READ ME File For 'Hybrid beamforming design for full-duplex millimeter wave communication dataset' Dataset DOI: 10.5258/SOTON/D0737 This is the dataset of the accepted paper (Nov, 2018): K. Satyanarayana, M. El-Hajjar, Ping-Heng Kuo, Alain Mourad and L. Hanzo, "Hybrid Beamforming Design for Full-Duplex Millimeter Wave Communication" * Paper Abstract: Harnessing the abundant availability of spectral resources at millimeter wave (mmWave) frequencies is an attractive solution to meet the escalating data rate demands. Additionally, it has been shown that full-duplex (FD) communication has the potential of doubling the bandwidth efficiency. However, the presence of significant residual self-interference (SI), which is especially more pronounced at mmWave frequencies because of the non-linearities in the hardware components, erodes the full potential of FD in practice. Conventionally, the residual SI is canceled in the baseband using digital processing with the aid of a transmit precoder. In this work, we propose a hybrid beamforming design for FD mmWave communications, where the SI is canceled by the joint design of beamformer weights at the radio-frequency (RF) and the precoder as well as combiner in the baseband. Our proposed design preserves the dimensions of the transmit signal, while suppressing the SI. We demonstrate that our joint design is capable of reducing the SI by upto 30 dB, hence performing similarly to the interference-free FD system while being computationally efficient. Our simulation results show that the proposed design significantly outperforms eigen-beamforming. * Project: The fiscal support of InterDigital as well as that of the European Research Council, Advanced Fellow Grant is gratefully acknowledged. * This DOI contains the datasets of Figures 5, 6, 7, 8, 9, 10, 11 and 12 of the aforementioned paper. Each folder is named according to its content, where the cruves of each figure are stored in text files. To regenerate the results, please use the Graphics Layout Engine (GLE), using the command "gle Figure.gle" * The embedded folders are as follows: - Figure-5: Contains the dataset of Figure 5. Digital precoding of the proposed design for SIR= -10 dB, -30 dB. The distance between the transmit and receive antenna arrays is D=0 and the angle between the arrays is T = 0 ? . The parameters of Table 1 are used. - Figure-6: Contains the dataset of Figure 6. Sum rate of hybrid beamforming relying on the proposed design for INR=0 dB as well as 10 dB. The distance between the transmit and receive antenna arrays is D=0 and the angle between the arrays is T = 10 ? . The parameters of Table 1 are used. - Figure-7: Contains the dataset of Figure 7. Sum rate comparison of the proposed design and of eigen beamforming using hybrid beamforming for different SIRs. The distance between the transmit and receive antenna arrays is D=0 and the angle between the arrays is T = 10. The parameters of Table 1 are used. - Figure-8: Contains the dataset of Figure 8. Comparison of the sum rate of the proposed design and of eigen beamforming using hybrid beamforming for different INRs. The distance between the transmit and receive antenna arrays is D=0 and the angle between the arrays is T = 0 ? . The parameters of Table 1 are used. - Figure-9: Contains the dataset of Figure 9. Comparison of the sum rate of FD and HD modes at different SI levels — (a) 128 × 32 (b) 64 × 32. The distance between the transmit and receive antenna arrays is D=0 and the angle between the arrays is T = 0 ? . The parameters of Table 1 are used. - Figure-10: Contains the dataset of Figure 10. Sum rate of the proposed design for different values of D when INR= 10 dB and angle between the transmit and receive antenna arrays T = 0 ? . - Figure-11: Constains the dataset of Figure 11. Sum rate performance vs angle between the arrays when D=0, SNR= -5 dB and INR =10 dB. -Figure 12: Contains the dataset of Figure 12. BER of the proposed design for FD under imperfect CSI. To obtain this curve, a single stream with QPSK modulation is transmitted using 128 × 32 element MIMO. The parameters of Table 1 are used. Dataset available under a CC BY 4.0 licence Publisher: University of Southampton, U.K. Date: December 2018