Intelligent reflecting surface assisted millimeter wave communication method based on information age perception

Abstract

The application provides an intelligent reflecting surface assisted millimeter wave communication method based on information age perception. The method comprises the following steps: an information age perception IRS assisted millimeter wave MIMO communication system model is established, and a hierarchical codebook is generated for a BS and each UE in the system; each UE is traversed, a block coordinate descent method is used to alternately optimize a BS and a transceiving end beamforming vector and an IRS phase shift of the UE, a scheduling strategy is determined based on a heuristic scheduling algorithm to maximize a system sum rate while meeting all user AoI constraints according to the transceiving end beamforming vector, the IRS phase shift, a received signal signal-to-noise ratio and a maximum achievable data rate of each UE, and the scheduled UE corresponding to each time slot and the transceiving end beamforming vector and the IRS phase shift of each time slot are obtained. The application realizes joint optimization of the transceiving end beamforming, the IRS phase shift and the scheduling strategy, and guarantees the user information timeliness and improves the system sum rate performance with low complexity.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a method for intelligent reflective surface-assisted millimeter-wave communication based on information age perception. Background Technology

[0002] In 6G communication systems, many applications such as holographic projection, smart healthcare, and autonomous driving rely on timely and effective information exchange. Once information is generated, it needs to be sent to the receiving end promptly for utilization; outdated information will severely impact the user's experience. Typically, Age of Information (AoI) is used as an indicator of information timeliness, representing its freshness. Considering each user's need for timely information, ensuring this need is crucial.

[0003] In recent years, the emergence of various new applications has led to a dramatic increase in mobile data traffic. This has placed higher demands on communication system capacity, further exacerbating the contradiction between communication capacity requirements and scarce spectrum resources. Intelligent Reflecting Surface (IRS), as a technology with broad development prospects, can reconfigure wireless channels according to wireless communication needs, providing great flexibility for the design of 6G communication systems and further stimulating the potential of communication systems. An IRS is a plane composed of a large number of passive reconfigurable reflective elements, each of which can independently control the amplitude and phase changes of the incident signal in a software-defined manner. With proper design, the passive reflections of all the reflective elements in the IRS can be coherently superimposed on the desired receiver to increase signal power. Simultaneously, the IRS creates a reliable reflection link, effectively combating signal blockage from direct-fire links, especially millimeter-wave direct-fire links. Furthermore, these passive reflections can also be destructively superimposed at undesired receivers to suppress interference. Compared with traditional communication technologies, the most significant feature of IRS is that it achieves artificial control of the radio propagation environment through the regulation of all reflective elements, making the uncontrollable wireless channel intelligent. Because IRS is a passive device, the reflecting unit only passively reflects the signal, eliminating the need for an RF chain and amplification / retransmission devices. Compared to massive MIMO, IRS significantly reduces hardware and energy costs; compared to repeaters, IRS can operate in full-duplex mode without introducing antenna noise amplification and self-interference. From an implementation and deployment perspective, IRS is small and lightweight, making it very easy to deploy on building facades, lampposts, walls, vehicle surfaces, etc., thus allowing it to be integrated into various communication systems, providing flexible and effective solutions for different communication needs in different scenarios. Therefore, IRS has extremely high research value and broad application prospects.

[0004] In IRS-assisted millimeter-wave multi-antenna communication systems, the joint optimization of active beamforming at the transceiver antennas and IRS phase shifting is a key research focus. One approach compares two optimization schemes in indoor IRS-assisted millimeter-wave environments with no line-of-sight (LOS) paths. Results show that joint optimization of the IRS reflector and transmitter precoding can effectively improve channel capacity. Another approach considers IRS-assisted downlink millimeter-wave systems with hybrid beamforming structures, developing a manifold optimization (MO)-based algorithm to jointly optimize the IRS reflection coefficient and hybrid beamforming at the base station, thereby maximizing spectral efficiency. A third approach designs a successive interference cancellation (SIC) method to address the bandwidth efficiency maximization problem, proposing a greedy algorithm for hybrid beamforming design and using a complex circle manifold (CCM)-based method to update IRS cells. One approach establishes a power minimization problem with signal-to-interference-plus-noise ratio (SINR) constraints in a multi-user scenario. It proposes a two-layer penalty-based algorithm to decouple the variables in the SINR constraint and designs three different methods to optimize the base station's simulated beamforming and IRS response matrix. Another approach studies an IRS-assisted high-speed rail communication scenario, where both the base station and the high-speed rail mobile relay employ multiple antennas. In this scenario, it investigates the minimization problem of system outage probability based on statistical CSI (Channel State Information), proposing an alternating optimization method to jointly optimize the transceiver beamforming vector and IRS phase shift.

[0005] Due to the extremely high requirements for information timeliness in emerging application scenarios such as the Industrial Internet of Things and autonomous driving, information age, as a performance indicator that can effectively characterize information timeliness, has received widespread attention from researchers. In these time-sensitive application scenarios, how to design scheduling strategies to minimize the system's information age or meet users' information age requirements is an urgent problem to be solved. One approach designs two scheduling strategies for minimizing information age in wireless networks: one constructs the optimization problem as an integer linear programming problem and uses existing optimization tools to solve for the global optimum on a small scale; the other is a suboptimal but more scalable steepest age descent algorithm, which achieves this by scheduling data packets that significantly reduce data age as early as possible. Another approach proposes an optimization problem that minimizes the network's expected weighted sum and information age while satisfying the real-time throughput constraints of nodes, and designs four low-complexity scheduling strategies for solving this problem. Theoretical analysis and simulation evaluation results show that the maximum weight strategy and the drift plus penalty strategy have better performance in terms of information age and throughput. One approach compares the minimization of information age problem with the minimization of maximum delay problem, demonstrating that any optimal solution to the latter is an approximation of the former. Based on this conclusion, a framework was developed that uses an approximation of the maximum delay problem as an approximation of the information age problem, significantly reducing the difficulty of solving it. Another approach considers the problem of maximizing long-term average throughput in fading channels, where the system's average information age requirement and average power limit are used as constraints. A stationary randomization strategy independent of information age is proposed to solve this problem, achieving near-optimal performance.

[0006] For optimizing the information age of IRS-assisted communication systems, some solutions have explored cooperative autonomous driving systems assisted by IRS. These solutions optimize resource blocks and IRS scheduling strategies to minimize the average information age of all data streams from the target node. Other solutions employ a joint optimization algorithm based on semi-definite relaxation (SDR) to optimize IRS phase shift and scheduling strategies, addressing the problem of minimizing the total information age in IRS-assisted wireless networks under single-antenna scenarios. Still others have investigated a wireless network using IRS-equipped unmanned aerial vehicles (UAVs) as passive relays to assist Internet-of-Things Devices (IoTDs). A deep reinforcement learning framework was developed to jointly optimize UAV altitude, IRS phase shift, and scheduling strategies to minimize the expected total information age of the system. Finally, a solution deploys the IRS between IoTDs and UAVs to overcome the obstruction of urban buildings, employing a deep reinforcement learning scheme to optimize UAV trajectories, IRS phase shift, and scheduling strategies to minimize the total information age of all devices. One approach uses deep reinforcement learning algorithms to jointly optimize the IRS phase shift and IoTD packet service time, addressing the issue of minimizing the average peak information age in IRS-assisted non-orthogonal multiple access (NOMA) networks.

[0007] The shortcomings of the information aging methods in the aforementioned existing IRS-assisted communication systems include: these methods primarily focus on minimizing the total information age. However, for real-world communication scenarios, these methods neglect the information freshness requirements of each user device. Only by meeting the information freshness requirements of user devices can the information sent by the transmitter be effectively utilized by the user devices, and only then does the information transmission have value. Therefore, optimizing the system design to meet the users' information freshness requirements is crucial for time-sensitive application scenarios.

[0008] These methods are designed for low-frequency communication systems. However, with the rapid development of mobile communication, low-frequency communication cannot provide sufficient channel capacity to handle the massive traffic of various emerging application scenarios, which is detrimental to providing users with a good communication experience. Using millimeter-wave band communication, which has a large available bandwidth, is an effective solution. However, significant path loss and susceptibility to blocking increase the information age of the system, hindering the acquisition of fresh information at the receiver. Therefore, conducting information age research in IRS-assisted millimeter-wave communication systems presents certain challenges.

[0009] Many studies on IRS-assisted millimeter-wave communication assume the perfect knowledge of all channels. However, obtaining a perfect channel index (CSI) through channel estimation is challenging for IRS-assisted millimeter-wave systems. On one hand, the IRS is a passive device without a radio frequency chain, unable to receive, transmit, or process signals. On the other hand, the large-scale antenna arrays and numerous IRS reflectors in the system result in a very large channel matrix, leading to significant overhead for channel estimation. Therefore, system design based on perfect CSI faces serious challenges. To address this issue, existing research utilizes beamforming techniques, selecting suitable codewords from the codebook through beam scanning for beamforming design; however, there is currently no research on time-sensitive scenarios. Summary of the Invention

[0010] Embodiments of the present invention provide a smart reflective surface-assisted millimeter-wave communication method based on information age perception, so as to achieve the guarantee of timeliness of user information and the improvement of system and rate performance with low complexity.

[0011] To achieve the above objectives, the present invention adopts the following technical solution.

[0012] A method for intelligent reflective surface-assisted millimeter-wave communication based on information age perception includes:

[0013] A model of an intelligent reflective surface IRS-assisted millimeter-wave MIMO communication system with information age awareness is established. In this model, by adjusting the discrete phase shift of each passive reflective unit of the IRS, time division multiple access is used to schedule only one UE to communicate with the base station BS in each time slot. Each UE has a specific information age AoI requirement.

[0014] Generate a hierarchical codebook for the BS and each UE in the system;

[0015] For each UE, the block coordinate descent method is used to alternately optimize the transceiver beamforming vector and IRS phase shift of the BS and the UE, and the optimized transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio and maximum achievable data rate of the UE are saved.

[0016] Based on each UE's transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate, a heuristic scheduling algorithm is used to determine a scheduling strategy that maximizes system performance and rate while satisfying all user AoI constraints.

[0017] Based on the scheduling strategy, the scheduled UE corresponding to each time slot is obtained, as well as the transceiver beamforming vector and IRS phase shift for each time slot.

[0018] Preferably, the aforementioned intelligent reflective surface IRS-assisted millimeter-wave MIMO communication system model for establishing information age awareness, in which the discrete phase shift of each reflective element of the IRS is adjusted, and time division multiple access is used to schedule only one UE to communicate with the base station BS in each time slot, and each UE has specific information age (AoI) requirements, including:

[0019] In a single-cell IRS-assisted millimeter-wave MIMO communication system, K UEs obtain information from a BS, and the BS is equipped with N... t Each UE is equipped with N antennas. r There are one antenna. The direct link between the BS and the UE is blocked by an obstacle. An IRS with M passive reflective elements is deployed in the system. The reflective link is provided to the UE by adjusting the discrete phase shift of each reflective element of the IRS.

[0020] The system time is divided into a series of non-overlapping superframes. Each superframe consists of two phases: a scheduling phase and a transmission phase. In the scheduling phase, the central controller located at the BS calculates the scheduling and network optimization scheme and sends the scheme to the IRS and UE. In the transmission phase, the BS communicates with the UE with the assistance of the IRS following the scheme. If the system changes during the transmission phase, causing a transmission failure, the UE will report the transmission failure to the BS. The BS will then redesign the scheme, and the system will advance to the next superframe. The transmission phase is further divided into T time slots. The system uses time division multiple access, and only one UE is allowed to communicate with the BS in each time slot. Each UE has specific Information Age (AoI) requirements.

[0021] Preferably, generating a hierarchical codebook for the BS and each UE in the system includes:

[0022] Design a layered codebook for the BS in the system. t Design a hierarchical codebook for each UE in the system. r In a hierarchical codebook, codewords in different layers have different beamwidths. Codewords in lower layers have wider beamwidths. The beamwidth of any codeword in a layer completely covers the beamwidths of two adjacent codewords in the next layer. The beamwidths of all codewords in each layer cover the entire angular domain. The usable angles of codewords in the last layer are within [-

[0023] The array response vector is represented by uniform sampling within [1,1]. To generate codewords for other layers, a joint subarray and deactivation method is used to broaden the beam.

[0024] Preferably, the step of traversing each UE and alternately optimizing the transceiver beamforming vector and IRS phase shift of the BS and the UE using a block coordinate descent method, and saving the optimized transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate of the UE, includes:

[0025] For each UE, the beamforming vectors and IRS phase shifts of the BS and the UE are alternately optimized using a block coordinate descent method. In each iteration, the beamforming vectors and IRS phase shift matrices of the BS and the UE are randomly initialized, the IRS phase shift matrix is ​​fixed, the BS is set to omnidirectional mode, and a hierarchical search is performed in the UE codebook to find the receive codeword that maximizes the data rate based on the UE's feedback. Then, the UE is set to the directional mode corresponding to the codeword, and a hierarchical search is performed in the BS codebook to find the transmit codeword that maximizes the data rate based on the UE's feedback. Finally, both the BS and the UE are set to the directional mode corresponding to the searched codeword, and a local search method is used to update the IRS phase shift based on the UE's feedback. If the ratio of the data rate difference between two consecutive iterations is less than a certain threshold, it indicates that the algorithm has converged, and the data rate value of the current iteration is the maximum achievable data rate value for the UE.

[0026] Preferably, the step of traversing each UE and alternately optimizing the transceiver beamforming vector and IRS phase shift of the BS and the UE using a block coordinate descent method, and saving the optimized transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate of the UE, includes:

[0027] In the absence of Channel State Information (CSI), the design of transceiver beamforming, IRS phase reflection coefficient matrix, and scheduling strategy is jointly optimized to model the system and rate maximization problem while ensuring the UE's information timeliness requirements as follows:

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035] Where U=[u1,u2,...,u T ] T Let W represent a set of T time slot scheduling strategies, where W = [w1, w2, ..., w T ] T F represents the set of beamforming vectors for T time slots at the BS end, where F = [f1, f2, ..., f T ] TRepresents the set of beamforming vectors for T time slot UEs, Φ = [Φ1, Φ2, ..., Φ T ] T R represents the set of phase reflection coefficient matrices for T time slot IRS. t =log2(1+γ) t () represents the achievable transmission rate of the time slot t system. This represents the average AoI over UEk T time slots. Indicates the maximum tolerable AoI, u for UE k. k,t ∈{0,1} indicates whether UE k is scheduled in time slot t. Let m be the phase reflection coefficient of the m-th IRS reflection unit in time slot t, derived from the discrete phase shift set. Take the value from;

[0036] M is the number of IRS reflection units, K is the total number of UEs in the system, T is the number of time slots, constraint (11) means that only one UE is scheduled in each time slot, constraint (12) ensures that the information timeliness of each UE meets the requirements, and constraint (13) means that the scheduling variable u k,t For 0-1 variables, constraint (14) indicates that the phase reflection coefficient of the IRS has discrete values, and constraints (15) and (16) indicate that the beamforming at the transmitting and receiving ends is obtained from the codebook;

[0037] The P1 problem is decomposed into K individual UE rate maximization problems and a scheduling strategy design problem. The problem of maximizing the data rate of each UE through joint optimization of beamforming and IRS reflection coefficients is expressed in the following form:

[0038]

[0039]

[0040] w k ∈Γ t (19)

[0041] f k ∈Γ r (20)

[0042] Where w k f k , and Φ k These correspond to the beamforming vector at the BS end when serving UE k, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix, respectively. Let m be the phase reflection coefficient of the m-th IRS reflection unit when serving UE k, from the discrete phase shift set The values ​​are taken from Γ, and the beamforming codebooks for BS and UE are represented as Γ. t and Γ rH k =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T For the transmitted signal power, σ 2 This represents the power of additive white Gaussian noise;

[0043] P1-1 is decomposed into a beamforming optimization subproblem and an IRS reflection coefficient optimization subproblem. In the beamforming optimization subproblem, with the IRS phase shift fixed, the problem of maximizing the data rate of each UE by optimizing the beamforming at the transceiver end is expressed as:

[0044]

[0045] stw k ∈Γ t , (twenty two)

[0046] f k ∈Γ r . (twenty three)

[0047] Where w k and f k The beamforming vectors of the BS and UE k respectively correspond to the beamforming vectors of UE k when serving UE k. The beamforming codebooks of the BS and UE are represented as Γ. t and Γ r H k =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T For the transmitted signal power, σ 2 This represents the power of additive white Gaussian noise;

[0048] In the IRS reflection coefficient optimization subproblem, with fixed transmit and receive beamforming, the problem of finding an effective IRS reflection coefficient matrix to maximize the data rate for each user is expressed as:

[0049]

[0050]

[0051] Where w k f k , and Φ k These correspond to the beamforming vector at the BS end when serving UE k, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix, respectively. Let m be the phase reflection coefficient of the m-th IRS reflection unit when serving UE k, from the discrete phase shift set Take the value from H k =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T For the transmitted signal power, σ 2 This represents the additive white Gaussian noise power, and M is the number of IRS reflector units.

[0052] Preferably, the method of determining a scheduling strategy based on a heuristic scheduling algorithm, which maximizes system performance and rate while satisfying all user AoI constraints, based on each UE's transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate, includes:

[0053] use Let this represent the set of UEs whose signal-to-noise ratio satisfies the constraints. This represents the set of data rates corresponding to a UE, where UEk is its data rate. The mapping between them is K indicates t To represent the UEs scheduled in time slot t, a heuristic scheduling algorithm is used to determine a scheduling strategy that maximizes system performance and rate while satisfying all user AoI constraints.

[0054] The scheduling algorithm's processing includes: allocating time slots to UEs that meet the SNR constraints in descending order of UE data rate, with each time slot scheduling set... The UE with the highest data rate is selected, and the data rate of the scheduled UE is reduced from... Remove from, if If it is an empty set, then... Reinitialize to To proceed with the next round of scheduling; traverse each UE in descending order of data rate. For each UE k, it is necessary to traverse all time slots. If UE k was not scheduled in time slot t, and UE k was scheduled in time slot t... t If the data rate of UE is less than the data rate of UEk, then the UE scheduled in time slot t will be set as UEk, and UEk K will be calculated. t The AoI is determined and it is determined whether it meets the AoI constraints. If it does, the adjustment is retained; otherwise, no adjustment is made. After the adjustment of all UEs is completed, the final scheduling strategy is obtained.

[0055] Preferably, obtaining the scheduled UE corresponding to each time slot, as well as the transceiver beamforming vector and IRS phase shift of the UE, according to the scheduling strategy includes:

[0056] Check whether the SNR of each user meets the requirements for reliable demodulation, and put the UEs that meet the requirements and their data rates into the set respectively. and In the middle, the collection and As input to the scheduling algorithm, the scheduling policy U is obtained. * Based on U * Obtained by block coordinate descent method and Obtain the beamforming vector and IRS phase reflection coefficient matrix corresponding to each time slot;

[0057] During the transmission phase, at the beginning of each time slot, the BS samples the state update information and sends state update packets. Old state update packets are replaced by newly arrived packets. k,t ∈{0,1} indicates whether UE k is scheduled in time slot t. If UE k is scheduled, u k,t =1, then the BS sends a state update packet to UE k; otherwise, u k,t =0, the scheduling strategy for time slot t that maximizes system performance and rate while satisfying all user AoI constraints is expressed as: u t =[u 1,t ,u 2,t ,...,u K,t ] T ;

[0058] If UE k successfully acquires and utilizes the status update information sent by the BS in time slot t, then UE k's AoI is reset to 1; otherwise, the AoI is incremented by 1. The evolution is given by the following formula:

[0059]

[0060] Assuming initial AoI Average AoI of UE k over T time slots Represented as

[0061]

[0062] Considering the timeliness requirements of each UE, the maximum tolerable AoI for UE k is denoted as: The AoI of each UE should satisfy

[0063] As can be seen from the technical solutions provided by the embodiments of the present invention above, the embodiments of the present invention consider IRS-assisted millimeter-wave downlink MIMO communication systems in time-sensitive scenarios. Multiple antenna elements are deployed at both the base station transmitter and the user equipment receiver. Beamforming technology is used to obtain antenna gain to compensate for path loss of millimeter-wave signals. Simultaneously, the phase of each reflection element of the IRS is dynamically adjusted to establish a reliable reflection path to counteract the blocking of the direct path. Considering the difficulty of channel estimation, the present invention designs a joint optimization scheme for transceiver beamforming, IRS reflection coefficients, and scheduling strategies in the case of unknown CSI, achieving both the guarantee of user information timeliness and the improvement of system and rate performance with lower complexity.

[0064] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of the invention. Attached Figure Description

[0065] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 A flowchart illustrating a method for IRS-assisted millimeter-wave communication based on information age awareness, provided in an embodiment of the present invention.

[0067] Figure 2 A model diagram of a single-cell millimeter-wave MIMO communication system provided in an embodiment of the present invention;

[0068] Figure 3 This is a schematic diagram of a hierarchical codebook provided in an embodiment of the present invention. Detailed Implementation

[0069] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0070] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or couplings. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0071] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0072] To facilitate understanding of the embodiments of the present invention, the following will provide further explanation and description with reference to the accompanying drawings and several specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0073] Information Age: The time elapsed since the latest information received by the receiver was generated, used to characterize the freshness of the information. Smart Reflective Surface: A two-dimensional plane composed of numerous passive metamaterial reflective elements, which can be reconfigured to change the radio wave propagation environment through software programming. Millimeter-Wave Communication System: A communication system operating in the millimeter-wave frequency band, capable of meeting the demands of high-volume, high-rate wireless communication. Beamforming: Concentrating antenna energy in a specific direction to form a directional beam, providing additional antenna gain to the communication system and thus improving received power. Phase Shift Optimization: Independently adjusting the phase parameters of each reflective element on the smart reflective surface through a controller, flexibly controlling the propagation direction of the incident signal reflected wave, thereby improving the system's transmission quality and reliability. Scheduling Strategy: Allocating limited time slot resources to user equipment in the system to meet user needs and improve overall system performance.

[0074] This invention focuses on ensuring users' timeliness needs in information age-aware communication systems. It considers the millimeter-wave band, covering a frequency range from 30 GHz to 300 GHz, which supports Gbps transmission rates. However, due to its higher frequency, millimeter waves experience greater path loss during propagation compared to lower frequency bands. To compensate for this propagation loss, large-scale antenna arrays and beamforming are typically used to concentrate signal energy in a specific direction to achieve antenna gain and increase received signal power. Therefore, this invention primarily focuses on millimeter-wave multiple-input multiple-output (MIMO) systems.

[0075] This invention decomposes the system and rate maximization problem under information age constraints into a single-user rate maximization problem and a scheduling strategy design problem. It utilizes a block coordinate descent algorithm to jointly optimize beamforming and IRS reflection coefficients, and proposes a low-complexity heuristic algorithm for scheduling strategy design.

[0076] The processing flowchart of an IRS-assisted millimeter-wave communication method based on information age awareness provided in this embodiment of the invention is as follows: Figure 1 As shown, the processing steps include the following;

[0077] Step S10: Establish an information age-aware IRS-assisted millimeter-wave MIMO communication system model. In this model, the direct path is blocked by obstacles in the environment. By adjusting the discrete phase shift of each passive reflective unit in the IRS, a reliable reflection link is created for the User Equipment (UE) to counteract the blockage of the direct path. This model employs time division multiple access, scheduling only one UE to communicate with the base station (BS) in each time slot. The timeliness of received information is crucial for the UE; in this model, each UE has specific AoI requirements.

[0078] Step S20: Generate a hierarchical codebook for the BS and each UE in the system.

[0079] Step S30: Traverse each UE and use the Block Coordinate Descent (BCD) method to alternately optimize the transceiver beamforming vector and IRS phase shift of the BS and the UE, and save the optimized transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio and maximum achievable data rate of the UE.

[0080] In each iteration, the beamforming vectors and IRS phase shift matrix of the BS and the UE are first randomly initialized. Then, the IRS phase shift matrix is ​​fixed, the BS is set to omnidirectional mode, and a hierarchical search is performed in the UE codebook to find the receive codeword that maximizes the data rate based on the UE's feedback. Next, the UE is set to the directional mode corresponding to this codeword, and a hierarchical search is performed in the BS codebook to find the transmit codeword that maximizes the data rate based on the UE's feedback. Finally, both the BS and the UE are set to the directional mode corresponding to the searched codeword, and a local search method is used to update the IRS phase shift based on the UE's feedback. If the ratio of the data rate difference between two consecutive iterations is less than a certain threshold, the algorithm has converged, and the data rate value of the current iteration is the maximum achievable data rate value for the UE.

[0081] Step S40: Based on the maximum achievable data rate of each UE, determine the scheduling strategy based on a heuristic scheduling algorithm to maximize system performance and rate while satisfying all user AoI constraints.

[0082] Step S50: Obtain the scheduled UE, the UE's transceiver beamforming vector, and the IRS phase shift corresponding to each time slot according to the scheduling strategy.

[0083] Consider a single-cell millimeter-wave MIMO communication system where K User Equipments (UEs) need to obtain the most up-to-date information from the Base Station (BS). For example, a UE needs real-time traffic information from the BS for route planning. Therefore, information timeliness is crucial for the UE. An embodiment of this invention provides a single-cell millimeter-wave MIMO communication system model as follows: Figure 2 As shown, BS is equipped with N t Each UE is equipped with N antennas. r There are one antenna. The direct link between the BS and the UE is blocked by some obstacles (such as tall buildings, trees, etc.). Therefore, an IRS with M passive reflective elements is deployed in the system. By adjusting the amplitude / phase of each reflective element, a reliable reflective link can be provided to the UE.

[0084] The system's time is divided into a series of non-overlapping superframes. Each superframe consists of two phases: a scheduling phase and a transmission phase. In the scheduling phase, the central controller located at the BS calculates a scheduling and network optimization scheme, which is then sent to the IRS and the UE. In the transmission phase, the BS communicates with the UE with the assistance of the IRS, which follows the scheme. If a change occurs during the transmission phase leading to a transmission failure, the UE will report the failure to the BS. The BS will then redesign the scheme, and the system will proceed to the next superframe ahead of schedule. The transmission phase can be further divided into T time slots. The system uses time division multiple access (TDMA), and only one UE is allowed to communicate with the BS per time slot, thus eliminating interference between different UEs.

[0085] This invention focuses on system performance during the transmission phase. In each time slot, the BS operates with the same signal power P. T Data transmission. The BS transmits data s to the scheduled UE in time slot t. t satisfy: The beamforming vectors at the BS and UE ends are denoted as follows: and The UE receives and processes the signal y in time slot t. t It can be represented as:

[0086]

[0087] in σ represents the additive white Gaussian noise introduced into the system. 2 This represents the Gaussian noise power. This represents the channel coefficient matrix. Because the direct link is blocked, the transmitted signal reaches the UE via the BS-IRS-UE reflection channel. Due to severe path loss, signals reflected twice or more by the IRS can be ignored; only the signal from the first reflection is considered. The BS-IRS channel is denoted as... IRS-UE channel denoted as Therefore, the channel matrix It can be represented as H t =G t Φ t H r,t .in Let represent the IRS reflection coefficient matrix, j denote the imaginary unit, and diag(·) denote the diagonal matrix function. β m,t ∈[0,1] and These correspond to the amplitude reflection coefficient and phase reflection coefficient of the m-th unit in time slot t, respectively. It is typically assumed that each IRS unit does not change the amplitude of the incident signal, i.e. For ease of hardware implementation, the IRS has discrete phase reflection coefficient values. We assume that each IRS unit can be implemented using b-bit quantization. bA set of distinct discrete phase shift values Represented as

[0088] Therefore, the received signal-to-noise ratio (SNR) γ of the UE scheduled in time slot t is... t It can be represented as:

[0089]

[0090] Where |·| represents the absolute value operation of a complex number. The achievable transmission rate R of the time-slot t system. t It can be represented as

[0091] R t =log2(1+γ) t (3)

[0092] It is worth noting that this invention assumes the CSI is unknown, meaning that channel parameters do not need to be obtained in advance through channel estimation, and the designed optimization scheme can be executed even with an unknown CSI. Therefore, this invention employs a beam training method during the scheduling phase. Specifically, the beam search space is represented by a codebook containing multiple codewords, with the codebooks for the BS and UE represented as Γ, respectively. t and Γ r Therefore, each time slot t has w t ∈Γ t and f t ∈Γ r During the scheduling phase, the BS continuously transmits beamforming signals to each UE via reflection from the IRS. Both the BS and each UE can scan the beamforming vectors in the pre-designed codebook, while simultaneously... Different phase shifts are selected for each IRS unit to change the direction of the reflected beam. Then, based on feedback from the UE, a combination of beamforming vector and IRS phase reflection coefficients that maximizes the UE's achievable transmission rate is selected.

[0093] Due to their short wavelengths, millimeter-wave channels exhibit limited scattering characteristics. Therefore, this invention considers all channels to conform to the SV (Saleh-Valenzuela) channel model. It is assumed that the channel remains constant within a superframe, and that the antennas at the BS and UE ends employ a uniform linear array (ULA), while the intelligent reflection unit at the IRS end uses a uniform planar array (UPA) deployment. In each time slot t, the BS-IRS channel G... t It can be represented as:

[0094]

[0095] Where P is the total number of paths between BS and IRS. This represents the complex gain of the LOS path. (i≠1) represents the complex gain of the i-th non-line-of-sight (NLOS) path, where denoted by , μ represents the Rice factor, which is defined as the ratio of the power of the LOS path to the sum of the power of all NLOS paths. and These represent the azimuth and elevation angles of the IRS end's angle of arrival, respectively. Indicates the departure angle at the BS end, a t (·) and a r (·) represent the normalized angular steering vector functions at the transmitter and receiver, respectively. For a ULA with N elements, this function is expressed as:

[0096]

[0097] Where d is the distance between adjacent antennas, and λ represents the signal wavelength.

[0098] For those with M = M a ×M b The UPA of each unit is expressed as follows:

[0099]

[0100] IRS-UE Channel H r,t It can also be represented using the SV channel model as follows:

[0101]

[0102] Where L is the total number of paths between BS and IRS. This represents the complex gain of the LOS path. (i≠1) represents the complex gain of the i-th NLOS path. and These represent the azimuth and elevation angles of the IRS departure angle, respectively. This represents the angle of arrival at the UE. Without channel estimation, these channel parameters are unknown during the scheduling phase.

[0103] This invention uses AoI to measure the timeliness of information received by the UE. It assumes that the BS follows a per-slot sampling strategy, meaning that at the beginning of each time slot, the BS samples the state update information and sends state update data packets. Simultaneously, older state update data packets are replaced by newly arriving packets; there is no data packet queuing process. k,t ∈{0,1} indicates whether UE k is scheduled in time slot t. If UE k is scheduled, uk,t =1, then the BS sends a state update packet to UE k; otherwise, u k,t =0. The overall scheduling strategy of the time slot t system is represented as u. t =[u 1,t ,u 2,t ,...,u K,t ] T However, for the UE to successfully acquire and utilize the status update information sent by the BS, in addition to factors related to the scheduling policy, the SNR of the UE's received signal must also exceed the reliable demodulation threshold γ. th , i.e. γ t >γ th If UE k successfully acquires and utilizes the state update information sent by the BS in time slot t, then UE k's AoI is reset to 1; otherwise, the AoI is incremented by 1. Therefore, UE k's AoI... The evolution is given by the following formula:

[0104]

[0105] Assuming initial AoI Average AoI of UE k over T time slots Represented as

[0106]

[0107] Considering the timeliness requirements of each UE, we represent the maximum tolerable AoI of UE k as follows: The AoI of each UE should satisfy

[0108] The problem of maximizing sum and rate with information age constraints is established and decomposed as follows: This invention considers maximizing system sum and rate while ensuring the information timeliness requirements of the UE. Under the condition of unknown CSI, it jointly optimizes the beamforming at the transceiver end, the IRS phase reflection coefficient matrix, and the scheduling strategy design. Therefore, the proposed system sum and rate maximization problem can be modeled as follows:

[0109]

[0110]

[0111]

[0112]

[0113]

[0114]

[0115]

[0116] Where U=[u1,u2,...,u T ] T Let W represent a set of T time slot scheduling strategies, where W = [w1, w2, ..., w T ] T F represents the set of beamforming vectors for T time slots at the BS end, where F = [f1, f2, ..., f T ] T Represents the set of beamforming vectors for T time slot UEs, Φ = [Φ1, Φ2, ..., Φ T ] T R represents the set of phase reflection coefficient matrices for T time slot IRS. t =log2(1+γ) t ) represents the achievable transmission rate of the time slot t system. This represents the average AoI over UEk T time slots. This represents the maximum tolerable AoI for UE k. k,t ∈{0,1} indicates whether UE k is scheduled in time slot t. Let m be the phase reflection coefficient of the m-th IRS reflection unit in time slot t, derived from the discrete phase shift set. Take the value from the middle.

[0117] The beamforming codebooks for BS and UE are represented as Γ, respectively. t and Γ r M is the number of IRS reflection units, K is the total number of UEs in the system, and T is the number of time slots. Among them, constraint (11) means that only one UE is scheduled in each time slot, constraint (12) ensures that the information timeliness of each UE meets the requirements, and constraint (13) means that the scheduling variable u k,t For 0-1 variables, constraint (14) indicates that the phase reflection coefficient of the IRS has discrete values, and constraints (15) and (16) indicate that the beamforming at the transmitting and receiving ends is obtained from the codebook.

[0118] Problem P1 is an integer nonconvex optimization problem, and the four optimization variables U, W, Φ, and F are related to the objective function and The problems are all coupled, and although an exhaustive search can yield the overall optimal solution, the computational complexity is quite high. Since only one user is scheduled per time slot, and the channel state remains unchanged within T time slots, P1 can be decomposed into K single-UE rate maximization problems and a scheduling strategy design problem for ease of solution. For the single-UE rate maximization problem, the goal is to maximize the data rate of each UE through joint optimization of beamforming and IRS reflection coefficients, which can be expressed in the following form:

[0119]

[0120]

[0121] w k ∈Γ t (19)

[0122] f k ∈Γ r (20)

[0123] Where w k f k , and Φ k These correspond to the beamforming vector of the BS end when serving UE k, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix, respectively. Let m be the phase reflection coefficient of the m-th IRS reflection unit when serving UE k, from the discrete phase shift set The value is taken from the middle. The beamforming codebooks of the BS and UE are represented as Γ respectively. t and Γ r H k =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T σ represents the transmitted signal power. 2 This represents the power of additive white Gaussian noise.

[0124] Considering w k ,Φ k and f k There is also coupling between them, further decomposing P1-1 into a beamforming optimization subproblem and an IRS reflection coefficient optimization subproblem. In the beamforming optimization subproblem, the IRS phase shift is fixed, and the data rate of each UE is maximized by optimizing the beamforming at the transceiver end. This problem is expressed as...

[0125]

[0126] stw k ∈Γ t , (twenty two)

[0127] f k ∈Γ r . (twenty three)

[0128] Where w k and f k These correspond to the beamforming vectors of the BS and UE k, respectively. The beamforming codebooks of the BS and UE are represented as Γ. t and Γ r Hk =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T σ represents the transmitted signal power. 2 This represents the power of additive white Gaussian noise.

[0129] In the IRS reflection coefficient optimization subproblem, with fixed transmit and receive beamforming, the goal is to find an efficient IRS reflection coefficient matrix to maximize the data rate for each user. This problem is expressed as follows:

[0130]

[0131]

[0132] Where w k f k , and Φ k These correspond to the beamforming vector of the BS end when serving UE k, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix, respectively. Let m be the phase reflection coefficient of the m-th IRS reflection unit when serving UE k, from the discrete phase shift set Take the value from H. k =GΦ k H r,k R represents the channel matrix from BS to UE k. k P represents the achievable transmission rate of UE k. T σ represents the transmitted signal power. 2 This represents the additive white Gaussian noise power. M is the number of IRS reflector units.

[0133] For the scheduling strategy design problem, which is based on solving the single UE rate maximization problem to obtain the maximum achievable data rate for each UE, the scheduling strategy is designed to maximize the system and rate over T time slots, and can be expressed in the following form:

[0134]

[0135]

[0136]

[0137]

[0138] Where U=[u1,u2,...,u T ] T R represents the set of T time slot scheduling strategies. k This indicates the achievable transmission rate of UE k. This represents the average AoI over k T time slots of the UE. This represents the maximum tolerable AoI for UE k. k,t ∈{0,1} indicates whether UE k is scheduled in time slot t. K is the total number of UEs in the system, and T is the number of time slots.

[0139] Data Rate Maximization Algorithm Design: To solve P1, algorithms can be designed to solve P1-1 and P1-2 separately. The core idea for solving P1-1 is to use a block coordinate descent algorithm to alternately solve the transmit / receive beamforming optimization subproblem P1-1-1 and the IRS phase reflection coefficient optimization subproblem P1-1-2 until the system converges. After solving P1-1, based on the maximum achievable data rate for each UE, a low-complexity heuristic algorithm is used to solve P1-2.

[0140] Optimized beamforming at the transceiver end: Since the system CSI is unknown, this invention uses beam training to obtain the beamforming vector at the transceiver end, thereby solving P1-1-1. Given the codebooks of the BS and UE, although exhaustively searching all transmit-receive beam pairs in the codebook can find the most efficient beam pair, this invention adopts a hierarchical search method to reduce complexity.

[0141] First, a layered codebook needs to be designed for the BS and UE. t and Γ r In a hierarchical codebook, codewords in different layers have different beamwidths, with codewords in lower layers having wider beamwidths. Figure 3 This is a schematic diagram of a hierarchical codebook provided in an embodiment of the present invention, such as... Figure 3 As shown, the beam of any codeword in a layer completely covers the beam of two adjacent codewords in the next layer, while the beam of all codewords in each layer covers the entire angular domain. Algorithm 1 provides details of the codebook design. w(l,n) represents the nth codeword in the l-th layer codebook. As shown in lines 2-3 of Algorithm 1, the array response vector uniformly sampled within the angle range [-1,1] can be used to represent the codewords in the last layer. Next, to generate codewords for other layers, a joint subarray and deactivation method is used to broaden the beam. Specifically, taking the first codeword of each layer as an example, the array containing N antennas is divided into Q subarrays, each subarray having N antennas. S As shown in lines 5-6 of Algorithm 1. Then, in lines 7-12, the number N of the activation subarrays is determined. A And the codeword w of each subarray q It can be given by the following formula:

[0142]

[0143] Where w qThis represents the codeword of the q-th subarray, where Q is the total number of subarrays, and N is the total number of subarrays. A N is the number of activated subarrays. S That is the number of antennas in each subarray. is the normalized angle steering vector function, where d is the distance between adjacent antennas and λ represents the signal wavelength.

[0144] After obtaining the first codeword w(l,1) of each layer, the nth codeword of that layer can be obtained by rotating w(l,1). The result is shown in line 13.

[0145]

[0146]

[0147] After obtaining the codebook t and Γ r Then, a hierarchical search is performed to find the optimal beam pair. The hierarchical search process is shown in Algorithm 2. First, the BS is fixed in omnidirectional mode, and then at Γ... r A binary tree search is performed to find the valid received codeword for the UE, as shown in lines 1-6 of Algorithm 2. Two adjacent codewords in the next layer of the codebook within the beam coverage area of ​​this codeword are used as candidate codewords for the next layer. The determination of the valid received codeword depends on the data rate information fed back by the UE. The UE is then fixed in the directional mode corresponding to the codeword, and... t Perform the same binary tree search to find the valid emission codeword of BS, as shown in lines 7-12 of Algorithm 2.

[0148]

[0149]

[0150] Optimizing IRS Phase Reflection Coefficient: To solve the subproblem P1-1-2 of optimizing the IRS phase reflection coefficient, it is necessary to select a suitable phase shift for each IRS unit from a finite set of discrete phase shifts. Considering the complexity, this invention employs a local search method, as shown in Algorithm 3. Specifically, this algorithm sequentially optimizes the phase shift of each IRS unit while keeping the phase shifts of the remaining M-1 units unchanged. For each IRS unit, the algorithm traverses all possible phase shifts and selects the phase shift that achieves the maximum UE transmission rate as the optimal phase shift for that unit based on UE feedback. This optimal phase shift is then used to optimize the phase shifts of other IRS units until the phase shifts of all units are optimized.

[0151]

[0152] Joint Optimization Based on Block Coordinate Descent Algorithm: To solve the single-user rate maximization problem P1-1, this invention employs the Block Coordinate Descent (BCD) method to alternately optimize the beamforming vector and IRS phase shift matrix, obtaining the maximum achievable data rate and SNR for each user. Specifically, as shown in Algorithm 4, the beamforming vector and IRS phase reflection coefficient matrix are first randomly initialized. In each iteration, the IRS reflection coefficient matrix is ​​first fixed to the result of the previous iteration, and Algorithm 2 is used to update the beamforming strategy. If the data rate of the UE after the beamforming strategy update is greater than the data rate before the update, the updated result is retained; otherwise, the beamforming vector is not updated. Then, using the latest beamforming vector, the IRS phase reflection coefficient matrix is ​​updated using Algorithm 3. If the ratio of the data rate difference between two consecutive iterations is less than a certain threshold, i.e. |R k (τ+1) -R k τ | / R k τ If the value is less than δ, we consider the algorithm to have converged.

[0153]

[0154] Optimized Scheduling Strategy: Based on the maximum achievable data rate for each UE obtained from Algorithm 4, this invention proposes a heuristic scheduling algorithm to solve the scheduling strategy design problem P1-2. Its main idea is to schedule as many UEs with higher data rates as possible while satisfying the AoI constraints of all UEs. For ease of explanation, [the following is used...] Let this represent the set of UEs whose signal-to-noise ratio satisfies the constraints. This represents the set of data rates corresponding to these UEs. UEk and its data rate The mapping between them is K indicates t Used to represent a UE scheduled in time slot t.

[0155] The pseudocode for this scheduling algorithm is shown in Algorithm 5. First, in lines 1-8, time slots are allocated to UEs that meet the SNR constraints in descending order of UE data rate. Specifically, as shown in lines 2-3, each time slot scheduling set... The UE with the highest data rate. In line 4, the data rate of the scheduled UE is changed from... Remove from the middle. In lines 5-7, if If it is an empty set, it means that all UEs have already been scheduled once, then... Reinitialize to To proceed with the next round of scheduling. Next, in lines 9-22, the scheduling strategy is further adjusted to maximize system performance and rate. Specifically, each UE is iterated through in descending order of data rate. For each UE k, all time slots need to be traversed. As shown in lines 12-13, if UE k is not scheduled in time slot t, and UE k is scheduled in time slot t... t If the data rate of UE is less than the data rate of UEk, then the UE scheduled in time slot t will be set as UEk. In lines 14-19, we calculate UEk. t The AoI (Aspect-Oriented Indicator) is determined, and it is verified whether it meets the AoI constraints. If it does, the adjustment is retained; otherwise, no adjustment is made. After adjusting all UEs, the final scheduling policy is obtained.

[0156]

[0157] Sum Rate Maximization: After proposing corresponding algorithms for solving each subproblem, this invention proposes a sum rate maximization algorithm to solve the original problem P1, as shown in Algorithm 6. First, a hierarchical codebook is designed for the BS and UE according to Algorithm 1. Then, Algorithm 4 is used to optimize the beamforming vector and IRS phase reflection coefficient matrix to obtain the maximum achievable data rate and SNR for each user. Next, as shown in lines 5-7, it is checked whether the SNR of each user meets the requirements for reliable demodulation, and UEs that meet the requirements and their data rates are respectively placed into sets. and In the middle. Then, the collection... and As input to Algorithm 5, the scheduling policy U is obtained. * Based on U * And obtained by Algorithm 4 and The beamforming vector and IRS phase reflection coefficient matrix corresponding to each time slot can be easily obtained, i.e., U... * W * and Φ * .

[0158]

[0159] In summary, this invention proposes an information age-aware IRS-assisted millimeter-wave MIMO communication system model. Considering the UE's information timeliness requirements, it establishes a system and rate maximization problem under information age constraints, jointly optimizing transceiver beamforming, IRS phase reflection coefficients, and scheduling strategies. This problem is further decomposed into a single-user rate maximization problem and a scheduling strategy design problem. For the former, with unknown CSI, hierarchical search and local search methods are used to optimize transceiver beamforming and IRS phase reflection coefficients respectively, and a block coordinate descent algorithm is used to iteratively update these two sets of variables until convergence. For the latter, a low-complexity heuristic algorithm is used to design the scheduling strategy, improving system and rate while ensuring that the UE's information timeliness requirements are met.

[0160] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0161] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0162] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0163] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for intelligent reflective surface-assisted millimeter-wave communication based on information age perception, characterized in that, include: A model of an intelligent reflective surface IRS-assisted millimeter-wave MIMO communication system with information age awareness is established. In this model, by adjusting the discrete phase shift of each passive reflective unit of the IRS, time division multiple access is used to schedule only one UE to communicate with the base station BS in each time slot. Each UE has a specific information age AoI requirement. Generate a hierarchical codebook for the BS and each UE in the system; For each UE, the block coordinate descent method is used to alternately optimize the transceiver beamforming vector and IRS phase shift of the BS and the UE, and the optimized transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio and maximum achievable data rate of the UE are saved. Based on each UE's transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate, a heuristic scheduling algorithm is used to determine a scheduling strategy that maximizes system performance and rate while satisfying all user AoI constraints. Based on the scheduling strategy, the scheduled UE corresponding to each time slot is obtained, as well as the transceiver beamforming vector and IRS phase shift for each time slot.

2. The method according to claim 1, characterized in that, The aforementioned model of an intelligent reflective surface IRS-assisted millimeter-wave MIMO communication system for information age awareness is described. In this model, by adjusting the discrete phase shift of each reflective element of the IRS, time division multiple access is used to schedule only one UE to communicate with the base station BS in each time slot. Each UE has specific information age (AoI) requirements, including: In a single-cell IRS-assisted millimeter-wave MIMO communication system Each UE obtains information from the BS, and the BS is equipped with... Each UE is equipped with one antenna. One antenna, the direct link between the BS and UE is blocked by an obstacle, and the system is deployed with The IRS of each passive reflective element provides a reflective link to the UE by adjusting the discrete phase shift of each reflective element in the IRS; The system time is divided into a series of non-overlapping superframes. Each superframe consists of two phases: a scheduling phase and a transmission phase. In the scheduling phase, the central controller located at the BS calculates a scheduling and network optimization scheme and sends the scheme to the IRS and UE. In the transmission phase, the BS communicates with the UE with the assistance of the IRS, which follows the scheme. If a change occurs in the system during the transmission phase, causing a transmission failure, the UE will report the failure to the BS. The BS will then redesign the scheme, and the system will advance to the next superframe. The transmission phase is further divided into evenly spaced segments. The system uses time division multiple access, and each time slot only allows one UE to communicate with the BS. Each UE has specific information age (AoI) requirements.

3. The method according to claim 2, characterized in that, The process of generating a hierarchical codebook for the BS and each UE in the system includes: Design a layered codebook for the BS in the system. Design a hierarchical codebook for each UE in the system. In a hierarchical codebook, codewords in different layers have different beamwidths. The beamwidth of codewords in lower layers is wider than that of codewords in higher layers. The beamwidth of any codeword in a layer completely covers the beamwidth of two adjacent codewords in the next layer. The beamwidth of all codewords in each layer covers the entire angular domain. The codewords in the last layer can be represented by an array response vector with angles uniformly sampled in the range [-1,1]. In order to generate codewords for other layers, a joint subarray and deactivation method is used to broaden the beamwidth.

4. The method according to claim 3, characterized in that, The process of traversing each UE and alternately optimizing the transmit / receive beamforming vector and IRS phase shift of the BS and the UE using a block coordinate descent method, and saving the optimized transmit / receive beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate of the UE, includes: For each UE, the beamforming vectors and IRS phase shifts of the BS and the UE are alternately optimized using a block coordinate descent method. In each iteration, the beamforming vectors and IRS phase shift matrices of the BS and the UE are randomly initialized, the IRS phase shift matrix is ​​fixed, the BS is set to omnidirectional mode, and a hierarchical search is performed in the UE codebook to find the receive codeword that maximizes the data rate based on the UE's feedback. Then, the UE is set to the directional mode corresponding to the codeword, and a hierarchical search is performed in the BS codebook to find the transmit codeword that maximizes the data rate based on the UE's feedback. Finally, both the BS and the UE are set to the directional mode corresponding to the searched codeword, and a local search method is used to update the IRS phase shift based on the UE's feedback. If the ratio of the data rate difference between two consecutive iterations is less than a certain threshold, it indicates that the algorithm has converged, and the data rate value of the current iteration is the maximum achievable data rate value for the UE.

5. The method according to claim 4, characterized in that, The process of traversing each UE and alternately optimizing the transmit / receive beamforming vector and IRS phase shift of the BS and the UE using a block coordinate descent method, and saving the optimized transmit / receive beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate of the UE, includes: In the absence of Channel State Information (CSI), the design of transceiver beamforming, IRS phase reflection coefficient matrix, and scheduling strategy is jointly optimized to model the system and rate maximization problem while ensuring the UE's information timeliness requirements as follows: P1： (10) (11) (12) (13) (14) (15) (16) in This represents a set of T time slot scheduling strategies. This represents the set of T time slot BS-end beamforming vectors. This represents the set of beamforming vectors for T time slot UEs. This represents the set of phase reflection coefficient matrices for T time slot IRS. Indicates time slot The system's achievable transmission rate, Indicates UE Average AoI per time slot, Indicates UE Maximum tolerable AoI Instruct UE In the time slot Whether it is scheduled For the first One IRS reflector unit in time slot The phase reflection coefficient, from the discrete phase shift set Take the value from; M is the number of IRS reflection units, K is the total number of UEs in the system, T is the number of time slots, constraint (11) means that only one UE is scheduled in each time slot, constraint (12) ensures that the information timeliness of each UE meets the requirements, and constraint (13) represents the scheduling variable. For 0-1 variables, constraint (14) indicates that the phase reflection coefficient of the IRS has discrete values, and constraints (15) and (16) indicate that the beamforming at the transmitting and receiving ends is obtained from the codebook; Decompose problem P1 into The problem of maximizing the data rate of a single UE and the problem of designing a scheduling strategy are combined and expressed as follows: P1-1： (17) (18) (19) (20) in , ,and Corresponding to the UEs The beamforming vector at the BS end, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix at that time. For the first One IRS reflection unit serves the UE Phase reflection coefficient at time, from discrete phase shift set The values ​​are taken from the middle, and the beamforming codebooks of the BS and UE are represented as follows: and , Indicates from BS to UE The channel matrix, Indicates UE The achievable transmission rate, For the transmitted signal power, This represents the power of additive white Gaussian noise; P1-1 is decomposed into a beamforming optimization subproblem and an IRS reflection coefficient optimization subproblem. In the beamforming optimization subproblem, with the IRS phase shift fixed, the problem of maximizing the data rate of each UE by optimizing the beamforming at the transceiver end is expressed as: P1-1-1： (21) (22) (23) in and Corresponding to the UEs The beamforming vectors at the BS and UE k are represented by the beamforming codebooks of the BS and UE, respectively. and , Indicates from BS to UE The channel matrix, Indicates UE The achievable transmission rate, For the transmitted signal power, This represents the power of additive white Gaussian noise; In the IRS reflection coefficient optimization subproblem, with fixed transmit and receive beamforming, the problem of finding an effective IRS reflection coefficient matrix to maximize the data rate for each user is expressed as: P1-1-2： (24) (25) in , ,and Corresponding to the UEs The beamforming vector at the BS end, the beamforming vector of UE k, and the IRS phase reflection coefficient matrix at that time. For the first One IRS reflection unit serves the UE Phase reflection coefficient at time, from discrete phase shift set Take the value from the middle. Indicates from BS to UE The channel matrix, Indicates UE The achievable transmission rate, For the transmitted signal power, This represents the additive white Gaussian noise power, and M is the number of IRS reflector units.

6. The method according to claim 5, characterized in that, The aforementioned scheduling strategy, which determines the scheduling policy based on a heuristic scheduling algorithm to maximize system performance and rate while satisfying all user AoI constraints, is based on each UE's transceiver beamforming vector, IRS phase shift, received signal-to-noise ratio, and maximum achievable data rate. This includes: use Let this represent the set of UEs whose signal-to-noise ratio satisfies the constraints. This represents the set of data rates corresponding to the UE. Its data rate The mapping between them is express, Used to indicate in time slot The scheduling strategy for UEs in the middle is determined based on a heuristic scheduling algorithm to maximize system performance and rate while satisfying all user AoI constraints. The scheduling algorithm's processing includes: allocating time slots to UEs that meet the SNR constraints in descending order of UE data rate, with each time slot scheduling set... The UE with the highest data rate is selected, and the data rate of the scheduled UE is reduced from... Remove from, if If it is an empty set, then... Reinitialize to To proceed with the next round of scheduling; traverse each UE in descending order of data rate, for each UE... It is necessary to traverse all time slots if the UE Not in the time slot It was scheduled in the middle, and UE with time slot scheduling Data rate less than UE The data rate will be in the time slot. The scheduled UE is set to UE Calculate UE The AoI is determined and it is determined whether it meets the AoI constraints. If it does, the adjustment is retained; otherwise, no adjustment is made. After the adjustment of all UEs is completed, the final scheduling strategy is obtained.

7. The method according to claim 6, characterized in that, The step of obtaining the scheduled UE corresponding to each time slot, as well as the transceiver beamforming vector and IRS phase shift of the UE according to the scheduling strategy, includes: Check whether the SNR of each user meets the requirements for reliable demodulation, and put the UEs that meet the requirements and their data rates into the set respectively. and In the middle, the collection and As input to the scheduling algorithm, the scheduling policy is obtained. ,based on Obtained by block coordinate descent method , and Obtain the beamforming vector and IRS phase reflection coefficient matrix corresponding to each time slot; During the transmission phase, at the beginning of each time slot, the BS samples the state update information and sends state update data packets. Old state update data packets are replaced by newly arrived data packets. Instruct UE In the time slot Whether it is scheduled, if the UE Scheduled Then BS to UE Send a status update packet; otherwise, This will maximize system and rate time slots while satisfying all user AoI constraints. The scheduling strategy is expressed as: ; If UE In the time slot If the UE successfully obtains and utilizes the status update information sent by the BS, then... The AoI is reset to 1; otherwise, the AoI is incremented by 1, and the UE... AoI The evolution is given by the following formula: (8) Assuming initial AoI UE exist Average AoI per time slot Represented as (9) Considering the timeliness requirements of each UE, the UE will be... The maximum tolerable AoI is expressed as Each UE's AoI should satisfy .

Images