F airness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast - - PowerPoint PPT Presentation

Elapsed time:

Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT

Luis F. Abanto-Leon

Co-author: Gek Hong (Allyson) Sim

Department of Computer Science Technical University of Darmstadt

IEEE International Conference on Communications (ICC 2020) SAC-IOT3: Internet of Things III (2nd Paper)

:

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Contents

2/ 26

1 Motivation 2 System Model 3 Problem Formulation 4 Proposed Solution 5 Simulation Results 6 Conclusions

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Motivation

3/ 26

In factories, multiple industrial devices are inherently hyper-connected via hard-wiring to ensure safety.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Motivation

3/ 26

In factories, multiple industrial devices are inherently hyper-connected via hard-wiring to ensure safety. Wired connections hinder automation deployment and constrain the mobile robotics mechanics.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Motivation

3/ 26

In factories, multiple industrial devices are inherently hyper-connected via hard-wiring to ensure safety. Wired connections hinder automation deployment and constrain the mobile robotics mechanics. Due to rapid densification of industrial devices, wired connections become less appealing for factories of the future.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Motivation

3/ 26

In factories, multiple industrial devices are inherently hyper-connected via hard-wiring to ensure safety. Wired connections hinder automation deployment and constrain the mobile robotics mechanics. Due to rapid densification of industrial devices, wired connections become less appealing for factories of the future. Wireless information transmission is a viable alternative for these environments.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Motivation

3/ 26

In factories, multiple industrial devices are inherently hyper-connected via hard-wiring to ensure safety. Wired connections hinder automation deployment and constrain the mobile robotics mechanics. Due to rapid densification of industrial devices, wired connections become less appealing for factories of the future. Wireless information transmission is a viable alternative for these environments. However, guaranteeing high performance in terms of fairness, spectral efficiency and reliability is a challenging task.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Problem Overview

4/ 26

We investigate dual-layer non-orthogonal transmissions for in- dustrial IoT millimeter-wave communications. Primary layer: ubiquitous multicast signal devised to serve all the devices with a common message Secondary layer: composite signal consisting of private uni- cast messages. We jointly optimize the hybrid precoder, analog combiners, power allocation, and fairness. The performance is evaluated in terms of the spectral efficiency, fairness, and bit error rate.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Solution Overview

5/ 26

We propose two solutions: PLDM-1: designs independently the multicast precoder from the unicast precoders PLDM-2: the multicast precoder is obtained as a combination

• f the unicast precoding vectors

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Non-Orthogonal Unicast/Multicast System

6/ 26

Figure: K-user Non-Orthogonal Unicast/Multicast System

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Hybrid Precoder

7/ 26

s m B NRF

tx

F Ntx s m B NRF

tx

= Ntx Ntx

Figure: Hybrid and fully-digital precoders

m ∈ CNRF

tx ×1: multicast digital

precoder B ∈ CNRF

tx ×NRF tx : unicast digital

precoder F ∈ FNtx×NRF

tx : analog precoder

F = √δtx, . . . , √δtxe

2π(Ltx−1) Ltx

• :

set of phase shifts Ntx: number of transmit antennas NRF

tx : number of RF chains

Ltx: number of phase shifts

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

System Model

8/ 26

The downlink signal is x = F [B|m] [s|z]T (1) where F = [f1, f2, . . . , fK] ∈ CNtx×K : analog precoder B = [b1, b2, . . . , bK] ∈ CK×K : digital unicast precoder m = [m1, m2, . . . , mK]T ∈ CK×1 : digital multicast precoder s = [s1, s2, . . . , sK]T ∈ CK×1 : unicast symbols z ∈ C : multicast symbol

Throughout the paper we assume that NRF

tx

= K

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

System Model

9/ 26

The received signal at user k ∈ K is yk = wH

k HkFmz

• common multicast signal

+ wH

k HkFbksk

• unicast signal for device k

+ wH

k HkF

• j=k

bjsj

• interference at device k

+ wH

k nk noise

, (2)

wk: combiner of the k-th user Hk: channel between the gNodeB and the k-th user K = {1, . . . , K}: set of users K: number of users

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

System Model

10/ 26

The multicast and unicast SINRs at k ∈ K are ˜ γk =

• wH

k HkFm

• 2
• j
• wH

k HkFbj

• 2 + σ2 wk2

2

(3) γk =

• wH

k HkFbk

• 2
• j=k
• wH

k HkFbj

• 2 + σ2 wk2

2 .

(4)

˜ γk: multicast SINR at the k-th user γk: unicast SINR at the k-th user

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Problem Formulation

11/ 26 P : max

{wk}K

k=1,{fk}K k=1,

{bk}K

k=1,m,∆

• k

log2 (1 + ˜ γk) + log2 (1 + γk) − C′∆ (5a) s.t. |˜ γk − γmin| ≤ ∆, ∀k ∈ K, (5b) ˜ γ1 ≥ ˜ γ2 ≥ . . . ≥ ˜ γK ≥ ˜ γ1, (5c) Fm2

2 /

• k

Fbk2

2 ≥ β,

(5d) Fm2

2 +

• k

Fbk2

2 ≤ Ptx,

(5e) [F]q,r ∈ F, q ∈ Q, r ∈ R, (5f) [wk]n ∈ W, n ∈ N, ∀k ∈ K, (5g) ∆ ≥ 0, (5h)

F: allowed phase shifts at the precoder W: allowed phase shifts at the combiners

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Proposed Solution

12/ 26 P0 : max

{wk}K

k=1,{fk}K k=1,

{pk}K

k=1,{vk}K k=1,m,∆

• k

˜ γk + γk − C∆ (6a) s.t. |˜ γk − γmin| ≤ ∆, ∀k ∈ K, (6b) ˜ γ1 ≥ ˜ γ2 ≥ . . . ≥ ˜ γK ≥ ˜ γ1, (6c) Fm2

2 /

• k

pk Fvk2

2 ≥ β,

(6d) Fm2

2 +

• k

pk Fvk2

2 ≤ Ptx,

(6e) [F]q,r ∈ F, q ∈ Q, r ∈ R, (6f) [wk]n ∈ W, n ∈ N, ∀k ∈ K, (6g) vk2

2 = 1, ∀k ∈ K,

(6h) pk ≥ 0, ∀k ∈ K, (6i) ∆ ≥ 0, ∀k ∈ K, (6j)

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Proposed Solution

13/ 26

P1 : max

{wk}K

k=1,{fk}K k=1,{vk}K k=1

• k

γk (7a) s.t. [F]q,r ∈ F, q ∈ Q, r ∈ R, (7b) [wk]n ∈ W, n ∈ N, ∀k ∈ K, (7c) vk2

2 = 1, ∀k ∈ K.

(7d) P2 : max

{pk}K

k=1,m,∆

• k

˜ γk + γk − C∆ (8a) s.t. |˜ γk − γmin| ≤ ∆, ∀k ∈ K, (8b) ˜ γ1 ≥ ˜ γ2 ≥ . . . ≥ ˜ γK ≥ ˜ γ1, (8c) Fm2

2 /

• k

pk Fvk2

2 ≥ β,

(8d) Fm2

2 +

• k

pk Fvk2

2 ≤ Ptx,

(8e) pk ≥ 0, ∀k ∈ K, (8f) ∆ ≥ 0. (8g)

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Optimization of {wk}K

k=1, {fk}K k=1, {vk}K k=1

14/ 26

P1,1 : max

{wk}K

k=1,{fk}K k=1

• k

pk

• wH

k HkFvk

• 2

(9a) s.t. [F]q,r ∈ F, q ∈ Q, r ∈ R, (9b) [wk]n ∈ W, n ∈ N, ∀k ∈ K. (9c) P1,2 : min

{vk}K

k=1

• k
• j=k

pj

• wH

k HkFvj

• 2

(10a) s.t. vk2

2 = 1, ∀k ∈ K.

(10b)

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Optimization of {pk}K

k=1, m, ∆

15/ 26

P2 : max

{pk}K

k=1,m,∆

• k

˜ γk + γk − C∆ (11a) s.t. |˜ γk − γmin| ≤ ∆, ∀k ∈ K, (11b) ˜ γ1 ≥ ˜ γ2 ≥ . . . ≥ ˜ γK ≥ ˜ γ1, (11c) Fm2

2 /

• k

pk Fvk2

2 ≥ β,

(11d) Fm2

2 +

• k

pk Fvk2

2 ≤ Ptx,

(11e) pk ≥ 0, ∀k ∈ K, (11f) ∆ ≥ 0. (11g)

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Optimization of {pk}K

k=1, m, ∆

16/ 26

• P2 :

max {pk}K

k=1,{µk}K k=1,{υk}K k=1,m,∆

• k

µk + υk − C∆ (12a) s.t.

• heff

k m

• 2 /
• pk |gk|2 + σ2

≥ µk, ∀k ∈ K, (12b) pk |gk|2 /σ2 ≥ υk, ∀k ∈ K, (12c) Fm2

2 /

• k

pk Fvk2

2 ≥ β,

(12d) Fm2

2 +

• k

pk Fvk2

2 ≤ Ptx,

(12e) µ1 ≥ µ2 ≥ . . . ≥ µK ≥ µ1, (12f) µk ≤ γmin + ∆, ∀k ∈ K, (12g) µk ≥ γmin − ∆, ∀k ∈ K, (12h) υk ≥ 0, ∀k ∈ K, (12i) pk ≥ 0, ∀k ∈ K, (12j) ∆ ≥ 0, (12k) Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Optimization of {pk}K

k=1, m, ∆

17/ 26

• P(t)

2

: max

m,p,µ,υ,∆

1T µ + 1T υ − C∆ (13a) s.t. 2Re

• diag
• Ap(t) + d
• I ⊗ m(t)H

C (1 ⊗ m)

• − diag
• Ap + d
• I ⊗ m(t)H

C

• 1 ⊗ m(t)

− diag

• Ap(t) + d
• diag
• Ap(t) + d
• µ 0,

(13b)

• A ⊙ (diag (d))−1

p υ, (13c) 2Re

• cT p(t)m(t)HFHFm
• −

cT pm(t)HFHFm(t) −

• cT p(t)2β ≥ 0,

(13d) Fm2

2 +

• k

pk Fvk2

2 ≤ Ptx,

(13e)

• I −

I

• µ 0,

(13f) µ (γmin + ∆) 1, (13g) µ (γmin − ∆) 1, (13h) υ 0, (13i) p 0, (13j) ∆ ≥ 0. (13k) Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results

18/ 26

Table: Simulation parameters

Description Symbol Value Units Number of users K 6

• Number of transmit antennas

Ntx 64

• Number of receive antennas

Nrx 4

• Number of RF chains (at the hybrid precoder)

N RF

tx

6

• Number of phase shifts at the precoder

Ltx 32

• Number of phase shifts at the combiner

Lrx 4

• Multicast QoS requirement

γmin 5 dB

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - Spectral Efficiency

19/ 26

−25 −10 5 5 10 15 20 25 30 35

Ptx/σ2 [dB]

SE [bps/Hz]

PLDM-0 PLDM-1 PLDM-2

• 30

0.10 0.15 0.20 0.25 0.30

• 25

0.40 0.50 0.60 0.70 0.80

• 20

1.25 1.45 1.65 1.85 2.05

• 15

3.20 3.70 4.20 4.70 5.20

Figure: Overall SE of the system

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - Spectral Efficiency

20/ 26

−30 −25 −20 −15 −10 −5 5 10 5 10 15 20 25 30 35 aggregate multicast SE

Ptx/σ2 [dB]

SE [bps/Hz]

PLDM-0 PLDM-1 PLDM-2

• 25

0.0 0.2 0.4 0.6 0.8

• 20

0.0 0.5 1.0 1.5 2.0

• 15

0.2 1.4 2.6 3.8 5.0

• 10

0.2 2.7 5.2 7.7 10 Multicast Unicast

Figure: Disaggregated SE of the system

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - Multicast Fairness

21/ 26

−30 −25 −20 −15 −10 −5 5 10 0.5 1 1.5 2 2.5 3 multicast target per device

Ptx/σ2 [dB]

SE [bps/Hz]

PLDM-0 PLDM-1 PLDM-2

Figure: Multicast SE per device

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - BER

22/ 26

−20 −10 10 10−4 10−3 10−2 10−1

Ptx/σ2 [dB] BER

PLDM-0 PLDM-1 PLDM-2

Figure: Aggregate

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - Multicast BER

23/ 26

−20 −10 10 10−4 10−3 10−2 10−1

Ptx/σ2 [dB] BER

PLDM-0 PLDM-1 PLDM-2

Figure: Multicast

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Simulation Results - Unicast BER

24/ 26

−20 −10 10 10−4 10−3 10−2 10−1

Ptx/σ2 [dB] BER

PLDM-0 PLDM-1 PLDM-2

Figure: Unicast

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Conclusions

25/ 26

We investigated the joint optimization of hybrid precoding, fairness, and power splitting in NOMA-LDM superimposed transmissions for industrial IoT scenarios. We proposed two solutions: one of them regarded as the superposition of two distinct precoders with different spatial and power signatures. The second approach is designed as a purely power-domain NOMA scheme. Through simulations we show that both proposed schemes, PLDM-1 and PLDM-2, are capable of providing remarkable fairness and high BER, which is relevant for the dissemination

• f critical control messages in this kind of scenarios.

Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :

Elapsed time: Motivation System Model Problem Formulation Proposed Solution Simulation Results Conclusions

Questions

26/ 26

Email: l.f.abanto@ieee.org Website: www.luis-f-abanto-leon.com

This work has been funded by the Deutsche Forschungsgemeinschaft (DFG) within the B5G-Cell project as part of the SFB 1053 MAKI. Luis F. Abanto-Leon Technical University of Darmstadt Fairness-Aware Hybrid Precoding for mmWave NOMA Unicast/Multicast Transmissions in Industrial IoT :