Secure NOMA method based on physical layer security

Abstract

Disclosed is a secure scheme that incorporates both internal and external eavesdroppers to secure all users' links in the downlink PD-NOMA. In particular, the proposed scheme implies that the BS induces a phase shift in each user's symbol based on its corresponding instantaneous channel phase. The phase shift of each user is restricted such that the original symbol is moved to the location of another symbol in the constellation diagram. Therefore, as each user is aware of its instantaneous channel phase only, it will be able to recover the actual phase of its corresponding symbol. Thus, the proposed scheme does not only protect the data against eavesdroppers, but it also guarantees confidentiality and privacy against all other users.

Claims

1. A secure downlink non-orthogonal multiple access (NOMA) method for use against an unknown internal eavesdropper, the secure downlink NOMA method comprising: inducing a phase shift on each symbol; merging and sending a superposed signal to a user from a base station; performing successive interference cancellation for each user by detecting a symbol of other users starting with a first user until reaching a symbol therefor by subtracting the detected symbol from a received symbol at each iteration and carrying a resulting signal to a next iteration; and performing another phase shift to detect a confidential signal.

2. The secure downlink NOMA method of claim 1, wherein the base station extracts a specific phase shift for each user in order to be used to rotate a corresponding symbol thereof.

3. The secure downlink NOMA method of claim 1, wherein the phase shift on each symbol of each user is in accordance with ${}_n=\left(\left[\frac{{}_n}{2/M}\right]+1\right)\frac{2}{M},$ wherein .sub.n is a phase of h.sub.n and M is a modulation level.

4. A secure uplink massive machine communication NOMA method for use against an external eavesdropper, the secure uplink massive machine communication NOMA method comprising: modulating a symbol of each user for delivery to a base station; shifting the symbol by a phase value; transmitting signals of the user in accordance with allocated power to the base station; receiving signals from all users simultaneously at the base station; detecting a signal of a strongest user and applying another phase shift in order to demodulate the signal of the strongest user; applying iterative successive interference cancellation to detect signals of remaining users; and applying a further phase shift until all signals of the users has been detected.

5. The secure uplink massive machine communication NOMA method of claim 4, wherein the phase shift on each symbol of each user is in accordance with ${}_n=\left(\left[\frac{{}_n}{2/M}\right]+1\right)\frac{2}{M},$ wherein .sub.n is a phase of h.sub.n and M is a modulation level.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows the flowchart of the secure downlink NOMA method against an unknown internal eavesdropper.

(2) FIG. 2 shows the flowchart of the secure uplink mMTC-NOMA method against the external eavesdropper.

DETAILED DESCRIPTION

(3) In this invention, a novel secure Non-Orthogonal Multiple Access (NOMA) scheme based on physical layer security concepts is proposed. The proposed scheme exploits the random and independent channel characteristics to induce a different phase shift in each users symbol. Based on the assumption that the instantaneous channel phase between a user and the base station is available only at both ends, other users will not be able to decode the right symbol. The proposed scheme does not impact the performance at users' ends, and at the same time, can guarantee data confidentiality of each user against both internal and external eavesdropping.

(4) In the first scenario, a secure downlink NOMA system against an unknown internal eavesdropper is proposed. In the BS, each user signal is modulated to be delivered.

(5) Then, different power levels for each user is assigned based on its channel conditions, where the best user (usually the nearest) is allocated the lowest power level, and the highest power level is allocated to the worst user (usually the farthest). Denoting the power coefficient of user n by .sub.n, the allocated power coefficients should satisfy:

(6) $\begin{array}{cc}{.Math.}_{n=1}^N{}_n=1,\mathrm{and}{}_1<{}_2<.Math.<{}_n.& \left(1\right)\end{array}$

(7) Based on the proposed scheme, the BS induces a phase shift on each symbol before merging the transmitted symbols. Specifically, the BS will exploit its knowledge of the CSI of all users in order to extract a specific phase shift for each particular user that will be used to rotate its corresponding symbol. However, the challenge in NOMA systems is that the BER at a user mainly relies on its ability to correctly detect the other users' symbols. As such, the induced phase shift on a user should be carefully selected not to affect other users' performance.

(8) Being a complex value, block fading channel h.sub.n can be expressed as h.sub.n=j.sub.ne.sup.in, where j.sub.n and .sub.n represent the magnitude and the phase of h.sub.n, respectively, and i is the imaginary unit. Accordingly, the phase shift on the symbol of user n, denoted by .sub.n, is given as follows:

(9) $\begin{array}{cc}{}_n=\left(\left[\frac{{}_n}{2/M}\right]+1\right)\frac{2}{M}.& \left(2\right)\end{array}$

(10) It should be noted the phase shift will rotate the corresponding symbol by a multiple of 2/M, and hence, it will appear as another symbol on the constellation diagram. Thus, users (other than the intended user) will detect it as one of the candidate symbols (constellation points), and only the intended user (that is aware of .sub.n) can recover the original phase of the symbol. FIG. 1 illustrates the proposed secure scheme for PD-NOMA systems.

(11) Following the proposed scheme, the received signal at the user n is now given by

(12) $\begin{array}{cc}{y}_n=\sqrt{P}{h}_n\left({.Math.}_{k=1}^N\sqrt{{}_k}{e}^{-i{}_{k_{S_k}}}\right)+w_n.& \left(3\right)\end{array}$ where w.sub.n is denote the additive white Gaussian noise (AWGN) with zero mean and .sup.2 variance w.sub.n, N(0, .sup.2) at the n.sup.th user's. Besides, all links are modeled by large-scale (path-loss) and small-scale fading

(13) $h_n=d_n^{-\frac{n}{2}}{f}_n$
where f.sub.n is modeled by the Rayleigh fading channel gain f.sub.nN(0, .sup.2). And d.sub.n is the separation distance between the transmitter and the receiver n, and n is the path loss exponent.

(14) Accordingly, the SIC detection to accommodate the induced phase as:

(15) $\begin{array}{cc}{\hat{s}}_n=\arg \underset{l=1:M}{\min }.Math.y_n-\sqrt{P}{h}_n\left({.Math.}_{k=n+1}^N\sqrt{{}_k}{\hat{s}}_k\right)-\sqrt{P{}_n}{h}_n{e}^{-i{}_{n_{S_l}}}.Math.& \left(4\right)\end{array}$

(16) Each user performs SIC. SIC implies that each user performs an iterative maximum likelihood detection. Specifically, given the power levels order, user n detects the other users' symbols starting with the first user (i.e., user N) until reaching its symbol. At each iteration, the detected symbol is subtracted from the received signal, and the result is passed to the next iteration. This process allows a user to detect only the symbols of the users that have higher power coefficients than its coefficient. In other words, the user N signal is detected and subtracted from the received signal (y.sub.n) to find the user (N1) signal and these procedures are repeated until the indented user detects his signal.

(17) In the second scenario, a secure uplink mMTC-NOMA scheme against an external eavesdropper is defined. The mMTC refers to provide a massive connection between a large number of devices that transmit a small amount of data traffic like IoT applications, healthcare sensors, smart homes, etc. In up-link NOMA-based mMTC networks, multiple MTC devices (MTCDs) utilized the subchannel for transmission at the same time but with different power allocation. And each user allocated power due to their channel condition. Thus, the MTCD with the worst channel is allocated more power compare with the MTCD with better channel conditions. By assuming that the allocated power coefficients .sub.1>.sub.2> . . . >.sub.n due to the channel gain condition |h.sub.1|.sup.2<|h.sub.n|.sup.2< . . . <|h.sub.N|.sup.2.

(18) As shown in FIG. 2, each MTCD will shift own original phase of the symbol by the amount of phase determined in Eq. (2) then transmit the signal to the BS. the received signals at legitimate BS and Eve-BS will be:

(19) $\begin{array}{cc}{y}_d={.Math.}_{k=1}^N{h}_n\sqrt{{}_n{P}_n}{s}_n{e}^{-i{}_k}+w_{d,n}& \left(5\right)\end{array}$$\begin{array}{cc}{y}_e={.Math.}_{k=1}^N{g}_n\sqrt{P_n}{s}_n{e}^{-i{}_k}+w_{e,n}& \left(6\right)\end{array}$
where w.sub.d,n and w.sub.e,n are denote the additive white Gaussian noise (AWGN) with zero mean and .sup.2 variance w.sub.d,n, w.sub.e,nN(0, .sup.2) between user n, legitimate BS, and Eve-BS, respectively. Besides, all links are modeled by large-scale (path-loss) and small-scale fading in such a way that h.sub.n and

(20) $g_n=d_n^{-\frac{n}{2}}{f}_n$
where f.sub.n is modeled by the Rayleigh fading channel gain f.sub.nN(0, .sup.2). And do is the separation distance between the MTCD and the BS, and n is the path loss exponent.

(21) Both legitimate BS and external eavesdropper exploit the SIC technology to decode the superposed received signals. SIC implies an iterative procedure where a BS first detects the signal of the strongest MTCD. Then subtracts this signal from the received signal until decode all MTCD receiver signals. At each iteration, the BS considers all other strong MTCD received signals as interference while weak MTCD received signals as noise. Since the CSI used to determine the phase shift (.sub.n) and secure the information is available only to the MTCDs and legitimate BS. The external eavesdropper will evaluate differently phase value

(22) ${}_n=\left(\left[\frac{{}_n}{2/M}\right]+1\right)\frac{2}{M}$
wherein .sub.n, the phase of g.sub.n. Thus, the legitimate BS will correctly detect the information for each user while external eavesdropper will not.

REFERENCES

(23) [1] N. Zhao, D. Li, M. Liu, Y. Cao, Y. Chen, Z. Ding, and X. Wang, Secure transmission via joint precoding optimization for downlink MISO NOMA, IEEE Trans. Veh. Tech., vol. 68, no. 8, pp. 7603-7615, 2019. [2] H. Bao, C. Zhang, L. Wu, and M. Li, Design of physical layer secure transmission scheme based on SWIPT NOMA systems, in IEEE ICCT, 2017, pp. 6-9. [3] B. Chen, Y. Chen, Y. Chen, Y. Cao, Z. Ding, N. Zhao, and X. Wang, Secure primary transmission assisted by a secondary full-duplex NOMA relay, IEEE Transactions on Vehicular Technology, vol. 68, no. 7, pp. 7214-7219, 2019. [4] Z. Qin, Y. Liu, Z. Ding, Y. Gao, and M. Elkashlan, Physical layer security for 5G non-orthogonal multiple access in large-scale networks, in IEEE ICC, 2016, pp. 1-6. [5] N. Rupasinghe, Y. Yapici, I. Guvenc, H. Dai, and A. Bhuyan, Enhancing physical layer security for NOMA transmission in mmWave drone networks, in 2018 52nd Asilomar Conference on Signals, Systems, and Computers. IEEE, 2018, pp. 729-733. [6] Y. Yapici, I. Guvenc, H. Dai, A. Bhuyan, and N. Rupasinghe, Physical layer security for mmwave drone links with noma. [7] Y. Yapici, N. Rupasinghe, I. Guvenc, H. Dai, and A. Bhuyan, \Physical layer security for noma transmission in mmwave drone networks, arXiv preprint arXiv:2001.04863, 2020. [8] G. Gomez, F. J. Martin-Vega, F. Javier Lopez-Martinez, Y. Liu, and M. Elkashlan, Physical layer security in uplink noma multi-antenna systems with randomly distributed eavesdroppers, IEEE Access, vol. 7, pp. 70422-70435, 2019. [9] H. Lei, J. Zhang, K.-H. Park, P. Xu, I. S. Ansari, G. Pan, B. Alomair, and M.-S. Alouini, On secure NOMA systems with transmit antenna selection schemes, IEEE Access, vol. 5, pp. 17450-17464, 2017. [10] Y. Liu, Z. Qin, M. Elkashlan, Y. Gao, and L. Hanzo, Enhancing the physical layer security of nonorthogonal multiple access in large-scale networks, IEEE Trans. Wireless Comm., vol. 16, no. 3, pp. 1656-1672, 2017. [11] L. Lv, Z. Ding, Q. Ni, and J. Chen, Secure MISO-NOMA transmission with articial noise, IEEE Trans. Veh. Tech., vol. 67, no. 7, pp. 6700-6705, 2018. [12] Y. Li, M. Jiang, Q. Zhang, Q. Li, and J. Qin, Secure beamforming in downlink MISO nonorthogonal multiple access systems, IEEE Trans. Veh. Tech., vol. 66, no. 8, pp. 7563-7567, 2017. [13] B. M. ElHalawany and K. Wu, Physical-layer security of NOMA systems under untrusted users, in 2018 IEEE GLOBECOM, 2018, pp. 1-6. [14] Y. Cao, N. Zhao, Y. Chen, M. Jin, Z. Ding, Y. Li, and F. R. Yu, Secure transmission via beamforming optimization for NOMA networks, IEEE Wireless Comm., 2019. [15] B. M. ElHalawany and K. Wu, Physical-layer security of noma systems under untrusted users, in 2018 IEEE Global Communications Conference (GLOBECOM), 2018, pp. 1-6. [16] K. Cao, B. Wang, H. Ding, T. Li, and F. Gong, Optimal relay selection for secure noma systems under untrusted users, IEEE Transactions on Vehicular Technology, vol. 69, no. 2, pp. 1942-1955, 2020. [17] Y. Li, M. Jiang, Q. Zhang, Q. Li, and J. Qin, Secure beamforming in downlink miso nonorthogonal multiple access systems, IEEE Transactions on Vehicular Technology, vol. 66, no. 8, pp. 7563-7567, 2017. [18] K. Cao, B. Wang, H. Ding, T. Li, J. Tian, and F. Gong, Secure transmission designs for noma systems against internal and external eavesdropping, IEEE Transactions on Information Forensics and Security, vol. 15, pp. 2930-2943, 2020. [19] R. M. Christopher and D. K. Borah, Physical layer security for weak user in miso noma using directional modulation (nomad), IEEE Communications Letters, vol. 24, no. 5, pp. 956-960, 2020.