CN-122001672-A - Physical layer safe transmission method based on multipath selective independent transmission

CN122001672ACN 122001672 ACN122001672 ACN 122001672ACN-122001672-A

Abstract

The invention provides a physical layer safe transmission method based on multipath selection independent transmission, which comprises the steps of firstly acquiring multipath channel state information of legal users by utilizing a large-scale MIMO channel estimation method, secondly, carrying out multipath selection based on the multipath channel state information, matching a group of additively unique resolvable constellation groups according to the multipath quantity and carrying out precoding design, eliminating symbol interference among the multipath, enabling a receiving signal of the legal users to meet the sum constellation design of the additively unique resolvable constellation groups, and further realizing accurate detection of receiving symbols. In a single user scenario, the method only needs to perform multipath selection to improve its secret capacity. In a multi-user scene, two-stage selection processes of multipath and users are needed, namely, multipath selection is firstly carried out on multiple users respectively to determine the optimal path combination of the multiple users, then user selection is carried out on the basis of the optimal path combination of each user, and the system combined secret capacity is improved.

Inventors

HAN GANGTAO
LI JIALU
LI JIAQI
WANG NING
PAN GAOFENG
ZHANG DALONG
ZHU ZHENGYU
HAO WANMING
TANG ZHIQING
WANG JUNJIE

Assignees

郑州大学

Dates

Publication Date: 20260508
Application Date: 20260319

Claims (8)

1. The physical layer safe transmission method based on multipath selective independent transmission is characterized by comprising the following steps: S1, constructing a communication system comprising a transmitting end and a receiving end, wherein the transmitting end is provided with a large-scale MIMO antenna, and the receiving end is a legal user and has a passive eavesdropper; S2, based on multipath channel state information, matching an additively unique resolvable constellation group for each legal user according to the multipath quantity, and mapping each path of input message bit stream to a sub-constellation of the additively unique resolvable constellation group respectively to obtain a symbol vector; S3, judging the current communication scene, if the current communication scene is a single-user scene, executing a greedy path selection strategy based on multipath channel state information, and screening out an optimal transmission path combination from candidate paths of the user; if the user is in a multi-user scene, executing a two-stage selection strategy, namely traversing all users in a user pool, executing a greedy path selection strategy on candidate paths of each user respectively, and screening out an optimal transmission path combination of the user; S4, based on an equivalent channel formed by the screened optimal transmission path combination or the optimal user combination, for different scenes, adopting different optimization methods to jointly design a digital precoding matrix and a DMA (direct memory access) simulation weight matrix under the condition of meeting the orthogonal constraint; S5, multiplying the symbol vector with the optimized digital precoding matrix and analog precoding matrix to obtain a transmitting signal vector; S6, transmitting signal vectors reach a receiving end through large-scale MIMO antenna radiation and wireless channel transmission to obtain receiving signals; s7, the receiving end detects the received signal and solves the most probable sum constellation symbol according to the minimum distance criterion; S8, decomposing the most probable sum constellation symbol by utilizing the unique decomposable characteristic of the additive unique decomposable constellation group, and recovering to a unique sub-constellation symbol vector; and S9, demodulating each symbol in the recovered sub-constellation symbol vector to obtain a final recovered message bit stream.
2. The physical layer security transmission method based on multipath selective independent transmission according to claim 1, wherein the method for acquiring multipath channel state information of a legal user in step S1 is as follows: The sender configuration comprises Microstrip each comprising Two-dimensional dynamic super-surface antenna of each radiation unit In successive sampling time slots, the phase shift of the dynamic subsurface antenna is reconstructed orthogonally, in the first In each sampling time slot, the first on each micro-band The phase of each radiating element is set to Make its equivalent weight be The rest of the unit phases are set to Making the equivalent weight of the mixture be 0; Collecting The received signals of the sampling time slots form a sampling matrix And calculating an estimated channel according to a minimum mean square error criterion by combining pilot signals, wherein the formula is as follows: , Wherein, the , In order to estimate the full channel vector of the channel, As a result of the pilot signal, Is that Is used for the conjugation of (a), The vectorization operation is represented by a vector, For compensating the phase shift of the optical disc, Is a power correlation coefficient.
3. The physical layer security transmission method based on multipath selective independent transmission according to claim 1, wherein the definition of the additively unique resolvable constellation group is: for any one If and only if Only when there is Then Called an additively unique set of decomposition constellations, i.e Wherein, the method comprises the steps of, Representing one constellation point in a sub-constellation, Representing another constellation point in the sub-constellation, Representing the sub-constellation of the symbols, Representing the number of sub-constellations.
4. The physical layer security transmission method based on multipath selective independent transmission according to claim 1, wherein the method for executing a greedy path selection policy in a single user scenario is as follows: candidate path set of single user Bob Comprises Initializing the selected path set as empty set and setting the selected path set as empty set The candidate path set is In the form of a gram-law As a measure, wherein For a set of selected paths A corresponding channel matrix; The total iteration number is In the first place In the iteration, the set determined based on the previous round Traversing the current remaining candidate set Each of the paths in (a) , wherein, Calculation is to New metric values generated after adding the current set: Selecting a path that maximizes the metric Adding a collection: ; adding the selected optimal path to the selected path set and removing from the candidate path set, i.e , ; When the maximum number of iterations is reached Stopping iteration and outputting the final selected optimal path combination Corresponding channel matrix 。
5. The physical layer secure transmission method based on multipath selective independent transmission according to claim 1, wherein the method of greedy user selection strategy based on subspace projection in a multi-user scenario is as follows: s31, traversing all users in the user pool Respectively executing greedy path selection strategies on candidate paths of each user, screening out the optimal transmission path combination of the user, and obtaining the user Is the optimal channel combination matrix of (1) ; S32, initializing a set, and setting the selected user set as an empty set The orthogonal base matrix is empty ; Traversing all candidate users, calculating the multipath total energy of each user Selecting the user with the largest energy as the first selected user to join Orthogonalization of schmitt to first selected user Strip path vector orthogonalization as initial orthogonalization basis ; S33, for each of the remaining unselected users in the user pool Defining a metric value For all of the users The strip path vector is on the current orthogonal base Sum of projection energies over orthogonal complement spaces of the tensed subspace: Wherein Is the first First level screening of individual users The channel vectors of the paths are used, Representing a user Is the first of (2) The strip path vector is relative to the current orthogonal base Is a component of the orthogonal component of (a); S34, selecting the measurement value Maximized user Joining a selected set of users The user is provided with A kind of electronic device The strip path vector is orthogonalized by schmitt, and the generated unit orthogonalization vector is extended to the current orthogonalization base In (a) and (b); s35, repeating the previous steps S31 to S34 until the selected user set The number of users in (a) reaches a preset value.
6. The physical layer secure transmission method based on multipath selective independent transmission according to claim 4, wherein in step S4, the optimization method adopted in the single user scenario is: constructing an objective function of maximum combining method channel secret capacity Obtaining a digital precoding matrix by solving a maximum secret capacity problem meeting the requirements of amplitude alignment constraint and interference elimination And DMA simulation weight matrix The objective function is: ; The constraint conditions are as follows: ; ; When (when) In the time-course of which the first and second contact surfaces, ; Wherein, the For the average power of the constellation set, In order for the desired receive gain to be achieved, For the additive white gaussian noise variance, For the DMA transfer response matrix, In order to transmit the signal vector(s), For the rated power of the electric motor, For the optimal channel vector to be a vector of channels, A digital precoding matrix vector; To decouple variables, a digital precoding matrix is used Expressed as pseudo-inverse of equivalent channel Wherein the equivalent channel matrix is Substituting the total transmit power constraint Obtaining the receiving gain ; Wherein DMA is simulated to weight matrix Expressed as a diagonal matrix, i.e , , , To be subjected to adjustable phase shift parameters The vector of the control is set to be, Is the sum of vectors The digital precoding matrix represented by the function concerned, Is the sum of vectors The function concerned represents the expected reception gain, Is the sum of vectors The pseudo-inverse version of the equivalent channel represented by the function concerned, Represents the conjugate transpose of the optimal channel matrix, For transmitting signal vectors Is used for the co-variance matrix of (a), , , For the number of multipaths, Is the number of DMA elements; The objective function is set Transformation of joint optimization of (c) into unconstrained black-box optimization with respect to DMA phase vector only The constraint is that Directly searching the optimal DMA simulation weight matrix in the phase space by utilizing the Nelder-Mead algorithm And calculate the digital pre-coding matrix according to the above 。
7. The physical layer security transmission method based on multipath selective independent transmission according to claim 6, wherein in step S4, after obtaining a legal optimal multipath channel matrix of a legal user in a multi-user scenario, a digital precoding matrix and a DMA analog weight matrix are obtained by adopting a fractional manifold optimization and orthogonality strategy: Firstly, analog domain tuning is carried out, and a directional channel matrix containing all multipath components of selected users is constructed Definition of equivalent channels And its Gram matrix To minimize Gram matrix The off-diagonal element energy of (2) is the objective function, i.e Satisfying the DMA radiation unit constant mode constraint Under the condition, adopting manifold optimization algorithm to solve optimal DMA simulation weight matrix ; Then carrying out zero forcing and power normalization in the digital domain Based on the determination, all the multipath channels of the users are stacked to form a total equivalent channel matrix Constructing an equivalent matrix equation according to homogeneity and orthogonality constraints required by additive unique resolvable constellation group transmission Calculating non-normalized zero forcing base by pseudo-inverse ; Finally, utilizing the boundary constraint of the total transmission power of the system Calculating an optimal reception gain Scaling the un-normalized substrate in equal proportion to obtain the optimal digital precoding matrix 。
8. The physical layer security transmission method based on multipath selective independent transmission according to claim 1, wherein the minimum distance criterion is: ; Wherein, the In order to receive the signal(s), Is that The sum of the individual transmitted symbols is used, Equivalent channel gain introduced for precoding.

Description

Physical layer safe transmission method based on multipath selective independent transmission Technical Field The invention relates to the technical field of communication safety, in particular to a physical layer safety transmission method of a large-scale MIMO antenna and a selection strategy, which utilizes a scene self-adaptive path or a user selection mechanism to assist an additively unique resolvable constellation group, and utilizes the multipath characteristic of a channel to carry out independent safety transmission at a physical layer, so as to maximize the confidentiality capacity in a passive eavesdropping scene. Background The existing physical layer security technology mainly utilizes the overall characteristics of a wireless channel, and establishes security advantages through methods such as beam forming, artificial noise injection and the like. However, most of the prior art modeling multi-antenna channels as a single overall channel for processing fails to effectively distinguish and exploit the rich multipath components in the channel. This simplified process ignores the independence of the individual paths and the unique channel characteristics that it contains, resulting in an inability to fully exploit the diversity gain and security potential provided by multipath channels. Meanwhile, the traditional large-scale antenna array depends on a complex phased array or an all-digital architecture, and each antenna unit needs to be provided with an independent radio frequency link, so that the hardware cost is high and the power consumption is huge. As an emerging technology, a dynamic super-surface antenna (DMA) has remarkable hardware advantages of low cost and low power consumption, but how to effectively apply the two-dimensional area array to the secure transmission of a multipath physical layer is still in need of intensive study. In addition, the secure transmission scheme based on the additively unique resolvable constellation set (UDCG) has extremely high requirements on channel orthogonality, and the transmitting end generally needs to forcedly eliminate symbol interference among multiple paths or users through precoding. In a practical rich scattering environment, the system would have to expend a large amount of transmit power to "force" the channels to be orthogonal if the spatial correlation of the selected channel combination is high, although a strategy of randomly picking multipath or users can also be used to achieve a substantially independent transmission. This suboptimal combination results in a significant waste of transmit power, greatly limiting the power resources that can be allocated to the active signal, and thus becomes a bottleneck in improving the security capacity of the system. Meanwhile, when facing to a fixed and fixed eavesdropper, the traditional physical layer security scheme cannot acquire eavesdropping channel state information, so that traditional artificial noise and other technologies are difficult to accurately align, and information security is effectively guaranteed. Disclosure of Invention Aiming at the technical problems that the prior physical layer security technology cannot fully utilize the hardware advantage of two-dimensional DMA and the independent transmission characteristic of multipath components of a channel when the passive wiretapping is handled, and a large amount of transmitting power is wasted on forced orthogonal caused by a traditional random channel combination mode, so that the improvement of secret capacity is limited, the invention provides a physical layer security transmission method based on multipath selective independent transmission. The method is characterized in that on the basis of obtaining legal user channel state information and matching an additively unique resolvable constellation group (UDCG), the hardware advantage of a two-dimensional DMA architecture is mainly utilized, and the scene self-adaptive multipath selection and user selection strategies are combined to optimize the power resource allocation of the system, so that the efficient and multipath independent safe transmission of the unique resolvable constellation group is realized. The greedy path selection method comprises the steps of executing greedy path selection strategies based on a gladhand determinant in a single-user scene, and executing two-stage strategies of 'path selection before user selection', namely greedy user selection algorithm combining greedy path screening and multipath subspace projection in a multi-user scene. The strategy can accurately pick out paths or user combinations with high gain and excellent mutual orthogonality from a rich scattering environment in a preprocessing stage. The core effect is that the condition number of the equivalent channel matrix is improved obviously from the physical level, so that the power cost of the system for meeting the strict orthogonality of the additivity UDCG is reduced greatly. Throu