Energy efficient resource allocation - Along with other 5G systems, 5G test network will be able to operate in a variety of frequency bands using different radio technologies. The future 5G system will include at least 2.3 GHz and 3.5 GHz bands, and it is very probable that so called cm wave bands (3 - 30 GHz) and mm wave bands (above 30 GHz) will be integrated to the 5G network, too. From network optimization point of view, it is essential to notice that the radio access differs highly depending on the frequency band used. Free space loss, radio wave absorption, available bandwidth, used duplexing technique (TDD or FDD), number of available tx/rx antennas, and even used waveform might differ depending on the used bandwidth. In this task, the optimization of different frequency band usage in 5G network will be studied from EE and SE point of view. Optimization algorithms will be developed, and simulations will be carried out to verify the performance of the resource allocation algorithms.
Uplink Virtual MIMO - In LTE/LTE-A system uplink transmission, multiple users (normally 2) can be paired to form a virtual MIMO. The current user pairing algorithms mainly consider the SE, i.e, to achieve the highest throughput. However, when we take the EE into account as well, we need to re-think the problems. In the first stage, we will mainly answer the following two questions:
1. With the constraint of the SE, where is the upper and the lower bound of the EE when a number of users are paired to form a virtual MIMO transmission? The same question stays for the SE when we take the EE as a constraint. Two-user case will be a good start point, and unlimited number of users will be studied for massive MIMO case in 5G mobile networks potentially.
2. After we know the performance bound, how can we design a practical user pairing algorithm to get closer to the bound? The obvious benchmarks would be at least the round-robin mechanism and the existing algorithms adopted by LTE/LTE-A systems. Moreover, the developed algorithms are expected to be tested or verified in the 5G testbed in Oulu and/or Shanghai.
Also, the triggering mechanism for enabling the user pairing is studied. When is the user pairing activated in the network, and how does the user pairing affect to the performance and energy efficiency of the whole network?
D2D Transmission - D2D transmissions are well studied for LTE-A systems, and have great potential to be used for future wireless social networks and content distributions. User pairing issues for D2D networks mainly include selecting a group of proper users to form a D2D cluster and selecting one or multiple proper user(s) to act as relay node(s). So far, few works have considered the EE aspect for the above pairing issue. In the second stage, we will answer at least the following questions:
1. How will the EE of the whole network change when a D2D cluster is formed or when we use a relay node, and where are the upper and the lower bounds? Many results proved that the SE will be increased under such a case. However the question for EE has not been clearly answered yet.
2. If the EE will be decreased during the process, how can we maintain a good trade-off between the SE and the EE? In other words, new pairing algorithms need to be designed to satisfy the different requirements of different scenarios in the future 5G mobile networks. The developed algorithms are expected to be tested or verified in the 5G testbed in Oulu and/or Shanghai.
Also in this task, the triggering mechanism for forming the D2D clusters is studied. When is the user pairing activated in the network, and how does the user pairing affect to the performance and energy efficiency of the whole network?
Based on the outcomes of the research tasks, we will set up the demonstration to showcase the joint testbed activities between VTT and WiCO. For example, 8K video or demanding on-line gaming could be demonstrated (examples, these will be defined during the project) over both WiCO and VTT testbeds simultaneously taking advantages of the properties of the testbeds as well as carrying out measurements over the data transmission.