Coordinated Multi-Point Transmission for Interference Mitigation in Cellular Distributed Antenna Systems M.A.Sc. Thesis Defence Talha Ahmad, B.Eng. Supervisor: Professor Halim Yanık¨ omero¯ glu July 20, 2011 Talha Ahmad (Carleton University) July 20, 2011 1 / 21
23
Embed
Coordinated Multi-Point Transmission for Interference ... · PDF fileCoordinated Multi-Point Transmission for Interference Mitigation in Cellular ... (ZF-DPC) scheme. ... for Interference
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Coordinated Multi-Point Transmission for InterferenceMitigation in Cellular Distributed Antenna Systems
M.A.Sc. Thesis Defence
Talha Ahmad, B.Eng.
Supervisor: Professor Halim Yanıkomeroglu
July 20, 2011
Talha Ahmad (Carleton University) July 20, 2011 1 / 21
4 Coordinated Port Selection and Beam Steering Optimization
5 Conclusion and Future Work
Talha Ahmad (Carleton University) July 20, 2011 2 / 21
Introduction
Introduction
Future cellular systems expected to provide ubiquitous high data ratecoverage to user terminals (UTs).
In conventional systems, UTs near the cell periphery experience lowrates due to distance-based attenuation and inter-cell interference.
Performance loss due to distance-based attenuation can be overcomeby dispersing the antennas over the coverage area.
Talha Ahmad (Carleton University) July 20, 2011 3 / 21
Introduction
Introduction (Cont.)
The distributed antenna system (DAS)architecture does not inherentlymitigate inter-cell interference.
Coordinated multi-point (CoMP)transmission techniques can be used toserve UTs with multiple distributedantenna ports and to mitigate bothintra-cell and inter-cell interference.
Distributed
Antenna PortBase Station
User
Terminal
A distributed antenna port used in theVideotron 3G network in Montreal,
Quebec (Source: Wikipedia).
Talha Ahmad (Carleton University) July 20, 2011 4 / 21
Introduction
Introduction (Cont.)
Which ports should be selected andwhat should be their antenna weights inorder to maximize
the aggregate spectral efficiency, orthe signal-to-interference-plus-noiseratio (SINR) of the UTs?
Distributed
Antenna PortBase Station
User
Terminal
A distributed antenna port used in theVideotron 3G network in Montreal,
Quebec (Source: Wikipedia).
Talha Ahmad (Carleton University) July 20, 2011 5 / 21
Related Literature
Related Literature: Distributed Antenna Systems
[Saleh et al., 1987, Yanikomeroglu & Sousa, 1993]: Early works onthe use of DAS for improving coverage in indoor wireless networks.
[Roh & Paulraj, 2002, Xiao et al., 2003]: Showed that MIMO DAScan improve capacity in cellular networks.
Distributed
Antenna Port
Base
Station
User
Terminal
Talha Ahmad (Carleton University) July 20, 2011 6 / 21
Related Literature
Related Literature: MIMO Broadcast Channel
System with coordination among multiple transmitters (BSs or ports) canbe modelled as a MIMO broadcast channel.
Tx
11
M
H1
... ...N Rx1
RxK
...
1
N
...
HK
[Costa, 1983, Weingarten et al., 2006]: Capacity-achieving scheme isdirty paper coding (DPC).
Obtain a close-to-optimal solution to each sub-problem using thesemidefinite relaxation (SDR) based Gaussian randomizationtechnique.
Talha Ahmad (Carleton University) July 20, 2011 14 / 21
Coordinated Port Selection and Beam Steering Optimization
Semidefinite Relaxation and Gaussian Randomization
1 Solve a relaxed version of the original problem.
S: Constraint set of the original problemU : Constraint set of the relaxed problem
2 Generate samples from a Gaussian distribution that is characterizedby the optimal solution to the relaxed problem.
Each sample represents a candidate solution to the original problem.
3 Choose candidate solution that yields the largest objective.
The SDR technique was used in [Karipidis et al., 2008]and [Chang et al., 2008] for the multicast transmit beamforming problem.
Talha Ahmad (Carleton University) July 20, 2011 15 / 21
Coordinated Port Selection and Beam Steering Optimization
Coordinated Port Selection and Beam SteeringOptimization (Cont.)
Two-stage approach for generating an approximate solution to the jointoptimization problem:
Select the J ≤ J1 candidate port state vectors out of the J1 vectorsgenerated in the first stage that yield the largest minimum SINR.
Generate a close-to-optimal beam steering coefficient vector for eachof the J port state vectors using the second stage.
Choose the approximate solution of the joint optimization problem tobe the pair of port state vector and beam steering coefficient vectorthat jointly yield the largest objective.
Talha Ahmad (Carleton University) July 20, 2011 16 / 21
Coordinated Port Selection and Beam Steering Optimization
Simulation Results (Stage 1)
−10 −5 0 5 10 15 20 25 30 35 40−10
−5
0
5
10
15
20
25
30
35
Transmit Power Per Port [dBm]
Min
imum
SIN
R [d
B]
Port Selection (Exhaustive Search)Algorithm 1 (First Stage)Single Port TransmissionAll Port Transmission
Distributed
Antenna PortBase Station
User
Terminal
Suburban macro-cell(NLoS)
dBS-BS = 1299 m
σs = 8 dB
σ2 = −114 dBm
fc = 2 GHz
M = 2
L = 7
drops = 500
J1 = 100
Talha Ahmad (Carleton University) July 20, 2011 17 / 21
Coordinated Port Selection and Beam Steering Optimization
Port Selection (Exhaustive Search)Proposed Port Selection (Stage 1)
July 18, 2011 DRAFT
Distributed
Antenna PortBase Station
User
Terminal
Suburban macro-cell(NLoS)
dBS-BS = 1299 m
σs = 8 dB
σ2 = −114 dBm
fc = 2 GHz
M = 2
L = 7
J1 = 100
J2 = 100
Talha Ahmad (Carleton University) July 20, 2011 18 / 21
Conclusion and Future Work
Summary
Integrated CoMP transmission schemes into the DAS architecture.
Chapter 3:
Developed extensions of the DAS ZF-DPC and DAS BD schemes toserve multiple UTs in a particular RB in each cell without intra-cellinterference.
Extended these schemes to the multi-cell processing scenario.
Chapter 4:
Considered coordinated joint selection and weighting of ports tomaximize the minimum SINR of the UTs (NP-hard problem).
Developed a novel polynomial-complexity two-stage approach, whichrelies on the SDR-based Gaussian randomization technique, to obtainan approximate solution to the joint optimization problem.
Talha Ahmad (Carleton University) July 20, 2011 19 / 21
Conclusion and Future Work
Future Work
Design transmission schemes such as DAS ZF-DPC and DAS BDwith a per-port or per-antenna power constraint.
Develop an algorithm that would allow the BSs to determine the portstates and beam steering coefficients in a distributed manner.
Joint optimization of the port states, beam steering coefficients, andthe UT scheduling for a given set of RBs.
Investigate the performance of CoMP schemes and algorithms in thepresence of imperfect channel state information.
Talha Ahmad (Carleton University) July 20, 2011 20 / 21
Conclusion and Future Work
Publications and Submissions
Chapter 3:
Talha Ahmad, Saad Al-Ahmadi, Halim Yanikomeroglu, and Gary Boudreau,“Downlink linear transmission schemes in a single-cell distributed antennasystem with port selection,” in Proc. IEEE Veh. Technol. Conf., May 2011.
Chapter 4:
Talha Ahmad, Ramy Gohary, Halim Yanikomeroglu, Saad Al-Ahmadi, andGary Boudreau, “Coordinated port selection and beam steering optimizationin a multi-cell distributed antenna system using semidefinite relaxation,”under review in IEEE Trans. Wireless Commun.
Talha Ahmad, Ramy Gohary, Halim Yanikomeroglu, Saad Al-Ahmadi, andGary Boudreau, “Coordinated max-min fair port selection in a multi-celldistributed antenna system using semidefinite relaxation,” submitted toIEEE Globecom Workshop on Distributed Antenna Systems for BroadbandMobile Communications, December 2011.
Talha Ahmad (Carleton University) July 20, 2011 21 / 21
Conclusion and Future Work
References I
[Caire & Shamai, 2003] Caire, G. & Shamai, S. (2003).49, 1691–1706.