JOINT RANDOM SUBCARRIER SELECTION AND CHANNEL-BASED ARTIFICIAL SIGNAL DESIGN AIDED PLS

Abstract

In the area of Joint Random Subcarrier Selection and Channel-Based Artificial Signal Design Aided PLS, a method for providing physical layer security (PLS) depending on the randomness of wireless channel is proposed. Specifically, a channel-based joint random subcarrier selection and artificial signal design are introduced to protect the communication in the presence of a passive eavesdropper which can be even stronger than the legitimate receiver. Our analysis assumes a window-based subcarrier selection method in which the strongest subcarriers in each window are selected. Chosen subcarriers are considered for secret sequence extraction. The generated channel dependent secret sequence is used for both random subcarrier selection and artificial signal design. We evaluate the efficiency of the proposed method through some representative metrics, such as secret sequence disagreement rate (SSDR), throughput and bit error rate (BER), in both perfect and imperfect channel estimation cases. Simulation results are presented and insightful discussions are drawn.

Claims

1. A joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) system comprising; two random subcarrier selection schemes as a window-based subcarrier selection using strongest subcarriers of each window for secret sequence extraction and a channel-based random subcarrier selection scheme for adding artificial signal to the information data to protect the transmitter's confidential data.

2. An operation method of a joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) system according to claim 1, comprising steps of; Finding strong subcarriers in a single-antenna legitimate receiver channel, Quantization on found strong subcarriers and secret sequence generation, Comparing minimum required length of secret sequence (L.sub.min) and L, If comparing minimum required length (L.sub.min) is bigger than L, reshaping of the secret sequence, then choose the first L.sub.min bits, If comparing minimum required length (L.sub.min) is not bigger than L then choose the first L.sub.min bits, After choosing the L.sub.min bits, make small packets of secret sequence with the length of M, Dividing the length of M bits to N for subcarrier number selection, After doing subcarriers number selection based on the generated secret sequence, adding artificial signal to the information data to protect the transmitter's confidential data where for designing artificial signal, the extracted secret sequence from the channel is modulated with the same order as information data, Sending the transmitted signal as X∈C.sup.1×K to the legitimate user (a single-antenna legitimate receiver) after applying cyclic prefix (CP) to the time-domain encrypted symbols to avoid inter symbol interference (ISI), Receiving the signal as a y.sub.b=h.sub.b*x+n.sub.b at a single-antenna legitimate receivers side, After removing CP and applying serial-to-parallel (S/P) conversion on the time domain received signal, using the fast Fourier transform (FFT) on the resulting signal by single-antenna legitimate receiver, Performing a zero-forcing channel equalization process, Receiving the transmitter's signal (a) at single-antenna legitimate receivers side after channel equalization process is found by element-wise division of the received signal and his channel, as a =Y.sub.b ØH.sub.b, After P/S conversion, legitimate receiver (b) generating the information signal by single-antenna legitimate receivers by subtracting the artificial signal from single-antenna legitimate receivers received signal, Extracting the information signal as a $S_b\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_b\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_b},\\ \left(k\right),& \mathrm{otherwise},\end{array}$ Capturing the signal as a y.sub.e=h.sub.e*x+n.sub.e from single-antenna eavesdroppers channel, Receiving the signal at the single-antenna eavesdropper side after channel equalization process as =Y.sub.e ØH.sub.e, Generating the single-antenna eavesdropper information signal by the single-antenna eavesdropper based on her channel by following the same algorithm as a single-antenna legitimate receiver, Extracting information signal in a single-antenna eavesdropper can be expressed as $S_e\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_e\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_e},\\ \left(k\right),& \mathrm{otherwise},\end{array}.$

3. A method for a joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) according to claim 2; wherein only strong subcarriers out of whole in each window have chosen to extract more reliable and random secret sequence at legitimate receiver.

Description

DEFINITION OF THE FIGURES OF THE INVENTION

[0026] The figures have been used in order to further disclose the joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) developed by the present invention which the figures have been described below:

[0027] FIG. 1: System model consisting of a single-antenna transmitter (Alice), a single-antenna legitimate receiver (Bob), and a single-antenna eavesdropper (Eve).

[0028] In FIG. 1:

[0029] A: Is the transmitter which is called Alice

[0030] B: Is the legitimate receiver which is called Bob

[0031] E: Is the eavesdropper which is called Eve

[0032] 1: Is the reference signal between transmitter and legitimate receiver which is used for channel estimation purpose.

[0033] FIG. 2: Flowchart of the proposed algorithm for random subcarrier selection.

[0034] FIG. 3: Effect of imperfect channel reciprocity on extracted secret sequence in the proposed technique.

[0035] In FIG. 3:

[0036] SNR(dB): Is the signal to noise ratio.

[0037] e: Is the noise power which is used to show the level of imperfect channel reciprocity.

[0038] Bob: Is the legitimate user.

[0039] FIG. 4: SSDR comparison between the proposed window-based subcarrier selection method and the scheme using all subcarriers for secret sequence extraction: a) e=1, b) e=0.1.

[0040] In FIG. 4 for both a and b part:

[0041] SSDR: Is the secret sequence disagreement rate.

[0042] Q: Is the quantization level which is applied on chosen strong subcarriers to generate secret sequence.

[0043] SNR(dB): Is the signal to noise ratio.

[0044] FIG. 5: Throughput performance of the proposed approach

[0045] In FIG. 5:

[0046] SNR(dB): Is the signal to noise ratio.

[0047] e: Is the noise power which is used to show the level of imperfect channel reciprocity.

[0048] Bob: Is the legitimate user.

[0049] Eve: Is the eavesdropper.

[0050] FIG. 6: BER performance of the proposed approach.

[0051] In FIG. 6:

[0052] BER: Is the bit error rate.

[0053] SNR(dB): Is the signal to noise ratio.

[0054] e: Is the noise power which is used to show the level of imperfect channel reciprocity.

[0055] Bob: Is the legitimate user.

[0056] Eve: Is the eavesdropper.

DETAILED DESCRIPTION OF THE INVENTION

[0057] The novelty of the invention has been described with examples that shall not limit the scope of the invention and which have been intended to only clarify the subject matter of the invention. The present invention has been described in detail below.

[0058] A novel method for physical layer security was presented. A channel-based random subcarrier selection and artificial signal design were proposed. High level of randomness due to window-based subcarrier selection achieved in the generated secret sequence as this subcarrier selection method implied in more uncorrelated subcarriers. Moreover, this method improved reliability for generated secret sequence only at legitimate node as well. Finally, the proposed window-based subcarrier selection method was compared with the scheme using all subcarriers for secret sequence extraction and results revealed the effectiveness of our method. For future works, an active eavesdropper instead of a passive eavesdropper can be considered. In the proposed method, this is challenging for finding the legitimate user while sending pilot signals for channel estimation purposes.

[0059] System Model:

[0060] System model consisting of a single-antenna transmitter (Alice), a single-antenna legitimate receiver (Bob), and a single-antenna eavesdropper (Eve).

[0061] The scenario considered in this patent contains a transmitter, called Alice, which sends a secret data and communicate confidentially with a legitimate user, called Bob, in the presence of a passive eavesdropper, called Eve. Eve's aim is to access the secret message content from the communication link between Alice and Bob through his/her own observations of the signals. Eve can be stronger than Bob in the sense of having multiple antennas, more power, off-line processing, hardware capabilities and better signal processing skills, and his/her location is not known by the transmitter. It is assumed that the channels of Bob and Eve are independent and uncorrelated from each other which means Eve is located at least half-wavelength away from Bob. Also, all received signals experience Rayleigh frequency-selective fading channel. It is assumed that the channel state information (CSI) of Bob is known at Alice by using the reciprocity property in a TDD system, but Alice doesn't have any knowledge about Eve's channel, since she/he is passive. Therefore, the channels between Alice and Bob, Alice-to-Bob and Bob-to-Alice, are assumed to be estimated as correlated with each other in TDD mode [8].

[0062] In considered scenario, Bob first transmits a reference signal to Alice. Then, Alice estimates the channel between herself and Bob using this reference signal. Exploiting the channel reciprocity property in TDD mode, in which the downlink channel is obtained from its uplink [9], there is no need to share the channel. The proposed method is based on using orthogonal frequency division multiplexing (OFDM) system. The bits, which are mapped by using BPSK modulation, are sent by Alice to Bob, in the presence of a passive Eve. The frequency-domain complex data symbols having the length of K is represented by S=[S.sub.1, S.sub.2, . . . , S.sub.K], where S∈.sup.1×K.

[0063] A. Secret Sequence Extraction:

[0064] The frequency response of the channel experienced by Bob and Eve are denoted by H.sub.b ∈.sup.K×1 and H.sub.e ∈.sup.K×1, respectively, where K denotes the channel length. The secret sequence is extracted by applying proposed window-based subcarriers selection method. In this method, firstly the length of window, W, is determined. This length shows the total number of subcarriers which are considered in each window. In addition, let P be the number of selected subcarriers (P<W) out of W points which are selected from the frequency response of Bob's channel whose gains are the highest ones among all subcarriers in each window, H.sub.b=[H.sub.b.sub.1, H.sub.b.sub.2, . . . , H.sub.b.sub.K].sup.T These P subcarriers corresponding to the strongest subcarriers in each window are considered by Alice to extract the secret sequence. The proposed window-based method selects only strong subcarriers from channel. The number P of selected subcarriers is assumed to be fixed in each window. Extracting secret sequence from these P points increases the randomness of the Bob's channel and makes his generated secret sequence more uncorrelated with eavesdropper's one. It is worth to mention that such a method decreases the correlation between Bob's channel and eavesdropper's channel, implying in more selective channel at Bob automatically. This is done without using any filter or any other techniques which has cost and complexity.

[0065] B. Random Subcarriers Selection:

[0066] Random subcarriers selection method is proposed to select those subcarriers which are considered to carry both artificial signal and secret data. After selecting strong subcarriers in each window of H.sub.b, Bob's channel is quantized by Alice and Bob to construct a secret sequence from these chosen strong subcarriers. Bob's channel gain measurements are equally divided into regions and each region is quantized into multibit quantization levels by both Alice and Bob. Each of the selected subcarriers corresponds to a bit stream. In the proposed method, a minimum length of secret sequence is required, and it is represented by L.sub.min. This minimum length is defined as L.sub.min=|I.sub.TSC|× N, where N shows the number of bits that are defining a subcarrier number and |I.sub.TSC| denotes the total number of subcarriers. The relation between W and N is defined by W=2.sup.N. The length of secret sequence is related to the number of chosen subcarriers. In other words, it is related to the length of window, W, and used subcarriers in each window, P. Considering these two parameters, there exist some cases in which the length of secret sequence, L, is less than the minimum required length, L<L.sub.min. In this case, the generated secret sequence is reshaped to reach the desired length which is L.sub.min. The length of secret sequence is subtracted from minimum length of secret sequence, L.sub.min then secret sequence samples from the head are added as a suffix to achieve the minimum required length, L.sub.min. On the other hand, if the length of secret sequence is higher than minimum required length, L>L.sub.min, only the first L.sub.min bits are considered for random subcarriers selection algorithm. In order to ensure minimum required length, L.sub.min the secret sequence is divided into small blocks with length M. Length of each sub-block, M, represents the total number of bits in each window. The number of bits that represents a subcarrier number, N, is defined by M=W×N. Since this secret sequence is channel dependent and the channel is random, the chosen subcarriers are selected randomly. Total number of subcarriers is expressed as |I.sub.TSC| and the number of selected subcarriers applying random subcarrier selection method as |I.sub.SSC|. The ratio between |I.sub.SSC| and |I.sub.TSC| is defined as

[00001]$0<\left(\frac{.Math.I_{\mathrm{SSC}}.Math.}{.Math.I_{\mathrm{TSC}}.Math.}\right)\le 1,$

which is random. As this ratio goes close to 1, the security level of system becomes higher. There exist some cases in which one specific subcarrier number is repeated more than one time in a window. In this case, this subcarrier number is counted one time and the repetition is not considered. FIG. 2 represents the flowchart of the proposed algorithm. It shows all steps and details of the proposed random subcarrier selection method.

[0067] C. Artificial Signal Design and How to Add It to the Information Signal:

[0068] After choosing subcarriers randomly, the artificial signal needs to be added to information signal in these specific subcarriers, while remaining subcarriers are carrying only information signal. For designing artificial signal, the extracted secret sequence from the channel is modulated with same order as information data. The proposed method for adding artificial signal is defined as

[00002]$\begin{array}{cc}X\left(k\right)=\{\begin{array}{cc}S\left(k\right)+C\left(k\right),& \mathrm{if}k\in I_{\mathrm{SSC}},\\ S\left(k\right),& \mathrm{otherwise},\end{array}& \left(1\right)\end{array}$

[0069] where S stands for the information signal and C is the generated secret sequence, and I.sub.SSC is the set of selected subcarriers. In (1), k refers to the index of each symbol, indicating whether it belongs to I.sub.SSC set. For those k which belong to I.sub.SSC, modulated secret sequence is added to information data with the same index. The transmitted signal, X∈.sup.1×K, is sent to the legitimate user, Bob, after applying cyclic prefix (CP) to the time-domain encrypted symbols to avoid intersymbol interference (ISI). The received signal at Bob's side can be written as

y.sub.b=h.sub.b*x+n.sub.b,  (2)

[0070] where h.sub.b is the Bob's channel in time-domain, x is the transmitted signal, and n.sub.b is the zero-mean complex additive white Gaussian noise (AWGN) at Bob. After removing CP and applying serial-to-parallel (S/P) conversion on the time domain received signal, y.sub.b, Bob uses fast Fourier transform (FFT) on the resulting signal. A zero-forcing channel equalization process is performed. The received signal at Bob's side after channel equalization process is found by element-wise division of the received signal and his channel, and it can be defined as

=Y.sub.bØH.sub.b,  (3)

[0071] where H.sub.b is the Bob's channel in frequency-domain and Y.sub.b is the frequency-domain received signal after S/P conversion. After P/S conversion, Bob generates the information signal by subtracting the artificial signal from his received signal, . The information signal is observed as

[00003]$\begin{array}{cc}{S}_b\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_b\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_b},\\ \left(k\right),& \mathrm{otherwise},\end{array}& \left(4\right)\end{array}$

[0072] where C.sub.b is the modulated secret sequence generated at Bob and S.sub.b is data symbols after subtraction from received symbols, . Moreover, I.sub.SSC.sub.b refers to the selected subcarriers set at Bob. Eve has also access to the transmitted signal, x and as she can be stronger than Bob and more capability, she follows the same steps as Bob to generate her own secret sequence from her channel. The signal which Eve captures from her channel, H.sub.e, can be defined as

y.sub.e=h.sub.e*x+n.sub.e  (5)

[0073] where h.sub.e denotes the Eve's channel in time-domain and n.sub.e means zero-mean complex AWGN noise. The received signal at Eve's side after channel equalization process can be defined as

=Y.sub.eØH.sub.e,  (6)

[0074] where Y.sub.e is the frequency-domain is received signal after S/P conversion and H.sub.e is the Eve's channel in frequency domain. Eve generates her information signal by following the same algorithm as Bob based on her channel. The extracted information signal in Eve can be expressed as

[00004]$\begin{array}{cc}{S}_e\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_e\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_e},\\ \left(k\right),& \mathrm{otherwise},\end{array}& \left(7\right)\end{array}$

[0075] where C.sub.e stands for the modulated secret sequence at Eve, S.sub.e is data symbols after subtracting from his/her received symbols, , and I.sub.SSC.sub.e is the selected subcarriers set at Eve. It is important to note that both Bob and Eve follow the same steps in order to achieve the secret data which they received from the transmitter, Alice. Since Alice transmits the data using the secret sequence which was extracted from Bob's channel, Bob can securely receive this signal in the presence of Eve. Although Eve can have multiple antennas and more skills than Bob, she cannot have access to the information data correctly even if she extracts her own secret sequence, and apply the proposed method for finding jammed subcarriers and artificial signal.

[0076] D. Illustrative Case:

[0077] In this section, an illustrative case is presented in order to gain further insights. The length of window is assumed to be W=4 and P=1. Based on the previous methodology, the parameters are provided in TABLE I.

TABLE-US-00001 TABLE 1 An example of proposed random subcarrier selection method. Parameters: Considered W = 4, P = 1, subcarrier M = 8, N = 2, Extracted numbers for I_TSC = 20, sequence in Corresponding adding LSSC = 14. each subcarrier artificial First window window numbers signal Second window 01110111 {1, 3, 1, 3} {1, 3} .fwdarw. {6, 8} Third window 11101100 {3, 2, 3, 0} {0, 2, 3} .fwdarw. {9, 11, 12} Fourth window 10001010 {2, 0, 2, 2} {0, 2} .fwdarw. {13, 15} Fifth window 00100111 {0, 2, 1, 3} {0, 1, 2, 3} .fwdarw. {17, 18, 19, 20}

[0078] In this example, the minimum required length for secret sequence is 40 and the length of generated secret sequence considering quantization level of 6 is 30. The first 10 bits from sequence are added as a suffix to the end, to reach the desired length. The resulting bits are divided into small sub-blocks with length 8, where each 2 bits represent a subcarrier number from 1 to 4. Finally, those specific modulated secret sequence symbols with the same index of the chosen subcarriers are considered to be added with the same index of information symbols. From the information provided in TABLE I, it can be concluded that the ratio between randomly selected subcarriers and total number of subcarriers is 0.7 which is high. This means that 70% of subcarriers are jammed so that they carry artificial signal plus information data and it is somehow impossible for Eve to find them. Due to the exploiting channel reciprocity property, channel between legitimate nodes are same and Bob knows which subcarriers are jammed so that he can decode his data correctly. However, even if Eve follows the same steps, her selected subcarriers are not same of Alice and Bob because of her different channel. This means that Eve cannot find the right jammed subcarriers to decode her data correctly.

[0079] Performance Analysis and Results:

[0080] Simulation results are presented to analyze the performance and prove the efficiency of the proposed method. The effectiveness of our proposed method is evaluated by means of bit error rate (BER) performance, secret sequence disagreement rate (SSDR), and throughput. In addition, comparison of the proposed window-based subcarrier selection method with the scheme using all subcarriers for secret sequence extraction is carried out in terms of SSDR, which represents the percentage of the number of different bits between Alice's and Bob's generated sequences for different variance values. In this particular study, BPSK modulation is used to map the bits to transmitted symbols of length 64. A Rayleigh fading channel with a total number of 5 taps with decaying power delay profile is generated for both Bob and Eve. The quantization level of secret sequence generation is determined as 64. Total number of subcarriers is considered to be 64. The window length is 4 and P=1. Imperfect channel reciprocity and estimation error are considered as well. Specifically, the estimated channel for Alice and Bob are expressed as Ĥ.sub.a=H.sub.a+ΔH.sub.a and Ĥ.sub.b=H.sub.b+ΔH.sub.b, respectively. H.sub.a=H.sub.b are the true channels of Alice and Bob, respectively. Also, ΔH.sub.a and ΔH.sub.b are independent Gaussian noise vectors at Alice's and Bob's sides with zero-mean and variance σ.sup.2=e×10.sup.−SNR(dB)/10 respectively. One of the most important metrics considered in secret sequence generation systems is randomness. The randomness of the proposed channel dependent secret sequence is checked by using a run test for randomness command, h=runstest(x), in MATLAB. The test results in h=0 for random sequence, as expected.

[0081] FIG. 3 plots the SSDR versus transmit SNR by assuming imperfect channel reciprocity. It is shown that when e decreases, SSDR also decreases. For example, at SNR=16 dB, from e=0.1 to e=0.0001, SSDR decreases from 0.3622 to 0.0421, respectively, which results in a gain of 88%. FIG. 4 compares the SSDR performance of the proposed method with the scheme using all subcarriers for extracting secret sequence. It is considered different quantization levels and channel estimation errors. As can be seen, the proposed method achieves a huge gap in worst case scenario, which is e=1. For example, for SSDR=0.48 in the case of Q=1024, FIG. (4a) shows that the proposed method achieves around 9 dB gain, which proves the efficiency of the proposed method. Same interpretation is valid for FIG. (4b), in which it is shown that the proposed method achieves around 5 dB gain, for the case of SSDR=0.457 and Q=128. Also, higher quantization level increases the SSDR due to the generation of more number of bits from each subcarrier.

[0082] FIGS. 5 and 6 plot throughput and BER, respectively, versus SNR considering different channel estimation errors. Note that the proposed method provides high throughput performance for considered system due to the usage of all the subcarriers for data transmission which cause to achieve high spectral efficiency. For example, in FIG. 5 at SNR=10 dB, for the worst case e=1, the value for throughput at Bob is 0.8443 while at Eve is 0.5139, which shows that the proposed technique achieves 40% gain in throughput at the legitimate user. Also in FIG. 6 at SNR=20 dB and assuming e=0.01, the value of BER at legitimate user is 0.0233 and at Eve is 0.5052, which proves that the proposed method provides 95% BER gain at Bob. The secret sequence which are used for throughput and BER calculation are extracted in high SNR values, SNR=40 dB.

[0083] In this patent, a novel method for physical layer security was presented. A channel-based random subcarrier selection and artificial signal design were proposed. High level of randomness due to window-based subcarrier selection achieved in the generated secret sequence. This subcarrier selection method implied in more uncorrelated subcarriers which improved reliability in extracted secret sequence only at legitimate user. The jammed subcarriers were chosen depends on the generated secret sequence based on Bob's channel. Lastly, those specific modulated secret sequence symbols with the same index of the chosen subcarriers were considered to be added with the same index of information symbols. As all the steps in proposed method were done based on Bob's channel, it was ensured both security and reliability only at Bob. Imperfect channel reciprocity conditions were also considered. Besides, simulation results showed a huge secrecy gap between Bob's and Eve's BER, which proves the efficiency of proposed method. Finally, the proposed window-based subcarrier selection method was compared with the scheme using all subcarriers for secret sequence extraction and results revealed the effectiveness of our method. For future works, an active eavesdropper instead of a passive eavesdropper can be considered. In the proposed method, this is challenging for finding the legitimate user while sending pilot signals for channel estimation purposes.

[0084] This invention is applicable to industrialization and the proposed method can be used to provide secure and reliable communication in wireless systems for the future communication networks and systems.

[0085] A joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) wherein the system is characterized by comprising; Two random subcarrier selection schemes as a window-based subcarrier selection which is using strongest subcarriers of each window for secret sequence extraction and a channel-based random subcarrier selection for the sake of adding artificial signal to the information data to protect the transmitter's confidential data.

[0086] An operation method of a joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) system wherein the method comprising; [0087] Finding strong subcarriers in a single-antenna legitimate receiver channel, [0088] Quantization on found strong subcarriers and secret sequence generation, [0089] Comparing minimum required length of secret sequence (L.sub.min) and L, [0090] If comparing minimum required length (L.sub.min) is bigger than L, reshaping of the secret sequence, then choose the first L.sub.min bits, [0091] If comparing minimum required length (L.sub.min) is not bigger than L then choose the first L.sub.min bits, [0092] After choosing the L.sub.min bits, make small packets of secret sequence with the length of M, [0093] Dividing the length of M bits to N for subcarrier number selection, [0094] After doing subcarriers number selection based on the generated secret sequence, adding artificial signal to the information data to protect the transmitter's confidential data, [0095] In this step; [0096] For designing artificial signal, the extracted secret sequence from the channel is modulated with the same order as information data, [0097] Sending the transmitted signal as X∈.sup.1×K to the legitimate user (a single-antenna legitimate receiver) after applying cyclic prefix (CP) to the time-domain encrypted symbols to avoid inter symbol interference (ISI), [0098] Receiving the signal as a y.sub.b=h.sub.b*x+n.sub.b at a single-antenna legitimate receivers side, [0099] After removing CP and applying serial-to-parallel (S/P) conversion on the time domain received signal, using the fast Fourier transform (FFT) on the resulting signal by single-antenna legitimate receiver, [0100] Performing a zero-forcing channel equalization process, [0101] Receiving the transmitter's signal (a) at single-antenna legitimate receivers side after channel equalization process is found by element-wise division of the received signal and his channel, as a =Y.sub.b ØH.sub.b, [0102] After P/S conversion, legitimate receiver (b) generating the information signal by single-antenna legitimate receivers by subtracting the artificial signal from single-antenna legitimate receivers received signal, [0103] Extracting the information signal as a

[00005]$S_b\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_b\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_b},\\ \left(k\right),& \mathrm{otherwise},\end{array}$ [0104] The transmitted signal can be accessed by single-antenna eavesdropper and eavesdropper can generate secret sequence from her single-antenna channel, [0105] Capturing the signal as a y.sub.e=h.sub.e*x+n.sub.e from single-antenna eavesdroppers channel, [0106] Receiving the signal at the single-antenna eavesdropper side after channel equalization process as =Y.sub.e ØH.sub.e, [0107] Generating the single-antenna eavesdropper information signal by the single-antenna eavesdropper based on her channel by following the same algorithm as a single-antenna legitimate receiver, [0108] Extracting information signal in a single-antenna eavesdropper can be expressed as

[00006]$S_e\left(k\right)=\{\begin{array}{cc}\left(k\right)-C_e\left(k\right),& \mathrm{if}k\in I_{{\mathrm{SSC}}_e},\\ \left(k\right),& \mathrm{otherwise},\end{array}.$

[0109] Embodiments of the invention are; [0110] A joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) system helps to alleviate the co-located attacks and temporal correlation problem in physical layer security techniques such as secret key generation from wireless channel characteristics. [0111] A method for a joint random subcarrier selection and channel-based artificial signal design aided physical layer security (PLS) wherein only strong subcarriers out of whole in each window has chosen to extract more reliable and high randomness secret sequence at legitimate receiver.

REFERENCES

[0112] [1] Z. Bouida, A. Stavridis, A. Ghrayeb, H. Haas, M. Hasna, and M. Ibnkahla, “Precoding-aided spatial modulation for the wiretap channel with relay selection and cooperative jamming,” Wireless Commun. Mob. Comp., 2018. [0113] [2] B. He, Y. She, and V. K. N. Lau, “Artificial noise injection for securing single-antenna systems,” IEEE Trans. Veh. Technol., vol. 66, no. 10, pp. 9577-9581, 2017. [0114] [3] J. M. Hamamreh, H. M. Furqan, and H. Arslan, “Classifications and applications of physical layer security techniques for confidentiality: A comprehensive survey,” IEEE Commun. Surveys & Tut., 2018. [0115] [4] E. Gvenkaya and H. Arslan, “Secure communication in frequency selective channels with fade-avoiding subchannel usage,” IEEE Int. Conf. Commun. (ICC), pp. 813-818, 2014. [0116] [5] A. Badawy, T. Elfouly, T. Khattab, C.-F. Chiasserini, A. Mohamed, and D. Trinchero, “Robust secret key extraction from channel secondary random process,” Wireless Commun. Mob. Comp., vol. 16, no. 11, pp. 1389-1400, 2016. [0117] [6] J. Zhang, T. Q. Duong, A. Marshall, and R. Woods, “Key generation from wireless channels: A review,” IEEE Access, vol. 4, pp. 614-626, 2016. [0118] [7] L. Jiao, N. Wang, P. Wang, A. Alipour-Fanid, J. Tang, and K. Zeng, “Physical layer key generation in 5G wireless networks,” IEEE Wireless Commun. Mag., 2019. [0119] [8] X. Zhou, L. Song, and Y. Zhang, Physical Layer Security in Wireless Communications. CRC Press, 2016. [0120] [9] A. Goldsmith, Wireless Communications. Cambridge University Press, 2005. [0121] [10] N. Zhao, Y. Cao, F. Yu, Y. Chen, M. Jin, and V. C. Leung, “Articial noise assisted secure interference networks with wireless power transfer,” IEEE Trans. Veh. Technol., vol. 67, no. 2, pp. 1087-1098, February 2018.