Non-orthogonal multiple access system transmission method based on dual-frequency index modulation

Abstract

The invention discloses a non-orthogonal multiple access system transmission method based on dual-frequency index modulation. According to the method, firstly, M-QAM modulation and dual-frequency index modulation of a specific system are respectively carried out on an input multi-user bit stream signal at a signal processing end of a base station, then OFDM-DFIM (Orthogonal Frequency Division Multiplexing-Distributed Frequency Index Modulation) modulation signals are generated by utilizing an orthogonal frequency division multiplexing technology, finally, channel conditions of all users are compared, power is allocated to all the users, and the signals are transferred to a time domain to be sent after superposition. Therefore, same-frequency simultaneous superposition transmission of signals of multiple users in a power domain is realized. And at a signal receiving end, performing signal separation by using a serial interference cancellation technology, and sequentially demodulating and restoring bit stream signals of each user through maximum likelihood estimation. According to the method, the spectrum efficiency is improved and the bit error rate is reduced by improving the proportion of the index information in all information, so that the system performance is further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and relates to a method for processing index modulation data in the communication field, and in particular to a non-orthogonal multiple access system transmission method based on dual-frequency index modulation. Background Art

[0002] As an alternative to orthogonal multiple access (OMA), non-orthogonal multiple access (NOMA) has great application prospects as a key technology in 5G wireless communications. NOMA introduces a new dimension - the power domain - in resource blocks through multiplexing. Different users can be allocated in the power domain, and multiple users can be transmitted in the same channel at the same time, using the same frequency domain and time domain resources, which produces significant gains in spectrum efficiency. In addition, in the application, the NOMA system can allocate different transmission powers to different user signals according to the channel conditions of each user detected. Reasonable power allocation between multiplexed users can effectively improve the system transmission performance. When receiving signals, the NOMA system uses the SIC (Serial Interference Cancellation) technology to detect the superimposed signal, and regards the user signal with low power as noise, directly decodes the high-power user signal, and then subtracts the high-power signal to achieve user signal separation.

[0003] Index Modulation OFDM (OFDM-IM) is a new type of multi-carrier transmission technology. The principle of IM is to encode data into a combination of activated / silent sub-elements in many different domains such as frequency domain, spatial domain, channel domain and time domain, so as to achieve additional transmission bit information without consuming energy. More specifically, it is by activating only a small part of certain index resource entities, such as subcarriers, antennas, time slots or channel states for data transmission, while the rest remain unused, and other information bits are implicitly conveyed through index usage or activation mode.

[0004] However, in the process of sending signals, there are a large number of silent elements in the IM index modulation technology that do not send data. The implicitly transmitted index information accounts for a small proportion of the total information, which reduces the spectrum efficiency and increases the demodulation complexity at the receiving end. The NOMA system has certain limitations. Although NOMA can achieve user fairness through appropriate power allocation, the performance improvement of weak users may lead to the deterioration of the performance of users with good channel conditions. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a non-orthogonal multiple access system transmission method based on dual-frequency index modulation, which divides the subcarrier block into two parts on average and divides the modulation symbol into an in-phase component and an orthogonal component, so that the input bit information bit and the index jointly confirm the activated subcarrier position, thereby implicitly transmitting additional signal bits, improving spectrum efficiency, increasing the index ratio in the user signal to reduce the bit error rate, and further improving the performance of the IM-NOMA system.

[0006] A non-orthogonal multiple access system transmission method based on dual-frequency index modulation is implemented based on the downlink non-orthogonal multiple access system NOMA-IM system. At the signal processing end of the base station, the bit stream signals of near users and far users are respectively subjected to M-QAM modulation and dual-frequency index modulation of a specific base system. Each user is allocated a different proportion of signal power at the base station, and the signal power is superimposed and sent to multiple users. At the signal reception end, the far user who is allocated a high-power signal ignores the influence of the low-power near user signal and directly uses ML (Maximum Likelihood Detection) to detect the signal. The near user signal uses SIC (Serial Interference Cancellation) technology to separate the direct signal from the superimposed signal and then uses ML to detect the signal. The method specifically includes the following steps:

[0007] Step 1: Perform serial-to-parallel conversion on the Q-bit signal input by a user according to the modulated M-QAM symbol. Determine the number of bits q required for each group according to the number of subcarriers to be activated and the modulation order M, and divide the group into k groups with q-bit signals, k= , Indicates rounding up. When Q < k*q, add a 0-bit signal to the end of the original signal. N is an integer. all The subcarriers are divided into g equal small blocks, each of which has a fixed number of subcarriers, 2N, g = N all / 2N.

[0008] Step 2: Map the information of the k groups to the g small blocks of subcarriers in order. The bit signal in each group is divided into three parts: q1, q2, and q3. The q1 part selects to map the M-QAM modulated signal to the corresponding activated subcarrier, including the subcarriers that need to be activated by the in-phase component index and the subcarriers that need to be activated by the orthogonal component index:

[0009] (1) N0 is the number of activated subcarriers in the first N subcarriers of each tile.

[0010] The q2 part is the M-QAM modulated bit signal carried on the corresponding activated subcarrier: (2) M is the base number of the quadrature amplitude modulation M-QAM symbol.

[0011] The q3 part is the dual-frequency compensation index signal bit. The number of bits required for q3 is 2N0. The activated subcarrier that needs to be changed in the q1 part is determined according to the q3 part. The subcarrier corresponding to 1 in the q3 part selects the Nth subcarrier after the subcarrier as the changed activated subcarrier to complete the dual-frequency compensation index.

[0012] The in-phase vector I that needs to be transmitted on the βth subcarrier block i and the orthogonal vector I q Respectively expressed as: (3) (4) where i i (r) and i q (r) represent the in-phase and quadrature transmission symbols of the rth subcarrier in the βth subcarrier block, r=1,...,2N, the superscript T represents transposition, β=1,...,g. The complete signal transmitted on the βth subcarrier block is generated by a group of bit signal index modulation, and the transmitted symbol vector I is:

[0013] (5) i(r) represents the data symbol generated by index modulation transmitted on the rth subcarrier in the βth subcarrier block. The g subcarrier blocks are combined to generate an OFDM-DFIM modulation symbol X:

[0014] (6) X(β) represents the modulation symbol generated by a group of bit signal index modulation carried by the βth subcarrier block. Then one user information needs to be sent through W=k / g OFDM symbols. When W is not an integer, a 0-bit signal needs to be added at the end of the last OFDM symbol to make it a complete OFDM symbol.

[0015] Step 3: Generate OFDM-DFIM modulation symbols X1, X2, .....X of each user in turn according to the method of steps 1 and 2. i , i is the number of users. Then power is allocated according to the channel status information of each user, and then the signals are superimposed:

[0016] (7) Among them, Y represents the frequency domain transmission signal after multi-user superposition, P BS is the total transmission power of the base station, α1, α2, .....α i is the power allocation coefficient of the user, and α1+α2+.....+α i =1,n~CN(0,σ 2 ) has a mean of 0 and a variance of σ 2 Complex baseband additive white Gaussian noise.

[0017] Construct a conjugate symmetric vector matrix for the superimposed frequency domain transmitted signal Y: (8) Where Y(·)* represents the conjugate matrix of the corresponding matrix, N zp Indicates a matrix size of 1×N zp The inverse Fourier transform is then performed to map the frequency domain signal Y into a time domain signal S and send it. This enables the simultaneous superposition and transmission of signals of multiple users in the power domain at the same frequency.

[0018] Step 4. When the near user receives the transmission signal S, it first subtracts the cyclic prefix to eliminate multipath interference, then performs channel equalization and inverse Fourier transform according to the pilot, and then divides the signal into an in-phase signal and an orthogonal signal. The q3 partial index bit information in each group signal is obtained according to the index set, and the q1 partial index bit information is determined according to the position of the activated subcarrier in the frequency domain before dual-frequency index modulation. Finally, the q2 partial bit information after M-QAM modulation is restored by combining the in-phase component and the orthogonal component to restore the signal before dual-frequency index modulation.

[0019] Step 5: For the signal restored in step 4, separate the superimposed signal using serial interference cancellation technology, and then demodulate the multi-user modulated signal through maximum likelihood estimation. The specific steps are as follows:

[0020] s5.1. According to the signal index set and the size of the power allocation, the following ML detection expression is used to detect the signal of the user with the maximum power: : (9) Where Y1 and H1 are the received signal and channel gain of the detected near user, respectively. b1 is the M-QAM modulation symbol vector of the far user, and B1 is the M-QAM modulation symbol set of the far user.

[0021] s5.2、The high power signal demodulated for the first time After reconstructing the signal X', the timing signal Y of the low-power near-user receiving signal is obtained through SIC (Serial Interference Cancellation)1-2 : (10) s5.3, for Y 1-2 Perform ML detection to modulate the bit stream signal of user 2 : (11) b2 is a near-user M-QAM modulation symbol vector, and B2 is a near-user M-QAM modulation symbol set.

[0022] s5.4. Consider the signal of low-power user 2 as noise and use the following ML detection expression to directly detect the received signal of the distant user: (12) Y2 and H2 are the received signal and channel gain detected at the far user, respectively. is the high-power distant user signal obtained after ML detection at the distant user.

[0023] The present invention has the following beneficial effects: By equally dividing the subcarrier block into two parts and dividing the modulation symbol into an in-phase component and an orthogonal component, the input bit information bit and the index jointly confirm the activated subcarrier position, thereby increasing the proportion of index information in the total information and the spectrum efficiency, and increasing the index ratio in the user signal to reduce the bit error rate and further improve the performance of the IM-NOMA system. Compared with the traditional NOMA system, the bit error rate of the near user is reduced, a certain user communication fairness is guaranteed, and the complexity of the ML detection required in the receiving section is reduced. The user's signal modulation order can also be adjusted according to different channel state information, thereby increasing the signal transmission speed while ensuring a certain bit error rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a block diagram of the system sending end of this method; Figure 2 It is a simulation curve diagram of the system of the present method and the NOMA system in the embodiment; DETAILED DESCRIPTION

[0025] The present invention will be further explained below with reference to the accompanying drawings; A non-orthogonal multiple access system transmission method based on dual-frequency index modulation, such as Figure 1 As shown, the specific steps are as follows: Step 1. In this embodiment, it is first assumed that the signal processing end of the base station first inputs the signals of the near user and the far user, each including Q=2000 random bits, and the number of subcarriers N all= 80. All subcarriers are equally divided into g = 10 subcarrier blocks, and the number of subcarriers in each block is 2N = 8.

[0026] The signal is modulated with 4-bit QAM and the input signal is grouped. The number of bits in each group is q=16, so it can be divided into k=Q / q=125 groups.

[0027] Step 2: Map the information of k groups to g small subcarriers in sequence. The g subcarrier groups are combined into one OFDM symbol. Then the user information that needs to be sent is W = k / g = 12.5 OFDM symbols. Since W is not an integer, a 0-bit signal is added to the end of the last OFDM symbol to supplement it into an OFDM symbol. For one user, a total of 13 OFDM symbols need to be sent.

[0028] In this embodiment, the subcarrier selected for activation is 3 among the 4 subcarriers, so the number of bits that can be carried by the q1 part in each bit signal group is 4 bits. Assuming that the q1 part is {0111}, the in-phase signal activates 1, 3, and 4 of the first 4 subcarriers, and the orthogonal signal activates 1, 2, and 3 of the first 4 subcarriers. The q2 part has a total of 6 bits of signal for QAM modulation, and the generated constellation symbol is: i i (1)+i q (1), i i (2)+i q (2), i i (3)+i q (3). Therefore, in this subcarrier block, the signal before dual-frequency index modulation is:

[0029] {i i (1)+i q (1), 0+i q (2), i i (2)+i q (3), i i (3)+0,0,0,0,0} Since the q1 part only selects to activate 3 subcarriers on the first 4 subcarriers in each subcarrier block to index the in-phase component and the orthogonal component of the 4-bit QAM modulation symbol, and the q3 part only selects the changed position in the last 4 subcarriers in each subcarrier block, each subcarrier has a bit signal to select whether to change the position. There are a total of 6 input random conditions for whether to change the subcarrier position, so the number of bits that q3 can carry is 6 bits. Assuming that the q3 part is {010110}, the second subcarrier position in the in-phase signal is changed to 7, and the first and second subcarrier positions in the orthogonal signal are changed to 5 and 6, so the signal after dual-frequency index modulation is:

[0030] {i i (1), 0, i q (3), i i (3), i q (1), i q (2), i i (2), 0} The mapping relationship between the in-phase and quadrature component signal indexes is shown in Table 1: Table 1 Step 3: After index modulation of the data of distant users and near users according to the methods in steps 1 and 2, power is allocated according to the channel state information, the signals after corresponding power allocation are superimposed to generate a total transmission signal, and conjugate construction and inverse Fourier transform are performed to obtain the time domain signal and then transmit it.

[0031] Step 4: After receiving the signal, the far user receiving end performs channel equalization, and then regards the superimposed low-power user signal as noise, and directly demodulates the high-power signal according to the corresponding bit index set to obtain the original signal. After receiving the signal, the near user receiving end first performs channel equalization according to the channel state information, and then reconstructs the high-power signal according to the SIC principle after demodulating the high-power signal. After subtracting the reconstructed signal from the total power of the receiving end, the low-power signal is left and then ML demodulates the low-power signal.

[0032] The bit error rates of the above methods with different signal-to-noise ratios in Rayleigh fading channels with multipath interference are calculated, and the performance of the theoretical NOMA system is compared. The results are as follows: Figure 2 As shown. In the range of 20-25 dB signal-to-noise ratio, this method achieves a reduction in the bit error rate of near-user signals. Since the number of signals that need to be sent is reduced and the system spectrum utilization is improved, the overall system power consumption of the index modulation NOMA system is significantly lower than that of the NOMA system without index modulation, and the system still maintains good communication quality in a multipath environment.

Claims

1. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation, implemented based on a downlink non-orthogonal multiple access system NOMA-IM system, characterized in that: The specific steps include: Step 1: Perform serial-to-parallel conversion on a Q-bit signal input by a user according to the modulated M-QAM symbol, determine the number of bits q required for each group according to the number of subcarriers to be activated and the modulation order M, and divide the group into k groups with q-bit signals; all The subcarriers are divided into g equal small blocks; Step 2: Map the information of the k groups to the g small blocks of subcarriers in order; the bit signal in each group is divided into three parts: q1, q2, and q3, where the q1 part selects to map the M-QAM modulated signal to the corresponding activated subcarrier: (1) N represents half of the number of subcarriers in each small block, and N0 is the number of activated subcarriers in the first N subcarriers of each small block; The q2 part is the M-QAM modulated bit signal carried on the corresponding activated subcarrier: (2) M is the base number of the quadrature amplitude modulation M-QAM symbol; The q3 part is used to determine the activated subcarrier that needs to be changed in the q1 part. For the subcarrier corresponding to 1 in the q3 part, the Nth subcarrier after the subcarrier is selected as the activated subcarrier after the change, and the dual-frequency compensation index is completed; The information transmitted by g subcarrier blocks is combined to generate an OFDM-DFIM modulation symbol X. After completing the information mapping of k groups, all OFDM-DFIM modulation symbols of a user are obtained. Step 3: Generate OFDM-DFIM modulation symbols X1, X2, .....X of each user in turn according to the method of steps 1 and 2. i , i is the number of users; power is allocated according to the channel status information of each user and then the signals are superimposed: （7） Among them, Y represents the frequency domain transmission signal after multi-user superposition, P BS is the total transmission power of the base station, α1, α2, .....α i is the power allocation coefficient of the user, and α1+α2+.....+α i =1,n~CN(0,σ 2 ) has a mean of 0 and a variance of σ 2 The complex baseband additive white Gaussian noise is generated; the frequency domain signal Y is mapped into a time domain signal S and sent; thereby realizing the simultaneous superposition transmission of signals of multiple users in the power domain at the same frequency; Step 4: When the near user receives the transmission signal S, the cyclic prefix is ​​first subtracted to eliminate multipath interference, and then channel equalization and inverse Fourier transform are performed according to the pilot. Then the signal is divided into an in-phase signal and an orthogonal signal, and the q3 part of the index bit information in each group signal is obtained according to the index set. The q1 part of the index bit information is determined according to the position of the activated subcarrier in the frequency domain before dual-frequency index modulation. Finally, the q2 part of the bit information after M-QAM modulation is restored by combining the in-phase component and the orthogonal component to restore the signal before dual-frequency index modulation; Step 5: For the signal restored in step 4, separate the superimposed signal using serial interference cancellation technology, and then demodulate in sequence through maximum likelihood estimation to obtain the modulated signals of multiple users.

2. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1, characterized in that: k= , Indicates rounding up. When Q < k*q, the signal length is increased by adding a 0-bit signal at the end of the original signal. is an integer.

3. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1, characterized in that: The in-phase vector I that needs to be transmitted on the βth subcarrier block i and the orthogonal vector I q Respectively expressed as: （3） （4） where i i (r) and i q (r) represent the in-phase and quadrature transmission symbols of the r-th subcarrier in the β-th subcarrier block, r=1,...,2N, the superscript T represents transposition, β=1,...,g; the complete signal transmitted on the β-th subcarrier block is generated by a group of bit signal index modulation, and the transmitted symbol vector I is: （5） i(r) represents a data symbol generated after index modulation and transmitted on the r-th subcarrier in the β-th subcarrier block.

4. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1 or 3, characterized in that: Combine g subcarrier blocks to generate an OFDM-DFIM modulation symbol X: （6） X(β) represents a modulation symbol generated by a group of bit signal index modulation carried by the β-th subcarrier block.

5. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1, characterized in that: One user information needs to be sent through W OFDM symbols. When W=k / g is not an integer, a 0-bit signal needs to be added at the end of the last OFDM symbol to make it a complete OFDM symbol.

6. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1, characterized in that: Construct a conjugate symmetric vector matrix for the superimposed frequency domain transmitted signal Y: (8) Where Y(·)* represents the conjugate matrix of the corresponding matrix, N zp Indicates a matrix size of 1×N zp All-zero matrix; Then perform inverse Fourier transform to convert the frequency domain signal into time domain signal.

7. A non-orthogonal multiple access system transmission method based on dual-frequency index modulation as claimed in claim 1, characterized in that: The specific steps of obtaining the modulated signal of multiple users through maximum likelihood estimation demodulation are as follows: s5.

1. According to the signal index set and the size of the power allocation, the following ML detection expression is used to detect the signal of the user with the maximum power: : (9) Among them, Y1 and H1 are the received signal and channel gain of the detected near user respectively; b1 is the M-QAM modulation symbol vector of the far user, and B1 is the M-QAM modulation symbol set of the far user; s5.2、The high power signal demodulated for the first time After reconstructing the signal X', the timing signal Y of the low-power near-user receiving signal is obtained through SIC 1-2 : (10) s5.3, for Y 1-2 Perform ML detection to modulate the bit stream signal of user 2 : (11) b2 is the near user M-QAM modulation symbol vector, B2 is the near user M-QAM modulation symbol set; s5.

4. Consider the signal of low-power user 2 as noise and use the following ML detection expression to directly detect the received signal of the distant user: (12) Y2 and H2 are the received signal and channel gain detected at the far user, respectively. is the high-power distant user signal obtained after ML detection at the distant user.

8. A non-orthogonal multiple access system transmission method based on dual frequency index modulation as claimed in claim 7, characterized in that: The signal index set is: 。

Images