SPEAKER: 

Prof. Zhisheng Niu, Tsinghua University, Beijing, China

REGISTRATIONS AND PRICE:

Participants are chosen in the registration order, as long as there is free space in the class room.
This course is free-of-charge for INFORTE member organisation's staff. For others, participation fee is 360 EUR.

COURSE DESCRIPTION:

The continuously growing demands for ubiquitous and broadband access to the Internet brings the explosive development of information and communication technology (ICT) industry, which has become one of the major sources (responsible for 2-10%) of worldwide energy consumption.  In the meantime, we have witnessed a consistent increase in the number of mobile terminals, especially high-end smart phones, which has triggered many new mobile broadband but more energy-consuming standards like 3G, WiMAX, LTE, and LTE-Advanced.  The coexistence of macro-, micro-, pico-, and femto-cells, i.e., heterogeneous cellular networks, has also consumed the scarce spectrum resources in an energy-inefficient way because each cellular network has to provide full coverage to their corresponding users by itself.  As a result, the contribution of wireless networks to the global carbon footprint is forecast to double over the next ten (10) years.

 According to the portfolio analysis of the total energy consumption in a typical mobile network[1], it is reported that nearly 75% comes from the base station (BS) side and, inside a BS, nearly 70% of energy is consumed by power amplifiers and air conditioners in order to keep the BS working (i.e., providing the coverage) even though there is no any traffic in the cell.  Therefore, only through the reduction of transmitting power does not help too much for the total energy savings, and so are the incremental approaches such as slim base stations or smart cooling technologies.  A more ambitious and system-wide solution is expected if some lightly loaded BSs can be turned into sleep mode or completely switched off so that the corresponding power amplifiers and air conditioners can also be shut down during that time.

 In contrary, the existing wireless networks are usually dimensioned for performance optimization without enough consideration of energy efficiency.  Specifically, the so-called worst-case network planning philosophy has been widely adopted in order to provide quality-of-services (QoS) guarantee even during the period with peak traffic.  As a result, during low traffic periods such as nights or holidays or in some sparse spots where the traffic load is temporarily getting very low due to the user mobility, many BSs are underutilized but still, by being active, consume a great amount of power.  Considering the fact that the non-working time (including holidays and night time) is in fact more than half of the year, the wasted energy of the existing cellular networks is remarkable.  This is even more severe for future mobile communication networks where the size of cells will be getting smaller and smaller (e.g., micro- or pico-cellular) in order to accommodate more high data rate users and increase the frequency reuse factor, which will further increase the dynamics of the traffic in a specific cell.  Therefore, it will be very important to have the transmitting power (and therefore energy consumed) of network nodes adapt to the traffic variation, including completely switching off some BSs when the traffic load is lower than a threshold.  However, this will bring a grand challenge to us: how to guarantee the cell coverage if the transmitting power is reduced or a BS is switched off or turned into sleep mode?  How long should the BSs sleep? etc.

 In this talk, we first claim that the networks planning and operation should be more energy-efficiency oriented and, in the meantime, the radio resources distributed over heterogeneous cellular networks should be optimized in a global way, i.e., Globally Resource-optimized and Energy-Efficient Networks (GREEN).  Then, we propose two new frameworks called CHORUS (Collaborative and Harmonious Open radio Ubiquitous Systems) and TANGO (Traffic-Aware Network planning and Green Operation) for GREEN, aiming at increasing the energy efficiency from the system point of view while guaranteeing the coverage and optimizing radio resources as well.  Some key technologies for the migration to CHORUS and TANGO are then presented and evaluated.  Theoretical modeling and simulation studies show that the CHORUS and TANGO schemes can greatly improve the energy-efficiency of the cellular networks, while the quality-of-services (QoS) can be kept at a satisfactory level.  In the end, this talk will give a comprehensive survey of the recent developments from algorithms to practical applications in green wireless communications, and summarize some open problems on critical energy-efficient design issues.

LOCATION

The event will be held in Hotel Arthur, Auditorio.   http://www.hotelarthur.fi/eng/yhteystiedot/

STUDY CREDITS FOR PhD STUDENTS

You can earn 3 credit points by actively participating to the event and  writing a home assignment. More details will be given at the event.