| |
|
|
 |
|
 |
Wireless Technologies Research Group
The design of innovative air interface technologies which fulfill the demands on future wireless communications is the key objective of the Wireless Technologies Research Group. Flexible and adaptive solutions with increased capacity and robustness are targeted.
Single User and Multiuser MIMO
The use of multiple antennas in wireless communication systems (multiple-input multiple-output, MIMO) provides significant capacity gains. Multiuser diversity efficiently increases the spectral efficiency and cell throughput where a scheduler allocates resources to the user with the best channel state. Our goal is to develop signal processing schemes for single user and multiuser MIMO which can exploit multiuser diversity and MIMO gains with feasible complexity. For single user MIMO, our activities include closed-loop methods with low feedback. For multiuser MIMO, we consider both capacity-approaching schemes based on non-linear transmit processing and low complexity linear pre-coding schemes.
MIMO Multihop Transmission
A major challenge for 4G systems is to provide coverage of high data rate services. An approach to achieve coverage is to introduce intermediate nodes that act as relays. Our main target is to compensate for the multihop inherent rate loss by simple means of network coding. We pursue two approaches: In two-way relaying, we combine the signals of uplink and downlink at the relay into a signal which is broadcasted to both base station and mobile terminal. In one-way relaying, we combine signals of different users into a single signal. Here, the combination with error control coding as joint network and channel coding is analyzed.
Ad-hoc MIMO Networking
Smart antennas improve the connectivity of ad-hoc networks and the performance due to improved link capacity and interference reduction. We apply a cross layer approach where physical layer and MAC layer are jointly designed. On the physical layer side, we investigate multiple access schemes which allow low-complexity interference cancellation in the nodes in order to cope with interference which might not be efficiently avoided by the MAC protocol. We investigate CDMA and interleave division multiple access (IDMA) as potential candidates and adapt those schemes to the needs of ad-hoc networks.
Decentralized Medium Access Techniques
In a multiple operator scenario where operators dynamically share the available resources, decentralized approaches for interference avoidance are promising. We consider dynamic timeslot allocation based on busy burst signaling. Its basic principle is that receivers upon data reception transmit a busy signal in an adjacent time multiplexed mini-slot. The busy burst serves as a reservation of a certain timeslot in the subsequent frame in the way that all potential transmitters which sense a strong busy burst must not use this timeslot. Thus, the busy burst effectively ensures that ongoing transmissions are protected against inter-cell interference.
Decentralized Network Synchronization
The deployment of decentralized wireless networks is more efficient if the network can be synchronized so that a time-slotted medium access scheme can be implemented. A biologically inspired slot synchronization algorithm is investigated that can accurately synchronize a network without a central controller.
Dynamic Spectrum Sharing
To exploit the available spectrum in an efficient manner, we investigate ways that allow to dynamically distribute resources between operators and to mitigate interference. One of our solutions is based on a spectrum trading approach that is motivated by the stock exchange. Spectrum trading by double auctions enables operators to buy and sell spectrum based on their short-term demands. Furthermore, spectrum trading allows operators to resell spectrum to virtual operators more efficiently.
Cross Layer Optimization
Cross layer optimization is considered between the physical/MAC layers and the application layer.
The objective is to optimize an adaptive multi-user scheduling algorithm such that the user perceived quality of service is maximized. We apply utility functions based on the mean opinion score (MOS), which allow to quantify user perceived quality. Means for parameter abstraction and for reducing the dimension of the problem space are investigated.
|
| |
|
|
|
