Edge sensing and task unloading cooperation method based on movable antenna

Abstract

The invention discloses an edge perception and task unloading cooperation method based on a movable antenna, and belongs to the technical field of wireless communication and perception integration, the method is applied to a base station equipped with a movable antenna array, and downlink environment perception and uplink multi-user calculation unloading are concurrently executed through the movable antenna array; constructing a joint optimization problem which takes optimization perception performance as a core and takes communication constraints into consideration; an alternative optimization framework is adopted, and receiving beam forming, user transmitting power and transmitting and receiving antenna positions are iteratively optimized, so that dynamic collaborative allocation of sensing and communication computing resources is realized. According to the method, the spatial degree of freedom can be fully mined, the sensing precision and the communication efficiency are improved, and the collaborative demand of high-precision sensing and multi-task concurrent unloading in an edge intelligent scene is effectively supported.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of wireless communication and perception integration, and particularly relates to an edge perception and task offloading coordination method based on a movable antenna. BACKGROUND

[0002] With the popularity of the Internet of Things and intelligent terminal devices, the requirements of edge perception scenarios for the cooperative performance of communication, perception and calculation are increasing. Traditional intelligent networks usually design communication (such as data transmission) and perception (such as radar and camera) as independent modules, which not only increases the system hardware and scheduling complexity, but also may cause competition for spectrum and computing resources, resulting in increased transmission delay and reduced energy efficiency. Therefore, the integrated sensing and communication (ISAC) technology emerges as the times require, which deeply integrates wireless communication and environmental perception functions by sharing hardware platforms and spectrum resources, thereby improving spectrum utilization while effectively reducing system power consumption and processing delay. Existing research has explored ISAC systems from the perspectives of waveform design, performance compromise and beamforming, such as transmitting communication signals using radar pulse intervals to reduce the perception blind area in full-duplex ISAC, or analyzing the trade-off between perception detection probability and communication rate. These works lay a theoretical foundation for the preliminary application of ISAC technology.

[0003] However, the existing ISAC scheme still has obvious limitations. On the one hand, related researches are mostly concentrated in single base station scenarios, which are limited by the signal coverage range, available spectrum resources and processing capacity of a single base station, making it difficult to support the cooperative needs of intelligent sensing and large-scale non-connected scenarios for high data rate and high-precision positioning; on the other hand, existing systems usually configure a fixed position antenna (FPA) array at the base station end, and the positions of the antenna units cannot be dynamically adjusted, which limits the system from fully exploiting the degrees of freedom (DoFs) in the continuous spatial array region, thereby restricting the overall performance of precise perception and efficient communication in complex and variable environments. SUMMARY

[0004] To solve the above technical problems, the application provides an edge perception and task offloading coordination method based on a movable antenna to solve the problems existing in the prior art.

[0005] To achieve the above purpose, the application provides an edge perception and task offloading coordination method based on a movable antenna, applied to a base station equipped with a movable antenna array, comprising: The base station transmits a downlink sensing signal through its movable transmit antenna array, senses a target in the environment, and receives uplink computing offloading signals from multiple user equipments; A joint optimization problem is constructed to optimize the sensing performance as a core objective and meet user communication demands and system constraints. An alternating optimization framework is used to solve the joint optimization problem, and the receive beamforming, user transmit power, transmit antenna position, and receive antenna position are iteratively optimized to achieve the cooperation of sensing and multi-user computing offloading.

[0006] Preferably, the sensing performance as a core objective means that the objective function of the joint optimization problem is to minimize the Cramer-Rao lower bound of angle estimation of the target.

[0007] Preferably, the meeting of user communication demands and system constraints includes that the uplink communication rate of each user equipment is not lower than its task demand rate, the transmit power of each user equipment does not exceed its maximum power budget, and the position of each movable antenna is located within a preset movable region and meets the minimum distance constraint.

[0008] Preferably, the iterative optimization of receive beamforming specifically refers to solving a sub-problem with the objective of maximizing the signal-to-interference-and-noise ratio (SINR) of each user equipment by using the semi-definite relaxation and Gaussian randomization method when other variables are given, to obtain the optimal receive beamforming vector.

[0009] Preferably, the iterative optimization of user transmit power specifically refers to solving a power control sub-problem with the objective of minimizing the Cramer-Rao lower bound and the constraint of ensuring that the SINR of each user equipment meets its demand rate by using a convex optimization tool when other variables are given.

[0010] Preferably, the iterative optimization of transmit antenna position specifically refers to obtaining the optimal position of each transmit antenna by solving a sub-problem with the objective of maximizing the antenna position related fraction under the condition that its position constraint is met, wherein there is a closed-form solution for part of the antenna positions.

[0011] Preferably, the iterative optimization of receive antenna position specifically refers to obtaining the optimal position of each receive antenna by solving a sub-problem with the objective of maximizing the antenna position related fraction by using a one-dimensional search or iterative method under the condition that its position constraint is met.

[0012] Preferably, the base station adopts a full-duplex working mode to concurrently perform the downlink sensing signal transmission and the uplink computing offloading signal reception on the same time-frequency resource.

[0013] Preferably, the movable antenna array includes a transmit uniform linear array composed of a plurality of movable antennas and a receive uniform linear array composed of a plurality of movable antennas.

[0014] Preferably, the target is at least one of the user equipment, or other intelligent agent within the base station communication coverage.

[0015] Compared with the prior art, the present application has the following advantages and technical effects: The present application adopts a movable antenna array as the core hardware architecture, and constructs a joint optimization problem with the core target of optimizing the sensing performance while meeting the communication and power constraints, and then solves it by using an alternating optimization framework of covering receive beamforming, user power, and transmit and receive antenna position. This scheme makes the position of the antenna unit a key variable that can be dynamically optimized, and can break through the fixed position antenna array's limitation in beamforming and spatial freedom degree utilization. By iteratively optimizing the antenna position, the system can actively adapt to the environment and user distribution, more fully exploit the freedom degree in continuous space, so as to realize more accurate environmental perception and more efficient multi-user communication under the same hardware scale, and the overall sensing-communication collaborative performance of the system is significantly improved.

[0016] The present application constructs an ISAC system framework supporting full-duplex working mode, so that the base station can concurrently perform downlink sensing and uplink computing offloading on the same time-frequency resource, and the sensing performance index (such as Cramer-Rao lower bound) and the communication service quality (user rate requirement) are jointly designed in a unified joint optimization problem. Through the alternating optimization framework, the key resources such as power, beam, and antenna position are collaboratively deployed, which fundamentally solves the resource competition problem between the communication and sensing modules in the traditional separate design, and realizes deep sharing and efficient multiplexing of hardware and spectrum resources. This effectively reduces the overall power consumption and signal processing delay of the system, and realizes the integration and efficient collaboration of sensing and communication functions.

[0017] The technical scheme of the present application explicitly supports multiple user equipment to simultaneously perform uplink intelligent task computing offloading, and guarantees the minimum communication rate requirement of each user through constraint conditions in the optimization model. Through the user power control and receive beamforming optimization sub-steps in the alternating optimization framework, the system can intelligently allocate power and spatial resources among multiple users, dynamically balance the communication needs of different users and the sensing task needs of the system. This makes the system of the present application particularly suitable for edge computing scenarios with dense intelligent terminals, and can reliably support concurrent computing task offloading of multiple devices while ensuring high-precision environmental perception, significantly improving the overall response capability and resource adaptation efficiency of the base station to heterogeneous service demands. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application and are incorporated in and constitute a part of this application. The illustrations, together with the description, serve to explain the application, but not to limit it. In the drawings: Figure 1 FD-ISAC assisted by a movable antenna (MA) of an embodiment of the present application; Figure 2 Convergence performance of different schemes of an embodiment of the present application; Figure 3 Relationship between Cramér-Rao Bound (CRB) and base station transmit power of an embodiment of the present application; Figure 4 Relationship between Cramér-Rao Bound (CRB) and the number of receive antennas of an embodiment of the present application; Figure 5 Relationship between Cramér-Rao Bound (CRB) and the size of the movable area of an antenna of an embodiment of the present application. DETAILED DESCRIPTION

[0019] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0020] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0021] Embodiment one An edge perception and task offloading collaborative method based on a movable antenna is provided in the embodiment, which is applied to a base station equipped with a movable antenna array, and includes the following steps: Step 1: The base station transmits a downlink perception signal through its movable antenna array, simultaneously performs perception on a target in the environment, and receives uplink computation offloading signals from a plurality of user equipment; Further, the base station adopts a full-duplex working mode to concurrently perform the downlink perception signal transmission and the uplink computation offloading signal reception on the same time-frequency resource.

[0022] Further, the movable antenna array includes a transmitting uniform linear array composed of a plurality of movable antennas and a receiving uniform linear array composed of a plurality of movable antennas.

[0023] Further, the target is at least one of the user equipment or other intelligent agents within the communication coverage of the base station.

[0024] Specifically, as Figure 1As shown, this embodiment considers a full-duplex integrated sensing and communication (FD-ISAC) mode assisted by a movable antenna (MA), where each base station includes a... A uniform linear array (ULA) of multiple emitters (MA) and a band The base station receives the ULA of each MA to fully explore the degrees of freedom in a continuous spatial region. Uplink intelligent task computation is offloaded for each single-antenna user, and downlink signals are continuously transmitted using the same time-frequency resources to sense a single target. Specifically, each user Send a signal Used for uplink communication. The average transmit power of each user should not exceed the maximum power budget, so the user transmit power should satisfy formula (1).

[0025] (1) in, For user transmission power, This is the user's maximum transmit power.

[0026] In this example of channel modeling, this embodiment models the channel between the BS and the user as follows: in, For complex channel gain coefficients, For the set of complex numbers, It refers to the number of receiving antennas. This is the base station's reception guidance vector for users.

[0027] In order to detect targets, the base station... Each snapshot continuously transmits sensing signals. Assume... Within each snapshot, the channel of the target base station link remains unchanged, and the base station in the... The transmission signal of a snapshot can be written as , This refers to the number of transmitting antennas, and the corresponding average power is... . This represents the expectation operation. Meanwhile, the base station transmit steering vector and receive steering vector used for target sensing are defined as follows: (2) (3) in, It is the angle of the target relative to the base station. Indicates the carrier wavelength. representing a user angle of arrival to the base station, representing a transpose, is a transmit steering vector, is a receive steering vector, let then the base station target The channel of the base station link is The sensed channel coefficients are given by where is a parameter related to the radar cross section of the target, where is the free path loss with a reference distance of 1 m in dB, representing the distance between the base station and the sensed target. Thus, the received signal at the base station in the th snapshot is: (4) where is the additive white Gaussian noise at the base station reception, is the communication signal, is the sensed echo signal, is the sensed signal transmitted by the base station in the th snapshot, is its average noise power. Thus, the received signal at the base station in the th snapshot is: (5) where the signal correlation quantities in different reception scenarios are and the received noise is To achieve uniform sensing performance for arbitrary angles of departure and angles of arrival, it is generally required that is a full-rank matrix (which requires ), so that generates an omnidirectional beam pattern in the angular domain to uniformly scan targets in all possible directions.

[0028] To recover the uplink user's signal, the present embodiment processes the received signal at the base station using a set of receive beamforming vectors To ensure the physical realizability of the receiver, the receive beamforming vectors need to satisfy .

[0029] Thus, the SINR of the user is: (6) where , . The user The maximum number of bits of error-free transmission information of the channel per unit time is: (7) wherein, is the transmission bandwidth.

[0030] In order to ensure the data transmission performance of the uplink communication of each user, the formula (8) should be met: (8) wherein, is the number of bits of the user to be transmitted per unit time.

[0031] Step 2: Construct a joint optimization problem with the core target of optimizing the sensing performance and meeting the user communication demand and system constraints; Further, the sensing performance as the core target means that the objective function of the joint optimization problem is to minimize the Cramer-Rao lower bound of the angle estimation of the target.

[0032] Further, the meeting of the user communication demand and system constraints includes that the uplink communication rate of each user equipment is not lower than the task demand rate thereof, the transmission power of each user equipment does not exceed the maximum power budget thereof, and the position of each movable antenna is located in the preset movable area and meets the minimum distance constraint.

[0033] Specifically, the main purpose of the embodiment is to optimize the sensing performance while ensuring the constraints of the user transmission rate and power, so the optimization problem can be modeled as: (9) wherein C3, C4, C5 and C6 represent the interval constraint between the position of the base station MA and the antenna, represents the maximum transmission power of the user, represents the position of the mth transmission antenna, represents the position of the nth receiving antenna, represents the maximum moving range of the movable antenna, represents the minimum distance between any two antennas.

[0034] Step 3: Solve the joint optimization problem by using the alternating optimization framework, and realize the cooperation of sensing and multi-user computing offloading by iteratively optimizing the receiving beamforming, user transmission power, transmission antenna position and receiving antenna position.

[0035] Further, the iterative optimization of the receiving beamforming is specifically: when other variables are given, the sub-problem with the target of maximizing the signal-to-interference-and-noise ratio of each user equipment is solved by using the semi-definite relaxation and Gaussian randomization method, and the optimal receiving beamforming vector is obtained.

[0036] Further, the iterative optimization of user transmit power is specifically: given other variables, using convex optimization tools to solve the power control sub-problem to ensure that the signal-to-noise ratio of each user equipment meets the demand rate as a constraint, and to minimize the Cramer-Rao lower bound as an objective.

[0037] Further, the iterative optimization of transmit antenna position is specifically: for each transmit antenna, under the condition of meeting its position constraint, the optimal position is obtained by solving the sub-problem with the antenna position related fraction as the objective, wherein there is a closed-form solution for part of the antenna positions.

[0038] Further, the iterative optimization of receive antenna position is specifically: for each receive antenna, under the condition of meeting its position constraint, the optimal position is obtained by one-dimensional search or iterative method to solve the sub-problem with the antenna position related fraction as the objective.

[0039] Specifically, due to the coupling between the above optimization variables, the problem is a non-convex optimization problem, and it is difficult to obtain an optimal solution. The embodiment adopts an alternating optimization algorithm to solve the iterative optimization scheme, and the optimization scheme specifically includes: (1) receive beamforming optimization; Under the condition of given transmit power, antenna position vector of the transmit MA array and antenna position vector of the receive MA array The receive beamforming design sub-problem can be expressed as: (10) wherein, , . Since are independent of each other, the optimization problem is equivalent to solving K sub-problems, and the optimization problem of can be expressed as: (11) Let and , at this time the optimization problem (11) can be transformed into: (12) Then the optimal is obtained through the semi-positive relaxation technique, and then satisfying the rank-one constraint is restored by using Gaussian randomization.

[0040] (2) user power control; Under the condition of given receive beamforming vector, antenna position vector of the transmit MA array and antenna position vector of the receive MA array The power control sub-problem can be expressed as: (13) in, For the optimization problem (13), the existing CVX toolbox can be used to solve it.

[0041] (3) Optimization of transmitting antenna position; According to equation (9), minimizing CRB is equivalent to maximizing it. This embodiment will optimize the position of each transmitting antenna one by one, given that each... One transmitting antenna location The optimization solution process is similar; the following will take the solution for the position of the i-th transmitting antenna as an example. Simplify to only contain The function, that is: (14) in, This is a constant term. Since ignoring the constant term does not affect the optimal solution, the subproblem of optimizing the transmit antenna location becomes: (15) Assumption ,make ,constraint and It can be written as: (16) Therefore, the closed-form solution for the position of the transmitting antenna can be obtained as follows: (17) in, express and Distance between the two More distant values ​​and .

[0042] (4) Optimization of receiving antenna position; Similarly, the following will be based on the first... Each receiving antenna position Taking the solution as an example. Simplify to only contain The function, that is: (18) in, This is a constant term. Therefore, the subproblem for optimizing the receiver antenna position is: (19) constraint and It can be written as: (20) Thus, the optimization problem (19) can be approximated as: (21) The optimization problem (19) can be solved by iteratively solving the optimization problem (21) for its local optimum. And the optimization problem (9) can be solved by alternately iteratively solving the above four optimization sub-problems.

[0043] Based on the AO algorithm, the non-convex problem is solved, and finally the relevant simulation result graphs are obtained, as shown in Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 .

[0044] Figure 2 The convergence performance of the scheme proposed in the embodiment and the benchmark scheme is given. As can be seen from Figure 2 , with the increase of the number of iterations, the CRB of the scheme proposed in the embodiment gradually decreases and finally tends to a stable value. In addition, the CRB of the scheme proposed in the embodiment is smaller than that of the benchmark scheme, because the scheme proposed in the embodiment optimizes both the transmitting antenna position and the receiving antenna position, while the benchmark scheme optimizes at most one of them.

[0045] Figure 3 The CRB change of the base station under different base station transmit powers is given. With the increase of the base station transmit power, the CRB decreases. It can be seen that the CRB is inversely proportional to the base station transmit power. From the physical meaning, the increase of the base station transmit power means that the energy of the transmitted sensing signal is enhanced, and the signal power is relatively enhanced compared with the noise, so that the signal-to-noise ratio is increased, and since the CRB is negatively correlated with the signal-to-noise ratio, the increase of the signal-to-noise ratio will reduce the CRB.

[0046] Figure 4 The change relationship of the CRB with the number of receiving antennas is given. The simulation results show that with the increase of the number of receiving antennas, the CRB decreases. Because the increase of the number of receiving antennas brings higher spatial diversity gain and information utilization rate. On the one hand, more receiving antennas can collect signals from different spatial positions, thereby obtaining more independent channel samples, so that the noise decreases due to spatial randomness. On the other hand, the superposition of multiple antennas makes the effective signal can carry more parameter information such as angle and distance, thereby increasing the trace of the Fisher information matrix. Since the CRB is the inverse of the Fisher information matrix, the greater the amount of information in the matrix, the smaller the CRB value corresponding to the inverse matrix. It can be seen that the number of receiving antennas directly acts on the denominator of the CRB, and the increase of the number of receiving antennas promotes the increase of the denominator, thereby reducing the CRB.

[0047] Figure 5The relationship curve between CRB and the size of the movable area of the antenna is given. The results show that, with the increase of the size of the movable area of the antenna, the CRB decreases. Because the expansion of the spatial coverage of the antenna gives the antenna array more spatial degrees of freedom, it can more flexibly adjust the phase difference of the signal arriving at different antennas, so as to realize more accurate parameter estimation by using more abundant phase gradient information. At the same time, with the increase of the size of the movable area of the antenna, the system approaches the resolution limit, and the amount of change of the CRB gradually decreases.

[0048] The above merely illustrates the preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any changes or replacements easily thought of by those skilled in the art within the technical scope disclosed by the present application shall be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A collaborative method for edge sensing and task offloading based on movable antennas, applied to a base station equipped with a movable antenna array, characterized in that, include: The base station transmits downlink sensing signals through its movable transmitting antenna array, while simultaneously sensing targets in the environment and receiving uplink computing offload signals from multiple user devices. Construct a joint optimization problem with the core objective of optimizing perception performance while satisfying user communication needs and system constraints; The joint optimization problem is solved using an alternating optimization framework. By iteratively optimizing the receive beamforming, user transmit power, transmit antenna position, and receive antenna position, the coordination of sensing and multi-user computing offloading is achieved.

2. The method according to claim 1, characterized in that, The core objective of perception performance means that the objective function of the joint optimization problem is to minimize the Cramer-Rao lower bound for angle estimation of the objective.

3. The method according to claim 1, characterized in that, The requirements for meeting user communication needs and system constraints include: the uplink communication rate of each user device is not lower than its task requirement rate, the transmit power of each user device does not exceed its maximum power budget, and the position of each movable antenna is located within a preset movable area and meets the minimum spacing constraint.

4. The method according to claim 1, characterized in that, The iterative optimization of receive beamforming specifically involves solving a subproblem aimed at maximizing the signal-to-interference-plus-noise ratio (SINR) of each user equipment, given other variables, through a positive semidefinite relaxation and Gaussian randomization method, to obtain the optimal receive beamforming vector.

5. The method according to claim 2, characterized in that, The iterative optimization of user transmit power specifically involves: given other variables, using convex optimization tools to solve a power control subproblem that ensures the signal-to-interference-plus-noise ratio of each user equipment meets its required rate, with the objective of minimizing the Cramer-Rao lower bound.

6. The method according to claim 1, characterized in that, The iterative optimization of the transmitting antenna position is as follows: for each transmitting antenna, under the condition of satisfying its position constraints, the optimal position is obtained by solving a subproblem that aims to maximize the fraction related to the antenna position. Some antenna positions have closed-form solutions.

7. The method according to claim 1, characterized in that, The iterative optimization of the receiving antenna position is as follows: for each receiving antenna, under the condition of satisfying its position constraints, a subproblem aimed at maximizing the fraction related to the antenna position is solved by a one-dimensional search or iterative method to obtain its optimal position.

8. The method according to claim 1, characterized in that, The base station adopts a full-duplex working mode, and concurrently transmits the downlink sensing signal and receives the uplink computing offload signal on the same time and frequency resources.

9. The method according to claim 1, characterized in that, The movable antenna array includes a transmitting uniform linear array composed of several movable antennas and a receiving uniform linear array composed of several movable antennas.

10. The method according to claim 1, characterized in that, The target is at least one of the user equipment, or other intelligent agents within the communication coverage area of ​​the base station.

Images