Efficient signal processing and Radio Resource Management (RRM) for coordinated multi-node downlink transmission in heterogeneous cellular networks
A significant challenge in the design of fourth generation (and beyond) cellular systems is the reliable delivery of high speed (up to 1 gigabit per second) wireless data to mobile users. The use of techniques involving multiple antennas at the transmitters and receivers is a key step toward this goal. However, interference from adjoining cells is a primary impediment to such techniques reaching their full potential, especially for users in regions near the cell boundaries. A promising way to solve this problem is for several transmitters in a cluster to act jointly and cooperate in their transmissions so as to reduce the interference to users in the cluster. However, such coordinated systems bring design challenges of their own. This is especially the case for emerging heterogeneous networks (HetNets), which may consist of overlapping cells of different sizes, transmitters with different available power, and so forth. The bulk of previous research has instead been on homogeneous networks, where all the transmission points and cells are identical.
One key challenge in coordinated HetNets is the issue of resource allocation. This includes selecting a subset of users to serve at any instant, as generally there are too many to serve all at once. With each user having multiple antennas, the number of data streams to each user must be chosen as well; a user could potentially receive up to as many streams as it has antennas. In a coordinated HetNet, the transmission points in each cluster can also be chosen dynamically. Lastly, the power assigned to each user and the precoding used at the transmission points must be configured to mitigate interference between the scheduled users in a cluster. A second key challenge is the sharing of information between transmission points. It is already a difficult problem to obtain accurate information regarding the current state of the wireless channel in current systems. In coordinated networks, the channel state information must be obtained for and shared among all the transmission points in the cluster. The data to be sent to the users generally must also be available at each transmission point. The design challenge is how much of this feedback is necessary and possible without inundating the system with control/signalling information, which reduces the bandwidth available for useful data transmission to the users.
The focus of this project is therefore threefold: 1) Investigate the design of coordinated cellular HetNets; 2) Design efficient algorithms for resource allocation within such networks; and 3) Investigate the amount and type of feedback necessary within HetNets, and analyze the effect of limited feedback on the network performance. This project is of great interest to the partners, as it will increase their expertise in advanced cellular systems, and allow them to contribute towards future wireless system standards and development.
