework, meeting both mobile network operators and subscriber's expectations of high quality of service with continuous cost decrease.

No abstract

In this article the spectrum occupancy of a GSM900 and a DCS1800 band as an analog power or binary quantized power is modeled. In the case of analog power it is presented histograms of the power distribution during one working day. In the case of quantized power it presents the two time statistics, the time period of opportunities distribution and the time between opportunities distribution. The measurement setup is standing in line of sight with the base station. Also, the measurement setup in terms of maximum sensitivity is described and analyzed. Spectrum non occupancy, for a working day, in terms of total time for the GSM900 band and for the DCS1800 band is given.

Abstract-Direct Sequence Code Division Multiple Access (DS-CDMA) signals exhibit cyclostationary properties which imply a redundancy between frequency components separated by multiples of the symbol rate. In this paper a Multiple Access Interference Canceller (Frequency Shift Canceller) that explores this property is presented. This linear frequency domain canceller operates on the spreaded signal in such way that the interference and noise at its output is minimized (Minimum Mean Squared Error Criterium). The Frequency Shift Canceller (FSC) performance was evaluated for a UMTS-TDD scenario and multisensor configurations, where the cases of diversity and beamforming were considered. All these configurations are evaluated concatenated with a parallel interference canceller (PIC-2D). The results are benchmarked against the performance of the conventional RAKE-2D detector, the conventional PIC-2D detector and single user scenario, and we observe considerable performance gains with the FSC specially for the diversity case and a performance close to the single user case when it was evaluated jointly with PIC-2D.

Increasingly mobile data traffic and high quality service demand has driven fast standards development and new mobile technologies deployment. Traffic demand in 5G networks is expected to rise unprecedentedly, bringing mobile network operators (MNOs) additional challenges, and added pressure regarding carbon footprint reduction. This work aims to study the environmental and financial feasibility of MNOs becoming carbon neutral, by developing biotic carbon dioxide sequestration programs. If feasibility exists, it would be extended and applied to future networks and other environmental scenarios. It is shown that achieving carbon neutrality is possible for heterogeneous deployments, especially when low energy powered base stations like femtocells exist and that the financial costs of such aim might represent little or negligible additional cost expenditure, with the added value of greener and environmental friendly network operation.

No abstract

Abstract-The cyclostationary properties of Direct Sequence Spread Spectrum signals are well known. These cyclostationary properties imply a redundancy between frequency components separated by multiples of the symbol rate. In this paper we present a Multiple Access Interference Canceller that explores this property and applies to UMTS-TDD. This frequency domain Canceller acts in the spreaded signal in such way that minimizes the interference and noise at its output (Minimum Mean Squared Error Criterium). The performance is evaluated in two detector configurations: one including the Frequency Shift Canceller (FSC) and the other plus a Parallel Interference Canceller (PIC). The results are benchmarked against the performance of the conventional RAKE detector and the conventional PIC detector.