2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 1 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks Antti T¨ olli with Praneeth Jayasinghe, Ganesh Venkatraman, Jarkko Kaleva, Markku Juntti, Matti Latva-aho and Le-Nam Tran, e-mail: atolli@ee.oulu.fi Centre for Wireless Communications, University of Oulu, Finland 2016 IEEE Communication Theory Workshop, Nafplio, Greece 16 May, 2016 G. Venkatraman, A. T¨ olli, L-N. Tran & M. Juntti, ”Traffic Aware Resource Allocation Schemes for Multi-Cell MIMO-OFDM Systems”, IEEE Transactions on Signal Processing, vol. 64, no. 11, pp. 2730–2745, June 2016. P. Jayasinghe, A. T¨ olli & M. Latva-aho, Bi-directional Signaling Strategies for Dynamic TDD Networks in Proc. IEEE SPAWC 2015, Stockholm, Sweden, July, 2015 G. Venkatraman, A. T¨ olli, M. Juntti & L-N. Tran ”Queue Aware Precoder Design via OTA Training”, in Proc. 2016 IEEE 17th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), Edinburgh, UK, July 3–6, 2016 � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 2 Heterogeneous Network Setting Heterogeneous network composed of ◮ Large macro cells with (massive) MIMO antenna arrays, ◮ Small cells and relays with (distributed) MIMO arrays, and ◮ D2D communication with base station coordination Backhaul / control Data � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 3 Dynamic TDD UL and DL Significant load control UL or DL data channels channels variation between adjacent cells Flexible UL/DL allocation provides large potential gains in spectral efficiency 2 frequency More challenging interference time management Figure: Flexible TDD frame structure 1 1 Nokia Networks, ”5G radio access system design aspects”, Nokia white paper, Aug. 2015. Available: http://networks.nokia.com/file/37611/5g-radio-access 2 3GPP TSG RAN WG1, ”Study on scenarios and requirements for next generation access technologies TR 38.913,” 3rd Generation Partnership Project 3GPP, www.3gpp.org, 2016 � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 4 Dynamic TDD Figure: UL-DL/DL-UL interference in Dynamic TDD Additional UL-to-DL and DL-to-UL interference associated with the dynamic TDD Interference mitigated by coordinated beamforming. More measurements and info exchange also at the terminal side Similar interference scenarios in underlay D2D transmission 3 3 A. T¨ olli, J. Kaleva & P. Komulainen, ”Mode Selection and Transceiver Design for Rate Maximization in Underlay D2D MIMO Systems”, in Proc. IEEE ICC 2015, London, UK, June, 2015 � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 5 System Model & Problem Formulation 𝑅 1 𝑅 2 𝑅 3 𝑅 4 𝑅 5 𝑅 6 OFDM system with N 𝑉 4 sub-channels and N B BSs, N T TX antennas 𝑉 3 per BS K users each with N R antennas 𝑉 1 𝑉 5 𝑉 2 𝑉 6 Desired signal Interference signal Goal: minimize the number of packets in BS queues via joint TX/RX design and resource allocation over spatial and frequency resources � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 6 Queueing Model Each user is associated with backlogged packets of size Q k packets. Queued packets Q k of each user follows dynamic equation at the i th instant as � + � Q k ( i + 1) = Q k ( i ) − t k ( i ) + λ k ( i ) (1) where t k = � N � L l =1 t l,k,n denotes the total number of n =1 transmitted packets corresponding to user k λ k represents the fresh arrivals of user k at BS b k � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 7 JSFRA Formulation 4 The optimization objective of joint space-frequency resource allocation (JSFRA) to design transmit precoders is q � � N L � � � � � minimize a k � Q k − t l,k,n (2) � � � � t l,k,n k ∈U n =1 l =1 � where a k are arbitrary weights used to control the priorities Exponent q = 1 , 2 , . . . , ∞ plays different role based on the value it assumes Inherent maximum rate constraint: � N � L l =1 t l,k,n ≤ Q k n =1 Special cases (when Q k > � N � L l =1 t l,k,n ∀ k ): n =1 ◮ q = 1 : Sum rate maximization ◮ q = 2 : Queue-Weighted Sum Rate Maximization (Q-WSRM) 4 G. Venkatraman, A. T¨ olli, L-N. Tran & M. Juntti, ”Traffic Aware Resource Allocation Schemes for Multi-Cell MIMO-OFDM Systems”, IEEE Transactions on Signal Processing, vol. 64, no. 11, pp. 2730–2745, June 2016. � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 8 JSFRA Formulation (MSE Reformulation) The queue minimization problem can be solved by utilizing the relation between the MSE and the SINR as ǫ l,k,n = (1 + γ l,k,n ) − 1 (3) Equivalence is valid only when the receivers are designed with the mean squared error (MSE) objective, i.e. , using MMSE receivers = − log 2 ( ǫ l,k,n ) (4a) t l,k,n � 2 ( d l,k,n − ˆ d l,k,n ) 2 � � 1 − w H � � � = = ǫ l,k,n E l,k,n H b k ,k,n m l,k,n � 2 + ` � � w H � � + (4b) l,k,n H b i ,k,n m j,i,n N 0 ( j,i ) � =( l,k ) � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 9 JSFRA Formulation (MSE Reformulation) Queue minimization via MSE reformulation minimize � ˜ v � q (5a) t l,k,n , m l,k,n , ǫ l,k,n , w l,k,n subject to t l,k,n ≤ − log 2 ( ǫ l,k,n ) ∀ l, k, n (5b) � 2 � � 1 − w H � ǫ l,k,n ≥ l,k,n H b k ,k,n m l,k,n � 2 + ` � � � w H � + N 0 ∀ l, k, n (5c) l,k,n H b i ,k,n m j,i,n ( j,i ) � =( l,k ) N L � � � tr ( m l,k,n m H l,k,n ) ≤ P max ∀ b. (5d) n =1 l =1 k ∈U b 1 k ( Q k − � N � L v k � a q where ˜ l =1 t l,k,n ) n =1 The nonconvex (difference of convex) rate constraints are approximated via successive convex approximation (SCA) method Receive beamformers are designed by the MMSE receivers using the converged TX precoders � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 10 Dynamic Traffic Scenario - Centralized Performance 140 Total backlogged bits after each tx slot, Σ k (Q k (i) - t k (i)) + JSFRA with q= ∞ JSFRA with q=2 120 JSFRA with q=1 Q-WSRM Q-WSRME 100 Sum arrivals Σ k λ k (i) 80 60 40 20 0 0 50 100 150 200 250 Time Slots Figure: Queue dynamics for { N, N B , K, N T , N R , A k } = { 4 , 2 , 12 , 4 , 1 , 6 } [G. Venkatraman, A. T¨ olli, L-N. Tran & M. Juntti, ”Traffic Aware Resource Allocation Schemes for Multi-Cell MIMO-OFDM Systems”, IEEE Transactions on Signal Processing, vol. 64, no. 11, pp. 2730–2745, June 2016.] � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 11 Distributed Methods Overhead of the 𝑅 1 𝑅 2 𝑅 3 𝑅 4 𝑅 5 𝑅 6 centralized design is large as the network size 𝑉 4 grows Distributed approaches 𝑉 3 based on primal decomposition or ADMM can be used to 𝑉 1 𝑉 5 𝑉 2 reduce the signaling 𝑉 6 Desired signal Interference signal Precoder design by solving the KKT expressions of the JSFRA problem (5) via MSE reformulation Practical approach to design precoders with minimal backhaul usage � Antti T¨ c olli, CWC
2016 IEEE Communication Theory Workshop, Nafplio, Greece , 16 May, 2016 Decentralized Resource Allocation and Effective CSI Signaling in Dense TDD Networks 12 KKT Expressions for (5) L � − 1 � � m ( i ) α ( i − 1) b k ,x,n w ( i − 1) y,x,n w H ( i − 1) α ( i − 1) b k ,k,n w ( i − 1) � y,x,n H H l,k,n H H l,k,n = H b k ,x,n + δ b I N T y,x,n l,k,n x ∈U y =1 L � − 1 � � w ( i ) � H b x ,k,n m ( i ) y,x,n m H ( i ) H b k ,k,n m ( i ) y,x,n H H l,k,n = b x ,k,n + N 0 I N R l,k,n y =1 x ∈U 2 2 � � � � ǫ ( i ) � 1 − w H ( i ) l,k,n H b k ,k,n m ( i ) � w H ( i ) l,k,n H b y ,k,n m ( i ) � + � w l,k,n � 2 N 0 l,k,n = + � � � � x,y,n l,k,n � � ( x,y ) � =( l,k ) ǫ ( i ) l,k,n − ǫ ( i − 1) � � t ( i ) l,k,n = − log 2 ( ǫ ( i − 1) l,k,n l,k,n ) − log(2) ǫ ( i − 1) l,k,n N L � ( q − 1) � + σ ( i ) � � t ( i ) a k q � � l,k,n = Q k − log(2) l,k,n n =1 l =1 � � σ ( i ) α ( i ) l,k,n = α ( i − 1) − α ( i − 1) l,k,n + ρ ( i ) l,k,n ǫ ( i ) l,k,n l,k,n � Antti T¨ c olli, CWC
Recommend
More recommend
