A Signal Space Diversity-Based TDBC Protocol in Two-Way Relay - PowerPoint PPT Presentation

A Signal Space Diversity-Based TDBC Protocol in Two-Way Relay Systems Hamza Umit Sokun Salama Ikki Mehmet Cagri Ilter Halim Yanikomeroglu Lakehead University Carleton University Canada Canada sikki@lakeheadu.ca {husokun, ilterm, halim}@sce.carleton.ca IEEE Globecom, Dec. 2015, San Diego, CA, 1 USA

Two-Way Relaying • Interest in terminal relaying (D2D) in 5G standards. R L • Traditional one-way relay systems enable R 1 spatial diversity at the expense of spectral A B efficiency due to half-duplex transmission. One-way Relaying 1. slot 2. slot R L 3. slot Two-way Relaying  higher spectral efficiency R 1 • – Time Division Broadcast Protocol (TDBC) A B (Using direct link  higher reliability) Two-way Relaying – Best-relay selection IEEE Globecom, Dec. 2015, San Diego, CA, 2 USA

Signal Space Diversity • SSD [1] is a type of diversity that is extracted in the modulation signal space. • In [2], the idea of SSD is applied to cooperative schemes (with single relay) and the constellation expansion method is proposed. Constellation expansion [2] • Using the constellation expansion method proposed in [2], the performance of multi-relay cooperative schemes is investigated in [3]. [1] J. Boutros and E. Viterbo, “Signal space diversity: a power-and-bandwidth-efficient diversity technique for the Rayleigh fading channels,” IEEE Trans. Info. Theory ., Jul. 1998. [2] S. A. Ahmadzadeh, S. A. Motahari, A. K. Khandani, “Signal space cooperative communication,” IEEE Trans. Wireless Comm ., Apr. 2010. [3] O. Amin, R. Mesleh, S. Ikki, M. Ahmed, and O. Dobre, “Performance analysis of multiple relays cooperative systems with signal space diversity,” IEEE Trans. Veh. Technol. , Aug. 2015. IEEE Globecom, Dec. 2015, San Diego, CA, 3 USA

Novelty/Contributions • Two-way relaying + Signal space diversity Good combination, because two end-sources exchange – Baseline: 2 symbols over 4 time-slots – Proposed: 4 symbols over 3 time-slots Adapt SSD signaling for two-way relaying (TDBC) • Obtained E2E error probability for arbitrary 2D constellations (as a function of SNR), which accounts for all non-uniform rectilinear constellation caused by constellation rotation. This allows – choosing the best rotation angle as a function of SNR. – the joint optimization of rotation angle, and transmit powers of all nodes. IEEE Globecom, Dec. 2015, San Diego, CA, 4 USA

System Model (1/4) • Original data symbols are rotated by a certain angle before being transmitted, and then the end-sources and the relay cooperate for transmitting in-phase and quadrature components of two consecutive rotated symbols. First symbol (belongs to the rotated constellation) Second symbol (belongs to the rotated constellation) The new constellation point that will be sent from source A (belongs to the expanded constellation), In the first time slot: In the second time slot: R L R L R 1 R 1 1. slot 2. slot A B A B IEEE Globecom, Dec. 2015, San Diego, CA, 5 USA

System Model (2/4) R L In the third time slot: R 1 A B Since each node knows their data, 3. slot the known parts will be removed: IEEE Globecom, Dec. 2015, San Diego, CA, 6 USA

System Model (3/4) Considering the direct and the cooperative links, the received signals at the end-source B: R L R 1 A B 1. slot To detect the original message, the end-source B 3. slot reorders the received components: IEEE Globecom, Dec. 2015, San Diego, CA, 7 USA

System Model (4/4) Finally, the end-source B applies ML detector on the reordered signals to detect the end-source messages: 1. 2. IEEE Globecom, Dec. 2015, San Diego, CA, 8 USA

Error Rate Performance End-to-End Average SER 1) (Difference of the 𝑚 -th and 𝑙 -th where symbols in the expanded constellation) 2) 3) (Complementary CDF of a bivariate Gaussian variable) 4) [4] L. Szczecinski, H. Xu, X. Gao, and R. Bettancourt, “Efficient evaluation of BER for arbitrary modulation and signalling in fading channels,” IEEE Trans. Comm ., vol. 55, no. 11, pp. 2061–2064, Nov. 2007. IEEE Globecom, Dec. 2015, San Diego, CA, 9 USA

Simulation Results (1/4) 0 10 -1 10 -2 10 P B (e) 7 -3 10 C-TDBC with 16-QAM (Simulation) P-TDBC ( θ = 10 ° ) with QPSK (Simulation) P-TDBC ( θ = 15 ° ) with QPSK (Simulation) -4 10 P-TDBC ( θ = 40 ° ) with QPSK (Simulation) P-TDBC ( θ opt ) with QPSK (Simulation) Analytical -5 10 0 5 10 15 20 25 30 E/ N 0 (dB) Fig. 1. SER performance of the proposed TDBC (P-TDBC) in compared to the conventional TDBC (C-TDBC) . (E A =E B =E R =E) IEEE Globecom, Dec. 2015, San Diego, CA, 10 USA

Simulation Results (2/4) 0 10 θ opt = 30.28 E / N 0 = 5 dB θ opt = 28.52 -1 E / N 0 = 10 dB 10 θ opt = 27.91 E / N 0 = 15 dB -2 10 E / N 0 = 20 dB θ opt = 27.66 P B ( e ) -3 10 7 θ opt = 27.56 E / N 0 = 25 dB -4 10 E / N 0 = 30 dB θ opt = 27.53 -5 10 E / N 0 = 35 dB θ opt = 27.5 -6 10 5 10 15 20 25 30 35 40 θ (deg) Fig. 2. The impact of different rotation angles on the system performance at the different SNR values. IEEE Globecom, Dec. 2015, San Diego, CA, 11 USA

Simulation Results (3/4) 0 10 -1 10 P A (e) + P B (e) -2 10 G AR /G RB =-30 dB R 1 A B -3 Rotation angle with fixed power, G AR /G RB =-30 dB 10 Joint rotation angle and power, G AR /G RB =-30 dB Rotation angle with fixed power, G AR /G RB =10 dB G AR /G RB =10 dB Joint rotation angle and power, G AR /G RB =10 dB -4 R 1 10 10 12 14 16 18 20 22 24 26 28 30 A B E T =N 0 (dB) Fig. 3. Impact of joint optimization of rotation angle, and transmit powers at all nodes on the system performance. (E A +E B +E R = E T , E max = 0.8E T ) IEEE Globecom, Dec. 2015, San Diego, CA, 12 USA

Simulation Results (4/4) Reactive Relay-Selection with Three Relays 0 10 -1 10 -2 10 -3 10 P B (e) -4 10 C-TDBC with 16-QAM (Simulation) -5 P-TDBC ( θ =10 ° ) with 4-QAM (Simulation) 10 P-TDBC ( θ =15 ° ) with 4-QAM (Simulation) P-TDBC ( θ =20 ° ) with 4-QAM (Simulation) -6 10 P-TDBC ( θ opt ) with 4-QAM (Simulation) Analytical -7 10 0 5 10 15 20 25 30 E T =N 0 (dB) Fig. 4. SER performance of the proposed TDBC (P-TDBC) in compared to the conventional TDBC (C-TDBC) with reactive relay selection, when the number of relays is 3. (E A =E B =E R = E T /3) IEEE Globecom, Dec. 2015, San Diego, CA, 13 USA

Summary • A signal space diversity-based TDBC protocol is proposed. – Higher spectral efficiency, – Higher spatial diversity. • Error rate performance analysis with arbitrary constellation is obtained. • Effect of rotation angle is investigated. • Joint effect of rotation angle and power allocation is shown. IEEE Globecom, Dec. 2015, San Diego, CA, 14 USA

Future Works • Imperfect channel estimation • Impact of coding rate • Cognitive radio IEEE Globecom, Dec. 2015, San Diego, CA, 15 USA

Thank you! This work is supported in part by Huawei Canada Co., Ltd., and in part by the Ontario Ministry of Economic Development and Innovation’s ORF-RE (Ontario Research Fund - Research Excellence) program. IEEE Globecom, Dec. 2015, San Diego, CA, 16 USA