Integration of 5G Networks and Internet of Things for Future Smart City
View this Special IssueResearch Article  Open Access
Jianguo Li, Xiangming Li, Aihua Wang, Neng Ye, "Performance Analysis for Downlink MIMONOMA in Millimeter Wave Cellular Network with D2D Communications", Wireless Communications and Mobile Computing, vol. 2019, Article ID 1914762, 11 pages, 2019. https://doi.org/10.1155/2019/1914762
Performance Analysis for Downlink MIMONOMA in Millimeter Wave Cellular Network with D2D Communications
Abstract
Enabling nonorthogonal multiple access (NOMA) in devicetodevice (D2D) communications under the millimeter wave (mmWave) multipleinput multipleoutput (MIMO) cellular network is of critical importance for 5G wireless systems to support low latency, high reliability, and high throughput radio access. In this paper, the closedform expressions for the outage probability and the ergodic capacity in downlink MIMONOMA mmWave cellular network with D2D communications are considered, which indicates that NOMA outperforms TDMA. The influencing factors of performance, such as transmission power and antenna number, are also analyzed. It is found that higher transmission power and more antennas in the base station can decrease the outage probability and enhance the ergodic capacity of NOMA.
1. Introduction
With the rapid growth of a variety of 5thgeneration wireless systems commercial requirements, like 8k video, cloud VR, unmanned driving, smart city, etc., the demands on high data rates, low latency, and high reliability of wireless communication are increasing rapidly. As the core technologies of 5G, the wide bandwidth and large scale antenna arrays can be used by mmWave MIMO to provide high data rates. Meanwhile, NOMA can increase the spectral efficiency of transmission so that more data is transmitted over the same bandwidth spectrum. At the same time, due to the short communication distance, D2D communication can obtain lower transmission power and lower delay to ensure the battery availability and system reliability. Therefore, combined with MIMONOMA and mmWave in D2D communications, the throughput of the entire cell system can be improved.
D2D communication was noticed as a viable candidate for some certain applications such as proximity services, content sharing, multiparty games, and coverage discovered with business purposes in the 3rdGeneration Partnership Project (3GPP) Longterm Evolution (LTE) 12th and 13th editions [1, 2]. By sharing the same resource blocks with downlink cellular users, D2D communication is a useful technology which can enhance the spectrum efficiency and system capacity successfully. Due to the short communication distance, D2D pairs can communicate with each other directly without being relayed by the base station which can reduce transmission power, increase transmission reliability, and extend system range [3, 4]. Since the cellular user and the D2D user coexist, if the D2D user and the cellular user are not well coordinated, it will not bring any benefits and may affect the communication of the normal cellular user. According to the literature [5], based on the overlapping coalition formation game theory, the authors have proposed a method to conduct joint interference management and resource allocation in the D2D communications. In addition, a new interference management strategy has been discussed to enhance the overall sum rate of cellular networks and D2D pairs which has combined the conventional mechanism and interference limited area control scheme [6]. Furthermore, a graph theoretic approach for transmissionorder optimization scheme in bidirectional D2D communications underlaying cellular TDD networks has been introduced in the [7]. Meanwhile, the authors in [8] have considered a continuous beamforming vector design for all cellular users and the D2D pairs association vector search algorithm to maximize the capacity of the cellular users and D2D pairs. In order to reduce the interference between the cellular users and D2D pairs, a combining call admission control and power control scheme under guaranteeing QoS of every users in the cellular network has been presented in [9]. Besides, it is significant that the NOMA and MUMIMO can also improve the overall system throughput in the cellular network with underlaid D2D communications, which have been proposed in [10].
As the core technology of 5G, NOMA can serve multiple users in the same resource block, such as a time slot, frequency channel, or spreading code. Compared to the orthogonal multiple access, NOMA can provide a set of visible benefits such as greater spectrum efficiency and higher system throughout [11–15]. There are two available NOMA technology categories, namely, powerdomain and codedomain NOMA, widely used in the cellular network. This paper focuses on the powerdomain NOMA which divides the users in powerdomain. On the transmitter sides, signals from multiple users are transmitted at the same resource blocks. On the receiver sides, the multiuser detection algorithms, such as successive interference cancellation, have been utilized to detect the signal, regarding the weak signal as the interference, and decode the strong signal while cancelling the strong signal to detect the weak signal [16, 17]. Aiming at enhancing system capacity and user fairness, the authors have presented a novel resource and power allocation technique which can provide a flexible balancing between capacity and fairness maximization in [18]. Meanwhile, the optimum received power levels of uplink NOMA signal have been proposed in [19]. Furthermore, comparing with OMA in massive connectivity, NOMA can supply more users than OMA because of the limited number of supported users by the amount of the available resources [20–22]. In addition, the analytical outage probability expression for each user has been derived in the downlink cooperative NOMA network over Nakagamim fading channels in [23]. Since the perfect channel state information at the transmitter side is nearly impractical for many communication scenarios, the literature [24] presented a practical downlink NOMA system over the Nakagamim fading channels with statistical CSI associated with each other. At the same time, the authors attempted to combine the NOMA and MIMO to accomplish high spectral efficiency [25, 26]. Additionally, the outage probability of the massive MIMONOMA system has been derived with perfect userordering and limited feedback [27]. Besides, the scheme discussed in [28] required that the number of antennas at the receiver is larger than that at the transmitter instead of the CSI at the transmitter. In [29], the optimal and low complexity suboptimal power allocation schemes have been proposed to maximize the ergodic capacity of MIMONOMA system with statistical channel state information in the transmitter over Rayleigh fading channel. Also, the authors in [30] have proposed a cluster beamforming strategy which can optimize beamforming vectors and power allocation coefficients in MIMONOMA system to decrease the total power. In addition, NOMA can improve the achievable rate greatly in the D2Daided cooperative relaying system in the literature [31].
Nowadays, as spectrum resources become increasingly scare, the mmWave band with a wider spectrum becomes a natural choice for largevolume content services [32]. Meanwhile, benefitting from the small wavelength of mmWave, the large scale antenna arrays can be adopted easily. Combining the high gain directional antenna and beamforming technology enabled by massive antennas, the high data rates can be reached within a 200meter mmWave cellular network which has high path loss in mmWave band [33]. In the literature [34], the authors have proposed that the base station which is equipped with a large antenna array can serve a set of users through the users precoding with massive MIMO. Furthermore, the literature [35] has proved that the orders of magnitude are increased in spectral efficiency with massive MIMO which can offer more multiuser gain. However, as for MIMO in conventional cellular frequency band, the beamforming precoding is totally achieved in the digital domain which can cancel the interference between different beams. In order to realize the precoder in the digital domain, a dedicated RF chain is needed by every antenna. But it is difficult for the mmWave base station with large scale antennas whose energy consumption is a large part of the total energy consumption at mmWave frequencies due to the wide bandwidth [36]. At the same time, the authors in [36] have discussed a successive interference cancellationbased hybrid precoding method with nearoptimal performance and low complexity which can maximize the achievable subrate of each subantenna array and avoid the need for singular value decomposition and matrix inversion. Additionally, the literature [37] has proposed an interferenceaware beam selection which can avoid serious multiuser interferences to reduce obvious performance loss. Besides, the authors in [38] have proposed a practical design of hybrid precoders and combiners with lowresolution phase shifters in mmWave MIMO systems adopting an iterative algorithm for hybrid precoders and combiners to optimize the spectral efficiency. Moreover, the MIMO capacity has been computed over the Nakagamim fading channel in [39] and the closedform expressions for outage probabilities achieved by NOMA users in a multicell downlink mmWave network have been obtained over a Poisson cluster point process. Moreover, a finegrained performance analysis over Poisson bipolar model of the mmWave D2D communication networks was provided in [40]. The literature [41] has presented an indepth capacity analysis for the integrated NOMA mmWavemassiveMIMO systems which can achieve significant capacity improvements.
Because of the large link attenuation and weak coverage in the MIMONOMA cellular network, the D2D communication can be used to enhance service for cell edge users. In this paper, the outage probability and the ergodic capacity are proposed in the downlink MIMONOMA cellular network with D2D communications. The remaining part of this paper is organized as follows. In Section 2, we introduce a system model of the downlink MIMONOMA cellular network with multiple direct D2D pairs underlaying communication. The closedform expressions of the performance analysis including the outage probability and the ergodic capacity are given in Section 3. Finally, the numerical results in Section 4 validate the theoretical analysis and demonstrate that the system capacity can be improved by the integrated MIMONOMA in the mmWave network with D2D communications.
2. System Model
The paper considers a downlink NOMA, MUMIMO mmWave cellular network with multiple direct D2D pairs underlaying communications where the cellular users are randomly distributed. The same mmWave resources are used by the cellular user and D2D pairs with NOMA. The mmWave MIMONOMA D2D communication system model is shown in Figure 1. The base station is equipped with multiple antennas, which can generate high directional and high gain beams for cellular users. There are cellular users with signal antenna and D2D users with signal antenna in the cellular network, which are denoted as and . The D2D pairs are randomly distributed in the edge of the cellular network and there is no direct link between the base station and the D2D pairs [10].
2.1. NOMA Signal
In beam , we assume that are scheduled on the same radio resource with NOMA, , where of in the beam is denoted as . is the transmitted signal by the base station in the beam , which is the sum of all K user signals.where is the power ratio of user, , and . is the total power in the beam , is the normalized transmitted signal of user in the beam , and .
2.2. Channel Model
As for the large scale fading, the mmWave link is similar to that used in [33]; the large scale fading , in dB, is modeled aswhere , is the carrier frequency, is the path loss exponents, and is the distance from transmitter to receiver.
As for the small scale fading, the Nakagamim fading is considered for each link. are donated as the link of base station to cellular user, D2D user to cellular user, and D2D user to D2D user whose modular square is normalized Gamma random variable [39]. And when is normalized Gamma random variable, which is denoted as , the probability density function and Cumulative Distribution Function of H are
2.3. Directional Beamforming
The base station with mmWave band is equipped with multiple antennas which can generate high directional and high gain beams. The actual antenna pattern is modeled as the sectorized antenna model approximately for the sake of mathematical tractability [42]. Generally the maximum power gain is adopted to replace the array gain within the halfpower beam width (main lobe gain), and the first minor maximum gain is used to replace the gains of the other DoAs (side lobe gain). According to the literature [39, 40], when the antenna pattern is a planar square, the total array gain from base station to the user is where where , and is the number of antennas.
2.4. Received Signal
For downlink MIMONOMA transmission, the cellular user will receive the sum of signal from base station and the signal from D2D transmitter at the same time. In addition to receiving the D2D transmitted signal, the D2D receiver will also receive the interference signal from the other D2D users. Without loss of generality, we assume there are beams and D2D pairs in the cellular network; then the received signal of cellular user and the received signal of D2D user can be formulated aswhere is the channel gain for downlink , , is the channel gain between and , and is the channel gain from transmitter to receiver. is the total beamforming array gain from base station to . is the superimposed signal by the total K in the beam . is the transmitted power of . is the signal transmitted by , and . Meanwhile and are the i.i.d. white Gaussian noise with zero mean and one variance at cellular user and D2D user , which is denoted as .
We denote the SINRs of and in the downlink NOMAMIMO cellular network as , . Without loss of generality, the interbeam interference is ignored in this paper. In the meantime, the perfect SIC is used to prevent error propagation in the NOMA users in the paper. Using (1) in (6) and (7), and can be formulated aswhere
3. Performance Analysis
In this section, we present the performance analysis of D2Daided mmWave MIMONOMA system. Specifically, the closedform expressions for the performance metrics (i.e., the outage probability and the ergodic capacity) are presented in the following. Without loss of generality, in the beam , th are adopted with NOMA in one beam, , and one is randomly distributed at the edge of the beam. Three events are considered in this system.
Event 1. According to the NOMA successive interference cancellation (SIC) principle, user 1 obtains the information intended for user 2 with and removes it. When decoding the information intended for user 2, user 1 cancels it successfully with .
Event 2. User 2 decodes the signal with , treating user 1 as interference.
Event 3. The DU receiver only receives signals from the DU transmitter, whose SINR is denoted as :
3.1. Outage Probability
In this section, we study the outage probability of and . The outage probability of the user 1, user 2, and is given by : where are the target rates of user 1, user 2, and .
Firstly, we consider the outage probability of user 1 ; then can be rewritten as , so we setAccording to (13),(20), we can get where , , , , and , ; then
In order to determine if , we set . When , , so ; therefore
When , which is ,
We set , and As for , thenWe set
Then by substituting (27) into (26), can be denoted as
Similarly, by adopting different parameters with , which is shown in Table 1, we can obtain .

In addition, when , . Hence is denoted as
Finally, the outage probability of user 1, user 2, and can be evaluated by , which are formulated as
3.2. Ergodic Capacity
The ergodic capacity is the average capacity of channel which can be defined as the instantaneous endtoend mutual information expectations and denoted as
The ergodic capacity of the system can be obtained by substituting (14), (15), and (16) into (33), which is formulated as
In order to compute (34), we first compute the first item of the formula. As we set before, , , , , , we can get
According to the literature [43, 44], we can get
Based on (36), we start with and , which can be written as
By substituting (37) and (38) into (35), we can obtain the first item of formula (34), which can be denoted as
Similarly, we can set a, b, c, d as different parameters to obtain the other items of the formula (34) and then pull everything together. The asymptotic result for the ergodic capacity of the consider system can be obtained.
4. Numerical Results
In this section, the outage probability and the ergodic capacity of MIMONOMA mmWave cellular network with D2D communications are investigated. The effects of different parameters on the probability of outage and ergodic capacity are analyzed such as the base station transmission power, the number of base station antennas, the power ratio of NOMA user, and the distance between D2D users. In order to verify the performance of the system, the traditional TDMA theme is adopted as the comparison between the two users of each beam. In particular, the time slot is equally divided by the two users. Hence the capacity of this theme is , which is denoted as
A simplified cellular network system is discussed for the performance analyzed here. The carrier frequency is 28GHz which is commonly used for wireless broadband service. There are 16 antennas in the base station whose coverage radius is 100m. There is single antenna with D2D user. And the distance between D2D users is 30m. Meanwhile, the transmission power of base station and D2D users is 5 dbm. In addition, there are 8 NOMA users and 4 D2D users in the cellular network. The path loss exponent is set as 3. Furthermore, the small scale fading is denoted as which is simplified for the simulation.
4.1. Outage Probability
In this section, we consider the outage probability of the NOMA far user and near user.
Figure 2 depicts the outage probability in the different base station transmission power with = 5 bit/s/Hz and = 3.32 bit/s/Hz. As the base station transmission power increases, it can be seen that the outage probability of the NOMA users decreases with the exponential form. Furthermore, the performance of each user’s outage probability in the NOMA scheme is significantly better than the TDMA, and the closedform solution obtained is consistent with the Monte Carlo simulation results.
In Figure 3, the impact of antenna number in the base station on the outage probability ( = 5.64 bit/s/Hz and = 4 bit/s/Hz) is presented. The simulation results effectively verify that the number of antennas of the base station can decrease the users’ outage probability in the MIMONOMA mmWave cellular network, thereby improving the throughput of the system under the limited timefrequency resources. As can be seen from Figure 3, the number of antennas has a greater impact on user 2 than user 1. When the number of antennas is 36, the outage probability of the system is satisfied which can balance the number of RF chains and the system performance. Then the NOMA scheme performs better than the traditional TDMA in the mmWave MIMO cellular network with D2D communications.
In Figure 4, we discuss the influence of the power ratio coefficient in the cellular network between the NOMA users. It can be seen that the outage probability ( = 4 bit/s/Hz and = 3 bit/s/Hz) of the two users in NOMA is balanced when the power ratio coefficient is approximately 0.2.
In Figure 5, the effect of the distance between D2D users is considered. The figure indicates that the outage probability ( = 5 bit/s/Hz and = 3.32 bit/s/Hz) of the NOMA users is reduced in the form of an exponent when the distance of D2D user is linear growth. Since the distance between D2D users is increasing, the interference from the D2D transmitter to the NOMA user is weak. Hence, the throughput of the NOMA users is improved, while the outage probability is dropping.
4.2. Ergodic Capacity
In this section, the total ergodic capacity is considered in the MIMONOMA mmWave cellular network with D2D communications. It can be seen that the numerical results are consistent with the closedform solution which is better than the traditional TDMA. Meanwhile, the transmission power and the number of base station antennas have a greater impact, and the power ratio coefficient and the distance between D2D users have less effect.
In Figure 6, the impact of the transmission power of base station on the ergodic capacity is considered in the MIMONOMA mmWave cellular network with D2D communications. It is shown that the ergodic capacity is growing linearly with the increase of transmission power. In addition, the ergodic capacity of the system we proposed is higher than the traditional TDMA. Hence, in order to improve the ergodic capacity of the system, we can increase the base station transmission power as much as possible without affecting others.
As the number of the base station antennas is increasing, it is indicated that the ergodic capacity can be improved better than the traditional TDMA in Figure 7. Benefiting from the length of mmWave, more and more antennas can be equipped for the base station. At the same time, we need to balance the improvement in ergodic capacity brought by the increasing of the number of antennas and the power consumption and hardware requirements of the increase in RF chains to determine the final number of antennas.
In Figure 8, the ergodic capacity is affected by the change of the power ratio coefficient of the NOMA users in the MIMONOMA mmWave cellular network with D2D communications. It can be seen that the total ergodic capacity changes slowly with the increase of power ratio.
In Figure 9, since the interference from the D2D users is decreasing, the total ergodic capacity is improved with the increase of the distance between the D2D users in the MIMONOMA mmWave cellular network. It can also be seen that the ergodic capacity in the MIMONOMA mmWave cellular network is always better than the traditional TDMA.
5. Conclusion
In this paper, the outage probability and the ergodic capacity of the NOMA in the MIMONOMA mmWave cellular network with D2D communications are studied. The closedform solutions of the outage probability and the ergodic capacity are obtained, which are consistent with the numerical results. Meanwhile, the performance of NOMA is shown to be better than traditional TDMA in the MIMO mmWave cellular network with D2D communications. Furthermore, the higher transmission power of base station and the larger antenna array can also improve system performance.
Data Availability
No data were used to support this study.
Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.
Acknowledgments
This work was supported by Advance Research Projects of 13th FiveYear Plan of Civil Aerospace Technology (B0105) and the National Natural Science Foundation of China (61771051).
References
 M. Tehrani, M. Uysal, and H. Yanikomeroglu, “Devicetodevice communication in 5G cellular networks: challenges, solutions, and future directions,” IEEE Communications Magazine, vol. 52, no. 5, pp. 86–92, 2014. View at: Publisher Site  Google Scholar
 S.Y. Lien, C.C. Chien, G. S.T. Liu, H.L. Tsai, R. Li, and Y. J. Wang, “Enhanced LTE devicetodevice proximity services,” IEEE Communications Magazine, vol. 54, no. 12, pp. 174–182, 2016. View at: Publisher Site  Google Scholar
 L. Lei, Z. D. Zhong, C. Lin, and X. M. Shen, “Operator controlled devicetodevice communications in LTEadvanced networks,” IEEE Wireless Communications Magazine, vol. 19, no. 3, pp. 96–104, 2012. View at: Publisher Site  Google Scholar
 A. Asadi and V. Mancuso, “Networkassisted outband D2Dclustering in 5G cellular networks: theory and practice,” IEEE Transactions on Mobile Computing, vol. 16, no. 8, pp. 2246–2259, 2017. View at: Publisher Site  Google Scholar
 J. Hu, W. Heng, Y. Zhu, G. Wang, X. Li, and J. Wu, “Overlapping coalition formation games for joint interference management and resource allocation in D2D communications,” IEEE Access, vol. 6, pp. 6341–6349, 2018. View at: Publisher Site  Google Scholar
 H. Min, J. Lee, S. Park, and D. Hong, “Capacity enhancement using an interference limited area for devicetodevice uplink underlaying cellular networks,” IEEE Transactions on Wireless Communications, vol. 10, no. 12, pp. 3995–4000, 2011. View at: Publisher Site  Google Scholar
 Z. Uykan and R. Jantti, “Transmissionorder optimization for bidirectional devicetodevice (D2D) communications underlaying cellular TDD networks—a graph theoretic approach,” IEEE Journal on Selected Areas in Communications, vol. 34, no. 1, pp. 1–14, 2016. View at: Publisher Site  Google Scholar
 L. L. Wei, R. Q. Hu, T. He, and Y. Qian, “Devicetodevice(d2d) communications underlaying MUMIMO cellular networks,” in Proceedings of the IEEE Global Communications Conference (GLOBECOM '13), pp. 4902–4907, IEEE, Atlanta, Ga, USA, December 2013. View at: Publisher Site  Google Scholar
 X. Li, W. Zhang, H. Zhang, and W. Li, “A combining call admission control and power control scheme for D2D communications underlaying cellular networks,” China Communications, vol. 13, no. 10, pp. 137–145, 2016. View at: Publisher Site  Google Scholar
 H. Sun, Y. Xu, and R. Q. Hu, “A NOMA and MUMIMO supported cellular network with underlaid D2D communications,” in Proceedings of the 2016 IEEE 83rd Vehicular Technology Conference (VTC Spring), pp. 1–5, Nanjing, China, May 2016. View at: Publisher Site  Google Scholar
 Z. Ding, X. Lei, G. K. Karagiannidis, R. Schober, J. Yuan, and V. K. Bhargava, “A survey on nonorthogonal multiple access for 5G networks: research challenges and future trends,” IEEE Journal on Selected Areas in Communications, vol. 35, no. 10, pp. 2181–2195, 2017. View at: Publisher Site  Google Scholar
 N. Ye, X. Li, H. Yu, A. Wang, W. Liu, and X. Hou, “Deep learning aided grantfree noma towards reliable lowlatency access in tactile internet of things,” IEEE Transactions on Industrial Informatics, vol. 15, no. 5, pp. 2995–3005, 2019. View at: Publisher Site  Google Scholar
 J. An, K. Yang, J. Wu, N. Ye, S. Guo, and Z. Liao, “Achieving sustainable ultradense heterogeneous networks for 5G,” IEEE Communications Magazine, vol. 55, no. 12, pp. 84–90, 2017. View at: Publisher Site  Google Scholar
 N. Ye, A. Wang, X. Li, H. Yu, A. Li, and H. Jiang, “A random nonorthogonal multiple access scheme for mmtc,” in Proceedings of the 2017 IEEE 85th Vehicular Technology Conference (VTC Spring), pp. 1–6, June 2017. View at: Google Scholar
 K. Yang, N. Yang, N. Ye, M. Jia, Z. Gao, and R. Fan, “Nonorthogonal multiple access: achieving sustainable future radio access,” IEEE Communications Magazine, vol. 57, no. 2, pp. 116–121, 2019. View at: Publisher Site  Google Scholar
 S. M. R. Islam, N. Avazov, O. A. Dobre, and K.S. Kwak, “Powerdomain nonorthogonal multiple access (NOMA) in 5G systems: potentials and challenges,” IEEE Communications Surveys & Tutorials, vol. 19, no. 2, pp. 721–742, 2017. View at: Publisher Site  Google Scholar
 N. Ye, A. Wang, X. Li, W. Liu, X. Hou, and H. Yu, “On constellation rotation of noma with sic receiver,” IEEE Communications Letters, vol. 22, no. 3, pp. 514–517, 2018. View at: Publisher Site  Google Scholar
 M. Hojeij, C. A. Nour, J. Farah, and C. Douillard, “Joint resource and power allocation technique for downlink powerdomain nonorthogonal multiple access,” in Proceedings of the 2018 IEEE Conference on Antenna Measurements & Applications (CAMA), pp. 1–4, September 2018. View at: Publisher Site  Google Scholar
 F. A. Rabee, K. Davaslioglu, and R. Gitlin, “The optimum received power levels of uplink nonorthogonal multiple access (NOMA) signals,” in Proceedings of the 18th IEEE Wireless and Microwave Technology Conference, WAMICON 2017, pp. 1–4, USA, April 2017. View at: Google Scholar
 Y. Li and G. A. A. Baduge, “Nomaaided cellfree massive mimo systems,” IEEE Wireless Communications Letters, vol. 7, pp. 950–953, 2018. View at: Google Scholar
 N. Ye, A. Wang, X. Li et al., “Rateadaptive multiple access for uplink grantfree transmission,” Wireless Communications and Mobile Computing, vol. 2018, Article ID 8978207, 21 pages, 2018. View at: Google Scholar
 N. Ye, H. Han, L. Zhao, and A.H. Wang, “Uplink nonorthogonal multiple access technologies toward 5G: a survey,” Wireless Communications and Mobile Computing, vol. 2018, Article ID 6187580, 26 pages, 2018. View at: Google Scholar
 Y. Liu, W.J. Lu, S. Shi et al., “Performance analysis of a downlink cooperative noma network over nakagamim fading channels,” IEEE Access, vol. 6, pp. 53034–53043, 2018. View at: Publisher Site  Google Scholar
 X. Wang, J. Wang, L. He, and J. Song, “Outage analysis for downlink noma with statistical channel state information,” IEEE Wireless Communications Letters, vol. 7, no. 2, pp. 142–145, 2018. View at: Publisher Site  Google Scholar
 A. J. Paulraj, D. A. Gore, R. U. Nabar, and H. Bölcskei, “An overview of MIMO communications—a key to gigabit wireless,” Proceedings of the IEEE, vol. 92, no. 2, pp. 198–217, 2004. View at: Publisher Site  Google Scholar
 W. Cai, C. Chen, L. Bai, Y. Jin, and J. Choi, “User selection and power allocation schemes for downlink NOMA systems with imperfect CSI,” in Proceedings of the 2016 IEEE 84th Vehicular Technology Conference (VTCFall), pp. 1–5, Montreal, QC, Canada, September 2016. View at: Publisher Site  Google Scholar
 Z. Ding and H. V. Poor, “Design of massiveMIMONOMA with limited feedback,” IEEE Signal Processing Letters, vol. 23, no. 5, pp. 629–633, 2016. View at: Publisher Site  Google Scholar
 Z. Ding, F. Adachi, and H. V. Poor, “The application of MIMO to nonorthogonal multiple access,” IEEE Transactions on Wireless Communications, vol. 15, no. 1, pp. 537–552, 2016. View at: Publisher Site  Google Scholar
 Q. Sun, S. Han, I. ChinLin, and Z. Pan, “On the ergodic capacity of MIMO NOMA systems,” IEEE Wireless Communications Letters, vol. 4, no. 4, pp. 405–408, 2015. View at: Publisher Site  Google Scholar
 J. Ding, J. Cai, and C. Yi, “An improved coalition game approach for MIMONOMA clustering integrating beamforming and power allocation,” IEEE Transactions on Vehicular Technology, vol. 68, no. 2, pp. 1672–1687, 2019. View at: Publisher Site  Google Scholar
 J.B. Kim, I.H. Lee, and J. Lee, “Capacity scaling for D2D aided cooperative relaying systems using NOMA,” IEEE Wireless Communications Letters, vol. 7, no. 1, pp. 42–45, 2018. View at: Publisher Site  Google Scholar
 S. Papaioannou, G. Kalfas, C. Vagionas et al., “5G mm Wave Networks Leveraging Enhanced FiberWireless Convergence for HighDensity Environments: The 5GPHOS Approach,” in Proceedings of the 2018 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB), pp. 1–5, Valencia, Spain, June 2018. View at: Publisher Site  Google Scholar
 S. A. Naqvi and S. A. Hassan, “Combining NOMA and mmWave technology for cellular communication,” in Proceedings of the 2016 IEEE 84th Vehicular Technology Conference (VTCFall), pp. 1–5, Montreal, QC, Canada, September 2016. View at: Publisher Site  Google Scholar
 F. Rusek, D. Persson, B. K. Lau et al., “Scaling up MIMO: opportunities and challenges with very large arrays,” IEEE Signal Processing Magazine, vol. 30, no. 1, pp. 40–60, 2013. View at: Publisher Site  Google Scholar
 T. Bai, A. Alkhateeb, and R. W. Heath, “Coverage and capacity of millimeterwave cellular networks,” IEEE Communications Magazine, vol. 52, no. 9, pp. 70–77, 2014. View at: Publisher Site  Google Scholar
 X. Gao, L. Dai, S. Han, I. ChihLin, and R. W. Heath, “Energyefficient hybrid analog and digital precoding for MmWave MIMO systems with large antenna arrays,” IEEE Journal on Selected Areas in Communications, vol. 34, no. 4, pp. 998–1009, 2016. View at: Publisher Site  Google Scholar
 X. Gao, L. Dai, Z. Chen, Z. Wang, and Z. Zhang, “Nearoptimal beam selection for beamspace mmwave massive mimo systems,” IEEE Communications Letters, vol. 20, no. 5, pp. 1054–1057, 2016. View at: Publisher Site  Google Scholar
 Z. Wang, M. Li, Q. Liu, and A. L. Swindlehurst, “Hybrid precoder and combiner design with lowresolution phase shifters in mmWave MIMO systems,” IEEE Journal of Selected Topics in Signal Processing, vol. 12, no. 2, pp. 256–269, 2018. View at: Publisher Site  Google Scholar
 Y. Sun, Z. Ding, and X. Dai, “On the performance of downlink NOMA in multicell mmWave networks,” IEEE Communications Letters, vol. 22, no. 11, pp. 2366–2369, 2018. View at: Publisher Site  Google Scholar
 N. Deng and M. Haenggi, “A finegrained analysis of millimeterwave devicetodevice networks,” IEEE Transactions on Communications, vol. 65, no. 11, pp. 4940–4954, 2017. View at: Publisher Site  Google Scholar
 D. Zhang, Z. Zhou, C. Xu, Y. Zhang, J. Rodriguez, and T. Sato, “Capacity analysis of NOMA with mmWave massive MIMO systems,” IEEE Journal on Selected Areas in Communications, vol. 35, no. 7, pp. 1606–1618, 2017. View at: Publisher Site  Google Scholar
 S. Singh, M. N. Kulkarni, A. Ghosh, and J. G. Andrews, “Tractable model for rate in selfbackhauled millimeter wave cellular networks,” IEEE Journal on Selected Areas in Communications, vol. 33, no. 10, pp. 2191–2211, 2015. View at: Publisher Site  Google Scholar
 X. Yan, H. Xiao, C.X. Wang, and K. An, “On the ergodic capacity of NOMAbased cognitive hybrid satellite terrestrial networks,” in Proceedings of the 2017 IEEE/CIC International Conference on Communications in China, ICCC 2017, pp. 1–5, China, October 2017. View at: Google Scholar
 Y. Huang, F. AlQahtani, C. Zhong, Q. Wu, J. Wang, and H. Alnuweiri, “Performance analysis of multiuser multiple antenna relaying networks with cochannel interference and feedback delay,” IEEE Transactions on Communications, vol. 62, no. 1, pp. 59–73, 2014. View at: Publisher Site  Google Scholar
Copyright
Copyright © 2019 Jianguo Li et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.