A Learning Approach to Cooperative Communication System Design Yuxin Lu ⋆ , Peng Cheng † , Zhuo Chen ∗ , Wai Ho Mow ⋆ and Yonghui Li † ⋆ The Hong Kong University of Science and Technology. † The University of Sydney. ∗ Data 61, CSIRO. The 45th IEEE ICASSP, 2020 Supported by the HK RGC (GRF 16233816) Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 1 / 32
Outline Background and Motivation 1 Relay-Assisted Cooperative Communication System 2 Learning the Cooperative System 3 Simulation Results 4 Conclusion 5 Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 2 / 32
Outline Background and Motivation 1 Relay-Assisted Cooperative Communication System 2 Learning the Cooperative System 3 Simulation Results 4 Conclusion 5 Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 3 / 32
Deep Learning Artificial Intelligence Deep learning (DL) is a branch Machine of machine learning ⇒ Learn to make own decisions Learning Structures algorithms in layers ⇒ Create an “artificial neural Deep network” Learning Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 4 / 32
Deep Learning in Communication Conventional communication system is optimized in a block-wise manner: source/channel coding, modulation, demodulation, source/channel decoding, equalization Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 5 / 32
Deep Learning in Communication Conventional communication system is optimized in a block-wise manner: source/channel coding, modulation, demodulation, source/channel decoding, equalization Deep learning techniques have been applied to replace certain blocks: channel coding/estimation Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 5 / 32
Deep Learning in Communication Conventional communication system is optimized in a block-wise manner: source/channel coding, modulation, demodulation, source/channel decoding, equalization Deep learning techniques have been applied to replace certain blocks: channel coding/estimation Individualized component-wise approach might not optimize the overall system function! Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 5 / 32
Deep Learning in Communication Can we optimize the communication system in a holistic manner? Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 6 / 32
Deep Learning in Communication Can we optimize the communication system in a holistic manner? Joint design of the transmitter and receiver over the channel Expand the optimization space ... Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 6 / 32
Deep Learning in Communication Can we optimize the communication system in a holistic manner? Joint design of the transmitter and receiver over the channel Expand the optimization space ... Yes. Communication Autoencoder ! Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 6 / 32
Deep Learning in Communication Can we optimize the communication system in a holistic manner? Joint design of the transmitter and receiver over the channel Expand the optimization space ... Yes. Communication Autoencoder ! Transmitter and receiver are represented by neural networks (NNs) Promising results have been obtained Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 6 / 32
Autoencoder General autoencoder (AE) learns data structure to compress (top) Encoder Decoder Image, Text, … Latent Vector Image, Text, … 00101101 Channel 00101101 01101001 Encoder Decoder 01111001 11010001 11010001 Bit Stream Bit Stream Distortion + Noise Dirty Noisy (Random) Latent Vector General autoencoder (top) v.s. Communication autoencoder. Figure Credit: Zhao, Vuran, Guo and Scott Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 7 / 32
Autoencoder General autoencoder (AE) learns data structure to compress (top) Encoder Decoder Image, Text, … Latent Vector Image, Text, … 00101101 Channel 00101101 01101001 Encoder Decoder 01111001 11010001 11010001 Bit Stream Bit Stream Distortion + Noise Dirty Noisy (Random) Latent Vector General autoencoder (top) v.s. Communication autoencoder. Figure Credit: Zhao, Vuran, Guo and Scott Communication AE learns the channel behavior to improve transmission accuracy Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 7 / 32
Autoencoder Most existing applications are for point-to-point communications Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 8 / 32
More Complicated Scenarios Can we design an AE to optimize more complicated communication scenarios? Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 9 / 32
More Complicated Scenarios Can we design an AE to optimize more complicated communication scenarios? Our focus: Relay-assisted cooperative communication system Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 9 / 32
Existing Works for AE + Relay Constellation design for two-way relay networks 1 ⇒ Focused on constellation optimization. No detection algorithm was addressed 1T.Matsumine, T.Koike-Akino, and Y.Wang, “Deep learning-based constellation optimization for physical network coding in two-way relay networks,” arXiv preprint arXiv:1903.03713 , Mar. 2019. Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 10 / 32
Existing Works for AE + Relay Constellation design for two-way relay networks 1 ⇒ Focused on constellation optimization. No detection algorithm was addressed Our focus: Joint optimization of the constellation and detection algorithm Start with a one-way relay network 1T.Matsumine, T.Koike-Akino, and Y.Wang, “Deep learning-based constellation optimization for physical network coding in two-way relay networks,” arXiv preprint arXiv:1903.03713 , Mar. 2019. Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 10 / 32
Outline Background and Motivation 1 Relay-Assisted Cooperative Communication System 2 Learning the Cooperative System 3 Simulation Results 4 Conclusion 5 Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 11 / 32
System Model MLD/Amplifier R : half-duplex R h SR h RD Source message: S D m S ∈ { 1 , 2 , · · · , 2 k } , encoded MLD/MRC h SD as x S of length n k/n bits/independent channel First phase DF/AF Second phase uses System model of a 3 -node relay network. Source ( S ), Relay ( R ), Destination ( D ) Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 12 / 32
System Model First Phase: � y SJ = E S h SJ x S + n SJ , J ∈ { R, D } , (1) E S : average source transmit energy h SJ : channel coefficient n SJ : Gaussian noise vector CN (0 , 2 σ 2 SJ I ) Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 13 / 32
System Model First Phase: � y SJ = E S h SJ x S + n SJ , J ∈ { R, D } , (1) E S : average source transmit energy h SJ : channel coefficient n SJ : Gaussian noise vector CN (0 , 2 σ 2 SJ I ) Second Phase: � y RD = E R h RD x R + n RD , (2) E R : average relay transmit energy h RD : channel coefficient n RD : Gaussian noise vector CN (0 , 2 σ 2 RD I ) Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 13 / 32
AF Relaying AF relay node: y SR √ Symbol-wise amplifying operation x R = SR , x R ∈ x R , P S | h SR | 2 +2 σ 2 y SR ∈ y SR , h SR ∈ h SR Drawback: noise amplification ⇐ y SR = √ E S h SR x S + n SR Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 14 / 32
AF Relaying AF relay node: y SR √ Symbol-wise amplifying operation x R = SR , x R ∈ x R , P S | h SR | 2 +2 σ 2 y SR ∈ y SR , h SR ∈ h SR Drawback: noise amplification ⇐ y SR = √ E S h SR x S + n SR Destination: Maximal-ratio combining (MRC) Optimal in the context of AF High complexity: O ( n · 2 k ) per block Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 14 / 32
DF Relaying DF relay node: Maximum-likelihood decoding (MLD) √ E S x � 2 , where C is code book, x R = arg min x ∈C � y SR − h SR |C| = 2 k . Drawback: hard decision ⇒ information loss Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 15 / 32
DF Relaying DF relay node: Maximum-likelihood decoding (MLD) √ E S x � 2 , where C is code book, x R = arg min x ∈C � y SR − h SR |C| = 2 k . Drawback: hard decision ⇒ information loss Destination: Near-optimal decoder (NOD) arg max x S ∈C Pr( y SD | x S ) � x R ∈C Pr( x S → x R ) Pr( y RD | x R ) Near-optimal in the context of DF High complexity: O ( n · 2 k · 2 k ) per block Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 15 / 32
Outline Background and Motivation 1 Relay-Assisted Cooperative Communication System 2 Learning the Cooperative System 3 Simulation Results 4 Conclusion 5 Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 16 / 32
A typical AE Transmitter Receiver Channel A typical AE for a point-to-point communication system Input: one-hot encoding, e.g., { 00 , 01 , 11 , 10 } �→ { 1000 , 0100 , 0010 , 0001 } Yuxin Lu (HKUST) Learning Cooperative Communication System ICASSP 2020 17 / 32
Recommend
More recommend
