Beamspace MIMO-NOMA for Millimeter-Wave Communications Using Lens - PowerPoint PPT Presentation

Beamspace MIMO-NOMA for Millimeter-Wave Communications Using Lens Antenna Arrays Bichai Wang † , Linglong Dai † , Xiqi Gao # , and Lajos Hanzo ∗ † Tsinghua University, Beijing, China # Southeast University, Nanjing, China ∗ University of Southampton, Southampton, U.K. Sep. 2017 1 / 24

Outline Technical Background System Model of Beamspace MIMO Proposed Beamspace MIMO-NOMA Simulation Results Conclusions 2 / 24

Technical Background 5G key performance indicators (KPIs) defined by ITU User experienced Peak data rate data rate (Gbit/s) (Mbit/s) 20 100 IMT-2020 10 1 Area traffic Spectrum capacity efficiency 2 (Mbit/s/m ) 10 3  1  1 0.1 1  350 10  400 100  500 IMT-advanced Mobility Network (km/h) energy efficiency 5 10 10 6 1 10 Connection density Latency 2 (ms) (devices/km ) M.2083-03 3 / 24

Technical Background Three technical directions for 5G More Antennas Higher/Wider Frequency Bands 4 / 24

Technical Background MmWave massive MIMO can combine the roadmaps of 5G in an unified form mmWave High frequency Short wavelength Serious path-loss Spectrum expansion Large antenna array Small cell 1000x data rates increase! 5 / 24

Technical Background Challenges of mmWave massive MIMO - Traditional MIMO: One dedicated RF chain for one antenna - Enormous number of RF chains due to large antenna array - Unaffordable energy consumption (250 mW per RF chain at 60 GHz) How to reduce the number of required RF chains? 6 / 24

Outline Technical Background System Model of Beamspace MIMO Proposed Beamspace MIMO-NOMA Simulation Results Conclusions 7 / 24

System Model of Beamspace MIMO Basic idea - Concentrate the signals from different directions (beams) on different antennas by lens antenna array - Transform conventional spatial channel into beamspace channel - Limited scattering at mmWave → beamspace channel is sparse - Select dominant beams to reduce the dimension of MIMO system - Negligible performance loss → significantly reduced number of RF chains High- dimensional digital signal processing RF chain Beamspace MIMO Conventional MIMO 8 / 24

System Model of Beamspace MIMO Sparsity � � - ˜ ˜ h 1 , ˜ h 2 , · · · , ˜ H = h K = UH = [ Uh 1 , Uh 2 , · · · , Uh K ] - ˜ h k with a small number of dominant elements - Approximately sparse Beam selection - Select a small number of dominant beams - Only a small number of RF chains Spatial channel Beamspace channel   H UH 9 / 24

System Model of Beamspace MIMO Fundamental limit of beamspace MIMO - A single beam can only support a single user in existing beamspace MIMO systems - The maximum number of users that can be supported cannot exceed the number of RF chains - Massive users cannot be supported with limited number of RF chains Selecting User 1 Network User 2 Dimension- Reduced Lens ... Digital Precoding RF Chains User K 10 / 24

Outline Technical Background System Model of Beamspace MIMO Proposed Beamspace MIMO-NOMA Simulation Results Conclusions 11 / 24

Proposed Beamspace MIMO-NOMA Non-Orthogonal Multiple Access (NOMA) - Superposition coding at the transmitter - Successive interference cancellation (SIC) at the receiver - Multiple users can be supported at the same time-frequency resources SIC of User 2 User 1 signal Power signal detection User 2 User 1 User 1 User 2 signal Frequency detection User 2 12 / 24

Proposed Beamspace MIMO-NOMA Basic principle - Selecting one beam for each user using beam selection algorithms, such as the maximum magnitude (MM) selection and SINR maximization based selection - Interfering users can be simultaneously served within the same beam - The number of supported users can be larger than the number of RF chains - Spectrum efficiency and connectivity density can be improved User 1,2 User 1 User 1,1 User 2,3 User 2 User 2,2 User 2,1 User ,1 N User K RF 13 / 24

Proposed Beamspace MIMO-NOMA System model - N RF beams, K users - The set of users in the n th beam is S n ( S i ∩ S j = Φ) - Beamspace channel vector between the BS and the m th user in the n th beam is denoted by h m , n - Uniform precoding vector for users in the n th beam is w n � � � � � � � � � h H � h H � h H - We assume that 1 , n w n 2 ≥ 2 , n w n 2 ≥ · · · ≥ | S n | , n w n � � � 2 - After intra-beam SIC, the remaining signal received at the m th user in the n th beam beam can be written as m − 1 m , n w n √ p m , n s m , n √ p i , n s i , n � y m , n = h H + h H ˆ m , n w n � �� � i =1 desired signal � �� � intra − beam interferences | S j | w j √ p i , j s i , j � � + h H + v m , n m , n ���� j � = n i =1 noise � �� � inter − beam interferences 14 / 24

Proposed Beamspace MIMO-NOMA System model - The SINR the m th user in the n th beam can be represented as � � � 2 � h H 2 p m , n m , n w n γ m , n = ξ m , n | S j | m − 1 � � � � � 2 � p i , n + � � 2 � p i , j + σ 2 � h H � h H where ξ m , n = m , n w n m , n w j 2 2 i =1 j � = n i =1 - The achievable rate of the m th user in the n th beam R m , n = log 2 (1 + γ m , n ) - Achievable sum rate | S n | N RF � � R sum = R m , n n =1 m =1 15 / 24

Proposed Beamspace MIMO-NOMA Precoding - Challenge: * The number of users is higher than the number of beams, which means that this system is underdetermined * Conventional linear precoding cannot be directly used - Solution: * An equivalent channel can be determined for each beam to generate the precoding vector * The beamspace channel vectors of different users in the same beam are highly correlated * we use the beamspace channel vector of the first user in each beam as the equivalent channel vector ˜ H = [ h 1 , 1 , h 1 , 2 , · · · , h 1 , N RF ] 16 / 24

Proposed Beamspace MIMO-NOMA Precoding - Precoding matrix: � � † � � − 1 H H ˜ ˜ ˜ = ˜ ˜ W = [ ˜ w 1 , ˜ w 2 , · · · , ˜ w N RF ] = H H H - After normalizing the precoding vectors, the precoding vector for the n th beam can be written as w n ˜ w n = � ˜ w n � 2 17 / 24

Proposed Beamspace MIMO-NOMA Power allocation - Problem formalization: | S n | N RF � � max R m , n { p m , n } n =1 m =1 s . t . C 1 : p m , n ≥ 0 , ∀ n , m , | S n | N RF � � C 2 : p m , n ≤ P n =1 m =1 C 3 : R m , n ≥ R min , ∀ n , m - The objective function is non-convex 18 / 24

Proposed Beamspace MIMO-NOMA Power allocation - Theorem 1: � � − a m , n e m , n 1 R m , n = max c m , n max + log 2 a m , n + ln 2 ln 2 a m , n > 0 � y m , n | 2 � where e m , n = E | s m , n − c m , n ˆ - The optimization problem can be reformulated as | S n | N RF � � a m , n > 0 ( − a m , n e m , n 1 max max c m , n max + log 2 a m , n + ln 2 ) ln 2 { p m , n } n =1 m =1 s . t . C 1 , C 2 , C 3 - Iteratively optimize { c m , n } , { a m , n } , { p m , n } (All of the three optimization problems are convex) 19 / 24

Outline Technical Background System Model of Beamspace MIMO Proposed Beamspace MIMO-NOMA Simulation Results Conclusions 20 / 24

Simulation Results Simulation parameters - N = 256, K = 32 - Channel: Saleh-Valenzuela multipath channel (1 LoS + 2 NLoS) 21 / 24

Outline Technical Background System Model of Beamspace MIMO Proposed Beamspace MIMO-NOMA Simulation Results Conclusions 22 / 24

Conclusions Propose the beamspace MIMO-NOMA to break the fundamental limit of beamspace MIMO The equivalent channel vector was determined for each beam for ZF-based precoding Propose to jointly optimize the power allocation of all users by maximizing the achievable sum rate An iterative optimization algorithm was developed for power allocation The proposed beamspace MIMO-NOMA achieves better performance than beamspace MIMO in terms of spectrum and energy efficiency 23 / 24

24 / 24