TY - JOUR
T1 - Stability region of an optimized bidirectional regenerative half-duplex relaying protocol
AU - Oechtering, Tobias J.
AU - Boche, Holger
N1 - Funding Information:
Paper approved by M. Zorzi, the Editor for Wireless Multiple Access of the IEEE Communications Society. Manuscript received July 6, 2006; revised February 4, 2007 and June 21, 2007. This work is supported by the STREP project No. IST-026905 (MASCOT) within the sixth framework programme of the European Commission.
PY - 2008
Y1 - 2008
N2 - In this work, we study a cross-layer design of a spectrally efficient bidirectional relay communication in a threenode network using superposition encoding at the relay node. On the physical layer, a half-duplex relay node decodes-andforwards the messages of two nodes in a two-phase protocol with optimal time-division. On the data link layer, we assume ergodic arrival processes at node 1 and 2 which have queues with infinite buffer length. At the beginning of each time-slot a centralized controller chooses the service rate pair which achieves the weighted rate sum maximum of the instantaneous achievable rate region for the block-fading channel state of the next time-slot with weights equal to the current buffer levels. To this end, the controller adjusts the time-division and relay power distribution. The policy is throughput optimal since the stability region is equal to the bidirectional ergodic rate region. This is because whenever the mean queue length is large, a negative drift of a quadratic Lyapunov function on the buffer levels can be proved.
AB - In this work, we study a cross-layer design of a spectrally efficient bidirectional relay communication in a threenode network using superposition encoding at the relay node. On the physical layer, a half-duplex relay node decodes-andforwards the messages of two nodes in a two-phase protocol with optimal time-division. On the data link layer, we assume ergodic arrival processes at node 1 and 2 which have queues with infinite buffer length. At the beginning of each time-slot a centralized controller chooses the service rate pair which achieves the weighted rate sum maximum of the instantaneous achievable rate region for the block-fading channel state of the next time-slot with weights equal to the current buffer levels. To this end, the controller adjusts the time-division and relay power distribution. The policy is throughput optimal since the stability region is equal to the bidirectional ergodic rate region. This is because whenever the mean queue length is large, a negative drift of a quadratic Lyapunov function on the buffer levels can be proved.
KW - Achievable rate region
KW - Bidirectional relaying
KW - Cross-layer design
KW - Stability region
KW - Throughput optimal policy
UR - http://www.scopus.com/inward/record.url?scp=52949106852&partnerID=8YFLogxK
U2 - 10.1109/TCOMM.2008.060399
DO - 10.1109/TCOMM.2008.060399
M3 - Article
AN - SCOPUS:52949106852
SN - 0090-6778
VL - 56
SP - 1519
EP - 1529
JO - IEEE Transactions on Communications
JF - IEEE Transactions on Communications
IS - 9
ER -