Partial Decode-and-Forward-Techniken für MIMO Relais-Kanäle
Final Report Abstract
A central aspect of modern wireless communication systems is to provide high-speed and highquality service to a steadily growing number of users without an increase of bandwidth or power. One means to accomplish this is to equip the communication devices with multiple antennas. Another option is to deploy relays, which support the communication between source(s) and destination(s) but do not have own information to transmit or receive. This research project addressed the performance limits of uni- and bidirectional communication in the Gaussian MIMO relay channel, a communication system that combines multi-antenna devices and the concept of relaying, under the assumption that perfect channel state information (CSI) is available at all nodes. For unidirectional communication in both the full-duplex and the half-duplex case, the main goals of this project were to evaluate the cut-set bound (CSB), the achievable decode-andforward (DF) rate, and the achievable partial decode-and-forward (PDF) rate, ideally by means of efficient convex optimization techniques. For the CSB and the DF rate, these goals could indeed be achieved by discovering the hidden convexity of the corresponding maximization problems. The optimal PDF rate, on the other hand, could not be determined for the general case. However, we developed algorithms that use convex programming techniques and find good suboptimal PDF rates. Furthermore, we identified special cases for which the optimal PDF rates can also be determined as the solutions of convex optimization problems. The key step to obtain the desired results for the half-duplex case was the derivation of a dual decomposition approach that allows to jointly optimize the input signals and the time shares assigned to the relay-receive and the relay-transmit phases. By means of this generic method, we were also able to generalize our results for unidirectional communication to bidirectional communication. In particular, we showed that both the cut-set outer bound region and the achievable DF rate region can be evaluated by means of convex optimization techniques and that the underlying dual decomposition approach may be used for designing optimal resource allocation protocols. Towards the end of the project, we also studied the Gaussian MIMO multiple-access relay channel (MARC) with two sources. More specifically, we evaluated the cut-set outer bound region as well as the achievable DF rate region by means of convex optimization techniques that are essentially the same as for the Gaussian MIMO relay channel. An achievable PDF rate region based on Gaussian inputs was also derived, and suboptimal PDF rate vectors were again obtained. Unfortunately, we discovered that the complexity of the optimization problems scales exponentially with the number of sources so that a detailed analysis of the Gaussian MIMO MARC is only possible for a very small number of sources. Finally, it should be mentioned that the thorough understanding of the MIMO relay channel with perfect CSI we gained from this project is an important building block for our follow-up work on relaying with imperfect (transmitter) CSI that is currently supported by the DFG.
Publications
- Optimized Capacity Bounds for the Half-Duplex Gaussian MIMO Relay Channel. In Proc. International ITG Workshop on Smart Antennas (WSA), February 2011
L. Gerdes and W. Utschick
- Optimized Capacity Bounds for the MIMO Relay Channel. In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), May 2011
L. Gerdes and W. Utschick
- On Achievable Rate Regions for Half- Duplex Gaussian MIMO Relay Channels: A Decomposition Approach. IEEE Journal on Selected Areas in Communications, 30(8):1319–1330, September 2012
L. Gerdes, M. Riemensberger, and W. Utschick
(See online at https://doi.org/10.1109/JSAC.2012.120903) - Utility Maximization in the Half-Duplex Two-Way Gaussian MIMO Relay Channel. In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), March 2012
L. Gerdes, M. Riemensberger, and W. Utschick
- A Zero-Forcing Partial Decode-and-Forward Scheme for the Gaussian MIMO Relay Channel. In Proc. IEEE International Conference on Communications (ICC), June 2013
L. Gerdes, L. Weiland, and W. Utschick
- Bounds on the capacity regions of half-duplex Gaussian MIMO relay channels. EURASIP Journal on Advances in Signal Processing, 2013(43):1–17, March 2013
L. Gerdes, M. Riemensberger, and W. Utschick
- Partial Decode-and-Forward Rates for the Gaussian MIMO Relay Channel: Inner Approximation of Non-Convex Rate Constraints. In Proc. IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC), June 2013
L. Weiland, L. Gerdes, and W. Utschick
- Optimal Partial Decodeand-Forward Rates for Stochastically Degraded Gaussian Relay Channels. In Proc. 48th Annual Conference on Information Sciences and Systems (CISS), March 2014
L. Gerdes, L. Weiland, M. Riemensberger, and W. Utschick
