Method, system and storage medium for capability programmable wireless network resource optimization

Abstract

The application relates to the technical field of wireless communication, and discloses a capacity programmable wireless network resource optimization method, a system and a storage medium. Network nodes are dynamically divided into different network node types according to energy reserves, computing capacity, loading functions and maximum communication radii; the optimal operation mode of the network nodes is dynamically determined by solving the problem of maximizing the network system life while ensuring that all terminals are covered by relays / base stations and all nodes in the system are connectable; the unloading proportion of the computing tasks of the terminal nodes to other computing nodes is determined by solving the problem of minimizing the system node computing resource occupation proportion under the limitations of task delay, per-hop delay and task computing amount; the source-terminal path considering link capacity demand and path hop count is obtained based on a filtered subgraph and by adopting BFS and Dijkstra algorithms; and the network nodes are controlled to sleep or wake up according to the network node function state, the residual energy and the task priority. The application significantly improves resource utilization and prolongs system working time through node function dynamic adaptation and heterogeneous collaboration.

Description

Technical Field

[0001] This invention relates to the field of wireless communication network technology, and specifically to a method, system, and storage medium for optimizing wireless network resources with programmable capabilities. Background Technology

[0002] In resource-constrained scenarios (such as remote areas and disaster environments), existing wireless networks typically employ fixed functional nodes and static resource allocation strategies. These methods struggle to adapt to dynamically changing communication demands and node resource status, leading to system inefficiency or task failures. For example, traditional cellular networks rely on fixed base stations and cannot dynamically adjust their operational modes based on node energy reserves, communication capabilities, and task requirements; resource allocation is often based on single-objective optimization (such as throughput or energy consumption), lacking comprehensive consideration of heterogeneous node collaboration. Furthermore, in extreme environments where infrastructure is scarce and node energy is limited, while existing multi-path routing schemes can improve reliability, they do not incorporate dynamic node function switching and global resource optimization, easily resulting in energy waste or communication interruptions. Existing Mesh networks for extreme dynamic environments generally use preset topologies and fixed node operating modes, unable to adjust base station / relay / terminal operating modes according to real-time status. Computational tasks require transmission back to the central node for processing, resulting in high latency and energy consumption. They also cannot dynamically optimize transmission paths and computational resource allocation based on task requirements. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method, system, and storage medium for optimizing programmable wireless network resources. It relates to methods for dynamic functional adaptation, resource allocation, and routing optimization of heterogeneous wireless networks under resource-constrained environments, and is particularly suitable for constructing private network communication systems in extreme environments or remote areas.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0005] In a first aspect, the present invention provides a method for optimizing wireless network resources with programmable capabilities, comprising:

[0006] Based on energy reserves, computing power, operating mode, and maximum communication radius, network nodes are dynamically divided into large network nodes, medium network nodes, and small network nodes.

[0007] By solving the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes can be connected, the optimal operating mode of network nodes is dynamically determined.

[0008] By addressing the problem of minimizing the proportion of network node computing resources occupied under constraints of task latency, per-hop latency, and task computation volume, the proportion of computing tasks of network nodes running in terminal mode to be offloaded to service nodes is determined; service nodes include network nodes running in base station mode in large network nodes, and network nodes running in relay mode in large and medium network nodes.

[0009] Based on the filtered subgraph, the breadth-first search algorithm and Dijkstra's algorithm are used to obtain the source-terminal path that takes into account link capacity requirements and path hop count.

[0010] The sleep or wake-up of network nodes is controlled based on their operating mode, remaining energy, and task priority.

[0011] In one embodiment, the dynamic division of network nodes into large, medium, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius specifically includes:

[0012] According to the energy reserves of network node n Computational ability Operating mode and maximum communication radius The network nodes are divided into large network nodes, medium network nodes, and small network nodes;

[0013] If network node n is a large network node , Given a large set of network nodes, network node n can operate in any of the following modes: base station, relay, or terminal, and satisfies: > ; The first energy reserve threshold, The first computing power threshold, Indicates the first communication distance threshold;

[0014] If network node n is a medium network node , Let n be the set of network nodes. Then network node n can operate in relay or terminal mode, and satisfy: ; This is the second energy reserve threshold. This is the second computing power threshold. Indicates the second communication distance threshold;

[0015] If network node n is a small network node , If n is a small set of network nodes, then network node n can only run in terminal mode and satisfies: .

[0016] In one embodiment, the step of dynamically determining the optimal operating mode of network nodes by solving the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes are interconnected, specifically includes:

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] The system lifetime is defined as the minimum survival time of the service nodes. A collection of service nodes , This represents the set of network nodes operating in base station mode within a large network node cluster. This refers to the set of network nodes operating in relay mode among large and medium network nodes. Network nodes operating in terminal mode are called user nodes. The set of user nodes... ; This represents the typical operating power of network node n;

[0024] Define decision variables , Represents network nodes Operating in base station mode Represents network nodes Operating in relay mode, Represents network nodes It operates in terminal mode, therefore it has constraints on large network nodes. There are constraints on network nodes. There are constraints on small network nodes. That is, the operating modes of the same node are mutually exclusive and the smaller nodes can only be used as terminals;

[0025] Define network nodes The coordinates are Network nodes Euclidean distance between The communication coverage area of ​​the base station is The relay's communication coverage is ; These are the intermediate variables corresponding to network node n and network node m, respectively;

[0026] A binary search combined with a feasibility check framework is used to make decisions on the operating modes of large and medium network nodes.

[0027] In one embodiment, the step of minimizing the proportion of network node computing resources occupied under constraints of task latency, per-hop latency, and task computational load, and determining the offloading ratio of network node computing tasks running in terminal mode to service nodes, specifically includes:

[0028] A set of service nodes is distributed in the geographic space. and user node set The computing power of the service nodes Greater than the computing power of user nodes The network nodes are distributed in a network that is not fully connected but where there is a path between any two nodes. User nodes can only communicate through service nodes.

[0029] Each network node Each has its own computing task ,in It refers to the computational load of the task. This is the maximum tolerable latency for the task; network nodes Including user nodes service nodes User node Capable of offloading some tasks to multiple service nodes The uninstallation ratio is recorded as , User node Unload to service node The computational cost is ; For user nodes The computational load of the task;

[0030] During unloading, the transmission delay is... Minimum number of node hops between Sum of coefficients The transmission delay is determined to be Therefore, at the service node The remaining processing time is , For service nodes Maximum tolerable latency for the task; required computational intensity for:

[0031] ;

[0032] service node Total computational intensity The computational intensity includes both the task itself and the unloading task:

[0033] ;

[0034] User Node Local computing intensity for:

[0035] ;

[0036] in User node The total uninstallation rate For user nodes The task has a maximum tolerable latency; the objective is to minimize the maximum computing capacity utilization ratio across all network nodes, resulting in the following problem:

[0037] ;

[0038] :

[0039] Formula 1: ;

[0040] Formula 2: ;

[0041] Formula 3: ;

[0042] Formula 4: ;

[0043] Formula 5: ;

[0044] For network nodes The calculated intensity, Network Nodes computing power For service nodes computing power For service nodes Computational power;

[0045] The constraints in the problem include Equations 1 to 5, where Equation 1 indicates that the computational intensity of the service node does not exceed the computational capacity of the service node; Equation 2 indicates that the local task processing latency of the user node meets the requirements; Equation 3 indicates the offloading ratio boundary; Equation 4 indicates the total offloading ratio limit; and Equation 5 indicates the transmission latency constraint.

[0046] By introducing auxiliary variables Let represent the maximum load ratio. Using the maximum load ratio as the objective variable, the problem is transformed into a standard linear programming problem, which is then solved using a standard LP solver.

[0047] In one embodiment, the process of obtaining a source-to-terminal path based on a filtered subgraph, using a breadth-first search algorithm and Dijkstra's algorithm, taking into account link capacity requirements and path hop count, specifically includes:

[0048] Communication between user nodes must be achieved through forwarding by service nodes, modeling the network as an undirected graph. , where the node set It includes all service nodes and user nodes; edge set It consists of two types of edges:

[0049] The edge between user nodes and service nodes: For each user node and each service node If the channel capacity between the two Then there exists an edge. ;

[0050] Edges between service nodes: For each pair of different service nodes If the channel capacity between the two Then there exists an edge. ;

[0051] Network Nodes and Channel capacity between ,in It's bandwidth. It's the transmission power. It is a network node and Channel coefficients between;

[0052] For a pair of network nodes that have a communication relationship and ,path Defined as a network node sequence , , There are edges between adjacent network nodes. Path capacity It is the minimum edge capacity on the path, i.e. ;

[0053] The goal is to find a path Make Minimize the number of hops at the same time The number of edges in the path is denoted by ; if no such path exists, then a path has been found. Make maximum; Represents network nodes and Communication capacity requirements between them;

[0054] When a path exists, a breadth-first search algorithm based on a filtered subgraph is used to find the path. When no path exists that meets the conditions, Dijkstra's algorithm is used to find the path with the maximum capacity.

[0055] In one embodiment, controlling the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority specifically includes:

[0056] If network node n is a large or medium network node, it will enter hibernation when the following conditions are met: there are no pending tasks, it is running in terminal mode and the duration of the state exceeds [a certain value]. Or the remaining energy of network node n ;in The low energy duration threshold, Let n be the sleep energy threshold for network node n;

[0057] Network node wake-up is triggered when the following events are detected: a neighboring network node's energy is below a threshold or the link is interrupted; or a new task arrives with high priority. ; Wake-up threshold for network nodes - task importance threshold.

[0058] In a second aspect, the present invention provides a capability-programmable wireless network resource optimization system, comprising:

[0059] Node partitioning module: Dynamically partitions network nodes into large network nodes, medium network nodes, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius;

[0060] Functional decision module: By solving the problem of maximizing network system lifetime while ensuring that all network nodes running in terminal mode can be covered by network nodes running in relay mode or base station mode, and that all network nodes can be connected, the module dynamically determines the optimal operating mode of network nodes.

[0061] The computation offloading module addresses the problem of minimizing the proportion of network node computational resource usage under constraints such as task latency, per-hop latency, and task computation volume. It determines the offloading ratio of network node computational tasks running in terminal mode to service nodes. Service nodes include network nodes running in base station mode in large network nodes, as well as network nodes running in relay mode in large and medium network nodes.

[0062] Path decision module: Based on the filtered subgraph, it uses breadth-first search algorithm and Dijkstra's algorithm to obtain the source-terminal path that takes into account link capacity requirements and path hop count;

[0063] Hibernation / Wake-up Module: Controls the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority.

[0064] In one embodiment, the node partitioning module, which dynamically partitions network nodes into large, medium, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius, specifically includes:

[0065] According to the energy reserves of network node n Computational ability Operating mode and maximum communication radius The network nodes are divided into large network nodes, medium network nodes, and small network nodes;

[0066] If network node n is a large network node , Given a large set of network nodes, network node n can operate in any of the following modes: base station, relay, or terminal, and satisfies: > ; The first energy reserve threshold, The first computing power threshold, Indicates the first communication distance threshold;

[0067] If network node n is a medium network node , Let n be the set of network nodes. Then network node n can operate in relay or terminal mode, and satisfy: ; This is the second energy reserve threshold. This is the second computing power threshold. Indicates the second communication distance threshold;

[0068] If network node n is a small network node , If n is a small set of network nodes, then network node n can only run in terminal mode and satisfies: .

[0069] In one embodiment, the functional decision module dynamically determines the optimal operating mode of network nodes by solving the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes are interconnected. This specifically includes:

[0070] ;

[0071] ;

[0072] ;

[0073] ;

[0074] ;

[0075] ;

[0076] The system lifetime is defined as the minimum survival time of the service nodes. A collection of service nodes , This represents the set of network nodes operating in base station mode within a large network node cluster. This refers to the set of network nodes operating in relay mode among large and medium network nodes. Network nodes operating in terminal mode are called user nodes. The set of user nodes... ; This represents the typical operating power of network node n;

[0077] Define decision variables , Represents network nodes Operating in base station mode Represents network nodes Operating in relay mode, Represents network nodes It operates in terminal mode, therefore it has constraints on large network nodes. There are constraints on network nodes. There are constraints on small network nodes. That is, the operating modes of the same node are mutually exclusive and the smaller nodes can only be used as terminals;

[0078] Define network nodes The coordinates are Network nodes Euclidean distance between The communication coverage area of ​​the base station is The relay's communication coverage is ; These are the intermediate variables corresponding to network node n and network node m, respectively;

[0079] A binary search combined with a feasibility check framework is used to make decisions on the operating modes of large and medium network nodes.

[0080] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any embodiment of the first aspect.

[0081] Compared with the prior art, the beneficial technical effects of the present invention are:

[0082] By dynamically adapting node functions and collaborating with heterogeneous architectures, resource utilization is significantly improved and system uptime is extended. Combined with multi-path routing and backup mechanisms, highly reliable communication is achieved under energy constraints. The allocation of computing and communication resources is jointly optimized to balance task performance and energy efficiency. Dynamic hibernation strategies further reduce the system's static energy consumption and adapt to the long-term operation requirements of extreme environments. Attached Figure Description

[0083] Figure 1 This is a flowchart of the method in an embodiment of the present invention;

[0084] Figure 2 This is a schematic diagram of the system architecture in an embodiment of the present invention;

[0085] Figure 3 These are schematic diagrams illustrating the effects of several typical network structures in embodiments of the present invention. Detailed Implementation

[0086] A preferred embodiment of the present invention will now be described in detail with reference to the accompanying drawings.

[0087] like Figure 1 As shown, the present invention provides a method for optimizing wireless network resources with programmable capability, comprising:

[0088] S1 dynamically divides network nodes into large network nodes, medium network nodes, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius.

[0089] S2, by solving the problem of maximizing the network system lifetime under the condition that all network nodes running in terminal mode can be covered by network nodes running in relay mode or network nodes running in base station mode, and all network nodes can be connected, dynamically determines the optimal operating mode of network nodes.

[0090] S3, by solving the problem of minimizing the proportion of network node computing resource occupancy under the constraints of task latency, per-hop latency, and task computation amount, determines the offloading ratio of network node computing tasks running in terminal mode to service nodes; service nodes include network nodes running in base station mode in large network nodes, and network nodes running in relay mode in large and medium network nodes.

[0091] S4, based on the filtered subgraph, uses breadth-first search and Dijkstra's algorithm to obtain the source-terminal path that takes into account link capacity requirements and path hop count;

[0092] S5 controls whether network nodes go into hibernation or wake up based on their operating mode, remaining energy, and task priority.

[0093] This invention achieves flexible function switching by dynamically adapting heterogeneous nodes to their functions. Nodes are categorized into large, medium, and small types based on energy, operating mode, and communication radius. The operating mode (e.g., base station mode, relay mode) is dynamically determined based on real-time status and task requirements. Based on computational workload and processing latency requirements, computational offloading is flexibly allocated from network nodes operating in terminal mode (referred to as user nodes) to service nodes, achieving computational load balancing within the system. A demand-guaranteed routing decision model is designed, comprehensively considering the throughput and latency requirements of communication tasks to find the shortest path that meets throughput requirements, achieving global coordination of communication resources. A dynamic sleep and wake-up mechanism is introduced, controlling the sleep of inactive nodes based on node energy and task priority, significantly reducing the system's static energy consumption.

[0094] Specifically, this invention establishes a closed-loop optimization framework for perception-decision-execution.

[0095] The global perception layer includes: real-time acquisition of the three-dimensional positions of network nodes. Node energy Channel State Information (CSI) Calculation task computational load Latency requirements Communication task throughput requirements .

[0096] The collaborative decision-making layer includes functional decision-making, computational offloading decision-making, routing, and dynamic hibernation and wake-up mechanisms.

[0097] In one embodiment, such as Figure 2 As shown, step S1, which dynamically divides network nodes into large, medium, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius, specifically includes:

[0098] According to the energy reserves of network node n Computational ability Operating mode and maximum communication radius The network nodes are divided into large network nodes, medium network nodes, and small network nodes;

[0099] If network node n is a large network node , Given a large set of network nodes, network node n can operate in any of the following modes: base station, relay, or terminal, and satisfies: > ; The first energy reserve threshold, The first computing power threshold, Indicates the first communication distance threshold;

[0100] If network node n is a medium network node , Let n be the set of network nodes. Then network node n can operate in relay or terminal mode, and satisfy: ; This is the second energy reserve threshold. This is the second computing power threshold. Indicates the second communication distance threshold;

[0101] If network node n is a small network node , If n is a small set of network nodes, then network node n can only run in terminal mode and satisfies: .

[0102] In one embodiment, step S2, by addressing the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes are interconnected, dynamically determines the optimal operating mode of the network nodes, specifically including:

[0103] ;

[0104] The constraints are as follows:

[0105] ;

[0106] ;

[0107] ;

[0108] ;

[0109] ;

[0110] The system lifetime is defined as the minimum survival time of the service nodes. A collection of service nodes , This represents the set of network nodes operating in base station mode within a large network node cluster. This refers to the set of network nodes operating in relay mode among large and medium network nodes. Network nodes operating in terminal mode are called user nodes. The set of user nodes... ; This represents the typical operating power of network node n;

[0111] Define decision variables , Represents network nodes Operating in base station mode Represents network nodes Operating in relay mode, Represents network nodes It operates in terminal mode, therefore it has constraints on large network nodes. There are constraints on network nodes. There are constraints on small network nodes. That is, the operating modes of the same node are mutually exclusive and the smaller nodes can only be used as terminals;

[0112] Define network nodes The coordinates are Network nodes Euclidean distance between The communication coverage area of ​​the base station is The relay's communication coverage is ; These are the intermediate variables corresponding to network node n and network node m, respectively;

[0113] A binary search combined with a feasibility check framework is used to make decisions on the operating modes of large and medium network nodes.

[0114] The first two constraints represent operating mode constraints, namely, large nodes can only activate one of the operating modes of base station, relay, and terminal; medium nodes can only activate one of the operating modes of relay and terminal; the third constraint indicates that each terminal will be covered by at least one service node; the fourth constraint indicates that any two service nodes can communicate with each other; and the fifth condition indicates that service nodes have different coverage ranges when they take different operating modes.

[0115] This problem is a runtime mode allocation optimization problem with geometric constraints, combining the coverage set problem (covering all nodes), the connectivity dominance set problem (service nodes are connected and dominate all), and max-min fair allocation (energy balance to maximize minimum lifetime). It is generally an NP-hard problem. Runtime mode allocation affects coverage, connectivity, and lifetime. Nodes are suitable as service nodes (capable of handling high power consumption), but their location is critical; base stations reduce the number of service nodes but lower costs. Relays, on the other hand, offer the opposite benefit. The trade-off lies in this: high coverage (using more base stations) extends coverage lifetime but may shorten system lifetime, while low coverage (using more relays) does the opposite. Connectivity constraints increase complexity, especially in sparse networks.

[0116] Since the problem is NP-hard, a binary search combined with a feasibility check framework is used for the solution. Binary search range , .in Represents a node Power consumption when operating in base station mode This represents the power consumption of a node operating in relay mode. Considering medium to large nodes, this is specifically for... Binary search. For a given... Check for the existence of run mode assignments and covered / connected subgraphs. The final output is the maximum feasible one. And its operating mode allocation. The detailed algorithm flow is as follows:

[0117] 1. Initialization , ;

[0118] 2. while ( (For precision), loop:

[0119] 2.1. ;

[0120] 2.2. If feasible ( Success: / / Feasibility check invoked

[0121] ;

[0122] Record operation mode allocation;

[0123] Else:

[0124] .

[0125] 3. Output Assignment to the corresponding operating mode.

[0126] The feasibility check algorithm in section 2.2 of the above algorithm is as follows:

[0127] For a given Define the set of candidate service nodes :

[0128] for Only when It can be used as a base station, only when It can be used as a relay.

[0129] for Only when It can be used as a relay.

[0130] Task: Choose form It satisfies coverage and connectivity constraints.

[0131] (1) Initialization: Uncovered terminal set (Initialize all nodes as terminal nodes); Service node set Unicom component set .

[0132] (2) Greedy selection of service nodes:

[0133] when :

[0134] A. For each candidate node Calculate its coverage potential:

[0135] a. If Can be used as a base station: Coverage set ;

[0136] b. If Can be used as a relay: Coverage set ;

[0137] c. If Dual operating modes are available: select the one with wider coverage (base station priority, because...). ), record the operating mode.

[0138] B. Select Node (Covering the most uncovered terminals), with high priority. (Long survival time).

[0139] C. Add arrive Assume its operating mode (base station or relay) and update. .

[0140] If it is not possible to cover all terminals, return "not feasible".

[0141] (3) Connectivity guarantee:

[0142] Build a service node graph :node ,when The edge exists at time.

[0143] examine Connectivity: If connected, continue. Otherwise, add a relay node bridging component:

[0144] A. Find the closest component pair and its node pairs ;

[0145] B. Select (Can be used as a relay) Priority ;

[0146] C. Add arrive ,renew ;

[0147] D. Repeat until connectivity or infeasibility (no candidate nodes).

[0148] (4) Output: If coverage and connectivity are satisfied, it is feasible and the running mode allocation is returned; otherwise, it is not feasible.

[0149] This heuristic, based on the principles of "maximizing coverage benefits first" and "highest survival time first," handles runtime mode dependencies and ensures connectivity. Complexity It is relatively efficient.

[0150] In one embodiment, step S3, by addressing the problem of minimizing the proportion of network node computing resource occupancy under constraints of task latency, per-hop latency, and task computational load, and determining the offloading ratio of network node computing tasks running in terminal mode to service nodes, specifically includes:

[0151] A set of service nodes is distributed in the geographic space. and user node set The computing power of the service nodes Greater than the computing power of user nodes The network nodes are distributed in a network that is not fully connected but where there is a path between any two nodes. User nodes can only communicate through service nodes.

[0152] Each network node Each has its own computing task ,in It refers to the computational load of the task. This is the maximum tolerable latency for the task; network nodes Including user nodes service nodes User node Capable of offloading some tasks to multiple service nodes The uninstallation ratio is recorded as , User node Unload to service node The computational cost is ; For user nodes The computational load of the task.

[0153] During unloading, the transmission delay is... Minimum number of node hops between Sum of coefficients The transmission delay is determined to be Therefore, at the service node The remaining processing time is , For service nodes Maximum tolerable latency for the task; required computational intensity for:

[0154] .

[0155] service node Total computational intensity The computational intensity includes both the task itself and the unloading task:

[0156] ;

[0157] User Node Local computing intensity for:

[0158] ;

[0159] in User node The total uninstallation rate For user nodes The maximum tolerable latency for the task.

[0160] The goal is to minimize the maximum computing capacity utilization ratio among all network nodes, resulting in the following problem:

[0161] ;

[0162] :

[0163] Formula 1: ;

[0164] Formula 2: ;

[0165] Formula 3: ;

[0166] Formula 4: ;

[0167] Formula 5: ;

[0168] For network nodes The calculated intensity, Network Nodes computing power For service nodes computing power For service nodes Its computing power.

[0169] The constraints in the problem include Equations 1 to 5, where Equation 1 indicates that the computing intensity of the service node does not exceed the computing capacity of the service node; Equation 2 indicates that the local task processing latency of the user node meets the requirements; Equation 3 indicates the offloading ratio boundary; Equation 4 indicates the total offloading ratio limit; and Equation 5 indicates the transmission latency constraint.

[0170] The core objective of the problem is to adjust node load through task offloading, minimizing the maximum load proportion among all nodes and ensuring system load balance. Given the intensity calculation formula, and The relationship is linear (because the denominator) (is a constant), therefore all and All about A linear function. Objective function It is a piecewise linear convex function, and all constraints are linear; therefore, the overall problem is a linear programming (LP) problem. Communication between nodes depends on the path (minimum hop count). However, network connectivity guarantees all pairs The path exists, and the impact of transmission latency on processing time needs to be considered when making a decision.

[0171] Since the problem is LP, it can be solved in polynomial time (e.g., using the interior-point method). The transmission delay is treated as a constant in the computational intensity (the denominator is negligible). This simplifies the problem; network connectivity and path information ( It can be obtained during the preprocessing stage.

[0172] By introducing auxiliary variables Let represent the maximum load ratio. Using the maximum load ratio as the objective variable, the problem is transformed into a standard linear programming problem, which is then solved using a standard LP solver.

[0173] Introducing auxiliary variables This represents the maximum load ratio (target variable). The objective function of the above problem can be transformed into...

[0174] ;

[0175] User node load ratio constraints (combining load ratio and original constraints):

[0176] ;

[0177] The second equation can be simplified to:

[0178] ;

[0179] Service node load ratio constraints (combining load ratio and original constraints):

[0180] ;

[0181] Unloading ratio boundary constraints:

[0182] ;

[0183] And nonnegativity constraints:

[0184] ;

[0185] This model is a standard linear programming problem and can be solved efficiently using standard LP solvers (such as the simplex method and the interior point method).

[0186] In one embodiment, step S4, based on the filtered subgraph, employs a breadth-first search algorithm and Dijkstra's algorithm to obtain a source-terminal path that considers link capacity requirements and path hop count, specifically including:

[0187] Communication between user nodes must be achieved through forwarding by service nodes, modeling the network as an undirected graph. , where the node set It includes all service nodes and user nodes; edge set It consists of two types of edges:

[0188] The edge between user nodes and service nodes: For each user node and each service node If the channel capacity between the two Then there exists an edge. ;

[0189] Edges between service nodes: For each pair of different service nodes If the channel capacity between the two Then there exists an edge. ;

[0190] Network Nodes and Channel capacity between ,in It's bandwidth. It's the transmission power. It is a network node and Channel coefficients between;

[0191] For a pair of network nodes that have a communication relationship and (These may all be user nodes, or may include service nodes), path Defined as a network node sequence , , There are edges between adjacent network nodes. Path capacity It is the minimum edge capacity on the path, i.e. ;

[0192] The goal is to find a path Make (Capacity requirements) while minimizing the number of hops. The number of edges in the path is denoted by ; if no such path exists, then a path has been found. Make Maximum (excluding jump count); Represents network nodes and Communication capacity requirements between them;

[0193] When a path exists, a breadth-first search algorithm (BFS) based on a filtered subgraph is used to find the path. When no path exists that meets the conditions, Dijkstra's algorithm is used to find the path with the maximum capacity.

[0194] Dynamic Execution Layer: Issues configuration commands based on decision results, including: node function mode switching (base station / relay / terminal); computation offloading; routing table update; node sleep and wake-up.

[0195] This problem falls under the category of constrained shortest path problems in graph theory, specifically the bottleneck-constrained shortest path problem. The scale may be large, but the number of service nodes is usually limited, so the algorithm needs to be efficient. The problem requires first satisfying the capacity requirement, then minimizing the number of hops; if there is no path that satisfies the requirement, then the path capacity is maximized.

[0196] The problem is solved using the following approach: For each communication star request (node) and ): 1) Construct a filter subgraph, Only the original image is included. Medium capacity not less than 1) Edges; 2) BFS to find the shortest path: in Run a breadth-first search (BFS) to find the source from arrive Find the shortest path (minimum number of hops). If found, return the path; 3) If no path meets the requirements: in the original graph Found from arrive The maximum bottleneck path (i.e. the path with the largest capacity) is determined using Dijkstra's algorithm.

[0197] In one embodiment, controlling the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority specifically includes:

[0198] If network node n is a large or medium network node, it will enter hibernation when the following conditions are met: there are no pending tasks, it is running in terminal mode and the duration of the state exceeds [a certain value]. Or the remaining energy of network node n ;in The low energy duration threshold, Let n be the sleep energy threshold for network node n;

[0199] Network node wake-up is triggered when the following events are detected: a neighboring network node's energy is below a threshold or the link is interrupted; or a new task arrives with high priority. ; Wake-up threshold for network nodes - task importance threshold.

[0200] Taking the collaborative operation scenario of a scientific expedition team in extreme environments as an example, the implementation process of the technical solution of this invention is explained in detail:

[0201] (1) Initial perception and configuration:

[0202] Scenario: Radius 4km; L1 / L2 (vehicle-mounted base station), M1-M4 (relay vehicle), S1-S8 (terminal).

[0203] Energy: L1 85%, L2 72%; M1-M4: 65 / 60 / 58 / 55%; S1-S8: 50 / 48 / 46 / 42 / 40 / 38 / 36 / 35%.

[0204] Channels: L1-L2 backbone is good; M1-M3 interconnection is good; S1-S3 are adjacent to M2; S4-S6 are adjacent to M3; S7-S8 are at the edge, with weak direct connectivity. Services: S1 reports 2Mbps; local image analysis ≤2s; S2 S5, S6 S7 requires 5Mbps passthrough.

[0205] (2) Dynamic functional adaptation:

[0206] Operating mode: L1 and L2 are set as base stations; M1, M2, and M3 are relays; M4 is for terminal power supply. Service nodes {L1, L2, M1, M2, M3} are connected and cover all S; S7-S8 are jointly covered by M3 / L2, satisfying the requirements of coverage, connectivity, and lifespan maximization.

[0207] (3) Calculate the unloading decision:

[0208] S1-S4: Image task 70%→L1, 20%→M2, 10% local; S5-S8: 60%→L2, 20%→M3, 20% local; M1-M3 each 10% local, 90%→nearest base station; all end-to-end latency ≤2s.

[0209] (4) Path routing planning:

[0210] 2Mbps reporting: S1→M2→L1; S2→M2→L1; S3→M1→L1; S4 / S5 / S6→M3→L2; S7 / 8→M3→L2.

[0211] 5Mbps passthrough: S2 S5 routes S2-M2-L2-M3-S5 (4 hops, bottleneck ≥5Mbps); S6 S7 follows S6-M3-L2-M2-S7 (4 hops). If the S8 link degrades, it follows S8-M3-L2-L1-M1-S1 (maximizing the bottleneck).

[0212] (5) Dynamic hibernation and wake-up:

[0213] Hibernation: M4 enters hibernation when it has no task and remains in terminal state for more than 10 minutes; S3 and S6 hibernate intermittently when idle to conserve power.

[0214] Wake-up: Upon detecting a drop in S7 link and the arrival of a new high-priority video task, M4 is triggered to wake up and temporarily switch to relay mode. L2 power is increased, and M4 goes back to sleep after the session is stably completed.

[0215] Comparison of effects:

[0216] System lifespan: Through node function adjustment and dynamic hibernation, total energy consumption is reduced and the continuous working time of the system is extended; Computation latency: The average latency of image analysis tasks is significantly reduced, meeting task requirements. Communication tasks: While ensuring the throughput of communication tasks, the number of wireless hops between the transmitter and receiver is minimized, thereby reducing system load and communication latency.

[0217] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple steps or stages, which are not necessarily completed at the same time, but may be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but may be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0218] Based on the description of the above method embodiments, this disclosure also provides a system. The system can employ software (applications), modules, components, servers, clients, etc., using the methods described in the embodiments of this specification, combined with necessary implementation hardware. Based on the same innovative concept, the systems in one or more embodiments provided by this disclosure are as described in the following embodiments. Since the implementation schemes and methods for solving the problem are similar, the specific system implementations in the embodiments of this specification can refer to the implementations of the foregoing methods, and repeated details will not be repeated. The terms "module" or "module" used below can refer to a combination of software and / or hardware capable of achieving a predetermined function. Although the systems described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0219] A programmable wireless network resource optimization system, comprising:

[0220] Node partitioning module: Dynamically partitions network nodes into large network nodes, medium network nodes, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius;

[0221] Functional decision module: By solving the problem of maximizing network system lifetime while ensuring that all network nodes running in terminal mode can be covered by network nodes running in relay mode or base station mode, and that all network nodes can be connected, the module dynamically determines the optimal operating mode of network nodes.

[0222] The computation offloading module addresses the problem of minimizing the proportion of network node computational resource usage under constraints such as task latency, per-hop latency, and task computation volume. It determines the offloading ratio of network node computational tasks running in terminal mode to service nodes. Service nodes include network nodes running in base station mode in large network nodes, as well as network nodes running in relay mode in large and medium network nodes.

[0223] Path decision module: Based on the filtered subgraph, it uses breadth-first search algorithm and Dijkstra's algorithm to obtain the source-terminal path that takes into account link capacity requirements and path hop count;

[0224] Hibernation / Wake-up Module: Controls the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority.

[0225] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory including instructions that can be executed by a processor to perform the above-described method. The storage medium may be a computer-readable storage medium, for example, a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0226] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0227] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for optimizing resources in a programmable wireless network, characterized in that, include: Based on energy reserves, computing power, operating mode, and maximum communication radius, network nodes are dynamically divided into large network nodes, medium network nodes, and small network nodes. By solving the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes can be connected, the optimal operating mode of network nodes is dynamically determined. By addressing the problem of minimizing the proportion of network node computing resources occupied under constraints of task latency, per-hop latency, and task computational load, the offloading ratio of computing tasks from network nodes running in terminal mode to service nodes is determined. Service nodes include network nodes running in base station mode within large network nodes, and network nodes running in relay mode within large and medium network nodes, specifically including: A set of service nodes is distributed in the geographic space. and user node set The computing power of the service nodes Greater than the computing power of user nodes The network nodes are distributed in a network that is not fully connected, but where paths exist between any two nodes; user nodes can only communicate through service nodes; each network node Each has its own computing task ,in It refers to the computational load of the task. This is the maximum tolerable latency for the task; Including user nodes service nodes ; Able to unload task components to multiple The uninstallation ratio is recorded as , ; Uninstall to The computational cost is ; for The computational load of the task; during unloading, the transmission latency is determined by... Minimum number of node hops between Sum of coefficients The transmission delay is determined to be ;exist The remaining processing time is , for Maximum tolerable latency for the task; required computational intensity for: ; Total computational intensity The computational intensity includes both the task itself and the unloading task: ; Local computing intensity for: ; yes The total uninstallation rate for The task has a maximum tolerable latency; the objective is to minimize the maximum computing capacity utilization ratio across all network nodes, resulting in the following problem: ； ： Formula 1: ; Formula 2: ; Formula 3: ; Formula 4: ; Formula 5: ; for The calculated intensity, for computing power for computing power for Computational power; The constraints in the problem include Equations 1 to 5. Equation 1 indicates that the computational intensity of the service node does not exceed its computational capacity; Equation 2 indicates that the local task processing latency of the user node meets the requirements; Equation 3 indicates the offloading ratio boundary; Equation 4 indicates the total offloading ratio limit; and Equation 5 indicates the transmission latency constraint. Auxiliary variables are introduced... Let the maximum load ratio be represented. Using the maximum load ratio as the objective variable, the above problem is transformed into a standard linear programming problem, which is then solved using a standard LP solver. Based on the filtered subgraph, the breadth-first search algorithm and Dijkstra's algorithm are used to obtain the source-terminal path that takes into account link capacity requirements and path hop count. The sleep or wake-up of network nodes is controlled based on their operating mode, remaining energy, and task priority.

2. The capability-programmable wireless network resource optimization method according to claim 1, characterized in that, The method of dynamically classifying network nodes into large, medium, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius specifically includes: According to the energy reserves of network node n Computational ability Operating mode and maximum communication radius The network nodes are divided into large network nodes, medium network nodes, and small network nodes; If network node n is a large network node , Given a large set of network nodes, network node n can operate in any of the following modes: base station, relay, or terminal, and satisfies: > ; The first energy reserve threshold, The first computing power threshold, Indicates the first communication distance threshold; If network node n is a medium network node , Let n be the set of network nodes. Then network node n can operate in relay or terminal mode, and satisfy: ; This is the second energy reserve threshold. This is the second computing power threshold. Indicates the second communication distance threshold; If network node n is a small network node , If n is a small set of network nodes, then network node n can only run in terminal mode and satisfies: .

3. The method for optimizing programmable wireless network resources according to claim 2, characterized in that, The method addresses the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode are covered by network nodes operating in relay mode or base station mode, and that all network nodes are interconnected. It dynamically determines the optimal operating mode for network nodes, specifically including: ； ； ； ； ； ； The system lifetime is defined as the minimum survival time of the service nodes. ; A collection of service nodes. , This represents the set of network nodes operating in base station mode within a large network node cluster. This represents the set of network nodes operating in relay mode among large and medium network nodes. Network nodes operating in terminal mode are called user nodes. A collection of user nodes. ; This represents the typical operating power of network node n; Define decision variables , Represents network nodes Operating in base station mode Represents network nodes Operating in relay mode, Represents network nodes It operates in terminal mode, therefore it has constraints on large network nodes. There are constraints on network nodes. There are constraints on small network nodes. That is, the operating modes of the same node are mutually exclusive and the smaller nodes can only be used as terminals; Define network nodes The coordinates are Network nodes Euclidean distance between , The communication coverage area of ​​the base station. The communication coverage area for relays; These are the intermediate variables corresponding to network node n and network node m, respectively; A binary search combined with a feasibility check framework is used to make decisions on the operating modes of large and medium network nodes.

4. The method for optimizing programmable wireless network resources according to claim 3, characterized in that, The process, based on a filtered subgraph, employs a breadth-first search algorithm and Dijkstra's algorithm to obtain a source-terminal path that considers link capacity requirements and path hop count. Specifically, this includes: Communication between user nodes must be achieved through forwarding by service nodes, modeling the network as an undirected graph. , where the node set It includes all service nodes and user nodes; edge set It consists of two types of edges: The edge between user nodes and service nodes: For each user node and each service node If the channel capacity between the two Then there exists an edge. ; Edges between service nodes: For each pair of different service nodes If the channel capacity between the two Then there exists an edge. ; Network Nodes and Channel capacity between ,in It's bandwidth. It's the transmission power. It is a network node and Channel coefficients between; For a pair of network nodes that have a communication relationship and ,path Defined as a network node sequence , , There are edges between adjacent network nodes. Path capacity It is the minimum edge capacity on the path, i.e. ; The goal is to find a path Make Minimize the number of hops at the same time The number of edges in the path is denoted by ; if no such path exists, then a path has been found. Make maximum; Represents network nodes and Communication capacity requirements between them; When a path exists, a breadth-first search algorithm based on a filtered subgraph is used to find the path. When no path exists that meets the conditions, Dijkstra's algorithm is used to find the path with the maximum capacity.

5. The capability-programmable wireless network resource optimization method according to claim 1, characterized in that, The method of controlling the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority specifically includes: If network node n is a large or medium network node, it will enter hibernation when the following conditions are met: there are no pending tasks, it is running in terminal mode and the duration of the state exceeds [a certain value]. Or the remaining energy of network node n ;in The low energy duration threshold, Let n be the sleep energy threshold for network node n; Network node wake-up is triggered when the following events are detected: a neighboring network node's energy is below a threshold or the link is interrupted; or a new task arrives with high priority. ; This is the task importance threshold used to trigger network node wake-up.

6. A capability-programmable wireless network resource optimization system, characterized in that, include: Node partitioning module: Dynamically partitions network nodes into large network nodes, medium network nodes, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius; Functional decision module: By solving the problem of maximizing network system lifetime while ensuring that all network nodes running in terminal mode can be covered by network nodes running in relay mode or base station mode, and that all network nodes can be connected, the module dynamically determines the optimal operating mode of network nodes. The computation offloading module addresses the problem of minimizing the proportion of network node computational resource usage under constraints such as task latency, per-hop latency, and task computation volume. It determines the offloading ratio of computational tasks from network nodes running in terminal mode to service nodes. Service nodes include network nodes running in base station mode within large network nodes, and network nodes running in relay mode within large and medium network nodes. Specifically, they include: A set of service nodes is distributed in the geographic space. and user node set The computing power of the service nodes Greater than the computing power of user nodes The network nodes are distributed in a network that is not fully connected, but where paths exist between any two nodes; user nodes can only communicate through service nodes; each network node Each has its own computing task ,in It refers to the computational load of the task. This is the maximum tolerable latency for the task; Including user nodes service nodes ; Able to unload task components to multiple The uninstallation ratio is recorded as , ; Uninstall to The computational cost is ; for The computational load of the task; during unloading, the transmission latency is determined by... Minimum number of node hops between Sum of coefficients The transmission delay is determined to be ;exist The remaining processing time is , for Maximum tolerable latency for the task; required computational intensity for: ; Total computational intensity The computational intensity includes both the task itself and the unloading task: ; Local computing intensity for: ; yes The total uninstallation rate for The task has a maximum tolerable latency; the objective is to minimize the maximum computing capacity utilization ratio across all network nodes, resulting in the following problem: ； ： Formula 1: ; Formula 2: ; Formula 3: ; Formula 4: ; Formula 5: ; for The calculated intensity, for computing power for computing power for Computational power; The constraints in the problem include Equations 1 to 5. Equation 1 indicates that the computational intensity of the service node does not exceed its computational capacity; Equation 2 indicates that the local task processing latency of the user node meets the requirements; Equation 3 indicates the offloading ratio boundary; Equation 4 indicates the total offloading ratio limit; and Equation 5 indicates the transmission latency constraint. Auxiliary variables are introduced... Let the maximum load ratio be represented. Using the maximum load ratio as the objective variable, the above problem is transformed into a standard linear programming problem, which is then solved using a standard LP solver. Path decision module: Based on the filtered subgraph, it uses breadth-first search algorithm and Dijkstra's algorithm to obtain the source-terminal path that takes into account link capacity requirements and path hop count; Hibernation / Wake-up Module: Controls the hibernation or wake-up of network nodes based on their operating mode, remaining energy, and task priority.

7. The capability-programmable wireless network resource optimization system according to claim 6, characterized in that, In the node partitioning module, the dynamic partitioning of network nodes into large, medium, and small network nodes based on energy reserves, computing power, operating mode, and maximum communication radius specifically includes: According to the energy reserves of network node n Computational ability Operating mode and maximum communication radius The network nodes are divided into large network nodes, medium network nodes, and small network nodes; If network node n is a large network node , Given a large set of network nodes, network node n can operate in any of the following modes: base station, relay, or terminal, and satisfies: > ; The first energy reserve threshold, The first computing power threshold, Indicates the first communication distance threshold; If network node n is a medium network node , Let n be the set of network nodes. Then network node n can operate in relay or terminal mode, and satisfy: ; This is the second energy reserve threshold. This is the second computing power threshold. Indicates the second communication distance threshold; If network node n is a small network node , If n is a small set of network nodes, then network node n can only run in terminal mode and satisfies: .

8. The capability-programmable wireless network resource optimization system according to claim 6, characterized in that, In the functional decision module, the optimal operating mode of network nodes is dynamically determined by solving the problem of maximizing network system lifetime while ensuring that all network nodes operating in terminal mode can be covered by network nodes operating in relay mode or base station mode, and that all network nodes are interconnected. Specifically, this includes: ； ； ； ； ； ； The system lifetime is defined as the minimum survival time of the service nodes. ; A collection of service nodes. , This represents the set of network nodes operating in base station mode within a large network node cluster. This represents the set of network nodes operating in relay mode among large and medium network nodes. Network nodes operating in terminal mode are called user nodes. A collection of user nodes. ; This represents the typical operating power of network node n; Define decision variables , Represents network nodes Operating in base station mode Represents network nodes Operating in relay mode, Represents network nodes It operates in terminal mode, therefore it has constraints on large network nodes. There are constraints on network nodes. There are constraints on small network nodes. That is, the operating modes of the same node are mutually exclusive and the smaller nodes can only be used as terminals; Define network nodes The coordinates are Network nodes Euclidean distance between , The communication coverage area of ​​the base station. The communication coverage area for relays; These are the intermediate variables corresponding to network node n and network node m, respectively; A binary search combined with a feasibility check framework is used to make decisions on the operating modes of large and medium network nodes.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images