Task unloading scheduling method based on RIS collaboration and mobile edge computing system

Abstract

The invention relates to a task unloading scheduling method based on RIS collaboration and a mobile edge computing system. The method provided by the invention comprises the following steps: acquiring cascade channel information and noise power between a user and an access point, and performing task unloading initialization by combining an optimal energy consumption objective function to generate an initial task unloading allocation iteration node and an optimal task unloading scheme; a lower bound branch algorithm based on user distribution is adopted to branch the current iteration node, the optimal task unloading scheme is updated, and the current iteration node is pruned according to the updated optimal task unloading scheme; then, judging whether the pruned iteration nodes meet iteration conditions or not: if yes, unloading and distributing the user calculation tasks according to the optimal task unloading scheme; and if not, continuing iteration. The RIS collaborative task unloading scheduling method provided by the invention has the advantage of greatly prolonging the endurance time of the user.

Description

Technical Field

[0001] The present invention relates to the field of resource scheduling in mobile edge computing, and particularly to a task offloading scheduling method, device, and mobile edge computing system based on RIS collaboration. Background Art

[0002] With the rapid development of mobile Internet and Internet of Things (IoT) devices, the demand for real-time performance, high throughput, and low power consumption in applications such as high-definition video, semantic recognition, virtual reality, and smart home is increasing day by day. However, the battery energy, computing resources, and processing speed of user terminals are limited, which easily leads to problems such as network latency, bandwidth bottlenecks, and concentrated computing loads, and it is difficult to meet the requirements of these applications. Accordingly, nowadays, mobile edge computing (MEC) is adopted, where computing servers are deployed at the network edge and connected through access points (APs) set by base stations, enabling user devices to offload some or all of their computing tasks through the access points of the base stations to the MEC. After the MEC completes the computing tasks, the task results are transmitted back to the user devices, so that the user devices can offload computing tasks at a closer physical distance, thereby reducing transmission latency, reducing core network load, and improving computing efficiency.

[0003] However, MEC is vulnerable to environmental factors such as building blockage, bad weather, and wireless signal fading, resulting in a decrease in the wireless link quality of MEC, and further causing a decrease in data transmission speed, task offloading failure, or an increase in system energy consumption. Based on this, the prior art adopts a reconfigurable intelligent surface (RIS) to optimize the wireless communication quality of MEC. The RIS is composed of a large number of tunable reflection units, and each unit dynamically adjusts the reflection phase through passive beamforming to optimize the propagation path of wireless signals. In particular, by intelligently regulating the wireless channel at the physical layer, the RIS can enhance signal strength, reduce multipath fading, and reduce interference, and significantly improve the communication performance of the MEC system. The connection between the user device and the base station or access point (AP) through the RIS can be referred to Figure 1 , Figure 1 as a simple schematic diagram of signal transmission using the RIS by the user device.

[0004] Among them, the RIS is usually used to assist wireless communication between user devices and access points (APs). Especially in the case of a non-line-of-sight (NLoS) channel, by dynamically adjusting the reflection phase, the channel gain is optimized, providing an additional propagation path, and significantly improving the reliability and transmission rate of the system.

[0005] However, due to the limited computing resources of the MEC access point (AP) and the limited maximum number of user devices that each AP can access, existing task allocation strategies are only based on static channel conditions or task priorities, resulting in low-computation tasks possibly over-occupying AP resources, while high-computation demand tasks are forced to be executed locally due to insufficient resources, leading to a significant increase in the energy consumption of user devices and severely reducing the battery life of user devices. Accordingly, there are problems in the prior art of insufficient coordination of task allocation-resource scheduling-RIS control in the MEC system, resulting in high energy consumption of user devices and low utilization rate of MEC resources. Summary of the Invention

[0006] Based on this, the object of the present invention is to provide a user task offloading scheduling method based on RIS collaboration.

[0007] A user task offloading scheduling method based on RIS collaboration includes the following steps:

[0008] S1: Obtain the cascaded channel information and noise power between the user and the access point;

[0009] S2: Perform task offloading initialization according to the cascaded channel information, noise power, and an optimal energy consumption objective function to generate an initial task offloading allocation iteration node and an optimal task offloading scheme; wherein, the initial task offloading allocation iteration node includes an initialized user allocation matrix, an initialized derivative scheme state, and an initialized minimum energy consumption estimate;

[0010] S3: Use a lower bound branch algorithm based on user allocation to branch the current task offloading allocation iteration node and update the optimal task offloading scheme to obtain an updated task offloading allocation iteration node and an optimal task offloading scheme;

[0011] S4: Perform a pruning operation on the task offloading allocation iteration node according to the target energy consumption value of the updated optimal task offloading scheme to obtain a refined task offloading allocation iteration node;

[0012] S5: Determine whether the refined task offloading allocation iteration node meets a branch and bound iteration condition: if so, perform offloading allocation of the user's computing tasks according to the current optimal task offloading scheme to complete the offloading scheduling of the user tasks; if not, continue to execute step S3.

[0013] The user task offloading and scheduling method based on RIS cooperation described in the present invention, compared with the prior art, through the multi-level optimization iteration of the task offloading and allocation iterative nodes, combined with the lower bound branch algorithm and the RIS phase optimization algorithm, realizes the joint optimization of the user allocation matrix and the RIS phase adjustment coefficient, thus effectively solving the problems of uneven task allocation, insufficient resource scheduling, and low cooperation rate of RIS control in the traditional MEC system. At the same time, it also significantly reduces the device energy consumption of users, improves the computing resource utilization rate of the MEC system, and enhances the task offloading efficiency of user devices. Accordingly, the present invention, through multi-level optimization iteration under limited computing resources, approaches the global optimal solution of the RIS phase adjustment coefficient and user allocation, enabling the RIS-assisted MEC system to maintain high efficiency and robustness in a dynamically changing network environment.

[0014] Further, the optimal energy consumption objective function E S (X OPT , θ OPT ) is specifically expressed as follows:

[0015]

[0016] In the formula, v OPT represents the target energy consumption value corresponding to the optimal task offloading scheme; X OPT represents the current optimal user allocation matrix, and its element x ij represents that the i-th user selects the j-th access point; θ OPT represents the phase adjustment coefficient vector of the current optimal RIS coordination unit;

[0017] is the energy consumption model of local computing. When x i0 = 1, it means that the i-th user selects to perform computing locally; k0 is the energy consumption related coefficient; α i is the computing complexity related coefficient; f i is the local computing frequency of the i-th user; len i represents the task length of the i-th user; M represents the total number of users;

[0018] is the remote computing model, β i is the uplink overhead coefficient, P T,i is the device transmission power of the i-th user; N represents the total number of access points; R ij (θ OPT ) represents the actual data transmission rate between the i-th user and the j-th access point under θ OPT , and its specific calculation is expressed as follows:

[0019]

[0020] where B W,j represents the bandwidth allocated by the j-th access point to the current user; is the channel gain vector between the i-th user and the j-th access point, and is Hermitian transposed; σ0 is the noise power;

[0021] The optimal energy consumption objective function includes several constraint conditions, and the several constraint conditions include user selection constraint conditions, access point maximum allocation constraint conditions, user allocation element constraint conditions, and RIS phase constraint conditions;

[0022] The specific representation of the user selection constraint condition is as follows:

[0023]

[0024] The specific representation of the access point maximum allocation constraint condition is as follows:

[0025]

[0026] where represents the total number of user connections of the j-th access point; L nk (j) represents the maximum number of connections of the j-th access point;

[0027] The specific representation of the user allocation element constraint condition is as follows:

[0028] x ij ∈{0,1},

[0029] The specific representation of the RIS phase constraint condition is as follows:

[0030] |θ l | = 1,

[0031] where θ l represents the phase adjustment coefficient of the l-th reconcilable unit of the RIS, that is, the phase angle; |θ l | = 1 means that the phase coefficient is a unit modulus complex number; L represents the total number of reconcilable units of the RIS, that is, the total number of reflection units.

[0032] For the task offloading process of users, the present invention simplifies the remote computing model, that is, it ignores the energy consumption of data backhaul and the MEC server computing part in the task offloading process; the reason is that the amount of data backhaul is usually small, and its energy consumption is negligible compared with the energy consumption of task offloading and computing, so it can be ignored; although there is computing energy consumption of the MEC server, since the access point (AP) is usually directly connected to the power supply, the energy consumption of the AP is provided by an external power supply and does not belong to the system energy consumption of the user side, so it can also be ignored;

[0033] Meanwhile, based on the joint optimization of the objective function, that is, by adjusting the user allocation matrix and the RIS phase adjustment coefficient, the system minimizes the target energy consumption value to reduce the energy consumption of user equipment, aiming to improve the battery life of user equipment. During the joint optimization process, the phase of the RIS reflection unit is mainly adjusted to reduce signal attenuation, interference, and multipath effects. The adjustment of the user allocation matrix focuses on the computing requirements of tasks and the computing resources of access points, avoiding low-computing tasks from occupying too many access point resources. Therefore, through the design and joint optimization of the above objective function, the task offloading efficiency and resource utilization rate of the system can be significantly improved, thereby reducing the computing burden of user equipment and effectively improving the battery life of user equipment.

[0034] Furthermore, the initialized minimum energy consumption estimation value lb0 is calculated using a minimum energy consumption estimation algorithm, which is specifically expressed as follows:

[0035]

[0036] In the formula, lb0 represents the initialized minimum energy consumption estimation value; represents the current user allocation matrix; Est(·) represents the minimum energy consumption estimation algorithm, and its specific steps are as follows:

[0037] S201. Generate a simulated user allocation index array and an initial energy consumption estimation value according to the size of the user allocation matrix of the current task offloading allocation iteration node;

[0038] S202. Generate a simulated user selection submatrix according to the current simulated user allocation index array;

[0039] S203. Concatenate the simulated user selection submatrix with the current user allocation matrix and determine whether it meets a maximum access point allocation constraint condition: If so, calculate the energy consumption estimation value of the concatenated user allocation matrix according to the user allocation energy consumption model and execute step S204; if not, execute step S205;

[0040] S204. Determine whether the energy consumption estimation value of the concatenated user allocation matrix is less than the current energy consumption estimation value: If so, update the current energy consumption estimation value and execute step S205; if not, directly execute step S205;

[0041] S205. Determine whether the current iteration number meets an energy consumption estimation condition: If so, use the current energy consumption estimation value as the minimum energy consumption estimation value of the current task offloading allocation iteration node; if not, update the current simulated user allocation index array and execute step S202.

[0042] The present invention calculates the minimum energy consumption estimation value corresponding to the iterative node, i.e., the node lower bound, through the minimum energy consumption estimation algorithm, so as to ensure that the RIS-based cooperative user task offloading and scheduling method described in the present invention can gradually approach the global optimal solution; at the same time, through the minimum energy consumption estimation algorithm, the system can effectively evaluate different task offloading schemes and resource allocations in each iteration process, ensure updates according to the scheme with the minimum energy consumption, and prevent falling into the local optimal solution.

[0043] Further, the generation of the simulated user selection sub-matrix is expressed as follows:

[0044] x ij ∈{0,1}

[0045] In the formula, X c is the simulated user selection sub-matrix, indicating that the size of the matrix is (M - n0) × (N + 1); n0 is the number of rows of the current user allocation matrix x ij represents each element x in the matrix ij , and the rule for the i-th user to select and access the j-th access point is expressed as follows:

[0046] j = select i ,

[0047] In the formula, represents traversing all users; select i is the simulated user allocation index corresponding to the i-th user, and its specific calculation is expressed as follows:

[0048]

[0049] In the formula, select i is the i-th element of the simulated user allocation index array; is used to represent traversing all elements in the simulated user allocation index array;

[0050] The judgment expression for the maximum allocation constraint condition of the access point is as follows:

[0051]

[0052] In the formula, is used to represent traversing all access points; is the energy consumption estimation value based on the spliced allocation matrix X, and its X is expressed as follows:

[0053]

[0054] is the upper bound of the data transmission rate between the \(i\)-th user and the \(j\)-th access point, and its specific calculation is as follows:

[0055]

[0056] The specific representation of updating the current simulated user allocation index array in step S205 is as follows:

[0057] select1 = select1 + 1

[0058] In the formula, select1 represents the first element in the current simulated user allocation index array.

[0059] The present invention effectively expands the exploitable space of the task offloading scheme by introducing a simulated user selection submatrix, thereby improving the accuracy of the minimum energy consumption estimation value calculated in step S204, ensuring that the minimum energy consumption estimation algorithm can more accurately evaluate the energy consumption of the task offloading scheme, and ensuring that each iteration approaches the optimal solution, avoiding premature abandonment of the current branch due to the local optimal solution, thus guaranteeing the optimization process of the RIS cooperative user task offloading scheduling method.

[0060] In addition, the access point maximum allocation constraint introduced in step S203 further enhances the compliance of the simulated user selection submatrix with the actual resource allocation limitations, that is, when excluding the non-essential exploration space, it conforms to the actual access situation of the current access point, ensuring the balance of task offloading and preventing unreasonable task offloading decisions.

[0061] Furthermore, the lower bound branch algorithm based on user allocation includes the following steps:

[0062] S301. Reset the branch of the user allocation matrix of the task offloading allocation iteration node to obtain the iteration node of the current branch;

[0063] S302. Determine whether the iteration node of the current branch is a leaf node: If not, use the minimum energy consumption estimation algorithm for calculation, update the minimum energy consumption estimation value of the iteration node of the current branch according to the calculation result, and execute step S305; if so, execute step S303;

[0064] S303. Use the RIS phase optimization algorithm to update the phase of the iteration node of the current branch to obtain the optimal task offloading scheme of the current branch and the corresponding target energy consumption value;

[0065] S304. Determine whether the target energy consumption value of the current branch is less than the target energy consumption value of the current optimal task offloading scheme: If so, update the current optimal task offloading scheme and execute step S305; if not, directly execute step S305;

[0066] S305. Determine whether the number of iterations of the current lower-bound branch algorithm assigned based on users is less than an iteration condition: If so, execute step S301; if not, complete the iteration of the lower-bound branch algorithm.

[0067] The present invention extends the explorability space of the lower-bound branch algorithm assigned based on users through branch reset, ensures that new branch nodes can be generated in each iteration, and provides an effective energy efficiency evaluation by calculating the minimum energy consumption estimation values of the parent node and the leaf node, thereby ensuring that each branch scheme advances towards the optimal solution and ensuring that the algorithm can update the optimal task offloading scheme in a timely manner.

[0068] In addition, the present invention calculates the phase angle of the corresponding RIS reflection unit according to the current user assignment matrix through the RIS phase optimization algorithm, thereby optimizing the signal transmission path, reducing signal attenuation, interference, and multipath effects, further improving the transmission efficiency of the system, and ensuring that the final signal quality reaches the optimal.

[0069] Further, the branch reset of step S301 is specifically represented as follows:

[0070]

[0071] wherein represents the user assignment matrix of the (N + 1)i'+k-th task offloading assignment iteration node, where i' represents the number of iterations of the current task offloading assignment iteration node; k is the number of iterations of the lower-bound branch algorithm assigned based on users;

[0072] represents the user assignment matrix of the i min -th transposed task offloading assignment iteration node, where i min represents the index of the iteration node with the minimum energy consumption estimation value;

[0073] represents the transposed identity matrix with size (N + 1), the element at the k-th position of which is 1 and other elements are 0;

[0074] The update of the minimum energy consumption estimation value of the iteration node of the current branch according to the calculation result in step S302 is specifically calculated as follows:

[0075]

[0076] wherein lb (N+1)i′+k represents the minimum energy consumption estimation value of the iteration node (N + 1)i'+k of the current branch; Est(·) is the minimum energy consumption estimation algorithm.

[0077] Accordingly, for the branch reset in the present invention, by splicing the user allocation matrix of the current iterative branch with the identity matrix, it is ensured that during each branch iteration, the minimum energy consumption estimation algorithm and the RIS phase optimization algorithm are used to jointly drive the lower bound branch algorithm to search for the global optimal energy efficiency solution in a wider solution space, thereby ensuring that the task offloading scheme can be effectively optimized and minimizing the energy consumption of user equipment to the greatest extent.

[0078] Furthermore, the specific calculation of the phase update in step S303 is expressed as:

[0079]

[0080] In the formula, θ (N+1)i′+k represents the phase adjustment coefficient vector of the RIS reconcilable unit corresponding to the optimal task offloading scheme of the most current branch; Opt RIS is the RIS phase update algorithm, and its specific steps are as follows:

[0081] S30301. Calculate the current phase update parameters according to the current phase adjustment coefficient vector of the RIS and the corresponding cascaded channel information; among them, the current phase update parameters include data transmission rate, signal-to-noise ratio gain, and channel gain correction factor;

[0082] The specific expression of the data transmission rate is:

[0083]

[0084] In the formula, R ij (θ (u) ) represents the data transmission rate corresponding to the phase adjustment coefficient vector of the u-th iteration; θ (u) is the phase adjustment coefficient vector of the u-th iteration. When u = 0, the modulus of each element of θ (0) is 1; where u represents the iteration number of the RIS phase update algorithm;

[0085] The specific expression of the signal-to-noise ratio gain is:

[0086]

[0087] In the formula, represents the signal-to-noise ratio gain of the u-th iteration; represents the matrix of the cascaded channel information between all users and all access points, and is subjected to Hermite transpose;

[0088] The specific expression of the channel gain correction factor is:

[0089]

[0090] wherein, represents the channel gain correction factor for the u-th iteration; represents the expected term of the signal gain; is the noise power term;

[0091] S30302. According to the current phase update parameter, use the coupled influence optimization algorithm to optimize the phases of all elements of the current phase adjustment coefficient vector of the RIS to obtain the optimized phase adjustment coefficient vector of the RIS;

[0092] S30303. Determine whether the current phase adjustment coefficient vector of the RIS converges: If not, continue to execute step S30301; if so, output the current phase adjustment coefficient vector of the RIS to complete the iteration of the lower bound branch algorithm based on user allocation;

[0093] wherein, the convergence is judged by using the target energy consumption value corresponding to the optimal energy consumption objective function, and its specific representation is as follows:

[0094]

[0095] wherein, represents the target energy consumption value corresponding to the phase adjustment coefficient vector for the (u + 1)-th iteration; δ is the convergence threshold of the RIS phase update algorithm.

[0096] Accordingly, the present invention uses the user allocation matrix after the foregoing branch reset as the input of the RIS phase update algorithm to find the phase angle of the RIS reflection unit that best matches the current user allocation matrix, so as to ensure that the calculated phase angle can greatly improve the signal transmission quality corresponding to the current user allocation matrix, thereby reducing the total energy consumption of the system.

[0097] Further, the phase optimization of the q-th element of the current phase adjustment coefficient vector of the RIS by the coupled influence optimization algorithm is represented as follows:

[0098]

[0099] wherein, represents the q-th element of the phase adjustment coefficient vector for the (t + 1)-th iteration in the coupled influence optimization algorithm, and its initial value φ (0) is the current phase adjustment coefficient vector θ (u) of the RIS; is the optimization direction vector of the q-th phase for the t-th iteration in the coupled influence optimization algorithm, and its specific representation is as follows:

[0100]

[0101] wherein, Gain effect for representing signal-to-noise ratio; Represents the coupling influence term, the coupling influence factor Γ for the p-th element and the q-th element of the phase adjustment coefficient vector pq , where Γ pq Is specifically represented as follows:

[0102]

[0103] In the formula, Represents the square of the channel gain correction factor; Represents the outer product of the channel gain matrix; where the superscript (*) represents the complex conjugate; the iteration condition of the coupling influence optimization algorithm is:

[0104] Judge whether ||φ (t+1) -φ (t) || is less than a threshold value ε: if so, take the phase adjustment coefficient vector δ of the RIS in the previous iteration (t) As the phase adjustment coefficient vector of the optimized RIS, that is, θ (u+1) =φ (t) , complete the phase optimization; if not, take φ (t+1) As the phase adjustment coefficient vector of the RIS in the current coupling influence optimization algorithm, continue to use the coupling influence optimization algorithm to optimize the phase of all elements; where the threshold value ε ∈ [0,1].

[0105] In the iterative process of the RIS phase update algorithm of the present invention, by combining the coupling influence optimization algorithm, the mutual influence between each reflecting unit and the currently adjusted reflecting unit is considered, that is, the phase change of each reflecting unit not only affects the signal transmission of its associated access point, but also has a chain reaction on other access points and users, thereby effectively avoiding the multi-path effect and signal interference ignored when optimizing the phase of a single reflecting unit, and ensuring that the global optimization conforms to the actual physical situation.

[0106] In particular, the present invention adopts the channel gain correction factor and the gain effect of signal-to-noise ratio (SNR) as the core calculation parameters for phase update, which are used to adjust the phase of each reflecting unit in real time to reduce signal attenuation, interference and multi-path effect, and significantly improve the signal quality and transmission rate of the system; in addition, the introduction of the coupling influence term further improves the coordination between the currently adjusted reflecting unit and other reflecting units, ensuring that the phase adjustment not only optimizes the signal of the current user, but also considers the overall performance of the entire system, achieving the optimal balance between system performance and energy efficiency.

[0107] A user task offloading scheduling device based on RIS cooperation, comprising a cascaded channel information acquisition unit for users, a task offloading scheme initialization unit, a task offloading scheme branching unit, a task offloading scheme pruning unit, and a task offloading scheme iterative judgment unit;

[0108] The cascaded channel information acquisition unit for users is used to acquire the cascaded channel information and noise power between the user and the access point;

[0109] The task offloading scheme initialization unit is used to perform task offloading initialization according to the cascaded channel information, noise power, and an optimal energy consumption objective function, and generate an initial task offloading allocation iterative node and an optimal task offloading scheme; wherein, the initial task offloading allocation iterative node includes an initialized user allocation matrix, an initialized derivative scheme state, and an initialized minimum energy consumption estimate;

[0110] The task offloading scheme branching unit is used to branch the current task offloading allocation iterative node by using a lower bound branching algorithm based on user allocation, and update the optimal task offloading scheme to obtain an updated task offloading allocation iterative node and an optimal task offloading scheme;

[0111] The task offloading scheme pruning unit is used to perform a pruning operation on the task offloading allocation iterative node according to the target energy consumption value of the updated optimal task offloading scheme to obtain a refined task offloading allocation iterative node;

[0112] The task offloading scheme iterative judgment unit is used to judge whether the refined task offloading allocation iterative node meets a branch and bound iteration condition: if so, the computing tasks of the user are offloaded and allocated according to the current optimal task offloading scheme to complete the offloading scheduling of the user tasks; if not, the task offloading scheme branching unit is called again.

[0113] A mobile edge computing system based on reconfigurable intelligent surface, comprising a reconfigurable surface, namely RIS, a base station, and a RIS cooperation-based task offloading scheduling device communicatively connected to the reconfigurable intelligent surface and the base station;

[0114] The reconfigurable surface is provided with a plurality of tunable units for receiving computing task offloading requests from user equipment, and adjusting the reflection phase through the tunable units to reflect and transmit the signal information to the base station; wherein, the user equipment is used to generate computing tasks and send the computing tasks and corresponding computing requirements to the reconfigurable surface;

[0115] The base station is provided with an access point to the edge computing server, which is used to calculate the cascaded channel information and noise power between the current user and the access point according to the computing task and the corresponding computing requirements, and input them into the RIS collaboration-based task offloading and scheduling device;

[0116] Among them, the RIS collaboration-based task offloading and scheduling device is used to execute the RIS collaboration-based task offloading and scheduling method described above.

[0117] For better understanding and implementation, the present invention will be described in detail below with reference to the accompanying drawings. Description of the Drawings

[0118] Figure 1 It is a simple schematic diagram of the user equipment using RIS signal transmission;

[0119] Figure 2 It is a simple schematic diagram of the signal transmission of the mobile edge computing system based on the reconfigurable intelligent surface;

[0120] Figure 3 It is a simple structural schematic diagram of the RIS collaboration-based task offloading and scheduling device described in the present invention;

[0121] Figure 4 It is a simple flow schematic diagram of the RIS collaboration-based task offloading and scheduling method described in the present invention;

[0122] Figure 5 It is a comparison schematic diagram of the joint RIS phase optimization and other phase selection methods of the present invention;

[0123] Figure 6 It is a simple reference schematic diagram of the pseudocode flow of the minimum energy consumption estimation algorithm described in the present invention;

[0124] Figure 7 It is a simple reference schematic diagram of the pseudocode flow of the RIS phase update algorithm described in the present invention;

[0125] Figure 8 It is a simple reference schematic diagram of the pseudocode flow of the coupling effect optimization algorithm described in the present invention. Detailed Embodiment

[0126] To solve the problem of insufficient coordination of task offloading - resource scheduling - RIS control in the existing MEC system, the present invention obtains the cascaded channel information and noise power between the user and the access point (AP), and combines the energy consumption model with the optimal energy consumption objective function to initialize task offloading, so as to generate an initial task offloading scheme and the target energy consumption value; and uses a lower - bound branch algorithm based on user allocation to branch the current task offloading scheme to generate several task offloading schemes to be allocated; then, updates the current target energy consumption value according to the several task offloading schemes to be allocated, and prunes the several task offloading schemes to be allocated based on the updated target energy consumption value to obtain a refined task offloading scheme; finally, determines whether the refined task offloading scheme meets the branch - and - bound iteration condition: if so, performs offloading allocation for the current user according to the current task offloading scheme; if not, then uses the lower - bound branch algorithm based on user allocation to perform a branching operation on the current refined task offloading scheme and continue the iteration.

[0127] Accordingly, the present invention optimizes the task offloading allocation scheme based on the branch - and - bound algorithm, and ensures that the system can achieve the optimal energy consumption in different network states by dynamically adjusting task offloading and RIS configuration, so as to minimize the total energy consumption of the user's task offloading, thereby significantly extending the user's battery life, and further improving the communication resource utilization efficiency of the MEC system and RIS control, and significantly enhancing the resource scheduling ability of the RIS - assisted MEC system.

[0128] Based on the above design, the present invention proposes a user task offloading scheduling method based on RIS cooperation, and proposes a user task offloading scheduling device based on RIS cooperation based on this method.

[0129] Please refer to Figure 2 , Figure 2 which is a simple schematic diagram of signal transmission for a mobile edge computing system based on a reconfigurable intelligent surface.

[0130] A mobile edge computing system based on a reconfigurable intelligent surface includes a user equipment 100, a reconfigurable intelligent surface (RIS) 101, a base station 102, and the task offloading scheduling device based on RIS cooperation of the present invention connected to the reconfigurable intelligent surface 101 and the base station 102.

[0131] The user equipment 100 is used to generate computing tasks, for example, data processing, video streaming, Internet of Things sensor data, etc., and send the computing tasks and the corresponding computing requirements to the reconfigurable intelligent surface 101.

[0132] The reconfigurable metasurface 101 is provided with a plurality of tunable units, which are used to receive the computing task offloading requests from the user equipment, and adjust the reflection phase through the tunable units to reflect and transmit the signal information to the base station 102.

[0133] The base station 102 is provided with an access point to the edge computing server, which is used to calculate the cascaded channel information and noise power between the current user and the access point (AP) according to the computing task and the corresponding computing requirements, and input them into the RIS collaborative-based task offloading scheduling device of the present invention.

[0134] Please refer to Figure 3 and Figure 4 , Figure 3 which is a schematic diagram of the simple structure of the RIS collaborative-based task offloading scheduling device of the present invention, Figure 4 and which is a schematic diagram of the simple process of the RIS collaborative-based task offloading scheduling method of the present invention.

[0135] The RIS collaborative-based task offloading scheduling device includes a cascaded channel information acquisition unit 1 for users, a task offloading scheme initialization unit 2, a task offloading scheme branching unit 3, a task offloading scheme pruning unit 4, and a task offloading scheme iterative judgment unit 5.

[0136] The cascaded channel information acquisition unit 1 for users is used to execute step S1: acquire the cascaded channel information and noise power between the user and the access point.

[0137] Specifically, the cascaded channel information between the user and the access point is specifically expressed as follows:

[0138]

[0139] In the formula, h ij represents the cascaded channel information between the i-th user and the j-th edge computing access point (AP), and i ∈ [1, M], h ∈ [1, N], where M and N respectively represent the total number of users and the total number of access points (AP); h a,il represents the channel information from the i-th user to the l-th tunable unit of the RIS; h b,lj represents the channel information from the l-th tunable unit of the RIS to the j-th edge computing access point of the base station; θ l represents the phase adjustment coefficient of the l-th tunable unit of the RIS, that is, the phase angle, which satisfies |θ i | = 1, that is, the phase coefficient is a unit modulus complex number, representing the phase adjustment of the RIS to the signal; L represents the total number of tunable units of the RIS, that is, the total number of reflection units;

[0140] Among them, for the convenience of processing and calculation, the present invention defines the cascaded channel information as a vector representation, then h ij is further defined as:

[0141]

[0142]

[0143] In the formula, represents the channel gain vector between the i-th user and the j-th edge computing access point (AP), that is, the cascaded channel information, and its specific size is That is, a complex vector of size 1×L, which is used to represent the sum of the contributions of the L tunable units of the RIS to the signal. The superscript H represents the Hermite transpose; represents the phase adjustment coefficient vector of all the RIS tunable units, that is, the phase adjustment amount (phase angle) of each reflection unit of the RIS. The superscript T is used to represent the ordinary transpose; The symbol represents the definition.

[0144] The task offloading scheme initialization unit 2 is used to execute step S2: perform task offloading initialization according to the cascaded channel information, noise power, and an optimal energy consumption objective function, and generate an initial task offloading allocation iteration node and an optimal task offloading scheme.

[0145] Specifically, the initial task offloading allocation iteration node includes an initialized user allocation matrix, an initialized derivative scheme state, and an initialized minimum energy consumption estimation value;

[0146] The specific representation of the initialized user allocation matrix is as follows:

[0147]

[0148] In the formula, represents the user allocation matrix of the 0th task offloading allocation iteration node, and its specific value is an empty set represents initialization; the specific representation of the initialized derivative scheme state is as follows:

[0149] status0 = 1

[0150] In the formula, status0 represents the derivative scheme state of the 0th task offloading allocation iteration node, and its value of 1 indicates that it can be derived. If it is 0, it means it cannot be derived; the initialized minimum energy consumption estimation value, that is, the node lower bound, is calculated by using a minimum energy consumption estimation algorithm, and the specific representation is as follows:

[0151]

[0152] Wherein, lb0 represents the initialized minimum energy consumption estimation value; represents the current user allocation matrix, and its initial value corresponds to the cascaded channel information; σ0 is the noise power; Est(·) represents the minimum energy consumption estimation algorithm, and its specific steps are as follows:

[0153] Please refer to Figure 6 , Figure 6 which is a schematic diagram of the pseudo-code simple process reference of the minimum energy consumption estimation algorithm described in the present invention;

[0154] S201. Generate a simulated user allocation index array and an initial energy consumption estimation value according to the size of the user allocation matrix of the iterative node for the current task offloading allocation.

[0155] Specifically, the simulated user allocation index array select is specifically represented as follows:

[0156]

[0157] Wherein, {0} represents an all-zero array, which is used to simulate the situation where users have not been assigned tasks. M is the total number of users, and n0 is the number of rows of the current user allocation matrix;

[0158] Among them, the initial energy consumption estimation value t min defaults to 1000, and its initial energy consumption estimation value is set to a relatively large value to ensure that the subsequent iterative process can proceed effectively, thereby preventing the loop process from being interfered by too small values.

[0159] S202. Generate a simulated user selection sub-matrix according to the current simulated user allocation index array.

[0160] Specifically, the generation of the simulated user selection sub-matrix is represented as follows:

[0161] x ij ∈{0,1}

[0162] Wherein, X c is the simulated user selection sub-matrix, represents that the size of the matrix is (M - n0)×(N + 1), where (M - n0) is the number of rows and (N + 1) is the number of columns, and N represents the total number of access points; x ij represents each element x in the matrix ij can only take the value of 0 or 1, which is used to represent whether the i-th user selects the j-th access point. If x ij = 1, it means that the i-th user selects the j-th access point, and the rule for selecting the j-th access point is represented as follows:

[0163] j = select i ,

[0164] where i = [1, M] represents all rows of the simulated user selection sub - matrix, that is, traversing all users; select i assigns an index to the simulated user corresponding to the i - th user, and its specific calculation is as follows:

[0165]

[0166] where select i is the i - th element of the index array for the simulated user, used to represent the access point number selected by the user; is used to represent traversing all elements in the index array of the simulated user.

[0167] S203. Concatenate the simulated user selection sub - matrix with the current user allocation matrix, and determine whether the maximum allocation constraint condition of an access point is satisfied: If so, calculate the energy consumption estimate value of the concatenated user allocation matrix according to the user allocation energy consumption model, and execute step S204; if not, execute step S205.

[0168] Specifically, the concatenated user allocation matrix is expressed as follows:

[0169]

[0170] where is the current user allocation matrix, with a size of M×(N + 1); X c is the simulated user selection sub - matrix, with a size of (M - n0)×(N + 1); X is the concatenated user allocation matrix, with a size of M×(N + 1).

[0171] The specific judgment rule for adopting the maximum allocation constraint condition of the access point is expressed as follows:

[0172]

[0173] where represents the total number of user connections of the j - th access point; L nk (j) represents the maximum number of connections of the j - th access point; is used to represent traversing all access points.

[0174] When the total number of user connections of all access points satisfies the maximum allocation constraint condition of the access point, calculate the energy consumption estimate value of the concatenated allocation matrix, and its specific calculation is as follows:

[0175]

[0176] In the formula, is the energy consumption estimation value based on the spliced allocation matrix, that is, the user allocation energy consumption model. Its input is the spliced allocation matrix, and the output is the energy consumption estimation value;

[0177] is the energy consumption model of local computing. When x i0 = 1, it means that the i-th user chooses to perform computing locally; k0 is the energy consumption correlation coefficient; α i is the computing complexity correlation coefficient; L i is the local computing frequency of the i-th user; len i represents the task length of the i-th user, and its unit is bit;

[0178] is the energy consumption model of remote computing, and β i is the uplink overhead coefficient, and its value β i ≥ 1, which is used to represent that there is overhead in the uplink; P t,i is the device transmission power of the i-th user; is the upper bound of the data transmission rate between the i-th user and the j-th access point, that is, the Shannon Capacity Formula. Its specific calculation is as follows:

[0179]

[0180] In the formula, B W,j represents the bandwidth allocated by the j-th access point to the current user; is the channel gain vector between the i-th user and the j-th access point; σ0 is the noise power.

[0181] S204. Determine whether the energy consumption estimation value of the spliced user allocation matrix is less than the current energy consumption estimation value: If so, update the current energy consumption estimation value and execute the next step; if not, directly execute the next step.

[0182] Specifically, the judgment expression for updating the current energy consumption estimation value is as follows:

[0183]

[0184] S205. Determine whether the current iteration number meets an energy consumption estimation condition: If so, use the current energy consumption estimation value as the minimum energy consumption estimation value of the current task offloading allocation iteration node; if not, update the current simulated user allocation index array and execute step S202.

[0185] Specifically, the specific judgment expression of the energy consumption estimation condition is as follows:

[0186]

[0187] Where count is the number of iterations of the current minimum energy consumption estimation algorithm, and its initial value is 0;

[0188] If the judgment of the energy consumption estimation condition is yes, output the current energy consumption estimation value and end the iteration;

[0189] If the judgment of the energy consumption estimation condition is no, update the current simulated user allocation index array through the following formula and continue to execute step S202 for iteration:

[0190] select1 = select1 + 1

[0191] Where select1 represents the first element in the current simulated user allocation index array, which is used to simulate the first user to select the next access point, promote the update of the simulated user selection submatrix, and thus explore more simulated user selection submatrices.

[0192] At the same time, use a feasible solution of the optimal energy consumption objective function as the initial optimal task offloading scheme {X OPT , θ OPT}, and calculate the corresponding target energy consumption value, which is specifically expressed as follows:

[0193]

[0194] Where v OPT represents the target energy consumption value corresponding to the optimal task offloading scheme, and is obtained by calculating the optimal energy consumption objective function E S (X OPT , θ OPT );

[0195] X OPT represents the current optimal user allocation matrix, and its initial value can be obtained by random allocation or by a heuristic algorithm (such as a greedy algorithm based on channel quality). Its element x ij represents the i-th user selecting the j-th access point;

[0196] θ OPT represents the current phase adjustment coefficient vector of the optimal RIS coordination unit, and its initial value can be obtained by random allocation or by a heuristic algorithm;

[0197] R ij (θ OPT ) is the actual data transmission rate between the i-th user and the j-th access point under θ OPT , and its specific calculation is expressed as follows:

[0198]

[0199] Among them, the optimal energy consumption objective function \(E\) S (X OPT , θ OPT ) has several constraint conditions; the several constraint conditions include user selection constraint conditions, maximum access point allocation constraint conditions, user allocation element constraint conditions, and RIS phase constraint conditions;

[0200] The user selection constraint condition is used to ensure that each user can only select local computing or select a certain access point for task offloading, and its specific representation is as follows:

[0201]

[0202] In the formula, \(x\) ij is an element in the user allocation matrix, indicating whether the \(i\)-th user has selected the \(j\)-th access point. When \(j = 0\), it is considered that the user selects local computing;

[0203] The maximum access point allocation constraint condition is used to ensure that the total number of user connections of each access point does not exceed its maximum connection number, and its specific representation is as follows:

[0204]

[0205] In the formula, \(L\) nk (j) represents the maximum connection number of the \(j\)-th access point;

[0206] The user allocation element constraint condition is used to ensure that the elements of the user allocation matrix are binary values, that is, yes or no, to prevent overflow, and its specific representation is as follows:

[0207] \(x\) ij ∈ {0, 1},

[0208] The RIS phase constraint condition is used to ensure that the absolute value of the phase adjustment coefficient of each adjustable unit in the RIS is 1, so as to conform to the physical properties of RIS phase adjustment, and its specific representation is as follows:

[0209] |\(θ\) l | = 1,

[0210] The task offloading scheme branch unit 3 is used to execute step S3: branch the current task offloading allocation iteration node by using the lower bound branch algorithm based on user allocation, and update the optimal task offloading scheme to obtain the updated task offloading allocation iteration node and the optimal task offloading scheme.

[0211] Specifically, the lower bound branch algorithm based on user allocation includes the following steps:

[0212] S301. Reset the branches of the user allocation matrix of the task offloading allocation iteration node to obtain the iteration node of the current branch.

[0213] Specifically, the branch reset is specifically expressed as follows:

[0214]

[0215] In the formula, represents the user allocation matrix of the (N + 1)i'+k-th task offloading allocation iteration node, where i' represents the number of iterations of the current task offloading allocation iteration node, that is, the number of loops of repeatedly executing the task offloading scheme branch unit 3 through step S5; k is the number of iterations of the lower bound branch algorithm based on user allocation;

[0216] represents the i min -th transposed user allocation matrix of the task offloading allocation iteration node, that is, the allocation matrix obtained from the current optimal solution node, where i min represents the index of the iteration node with the minimum energy consumption estimation value (lowest lower bound);

[0217] represents the transposed identity matrix with size (N + 1), the element at the k-th position of which is 1 and other elements are 0, and is used to represent adding a new access point allocation on the basis of the current iteration node;

[0218] Accordingly, through branch reset and updating the access point allocation on the basis of the existing optimal solution, the branch algorithm generates a new task offloading scheme, thereby promoting the optimization process, further realizing branch exploration, effectively reducing the search space and gradually approaching the minimum energy consumption solution.

[0219] S302. Determine whether the iteration node of the current branch is a leaf node: if not, calculate using the minimum energy consumption estimation algorithm, update the minimum energy consumption estimation value of the iteration node of the current branch according to the calculation result, and execute step S305; if so, execute step S303.

[0220] Specifically, the specific expression of the minimum energy consumption estimation value of the iteration node of the current branch is as follows:

[0221]

[0222] In the formula, lb (N+1)i′+k represents the minimum energy consumption estimation value of the iteration node (N + 1)i'+k of the current branch.

[0223] S303. Use the RIS phase optimization algorithm to update the phase of the iterative nodes in the current branch, and obtain the minimum energy consumption estimation value and the optimal task offloading scheme of the iterative nodes in the current branch.

[0224] Specifically, the minimum energy consumption estimation value of the iterative nodes in the current branch is expressed as:

[0225]

[0226] In the formula, θ (N+1)i′+k represents the phase adjustment coefficient vector of the RIS co - ordination unit corresponding to the optimal task offloading scheme of the current branch, which is obtained through the phase update, and its specific calculation is expressed as:

[0227]

[0228] In the formula, Opt ROS is the RIS phase update algorithm, which is used to update the RIS phase according to the user allocation matrix of the iterative nodes in the current branch. The specific steps are as follows:

[0229] Please also refer to Figure 7 , Figure 7 , which is a schematic diagram of the simplified pseudo - code process of the RIS phase update algorithm described in the present invention;

[0230] S30301. Calculate the current phase update parameters according to the phase adjustment coefficient vector of the current RIS and the corresponding cascaded channel information.

[0231] Specifically, the current phase update parameters include data transmission rate, signal - to - noise ratio gain, and channel gain correction factor;

[0232] The data transmission rate is used to measure the communication performance between the i - th user and the j - th access point in the u - th iteration, and its specific expression is:

[0233]

[0234] In the formula, R ij (θ (u) ) represents the data transmission rate corresponding to the phase adjustment coefficient vector of the u - th iteration; θ (u) is the phase adjustment coefficient vector of the u - th iteration. When u = 0, the modulus of each element of θ (0) is 1, which is used to constrain the randomly initialized phase adjustment coefficient vector; where u represents the number of iterations of the RIS phase update algorithm;

[0235] The signal - to - noise ratio gain is used to measure the enhancement effect of the RIS on the signal, and its specific expression is:

[0236]

[0237] In the formula, represents the signal-to-noise ratio gain of the \(u\)th iteration; represents the matrix of the cascaded channel information between all users and all access points, and is used to comprehensively represent the channel gain between all users and access points through Hermite transpose;

[0238] The channel gain correction factor is used to guide the optimization direction of the phase during the iterative update process, and is specifically expressed as:

[0239]

[0240] In the formula, represents the channel gain correction factor of the \(u\)th iteration; represents the expected term of the signal gain, which is used to represent the amplification factor of the channel gain on the signal; is the noise power term, which is used to ensure the stability of the signal energy calculation and avoid unreasonable signal amplification.

[0241] S30302. According to the current phase update parameter, use the coupled influence optimization algorithm to optimize the phases of all elements of the current phase adjustment coefficient vector of the RIS, and obtain the optimized phase adjustment coefficient vector of the RIS.

[0242] Please also refer to Figure 8 , Figure 8 which is a schematic diagram of the simplified pseudo-code process of the coupled influence optimization algorithm described in the present invention;

[0243] Specifically, for the \(q\)th element of the current phase adjustment coefficient vector of the RIS, its phase optimization in the coupled influence optimization algorithm at the \((t + 1)\)th iteration is expressed as follows:

[0244]

[0245] In the formula, represents the \(q\)th element of the phase adjustment coefficient vector at the \((t + 1)\)th iteration in the coupled influence optimization algorithm, and its initial value \(\varphi\) (0) is the current phase adjustment coefficient vector \(\theta\) of the RIS (u) ; is the optimization direction vector of the \(q\)th phase at the \(t\)th iteration in the coupled influence optimization algorithm, which is used to determine the optimization direction of the current RIS reflection unit, and adjusts its phase angle according to the optimization direction to maximize the signal gain, and is specifically expressed as follows:

[0246]

[0247] In the formula, The gain effect for representing the signal-to-noise ratio, that is, enhancing the signal quality by increasing the signal-to-noise ratio;

[0248] Denote the coupling influence term, which is used to reflect the interaction with other reflection units p when adjusting the q-th element of the phase adjustment coefficient vector of the RIS, so as to ensure that the optimization process is not limited to the local optimum, and for the coupling influence factor Γ between the p-th element and the q-th element of the phase adjustment coefficient vector pq , where Γ pq The specific representation is as follows:

[0249]

[0250] In the formula, Denote the square of the channel gain correction factor, which is used to quantify the enhancement effect of the signal; Denote the outer product of the channel gain matrix, which is used to reflect the channel quality from all users to all access points, so as to characterize the interactive channel influence between users and access points;

[0251] Among them, the superscript (*) represents the complex conjugate, which is used to ensure that the phase adjustment is within the unit modulus range and conforms to the physical constraints.

[0252] Next, the iteration condition of the coupling influence optimization algorithm is:

[0253] Judge whether ||φ (t+1) - φ (t) || is less than a threshold ε: if so, take the phase adjustment coefficient vector φ of the RIS in the previous round of iteration (t) as the phase adjustment coefficient vector of the optimized RIS, that is, θ (u+1) = φ (t) , and complete the phase optimization; if not, take φ (t+1) as the phase adjustment coefficient vector of the RIS in the current coupling influence optimization algorithm, and continue to use the coupling influence optimization algorithm to optimize the phases of all elements, which can be specifically expressed as:

[0254]

[0255] Among them, the threshold ε ∈ [0, 1], and the default setting is 0.01, which is used to ensure that there is no excessive fluctuation in the phase optimization process; ‖·‖ represents the norm of the vector, and the L2 norm is default adopted.

[0256] Accordingly, through the coupled influence optimization algorithm, in the process of RIS phase optimization, the present invention adopts a nested iteration method to consider the mutual influence between reflection units, thereby effectively avoiding the performance degradation of the RIS after phase optimization caused by local optimization, ensuring the global optimization of the system, and further improving the system energy efficiency and task offloading efficiency by maximizing the global signal gain.

[0257] S30303. Determine whether the phase adjustment coefficient vector of the current RIS converges: If not, continue to execute step S30301; if so, output the phase adjustment coefficient vector of the current RIS, and complete the iteration of the RIS phase update algorithm.

[0258] Specifically, for the determination of convergence, the target energy consumption value corresponding to the optimal energy consumption objective function is used, and its specific representation is as follows:

[0259]

[0260] In the formula, represents the target energy consumption value corresponding to the phase adjustment coefficient vector in the (u + 1)-th iteration; δ is the convergence threshold of the RIS phase update algorithm, and δ ∈ [0, 1], which is used to determine whether the phase adjustment coefficient vector of the current RIS phase update algorithm converges.

[0261] In addition, the vector variation method or other conditions can also be used to determine whether the current RIS phase update algorithm converges. Therefore, the present invention does not specifically limit the selection of its convergence conditions here.

[0262] S304. Whether the minimum energy consumption estimation value of the iterative node of the current branch is less than the target energy consumption value of the current optimal task offloading scheme: If so, update the current optimal task offloading scheme and execute step S305; if not, directly execute step S305.

[0263] Specifically, the update of the current optimal task offloading scheme is specifically represented as:

[0264] lflb (N+1)i ′ +k <v OPT then

[0265] v OPT =lb (N+1)i ′ +k

[0266]

[0267] θ OPT =θ (N+1)i ′ +k

[0268] Accordingly, during the iteration process of the lower bound branch algorithm, when using the coupled influence optimization algorithm instead of the minimum energy consumption estimation algorithm, it can be directly determined whether the minimum energy consumption estimation value obtained by the coupled influence optimization algorithm is less than the target energy consumption value of the current optimal task offloading scheme. If so, the current optimal task offloading scheme is directly updated to ensure that the scheme with the lowest energy consumption can be selected in each iteration, promoting the optimization process to gradually approach the global optimal solution.

[0269] S305. Determine whether the iteration count of the current lower bound branch algorithm based on user allocation is less than an iteration condition: If so, execute step S301; if not, complete the iteration of the lower bound branch algorithm.

[0270] Among them, the iteration condition is k < (N + 1). By taking the number of access points as the maximum iteration count of the lower bound branch algorithm, it is ensured that the algorithm can branch according to all access points to fully explore the corresponding task offloading schemes, thereby finding the optimal resource allocation scheme.

[0271] The task offloading scheme pruning unit 4 is used to execute step S4: According to the target energy consumption value of the updated optimal task offloading scheme, perform pruning operations on the task offloading allocation iteration nodes to obtain refined task offloading allocation iteration nodes.

[0272] Specifically, the specific representation of the pruning operation is as follows:

[0273]

[0274] In the formula, status i represents the derivative scheme status of the i-th task offloading allocation iteration node; lb i represents the minimum energy consumption estimation value of the i-th task offloading allocation iteration node, and indicates traversing n task offloading allocation iteration nodes, where n represents the total number of current task allocation iteration nodes.

[0275] If there exists lb i ≥v OPT then it means that the minimum energy consumption estimation value of the current node is greater than the existing optimal solution, indicating that this node cannot generate a better solution. Therefore, the derivative scheme status status i can be set to 0 so that its iteration node does not continue to participate in the branch iteration.

[0276] The task offloading scheme iteration judgment unit 5 is used to execute step S5: Determine whether the refined task offloading allocation iteration nodes meet a branch and bound iteration condition: If so, perform offloading allocation of the user's computing tasks according to the current optimal task offloading scheme to complete the offloading scheduling of the user tasks; if not, continue to call the task offloading scheme branching unit 3.

[0277] Specifically, the specific judgment representation of the branch and bound iteration condition is as follows:

[0278]

[0279] Among them, if the status of all derivative solutions i is 0, it can be considered that all branches have been traversed, and no new better solution is generated. The algorithm can directly perform offloading scheduling according to the existing optimal task offloading solution {X OPT , θ OPT}, that is, adjust the adjustable units of the reconfigurable metasurface according to the phase adjustment coefficient vector θ OPT of the current optimal RIS reconcilable unit, and allocate access points to user equipment according to the current optimal user allocation matrix X OPT to complete the task offloading scheduling of user equipment.

[0280] Compared with the prior art, the present invention ensures that the user allocation and the adjustment of RIS reflection units can make the task offloading allocation reach the global optimal solution through multi-layer iterative optimization. In particular, a lower bound branch algorithm based on user allocation is adopted to effectively expand the search space through branch reset and minimum energy consumption estimation algorithms, ensure the full exploration of the user allocation matrix, and avoid falling into local optimal solutions, thereby significantly improving the task offloading efficiency and system resource utilization rate.

[0281] In addition, in the RIS phase optimization algorithm of the present invention, the phase of the reflection units of RIS is optimized by using the user allocation matrix corresponding to the branch reset of the iterative node, effectively improving the signal quality, reducing the interference and attenuation between reflection units, thereby significantly improving the communication performance of the system and further reducing the system energy consumption;

[0282] At the same time, the present invention also ensures that the adjusted reflection units avoid multipath effects during the RIS phase optimization process through the coupling influence optimization algorithm, thereby significantly enhancing the signal quality control ability during the RIS-assisted task offloading process, ensuring the efficiency and stability of task offloading, and ultimately effectively improving the battery life of user equipment.

[0283] Please refer to Figure 5 , Figure 5 which shows a comparison schematic diagram of the joint RIS phase optimization of the present invention and other phase selection methods. Among them, local computing means that the user only uses local devices for computing and does not offload tasks to a remote server; fixed phase means that the algorithm only optimizes the user allocation matrix X, and the phase coefficient θ = 1 16×1 ; random selection means a feasible solution of the optimal energy consumption objective function under certain constraint conditions;

[0284] Compared with the solutions that adopt local computing, fixed phase selection, and random phase selection, the method of the present invention not only significantly reduces the total energy consumption compared to other methods, but also has a lower energy consumption growth rate when the number of users increases. That is to say, as the number of users increases, the method described in the present invention can effectively control the growth of the system energy consumption, thereby significantly reducing the energy consumption of user devices, improving the computing resource utilization rate of the MEC system, and effectively extending the battery life of user devices.

[0285] Based on the same inventive concept, the present application also provides an electronic device, which can be a server, a desktop computing device, or a mobile computing device (such as a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.) and other terminal devices. The device includes one or more processors and a memory, where the processor is used to execute a program to implement the user task offloading and scheduling method based on RIS collaboration in the embodiments of the present invention; the memory is used to store a computer program executable by the processor.

[0286] Based on the same inventive concept, the present application also provides a computer-readable storage medium, corresponding to the embodiments of the above-mentioned user task offloading and scheduling method based on RIS collaboration. The computer-readable storage medium stores a computer program thereon, and when the program is executed by a processor, it implements the steps of the user task offloading and scheduling method based on RIS collaboration described in any of the above embodiments.

[0287] The present application can be in the form of a computer program product implemented on one or more storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) that contain program codes. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device.

[0288] The above-described embodiments merely represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and the present invention also intends to include these modifications and variations.

Claims

1. A user task offloading and scheduling method based on RIS collaboration, characterized in that, It includes the following steps: S1: Obtain the cascaded channel information and noise power between the user and the access point; S2: Perform task offloading initialization according to the cascaded channel information, noise power, and an optimal energy consumption objective function to generate an initial task offloading allocation iteration node and an optimal task offloading scheme; wherein, the initial task offloading allocation iteration node includes an initialized user allocation matrix, an initialized derivative scheme state, and an initialized minimum energy consumption estimation value; S3: Use a lower bound branch algorithm based on user allocation to branch the current task offloading allocation iteration node and update the optimal task offloading scheme to obtain an updated task offloading allocation iteration node and an optimal task offloading scheme; S4: Perform a pruning operation on the task offloading allocation iteration node according to the target energy consumption value of the updated optimal task offloading scheme to obtain a refined task offloading allocation iteration node; S5: Determine whether the refined task offloading allocation iteration node meets a branch and bound iteration condition: if so, perform offloading allocation of the user's computing tasks according to the current optimal task offloading scheme to complete the offloading scheduling of the user tasks; if not, continue to execute step S3.

2. The user task offloading and scheduling method based on RIS collaboration according to claim 1, wherein The optimal energy consumption objective function E s (X OPT , θ OPT ) is specifically expressed as follows: where, v OPT represents the target energy consumption value corresponding to the optimal task offloading scheme; X OPT represents the current optimal user allocation matrix, and its element x ij indicates that the i-th user selects the j-th access point; θ OPT denotes the phase adjustment coefficient vector of the current optimal RIS coordination unit; is the energy consumption model for local computing, where when x i0 = 1, it means that the i-th user chooses to perform computing locally; k0 is the energy consumption related coefficient; α i is the computing complexity related coefficient; f i is the local computing frequency of the i-th user; len i represents the task length of the i-th user; M represents the total number of users; is a remote computing model, and β i is the uplink overhead coefficient, and P T,i is the device transmit power of the i-th user; N represents the total number of access points; R ij (θ OPT ) represents the actual data transmission rate between the i-th user and the j-th access point at θ oPT as follows: Where B W,j represents the bandwidth allocated by the j-th access point to the current user; is the channel gain vector between the i-th user and the j-th access point and is Hermitian transposed; σ0 is the noise power; The optimal energy consumption objective function includes several constraint conditions, and the several constraint conditions include user selection constraint conditions, access point maximum allocation constraint conditions, user allocation element constraint conditions, and RIS phase constraint conditions; The specific representation of the user selection constraint condition is as follows: The specific representation of the access point maximum allocation constraint condition is as follows: wherein, represents the total number of user connections of the j-th access point; L nk (j) represents the maximum number of connections of the j-th access point; The specific representation of the user allocation element constraint condition is as follows: The specific representation of the RIS phase constraint condition is as follows: where θ l represents the phase adjustment coefficient of the l-th reconcilable unit of the RIS, that is, the phase angle; |θ l | = 1 means that the phase coefficient is a unit modulus complex number; L represents the total number of reconcilable units of the RIS, that is, the total number of reflection units.

3. The user task offloading scheduling method based on RIS collaboration according to claim 2, wherein The initialized minimum energy consumption estimation value lb0 is calculated by using a minimum energy consumption estimation algorithm, and the specific representation is as follows: where \(lb_0\) represents the initialized minimum energy consumption estimation value; represents the current user allocation matrix; \(Est(\cdot)\) represents the minimum energy consumption estimation algorithm, and its specific steps are as follows: S201: Generate a simulated user allocation index array and an initial energy consumption estimation value according to the size of the user allocation matrix of the current task offloading allocation iteration node; S202: Generate a simulated user selection submatrix according to the current simulated user allocation index array; S203: Concatenate the simulated user selection submatrix with the current user allocation matrix and determine whether it meets an access point maximum allocation constraint condition: if so, calculate the energy consumption estimation value of the concatenated user allocation matrix according to the user allocation energy consumption model and execute step S204; if not, execute step S205; S204: Determine whether the energy consumption estimation value of the concatenated user allocation matrix is less than the current energy consumption estimation value: if so, update the current energy consumption estimation value and execute step S205; if not, directly execute step S205; S205: Determine whether the current iteration number meets an energy consumption estimation condition: if so, use the current energy consumption estimation value as the minimum energy consumption estimation value of the current task offloading allocation iteration node; if not, update the current simulated user allocation index array and execute step S202.

4. The method for user task offloading scheduling based on RIS collaboration according to claim 3, wherein The generation of the simulated user selection submatrix is represented as follows: Wherein, X c is a simulated user selection sub-matrix, indicating that the size of the matrix is (M - n0) × (N + 1); n0 is the number of rows of the current user allocation matrix ; x ij represents each element x in the matrix ij , and the rule for the i-th user to select and access the j-th access point is expressed as follows: In the formula, represents traversing all users; select i assigns an index to the simulated user corresponding to the i-th user, and its specific calculation is shown as follows: where select i is the i-th element of the index array assigned to the simulated user; is used to represent traversing all elements in the index array assigned to the simulated user; The judgment expression of the access point maximum allocation constraint condition is as follows: In the formula, is used to represent traversing all access points; is the energy consumption estimation value based on the spliced allocation matrix X, and X is represented as follows: is the upper bound of the data transmission rate between the \(i\)-th user and the \(j\)-th access point, and its specific calculation is as follows: The specific representation of updating the index array assigned to the current simulated user in step S205 is as follows: select1 = select1 + 1 In the formula, select1 represents the first element in the index array assigned to the current simulated user.

5. The user task offloading scheduling method based on RIS collaboration according to claim 4, wherein The lower bound branch algorithm based on user assignment includes the following steps: S301. Reset the branch of the user assignment matrix of the task offloading assignment iteration node to obtain the iteration node of the current branch; S302. Determine whether the iteration node of the current branch is a leaf node: If not, calculate using the minimum energy consumption estimation algorithm, update the minimum energy consumption estimation value of the iteration node of the current branch according to the calculation result, and execute step S305; if so, execute step S303; S303. Update the phase of the iteration node of the current branch using the RIS phase optimization algorithm to obtain the optimal task offloading scheme of the current branch and the corresponding target energy consumption value; S304. Determine whether the target energy consumption value of the current branch is less than the target energy consumption value of the current optimal task offloading scheme: If so, update the current optimal task offloading scheme and execute step S305; if not, directly execute step S305; S305. Determine whether the iteration times of the current lower bound branch algorithm based on user assignment are less than an iteration condition: If so, execute step S301; if not, complete the iteration of the lower bound branch algorithm.

6. The RIS collaboration-based user task offloading and scheduling method according to claim 5, wherein The branch reset in step S301 is specifically represented as follows: In the formula, represents the user allocation matrix of the (N + 1)i ′ + k task offloading and allocation iteration nodes, where i ′ represents the iteration number of the current task offloading and allocation iteration node; k is the iteration times of the lower bound branch algorithm based on user assignment; Denote the user allocation matrix of the i min -th transposed task offloading allocation iteration node, where i min represents the index of the iteration node with the minimum energy consumption estimation value; Denotes the transpose of the identity matrix, with size (N + 1), where the element at the k-th position is 1 and the other elements are 0; The update of the minimum energy consumption estimation value of the iteration node of the current branch according to the calculation result in step S302 is specifically calculated as follows: where, lb (N+1)i′+k represents the iterative node (N + 1)i of the current branch ′ + the minimum energy consumption estimation value of k; Est(·) is the minimum energy consumption estimation algorithm.

7. The method for user task offloading scheduling based on RIS collaboration according to claim 6, wherein, The specific calculation representation of the phase update in step S303 is: where θ (N+1)i′+k represents the phase adjustment coefficient vector of the RIS reconcilable unit corresponding to the optimal task offloading scheme of the most current branch; Opt RIS is the RIS phase update algorithm, and its specific steps are as follows: S30301. Calculate the current phase update parameter according to the current phase adjustment coefficient vector of the RIS and the corresponding cascaded channel information; wherein, the current phase update parameter includes data transmission rate, signal-to-noise ratio gain, and channel gain correction factor; The specific representation of the data transmission rate is: where, R ij (θ (u) ) represents the data transmission rate corresponding to the phase adjustment coefficient vector of the u-th iteration; θ (u) is the phase adjustment coefficient vector of the u-th iteration. When u = 0, the modulus of each element of θ (0) is 1; where, u represents the iteration number of the RIS phase update algorithm; The specific representation of the signal-to-noise ratio gain is: wherein, represents the signal-to-noise ratio gain of the u-th iteration; represents the matrix of the cascaded channel information between all users and all access points, and is transposed by Hermite; The specific representation of the channel gain correction factor is: wherein, represents the channel gain correction factor for the u-th iteration; represents the expected term of the signal gain; is the noise power term; S30302. Optimize the phases of all elements of the current phase adjustment coefficient vector of the RIS using the coupled influence optimization algorithm according to the current phase update parameter to obtain the optimized phase adjustment coefficient vector of the RIS; S30303. Determine whether the current phase adjustment coefficient vector of the RIS converges: If not, continue to execute step S30301; if so, output the current phase adjustment coefficient vector of the RIS to complete the iteration of the lower bound branch algorithm based on user assignment; Among them, for the convergence, it is judged using the target energy consumption value corresponding to the optimal energy consumption objective function, and the specific representation is as follows: wherein represents the target energy consumption value corresponding to the phase adjustment coefficient vector of the (u + 1)-th iteration; δ is the convergence threshold of the RIS phase update algorithm.

8. The user task offloading scheduling method based on RIS collaboration according to claim 7, wherein The coupling influence optimization algorithm optimizes the phase of the q-th element of the current phase adjustment coefficient vector of the RIS as follows: The phase optimization is as follows: wherein, represents the q-th element of the phase adjustment coefficient vector for the (t + 1)-th iteration in the coupled influence optimization algorithm, and its initial value φ (0) is the phase adjustment coefficient vector θ of the current RIS (u) ; is the optimization direction vector of the q-th phase for the t-th iteration in the coupled influence optimization algorithm, and its specific representation is as follows: In the formula, used to represent the gain effect of the signal-to-noise ratio; represents the coupling influence term, which is the coupling influence factor Γ for the p-th element and the q-th element of the phase adjustment coefficient vector pq , where Γ pq is specifically expressed as follows: In the formula, represents the square of the channel gain correction factor; represents the outer product of the channel gain matrix; where the superscript (*) represents the complex conjugate; the iterative condition of the coupling influence optimization algorithm is: Determine || φ (t+1) - φ (t) || whether it is less than a threshold value ε: If so, the phase adjustment coefficient vector φ of the RIS in the previous iteration (t) is used as the phase adjustment coefficient vector of the optimized RIS, that is, θ (u+1) = φ (t) , and the phase optimization is completed; if not, then φ (t+1) is used as the phase adjustment coefficient vector of the RIS in the current coupling influence optimization algorithm, and the coupling influence optimization algorithm is continued to perform phase optimization on all elements; where the threshold value ε ∈ [0, 1].

9. A user task offloading and scheduling device based on RIS collaboration, characterized in that, It includes a cascaded channel information acquisition unit for users, a task offloading scheme initialization unit, a task offloading scheme branching unit, a task offloading scheme pruning unit, and a task offloading scheme iteration judgment unit; The cascade channel information acquisition unit of the user is used to acquire the cascade channel information and noise power between the user and the access point; The task offloading scheme initialization unit is used to perform task offloading initialization according to the cascade channel information, noise power, and an optimal energy consumption objective function, and generate an initial task offloading allocation iteration node and an optimal task offloading scheme; wherein, the initial task offloading allocation iteration node includes an initialized user allocation matrix, an initialized derivative scheme state, and an initialized minimum energy consumption estimation value; The task offloading scheme branching unit is used to branch the current task offloading allocation iteration node by using a lower bound branching algorithm based on user allocation, and update the optimal task offloading scheme to obtain an updated task offloading allocation iteration node and an optimal task offloading scheme; The task offloading scheme pruning unit is used to perform a pruning operation on the task offloading allocation iteration node according to the target energy consumption value of the updated optimal task offloading scheme to obtain a refined task offloading allocation iteration node; The task offloading scheme iteration judgment unit is used to judge whether the refined task offloading allocation iteration node meets a branch and bound iteration condition: if so, perform task offloading allocation on the user's computing tasks according to the current optimal task offloading scheme to complete the offloading scheduling of the user tasks; if not, continue to call the task offloading scheme branching unit.

10. A mobile edge computing system based on reconfigurable intelligent surfaces, characterized in that, It includes a reconfigurable metasurface, i.e., RIS, a base station, and a RIS-based collaborative task offloading scheduling device communicatively connected to the reconfigurable intelligent metasurface and the base station; The reconfigurable metasurface is provided with a plurality of tunable units, which are used to receive the computing task offloading requests from user equipment, adjust the reflection phase through the tunable units, and reflect and transmit the signal information to the base station; wherein, the user equipment is used to generate computing tasks and send the computing tasks and corresponding computing requirements to the reconfigurable metasurface; The base station is provided with an access point to an edge computing server, which is used to calculate the cascade channel information and noise power between the current user and the access point according to the computing tasks and corresponding computing requirements, and input them into the RIS-based collaborative task offloading scheduling device; Wherein, the RIS-based collaborative task offloading scheduling device is used to execute the RIS-based collaborative task offloading scheduling method according to any one of claims 1-7.

Images