Method and system for resource allocation in a converged network

Abstract

The application provides a resource allocation method and system of a sensing and communication integrated network. The method is applied to a fusion center computing device and comprises the following steps: acquiring channel estimation parameters reported by each sensing and communication integrated node; based on the channel estimation parameters, a minimum relative square position error boundary is taken as a target, a greedy communication scheduling strategy is used to allocate a communication data rate for each link under a limited code length transmission condition, a projection gradient descent-based strategy is used to allocate a transmission power of each sensing and communication integrated node, and an integer discrete search strategy is used to determine a sub-code word length of each link to obtain a resource allocation strategy; the resource allocation strategy comprises an optimal communication data rate, an optimal transmission power and an optimal sub-code word length; and then the resource allocation strategy is sent to each sensing and communication integrated node to control the communication and sensing resource allocation of each sensing and communication integrated node. The application significantly improves the communication capacity and positioning accuracy of the sensing and communication integrated network in an anchor-free ad hoc network scene.

Description

Technical Field

[0001] This invention relates to the field of cross-disciplinary technology of wireless communication and wireless network positioning, and in particular to a resource allocation method and system for an integrated sensing network. Background Technology

[0002] High-quality communication and high-precision positioning are core requirements for wireless network applications such as drone swarms, autonomous driving, and multi-agent collaboration. Traditional Integrated Sensing and Communication (ISAC) link technology typically assumes the existence of fixed base stations (anchor points) and uses a co-located transmit / receive mode to achieve both communication and positioning functions. However, in actual deployments of anchor-free self-organizing networks (such as drone swarms), nodes can only establish transmit / receive separated links, and communication and relative positioning can only be achieved through relative measurements. The lack of absolute reference points poses a significant challenge to high-speed communication and high-precision positioning. Cooperative integrated sensing and communication networks based on transmit / receive separated links are considered an effective alternative for the above scenarios due to their advantages such as flexible networking and scalable positioning accuracy.

[0003] Currently, research on resource allocation in collaborative sensor networks still has the following significant shortcomings: First, communication performance evaluation is detached from actual physical layer transmission characteristics. Most existing methods use weighted summation of Shannon capacity as the objective function or constraint for network communication performance. Shannon capacity is based on the assumption of infinitely long codewords and cannot truly reflect the communication capability of the physical layer under finite codeword length (i.e., short packet transmission) conditions. In real-world systems, data packet length is finite, and there is a non-linear trade-off between coding rate and bit error rate. Using Shannon capacity will cause resource allocation results to deviate from actual achievable performance, resulting in wasted communication resources or unreliable links.

[0004] Second, current positioning resource allocation methods are limited to link-level or single-target evaluation, failing to fully utilize network-level cooperative gains. Existing sensing resource allocation methods often only consider the mean square error of channel estimation for a single link, or only optimize the positioning error of a single target, failing to explore the geometric cooperative gains between multiple nodes from a global network perspective. For example, the contribution of channel estimation accuracy to the final relative positioning accuracy varies among different links, but existing methods lack modeling and utilization of the coupling relationships between links across the entire network, resulting in low efficiency in multi-node sensing resource scheduling and an inability to achieve optimal network-level positioning accuracy.

[0005] Third, the resource allocation for communication and sensing is separated, ignoring the information coupling and cooperative gains between the two. Existing technologies generally adopt a resource allocation strategy that separates communication and sensing functions, i.e., optimizing communication and sensing metrics separately and then simply compromising or weighted summation. This decoupling method does not fully consider the coupling mechanism between communication and sensing at the underlying signal processing level in an integrated communication and sensing network—for example, the decoded communication subcodewords can serve as prior auxiliary information for sensing, while the channel estimation accuracy directly affects the effective data rate of the communication link. The separate allocation method cannot capture this bidirectional coupling gain, resulting in limited resource utilization efficiency.

[0006] Fourth, there is a lack of a joint resource allocation framework adapted to anchor-free relative positioning scenarios. Most existing collaborative sensing integration research assumes the existence of base stations or anchors with known locations, and their resource allocation methods directly rely on absolute position references. In anchor-free ad hoc network scenarios, network positioning depends on relative measurements between nodes, and its positioning error exhibits rotation and translation invariance, rendering traditional single-target positioning error metrics inapplicable. Currently, there is no joint resource allocation method specifically designed for this scenario that simultaneously considers finite code length transmission characteristics, global network geometric cooperation, and communication-sensory coupling.

[0007] In summary, existing resource allocation technologies for integrated sensing networks have shortcomings in communication indicator selection, sensing cooperation mechanisms, joint optimization of communication and sensing, and adaptation to anchorless scenarios. There is an urgent need for an allocation method that can fully utilize the network-wide cooperation gain and achieve joint optimization of communication and sensing resources under limited code length transmission conditions, so as to improve the overall network performance. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a resource allocation method and system for a sensor-integrated network, which significantly improves the communication capacity and positioning accuracy of the sensor-integrated network in an anchorless self-organizing network scenario.

[0009] In a first aspect, the present invention provides a resource allocation method for a sensor-integrated network, applied to a converged central computing device; the method includes the following steps: The channel estimation parameters reported by each integrated sensing node are obtained; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; Based on the aforementioned channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each of the aforementioned integrated sensing nodes, and an integer discrete search strategy is used to determine the sub-codeword length of each of the aforementioned links, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; The resource allocation strategy is distributed to each of the integrated sensing nodes to control the allocation of communication and sensing resources for each integrated sensing node.

[0010] According to a resource allocation method for an integrated sensing network provided by the present invention, the channel estimation parameters are determined through the following steps, including: Each of the aforementioned integrated sensing nodes is controlled to split the total codeword to be transmitted into multiple orthogonal sub-codewords according to the link. Each sub-codeword corresponds to a neighbor node, and is broadcast to different neighbor nodes respectively. Each of the integrated sensing nodes is controlled to extract and decode the sub-codewords allocated to each of the integrated sensing nodes from the received broadcast signal. The decoded sub-codewords corresponding to each integrated sensing node are used as the known auxiliary sequence for channel parameter estimation to obtain the channel estimation parameters of each integrated sensing node.

[0011] According to a resource allocation method for a sensor-integrated network provided by the present invention, the method of allocating communication data rates to each link using a greedy communication scheduling strategy includes: Fix the transmit power and subcodeword length of the current node in each of the aforementioned integrated sensing nodes, and initialize the communication data rate of each link to zero; Calculate the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate, and determine the marginal impact value as the loss function value of the link; the loss function value of the link characterizes the degree of impact of increasing the coding rate of the link on network positioning performance; Traverse all links in the network and mark the links whose communication load has reached its limit as fully loaded links; Select the link with the smallest marginal impact value from the links that have never been marked as fully loaded, allocate the communication load increment to the link with the smallest marginal impact value first, and update the communication data rate corresponding to the link with the smallest marginal impact value. Repeat the above allocation process until the total transmission capacity of all links meets the preset network communication constraints or reaches the maximum number of iterations, and obtain the communication data rate corresponding to each link.

[0012] According to a resource allocation method for a sensor-integrated network provided by the present invention, the step of calculating the marginal impact of increasing the communication load per unit on the network positioning performance of each link at the current communication data rate includes: The loss function value of the link is calculated based on the derivative of the relative squared position error bound with respect to the mean square error of channel estimation, the derivative of the mean square error of channel estimation with respect to the data rate, and the derivative of the data rate with respect to the transmission capacity. The relative squared position error bound is calculated based on the mean square error of channel estimation for each link. The mean square error of channel estimation is related to the communication data rate, node transmit power, and subcodeword length of the link.

[0013] According to a resource allocation method for a sensor-integrated network provided by the present invention, the method includes allocating the transmit power of each sensor-integrated node using a projection gradient descent strategy and determining the subcodeword length of each link using an integer discrete search strategy, comprising: Fix the current communication data rate of each link, and initialize the node transmit power and subcodeword length; Determine the gradient direction of the loss function of the network under the node's transmit power, and search the iteration step size by backtracking straight line method to initially update the node transmit power allocation; The initially updated node transmit power allocation is projected onto the feasible region that satisfies the total power constraint of the entire network and the minimum energy requirement of each link to obtain the optimized node transmit power allocation. With the optimized node transmit power fixed, and aiming to maximize the transmission capacity contributed by each broadcast node, the optimal subcodeword length for each link is searched using the discrete grid descent method. Repeat the above steps until the improvement value of the positioning error is less than the threshold or the maximum number of iterations is reached, to obtain the transmit power allocated to each of the integrated sensing nodes and the subcodeword length of each of the links.

[0014] According to a resource allocation method for a sensor-integrated network provided by the present invention, the step of distributing the resource allocation strategy to each sensor-integrated node to control the communication and sensing resource allocation of each sensor-integrated node includes: The resource allocation strategy is encapsulated as control signaling; The control signaling is transmitted to each of the integrated sensing nodes via a wireless communication network, so that each integrated sensing node can adjust the sub-codeword segmentation method, transmission power, and encoding data rate of the broadcast encoder according to the control signaling.

[0015] Secondly, the present invention also provides a resource allocation system for a sensor-integrated network, the system comprising the following modules: Multiple integrated sensing nodes, each of which includes a broadcast coding module and a decoding estimation module; the broadcast coding module is used to divide the total codeword to be transmitted into multiple orthogonal sub-codewords for broadcasting to different neighboring nodes; the decoding estimation module is used to decode the received sub-codewords and use the decoded sub-codewords as a known auxiliary sequence for channel parameter estimation and extraction; A converged central computing device is connected to each of the integrated sensing nodes via a wireless communication network. It is used to collect channel estimation parameters reported by each integrated sensing node and execute the resource allocation method of the integrated sensing network described in any one of the claims to control the communication and sensing resource allocation of each integrated sensing node.

[0016] According to the resource allocation system of the integrated sensing network provided by the present invention, each integrated sensing node further includes a transceiver antenna measurement module; The transceiver antenna measurement module is used to transmit wireless signals according to the communication measurement instructions issued by the broadcast coding module, and to receive wireless signals from different neighboring nodes, and to transmit the received signals to the decoding estimation module.

[0017] Thirdly, the present invention also provides a resource allocation device for a sensor-integrated network, applied to a converged central computing device; the device includes the following modules: The acquisition module is used to acquire the channel estimation parameters reported by each integrated sensing node; the channel estimation parameters are obtained by each integrated sensing node by decoding the received broadcast signal and using the decoded subcodewords as an auxiliary sequence for channel estimation; The resource allocation module is used to allocate communication data rates to each link based on the channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, using a greedy communication scheduling strategy, allocating the transmit power of each of the integrated sensing nodes using a strategy based on projection gradient descent, and determining the sub-codeword length of each link using an integer discrete search strategy, thereby obtaining a resource allocation strategy; the resource allocation strategy includes optimal communication data rate, optimal transmit power, and optimal sub-codeword length; The distribution module is used to distribute the resource allocation strategy to each of the integrated sensing nodes in order to control the communication and sensing resource allocation of each of the integrated sensing nodes.

[0018] Fourthly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the resource allocation method of any of the above-described sensor network.

[0019] Fifthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the resource allocation method of the integrated sensing network as described above.

[0020] In a sixth aspect, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the resource allocation method of any of the above-described sensor-integrated networks.

[0021] This invention provides a resource allocation method and system for a sensor-integrated network. The method is applied to a fusion center computing device. The method includes: first, acquiring channel estimation parameters reported by each sensor-integrated node. The channel estimation parameters are obtained by each sensor-integrated node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; then, based on each channel estimation parameter, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, allocating communication data rates to each link using a greedy communication scheduling strategy, allocating the transmit power of each sensor-integrated node using a strategy based on projective gradient descent, and determining the sub-codeword length of each link using an integer discrete search strategy, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, optimal transmit power, and optimal sub-codeword length; and finally, distributing the resource allocation strategy to each sensor-integrated node to control the allocation of communication and sensing resources for each sensor-integrated node.

[0022] This invention addresses the shortcomings of existing technologies by introducing data-assisted sensing with limited code length, network-level joint optimization using relative squared position error bound (rSPEB) as the sensing index, and a continuous-discrete hybrid optimization strategy. This achieves resource allocation that is realistic at the physical layer, collaborative at the network level, integrated with sensing, and deployable in engineering, significantly improving the communication capacity and positioning accuracy of the integrated sensing network in anchorless self-organizing network scenarios. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the resource allocation method for the integrated sensor network provided by the present invention.

[0025] Figure 2 This is a schematic diagram of communication measurement between any integrated sensing node and its neighboring nodes provided by the present invention.

[0026] Figure 3 This is a flowchart illustrating the network link communication load scheduling strategy provided by the present invention.

[0027] Figure 4 This is a schematic diagram of the node power allocation and subcodeword length allocation process provided by the present invention.

[0028] Figure 5 This is a comparative diagram of the network-level communication perception performance curves provided by the present invention.

[0029] Figure 6 This is a schematic diagram of the resource allocation system of the integrated sensor network provided by the present invention.

[0030] Figure 7 This is a schematic diagram of the integrated sensing node provided by the present invention.

[0031] Figure 8 This is a schematic diagram of the resource allocation device for the integrated sensor network provided by the present invention.

[0032] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0034] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, a first node can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0035] This invention provides a sensor-integrated network resource allocation system and method. First, through the collaborative work of multiple modules within the sensor-integrated nodes, signal transmission and reception, codeword segmentation and transmission, and accurate estimation of channel parameters are achieved. Simultaneously, relying on a fusion center computing device, the entire network parameters are collected and centrally optimized, achieving global coordination of resource allocation strategies. The proposed method employs a two-layer iterative optimization strategy combined with heuristic discrete search. First, greedy communication scheduling achieves accurate data rate allocation. Then, projection gradient descent is used to allocate node transmit power. Finally, discrete search determines the engineered sub-codeword length parameters, and iterative convergence ensures the optimality of the allocation strategy. Furthermore, the system and method of this invention are adaptable to anchorless multi-agent collaborative network scenarios, can adapt to communication characteristics under limited code length transmission, fully exploit the geometric collaborative gain of the global network, improve the efficiency of multi-node sensing resource scheduling, have high network adaptability, and can be stably and efficiently deployed in multi-agent collaborative networks.

[0036] The following is combined with Figures 1 to 9 The present invention describes a resource allocation method and system for a sensor-integrated network.

[0037] Figure 1 This is a flowchart illustrating the resource allocation method for an integrated sensing network provided by the present invention, which is applied to a convergence center computing device; such as Figure 1 As shown, the method includes the following: Step 101: Obtain the channel estimation parameters reported by each integrated sensing node; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters.

[0038] The execution subject of the resource allocation method for the integrated sensing network provided by the present invention can be the fusion center computing device in the resource allocation system of the integrated sensing network, or it can be the electronic device on the fusion center side.

[0039] In this integrated sensing network, each node is a single node within the network, comprising a transceiver antenna measurement module, a broadcast coding module, and a decoding and estimation module. The transceiver antenna module transmits and receives physical wireless signals; the broadcast coding module divides the transmitted total codeword into multiple orthogonal sub-codewords for transmission to different neighboring nodes; and the decoding and estimation module decodes the received sub-codewords and uses them as known auxiliary sequences for channel parameter estimation. Each node in the integrated sensing network broadcasts to each other, obtaining channel estimation parameters reported by each node. Specifically, after receiving a broadcast message from a neighboring node, each node extracts its corresponding assigned sub-codeword portion for decoding, uses a data-assisted mechanism to apply the decoding result (the decoded sub-codeword) to location parameter estimation, and sends the parameter estimation result (channel estimation parameters) back to the fusion center via the transceiver antenna measurement module. Through this method, any node in the integrated sensing network can complete communication and location parameter measurement with all neighboring nodes. The numerical results obtained from the above estimation process, such as link... k , j Channel coefficient estimates and its mean square error These parameters are the core input data for the fusion center to perform network positioning and resource allocation.

[0040] In practical applications, the channel estimation parameters obtained by the fusion center computing equipment mainly include the mean square error of channel estimation for each link and the channel gain (channel estimation), which are used in subsequent resource allocation.

[0041] Step 102: Based on the estimated parameters of each channel, with the goal of minimizing the relative squared position error bound, under the condition of finite code length transmission, a greedy communication scheduling strategy is used to allocate the communication data rate to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each integrated sensing node, and an integer discrete search strategy is used to determine the sub-codeword length of each link, thus obtaining the resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length.

[0042] Specifically, the channel estimation parameters (especially the mean square error of each link) received by the fusion center's computing equipment form the basis for network positioning performance evaluation. The fusion center uses these parameters to build an equivalent sensing model of the current network, and then calculates the sensitivity of this model to changes in different link parameters.

[0043] Channel estimation parameters (especially mean square error) serve as a bridge between "physical sensing" and "resource allocation." They are used in two ways: 1. Directly as a measure of sensing performance: determining the current relative positioning accuracy. 2. As a "sensitivity" indicator to the impact of communication load: predicting how much positioning accuracy will deteriorate if the data rate of a link is increased (i.e., the loss function in greedy scheduling).

[0044] The specific logic for using channel estimation parameters (detailed step-by-step explanation) is as follows: 1. Used to build a network positioning performance benchmark (establish optimization targets) The fusion center collects reports from all nodes. Then, utilize channel gain. Calculate the direction vector between nodes (Perceived geometric relationships). Utilizing mean square error Construct the Equivalent Fisher Information Matrix (EFIM).

[0045] For example, the fusion center computing device establishes an equivalent network model that includes network topology, transmit power constraints for each node, and network transmission capacity constraints. Based on the network transmission capacity constraints, a cost function is constructed with the objective of minimizing the relative squared position error bound, as follows: s.t. in, This indicates the communication data rate of link kj. For node transmit power, The subcodeword length of link kj, inverse function Indicates based on Calculate the link channel estimation error , This represents the position vector of each synesthetic node in the network. The Fisher Information Matrix (FIM) represents the location information contained in the channel estimate. This represents finding the trace of a matrix. The link communication weight of link kj (a parameter that characterizes the importance of each link communication in the network). This indicates network transmission capacity constraints. For the total power constraint of the network, This is the total codeword length. For a set of nodes, It is the set of neighboring nodes.

[0046] Purpose: To calculate the relative squared position error bound (rSPEB) under the current network state, with the goal of minimizing the rSPEB. The subsequent goal of the optimization algorithm is to further reduce this error bound when adjusting resources (power, code length).

[0047] 2. Used for greedy scheduling decisions (determining who to allocate communication load to). This is the first step in starting the algorithm (fixed power and code length, only the data rate is adjusted).

[0048] How to use: Calculating marginal impact: The algorithm calculates the loss function value for each link. That is, calculating when the link... kj estimation error If it gets a little worse, how much will the overall positioning accuracy deteriorate? When we want to give the link kj Increase communication data rate At that time, the channel will deteriorate. It will increase.

[0049] Logical chain: Input: Initial (Provided by reported parameters).

[0050] Decision: Choose the option that "increases the data rate and leads to..." The link that "increases and thus minimizes the negative impact on rSPEB" is the one that is added.

[0051] Effect: It ensures that the increase in communication capacity minimizes the damage to positioning accuracy.

[0052] 3. For compromise relation mapping under finite code length (establishing physical constraints) Under finite code length, communication data rate and estimation error It is a pair of contradictions (i.e.) function).

[0053] How to use: The fusion center knows the currently reported data. (Perceiving the current situation). When the algorithm wants to adjust the power or data rate, it must do so through... ↔ The relationship is used for verification.

[0054] Function: To ensure that the newly allocated power or data rate is physically achievable and that the changes in budget estimate error are real, rather than arbitrary mathematical alterations.

[0055] In practical applications, the fusion center calculates based on the estimated parameters of each channel, with the goal of minimizing the relative squared position error bound. Under the condition of finite code length transmission, it uses a greedy communication scheduling strategy to allocate communication data rates to each link, uses a strategy based on projection gradient descent to allocate the transmit power of each integrated sensing node, and uses an integer discrete search strategy to determine the sub-codeword length of each link, thus obtaining the resource allocation strategy.

[0056] Greedy communication scheduling strategy is an iterative resource allocation method for allocating communication data rates to each link. Under the premise of fixed node transmit power and subcodeword length, this strategy gradually increases network transmission capacity from zero to meet preset communication constraints by adding only a small amount of communication load to one link in each iteration, while minimizing the negative impact on network positioning performance caused by the increased load.

[0057] The core mechanisms include: Loss function: Defines the "marginal impact value" of each link, which represents the increase in the relative squared positional error bound of the network when a unit communication load is added to the link (i.e., the data rate is increased).

[0058] Greedy selection: In each iteration, select the link with the smallest marginal impact value from all unloaded links and increase its communication load.

[0059] Iteration termination: When the total network transmission capacity reaches a preset threshold (C0) C The iteration terminates when the number of iterations reaches 0, all links are fully loaded, or the maximum number of iterations is reached.

[0060] This strategy avoids the high complexity of exhaustive search, achieves near-optimal allocation of communication load on each link of the network with low computational overhead, and prioritizes the allocation of communication resources to the links that have the least impact on positioning accuracy, thereby protecting sensing performance while meeting communication needs.

[0061] Projected gradient descent is an optimization method for allocating transmit power among nodes, specifically designed for continuous variable optimization problems with total power constraints. This strategy updates the link power variables along the negative gradient direction of the objective function (relative squared position error bound) while keeping the communication data rate and subcodeword length fixed for each link. The updated result is then "projected" back into the feasible region that satisfies the total power constraint, ensuring the physical realizability of the allocation result.

[0062] The core mechanisms include: Gradient calculation: Calculate the relative squared position error bound for each link length-power product ( The partial derivative of ) is used to determine the direction in which the positioning error decreases the fastest.

[0063] Step size search: Use methods such as backtracking line search to determine a suitable iteration step size, so as to avoid the objective function from oscillating or diverging due to excessively large update step size.

[0064] Projection operation: Maps the power variables after gradient descent back to satisfy the total power constraint of the entire network. ) and minimum power constraints for each link ( The feasible region of the power allocation results is defined to ensure the executability of the power allocation results.

[0065] This strategy can efficiently handle non-convex optimization problems with linear constraints. Under the premise of ensuring that the total power does not exceed the network capacity, it tilts power resources toward links that contribute more to positioning accuracy, achieves a near-optimal solution for power allocation, and has a fast convergence speed, making it suitable for online computation.

[0066] Integer-based discrete search strategy is a discrete optimization method for determining the sub-codeword length of each synesthetic node. Because the sub-codeword length ( In a physical sense, it is an integer (e.g., the number of symbols occupied) and satisfies that the sum of the subcodeword lengths of each node equals the total codeword length. This strategy, under the premise of fixed communication data rate and node transmit power, uses discrete grid search to find the optimal combination of integer subcodeword lengths.

[0067] Core Mechanism: Real number relaxation solution: Relax the integer constraints to real number constraints, and use the gradient descent method to find the optimal solution for the subcodeword length of real numbers.

[0068] Neighborhood Discrete Search: Within the neighborhood of the optimal real-number solution (e.g., the combination of nearby integers after rounding), a coordinate descent method or a local enumeration method is used to search for feasible solutions that satisfy the integer constraints.

[0069] Goal-oriented: The goal is to maximize the transmission capacity contributed by each broadcast node, while ensuring that the perceived mean square error of each link does not exceed the preset lower bound (i.e., communication quality requirements).

[0070] This strategy addresses the practical engineering problem that codeword lengths must be integers, avoids the performance loss caused by rounding after relaxing integer variables, and approaches the theoretical upper bound of continuous optimization while ensuring the physical feasibility of the sub-codeword segmentation scheme.

[0071] In practical applications, the synergistic relationship of these three strategies is shown in Table 1 below: Table 1:

[0072] The final calculated resource allocation strategy refers to the optimal set of parameters output by the fusion center computing device after executing the above resource allocation method, including the optimal communication data rate, optimal transmit power, and optimal subcodeword length, as shown in Table 2 below: Table 2:

[0073] Step 103: Distribute the resource allocation strategy to each integrated sensing node to control the allocation of communication and sensing resources for each integrated sensing node.

[0074] Specifically, in the aforementioned integrated sensing network, each integrated sensing node allocates network resources according to resource allocation instructions provided by the fusion center. Specifically, the network resources to be allocated include the sub-codeword lengths of each link. Communication data rate of each link and the transmission power of each node. .

[0075] Subsequently, the resource allocation strategy can be encapsulated into standardized information, such as control signaling. The control signaling is then sent to each integrated sensing node to instruct the node to adjust its physical layer operating state in order to control the allocation of communication and sensing resources for each integrated sensing node. For example, a node may adjust the operating parameters of the broadcast encoder according to the control signaling, update the operating state of the antenna RF front end, or execute communication and sensing tasks according to the new strategy.

[0076] The method provided in this invention is applied to a converged computing center. The method includes: first, acquiring channel estimation parameters reported by each integrated sensing node. The channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; then, based on each channel estimation parameter, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, allocating communication data rates to each link using a greedy communication scheduling strategy, allocating the transmit power of each integrated sensing node using a strategy based on projection gradient descent, and determining the sub-codeword length of each link using an integer discrete search strategy to obtain a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; and then distributing the resource allocation strategy to each integrated sensing node to control the allocation of communication and sensing resources for each integrated sensing node.

[0077] The technical effects achieved by this invention are as follows: 1. Under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link. This strategy is based on the real trade-off between channel coding rate and bit error rate under finite code length (as characterized by DA communities), rather than the idealized Shannon capacity. Therefore, the allocated data rate is actually achievable by the physical layer in short packet transmission, avoiding resource waste or link unreliability caused by theoretical assumptions deviating from reality. This improves the matching degree between resource allocation results and actual physical layer transmission capabilities, ensuring the reliability and efficiency of communication links.

[0078] 2. With the optimization objective of "minimizing the relative squared position error bound (rSPEB)," a "greedy communication scheduling strategy" is employed to prioritize load allocation to links with the least impact, and a "projection gradient descent-based strategy" is used to allocate transmit power. These operations are guided by the sensitivity of rSPEB to link parameters, thus prioritizing the allocation of scarce power and codeword resources to links that contribute the most to the overall network positioning accuracy. This achieves coordinated scheduling of network-level sensing resources, fully leveraging the geometric complementarity between multiple nodes, and significantly improving relative positioning accuracy compared to single-link or single-objective optimization. The method of simultaneously determining communication parameters (communication data rate) and sensing parameters (transmit power, sub-codeword length) in the same optimization problem ensures that an increase in communication load affects positioning accuracy through channel estimation errors, while adjustments to power / codeword resources also inversely affect the communication data rate, forming a closed-loop coupling. This breaks the limitation of separating communication and sensing functions, achieving unified allocation of underlying signal processing resources, minimizing positioning errors while meeting communication requirements, and obtaining joint performance gains.

[0079] 3. The "relative squared position error bound" is adopted as the positioning index. This index naturally possesses rotation and translation invariance, making it suitable for relative positioning networks without absolute reference points. Furthermore, an "integer discrete search strategy" is used to determine the sub-codeword length, directly reflecting the requirement that codeword lengths must be integers in engineering practice. Power allocation, on the other hand, handles continuous variables through a "projection gradient descent-based strategy." This continuous-discrete hybrid optimization path, compared to exhaustive search or rounding after complete relaxation, has lower computational complexity and higher engineering feasibility. The technical effect is: it provides a resource allocation method specifically designed for anchor-free cooperative networks, capable of outputting the directly executable integer codeword length and continuous power value for each node without requiring prior absolute coordinates.

[0080] In summary, the technical solution of this invention addresses the shortcomings of existing technologies by introducing data-assisted sensing under finite code length, network-level joint optimization with relative squared position error bound (rSPEB) as the sensing index, and a continuous-discrete hybrid optimization strategy. This achieves resource allocation that is realistic at the physical layer, collaborative at the network level, integrated with sensing, and deployable in engineering, significantly improving the communication capacity and positioning accuracy of the integrated sensing network in anchorless self-organizing network scenarios.

[0081] It should be noted that each implementation method of this application can be freely combined, rearranged, or executed individually, and does not need to rely on or depend on a fixed execution order.

[0082] According to the resource allocation method for a sensing-integrated network provided by the present invention, the channel estimation parameters are determined through the following steps, including: Each integrated sensor node is controlled to split the total codeword to be sent into multiple orthogonal sub-codewords according to the link. Each sub-codeword corresponds to a neighbor node and is broadcast to different neighbor nodes respectively. Each integrated sensing node is controlled to extract and decode the sub-codewords assigned to it from the received broadcast signal. The decoded sub-codewords corresponding to each integrated sensing node are used as known auxiliary sequences for channel parameter estimation to obtain the channel estimation parameters of each integrated sensing node.

[0083] Specifically, in some embodiments, the channel estimation parameters are determined by the following steps: Each integrated sensing node divides the total codeword to be transmitted into multiple orthogonal sub-codewords according to the link. The total codeword to be transmitted refers to the physical layer signal sequence that a single integrated sensing node is prepared to send to all neighboring nodes within a complete broadcast cycle. The total codeword has a fixed symbol length (i.e., block length, denoted as ). N The total codeword (MCC) is the original data unit for subsequent codeword segmentation. The total codeword carries the set of message bits that the node sends to each of its neighboring nodes. A sub-codeword is a continuous sequence of symbols partitioned from the total codeword, specifically used to transmit messages to a particular neighboring node; that is, each sub-codeword corresponds to one neighboring node. Each sub-codeword occupies non-overlapping symbol positions within the total codeword, thus exhibiting orthogonality (orthogonality in the time domain or code domain), avoiding mutual interference between different links. The length of the sub-codeword (denoted as ) L kj () is a configurable resource allocation variable that satisfies the condition that the sum of the lengths of all sub-codewords on the same node equals the total codeword length. N .

[0084] In practical applications, each integrated sensing node broadcasts the resulting multiple orthogonal sub-codewords to different neighboring nodes. Specifically, a single integrated sensing node simultaneously transmits multiple orthogonal sub-codewords on the same time-frequency resource. Since the sub-codewords occupy different symbol positions in the total codeword (e.g., different time slots or different spreading codes), they are essentially orthogonally multiplexed. All neighboring nodes can receive the entire broadcast signal, but each neighboring node only needs to extract its assigned sub-codeword segment for decoding.

[0085] Furthermore, each sensing node extracts and decodes the sub-codewords assigned to it from the received broadcast signal. The decoded sub-codewords (i.e., successfully decoded communication data) for each sensing node are used as a known auxiliary sequence for channel parameter estimation, equivalent to a reference signal known to the receiver. In channel estimation, the reference signal known to both the transmitter and receiver is called the pilot sequence. Traditional methods require inserting pilot symbols separately into the transmitted signal. This invention's "communication data-assisted sensing" method eliminates the need for additional pilot overhead.

[0086] For example, Figure 2 This is a schematic diagram illustrating the communication measurement between any integrated sensing node and its neighboring nodes provided by this invention, achieved by all nodes taking turns broadcasting. For example... Figure 2 As shown, all nodes in the network constitute a node set. Node set It includes node k, neighbor node i, and neighbor node j Each node is equipped with a single transmit antenna and a single receive antenna. Any node k in the network is designated as the current broadcasting node, and its center coordinates in the local coordinate system are... Its neighbors constitute the set of neighbor nodes. Node k segments the broadcast codeword to be sent into... Individual character encoding, i.e. Each subcodeword carries a message sent by that node to its corresponding neighbor node; among them, the subcodeword The length is The data rate is Broadcast code The total transmit power is For broadcast sending nodes any neighboring node Its reception comes from the node The wireless signals have the same codeword segmentation form, that is... .node Extract the subcodeword fragment allocated to this link, and decode the corresponding allocated subcodeword. And based on the decoding results, data-assisted channel estimation is obtained. Similarly, nodes Extract the sub-codeword fragments allocated to this link, decode the corresponding allocated sub-codewords, and obtain data-assisted channel estimation based on the decoding results. The channel estimation results are uploaded to the fusion center for centralized calculation of network node locations.

[0087] It should be noted that during the communication process from node k to node i, another neighbor node of node k... It is also possible to obtain the subcodeword assigned by node k to node i. However, the received signal is not transmitted to the node. Decoding estimation module, node j The receive subcodeword is not needed for parameter estimation during network communication and positioning; however, the reception result will be stored on the node. j Locally, the signal is sent to the fusion center as needed after the communication measurement cycle ends for signal enhancement.

[0088] Furthermore, after a complete communication measurement cycle, any integrated sensing node completes communication and measurement with all neighboring nodes. At this point, the node... The available measurement data includes: the decoding results of the subcodewords sent to this node by all neighboring nodes, i.e. It enables information transmission between nodes; and channel estimation for all links of that node, i.e. This estimate is uploaded to the fusion center at the end of each communication cycle for network positioning.

[0089] The method provided in this invention controls each integrated sensing node to split the total codeword into orthogonal sub-codewords for broadcast transmission, and extracts the corresponding sub-codewords at the receiving end and decodes them, using them as known auxiliary sequences for channel estimation. This realizes the direct reuse and assistance of communication decoding results for the sensing process, avoiding additional pilot overhead and improving the channel estimation accuracy by utilizing the known signals after decoding, thereby deeply integrating communication and sensing functions at the physical layer.

[0090] According to a resource allocation method for a sensor-integrated network provided by the present invention, a greedy communication scheduling strategy is used to allocate communication data rates to each link, including: Fix the transmit power and subcodeword length of the current node in each integrated sensing node, and initialize the communication data rate of each link to zero; Calculate the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate, and determine the marginal impact value as the link's loss function value; the link's loss function value characterizes the degree of impact of increasing the link's coding rate on network positioning performance; Traverse all links in the network and mark links whose communication load has reached its limit as fully loaded links; Among the links that have never been marked as fully loaded, select the link with the smallest marginal impact value, allocate the communication load increment to the link with the smallest marginal impact value first, and update the communication data rate corresponding to the link with the smallest marginal impact value. Repeat the above allocation process until the total transmission capacity of all links meets the preset network communication constraints or reaches the maximum number of iterations, and obtain the communication data rate corresponding to each link.

[0091] Specifically, in some embodiments, the implementation of allocating communication data rates to each link using a greedy communication scheduling strategy includes the following steps: First, fix the transmit power and subcodeword length of the current node in each integrated sensing node, initialize the communication data rate of each link to zero, and record the iteration round number. The total number of algorithm iterations is .

[0092] Furthermore, the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate is calculated, and this marginal impact value is determined as the link's loss function value. The link's loss function value characterizes the degree of impact of increasing the link's coding rate on network positioning performance; that is, it characterizes the degree of impact on network positioning performance when, under the current state, the total network communication capacity is increased by a micro-element by increasing the link's coding rate.

[0093] Furthermore, iterate through all links in the network and mark links whose communication load has reached its limit as fully loaded links. Specifically, if the current link's communication load has reached its transmission resource limit... If the maximum load can be reached, the link will be marked as a full-load link, meaning no further communication load will be allocated to it.

[0094] Furthermore, among the links that were never marked as fully loaded, the link with the smallest marginal impact value (loss function value) is selected and denoted as . Prioritize allocating incremental communication load to the link with the smallest marginal impact. And update the communication data rate corresponding to the link with the smallest marginal impact value.

[0095] Specifically, the incremental element of the current network communication load is determined based on the total number of algorithm iterations. And update the link with that increment: in, The link with the minimum marginal impact value The communication data rate corresponding to the i-th iteration. The link with the minimum marginal impact value The communication data rate corresponding to the (i-1)th iteration This represents the incremental micro-element of the current network communication load. This indicates the upper limit of the communication load for this link; simultaneously, when the communication load of this link changes, its link channel estimation error is calculated and updated. The communication load of the remaining links remains unchanged.

[0096] Furthermore, the above allocation process is repeated until the total transmission capacity of all links meets the preset network communication constraints or the maximum number of iterations is reached, thus obtaining the communication data rate corresponding to each link. Specifically, the fusion center checks whether the currently allocated communication load of all links meets the total network constraints. or number of algorithm iterations Has the maximum number of iterations been reached? If none of these conditions are met, repeat the above steps to calculate the loss function value, mark fully loaded links, and update the communication data rate of the link with the smallest marginal impact value among the unloaded links, while simultaneously resetting the algorithm iteration rounds. If the conditions are met, the iteration process ends. Check the iteration termination condition. If the communication load does not meet the network constraints, distribute the remaining load to the links through average or random allocation. Note that the allocation result must not exceed the link's communication load limit, otherwise it may cause the network positioning function to fail.

[0097] For example, Figure 3 This is a flowchart illustrating the network link communication load scheduling strategy provided by the present invention, as shown below. Figure 3 As shown, the method includes: Step 31: Initialize the communication load of the network link; Step 32: Calculate the link loss function value under the current network communication load distribution; Step 33: Traverse all links in the network. If the current link's communication load reaches the current link's communication capacity limit, mark the link as fully loaded. Step 34: Select the link with the smallest loss function that is not marked as fully loaded, calculate the communication load increment element and update it to the link, and update the channel estimation mean square error of the link at the same time. The communication load of the remaining links remains unchanged. Step 35: Check whether the communication load meets the network constraints and whether the algorithm iteration count has reached the maximum number of iteration rounds; Step 36: Check the iteration termination condition. If the communication load does not meet the network constraints, the remaining load is distributed to the links by average distribution or random distribution.

[0098] The method provided in this invention gradually increases the load from zero after fixing the power and code length, calculates the marginal impact of increasing the load by one unit on the positioning accuracy of each link, and prioritizes allocating the load to the non-fully loaded links with the smallest marginal impact value. This achieves a near-optimal allocation of communication data rate with minimal positioning performance loss while satisfying network communication constraints, thereby obtaining a near-globally optimal trade-off between communication and perception with low computational complexity.

[0099] According to a resource allocation method for a sensor-integrated network provided by the present invention, the marginal impact of increasing the communication load per unit on the network positioning performance of each link at the current communication data rate is calculated, including: The loss function value of the link is calculated based on the derivative of the relative squared position error bound with respect to the mean square error of channel estimation, the derivative of the mean square error of channel estimation with respect to the data rate, and the derivative of the data rate with respect to the transmission capacity. The relative squared position error bound is calculated based on the mean square error of channel estimation for each link. The mean square error of channel estimation is related to the communication data rate of the link, the node transmit power, and the subcodeword length.

[0100] Specifically, in some embodiments, the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate is calculated, including: The link loss function value is calculated based on the derivatives of the relative squared position error bound with respect to the mean square error of channel estimation, the mean square error of channel estimation with respect to the data rate, and the data rate with respect to the transmission capacity. Specifically, calculate the link loss function value under the current network communication load distribution: in, for kj The link's loss function value. The relative squared position error bound for network positioning serves as an evaluation metric for the positioning performance of integrated sensing networks. It is derived from the mean square error of channel estimation for each link. Decide. for kj Link communication data rate, Network transmission capacity is an evaluation indicator for the total communication capacity of the integrated sensing network.

[0101] Due to the differences in the underlying information processing mechanisms of communication and sensing functions, when transmission resources are fixed, the higher the signal-coded data rate (communication data rate), the worse the channel estimation result, which can be expressed as: in, expresskj Mean square error of channel estimation for the link. for kj Link communication data rate, node transmit power Subcode word length This constitutes a constraint on transmission resources.

[0102] The relative squared position error bound is calculated based on the mean square error of channel estimation for each link. The mean square error of channel estimation is related to the communication data rate of the link, the node transmit power, and the subcodeword length.

[0103] The method provided in this invention quantifies the marginal impact of a link by utilizing the derivative of the relative squared position error bound with respect to the mean square error of channel estimation, the derivative of the mean square error of channel estimation with respect to the data rate, and the derivative of the data rate with respect to the transmission capacity. This achieves precise quantification of the degree of deterioration in positioning performance when each link increases by a unit of communication load, providing an accurate ranking basis for "prioritizing the link with the least marginal impact" in greedy scheduling, thereby theoretically ensuring that the loss of positioning performance due to communication load allocation is minimized.

[0104] According to the resource allocation method of the present invention for a sensor network, the method allocates the transmit power of each sensor node using a projection gradient descent strategy and determines the sub-codeword length of each link using an integer discrete search strategy, including: Fix the current communication data rate of each link, and initialize the node transmit power and subcodeword length; Determine the gradient direction of the loss function of the network under the node transmit power, and search the iteration step size by backtracking straight line method to initially update the node transmit power allocation; The initially updated node transmit power allocation is projected onto the feasible region that satisfies the total power constraint of the entire network and the minimum energy requirement of each link to obtain the optimized node transmit power allocation. With the optimized node transmit power fixed, and aiming to maximize the transmission capacity contributed by each broadcast node, the discrete grid descent method is used to search for the optimal subcodeword length for each link. Repeat the above steps until the improvement value of the positioning error is less than the threshold or the maximum number of iterations is reached, to obtain the transmit power allocated to each integrated sensing node and the subcodeword length of each link.

[0105] Specifically, in some embodiments, the transmit power of each integrated sensing node is allocated using a projection gradient descent-based strategy, and the sub-codeword length of each link is determined using an integer discrete search strategy through the following steps: First, fix the current communication data rate of each link, and initialize the node transmit power and subcodeword length as follows: in, Let k be the transmit power of node k. To constrain the total network transmit power, express The tall typewriter Represents a set of nodes.

[0106] Record the number of iterations The total number of algorithm iterations is Iteration threshold .

[0107] Furthermore, determine the gradient direction of the loss function of the network at the (current) node's transmit power: in, Let be the gradient of the loss function for the node's transmit power. This is the relative squared position error bound for network positioning. For link kj The mean square error of channel estimation, Let be the transmit power of node k.

[0108] The iteration step size is searched using the backtracking line method. Initial update of node transmit power allocation: in, For the initial update of node transmit power allocation, Allocate node transmit power for the (i-1)th round (the previous round). Let be the gradient of the loss function for the transmit power of the node in the i-th round. This is the iteration step size.

[0109] In this process, after updating the power allocation in favor of improving network positioning performance, the node power allocation results may not meet network power constraints. Therefore, the initial update results cannot be directly used to update network parameters. Furthermore, the initially updated node transmit power allocation is projected onto a feasible region that satisfies the total network power constraint and the minimum energy requirements of each link. To ensure that the network power constraints are met, the optimized node transmit power allocation is calculated.

[0110] Therefore, this invention designs the following node power optimization problem: in, Allocate final node power (optimized node transmit power). For the initial update of node transmit power allocation, For nodes kThe node's transmit power.

[0111] By solving the problem This ensures that the node power allocation meets the network's total power constraint, resulting in the final node power allocation (optimized node transmit power). .

[0112] Furthermore, the subcodeword length is redistributed through an integer-type discrete grid search. Specifically, the optimized node transmit power is fixed at... With the goal of maximizing the transmission capacity contributed by each broadcast node, the discrete grid descent method is used to search for the optimal subcodeword length for each link. .

[0113] Repeat the above steps until the improvement value of the positioning error is less than the threshold or the maximum number of iterations is reached, to obtain the transmit power allocated to each integrated sensing node and the subcodeword length of each link.

[0114] Specifically, the fusion center checks positioning errors. Has the improvement value compared to the previous round exceeded the threshold? Or, has the algorithm reached its maximum number of iterations? If none of these conditions are met, repeat the above steps (determining the gradient direction of the loss function, projecting the feasible region, and searching the discrete grid with integers), while simultaneously increasing the number of algorithm iterations. If the condition is met, the iteration process ends. Check the iteration termination condition. If the iteration process has not converged, you can choose to output the current result as the network node power and subcodeword length allocation, or continue to return to the step loss function gradient direction determination for iterative optimization.

[0115] For example, Figure 4 This is a schematic diagram of the node power allocation and subcodeword length allocation process provided by the present invention, as shown below. Figure 4 As shown, the method includes: Step 41: Initialize node power, with subcodeword length evenly distributed; Step 42: Determine the gradient direction of the loss function under the current network node power allocation, search the iteration step size by backtracking straight line method, and initially update the node power allocation; Step 43: Perform a projection operation on the initially updated node power allocation, and ensure that the node power allocation meets the total network power constraint based on the given node power optimization problem; Step 44: Reallocate the subcodeword length using an integer discrete grid search; Step 45: Check whether the improvement of the positioning error in two adjacent iterations exceeds the threshold, and whether the number of algorithm iterations has reached the maximum number of iterations; Step 46: Check the iteration termination condition. If the iteration process has not converged, the current node power and subcodeword length allocation results can be output, or iterative optimization can continue.

[0116] The method provided in this invention uses projection gradient descent to project power into the feasible region after fixing the data rate to meet the total power and minimum energy constraints. Then, it optimizes the integer subcodeword length through discrete grid search, achieving efficient coordination between continuous power optimization and discrete codeword length optimization. This ensures that the power and codeword allocation results satisfy both physical realizability and approach the joint optimal communication sensing performance.

[0117] According to a resource allocation method for a sensor-integrated network provided by the present invention, a resource allocation strategy is distributed to each sensor-integrated node to control the allocation of communication and sensing resources for each sensor-integrated node, including: Encapsulate resource allocation strategies into control signaling; Control signals are sent to each integrated sensing node via a wireless communication network, allowing each node to adjust the sub-codeword segmentation method, transmission power, and encoding data rate of the broadcast encoder according to the control signals.

[0118] Specifically, in some embodiments, step 103, which involves distributing the resource allocation strategy to each integrated sensing node, is achieved through the following steps: The resource allocation strategy refers to the optimal set of parameters output by the convergence center computing device after executing the resource allocation method. It includes at least: the optimal transmit power of each integrated sensing node, the sub-codeword length of each link, and the optimal communication data rate of each link. The sub-codeword length of each link includes the sub-codeword length allocated by the integrated sensing node to each neighboring link, and the sum of all sub-codeword lengths of the same node equals the total codeword length. These parameters collectively determine the transmission behavior of network nodes in the next communication measurement cycle. Control signaling is a formatted message generated by the convergence center computing device and broadcast or unicast to each integrated sensing node via the wireless communication network.

[0119] In practical applications, the fusion center computing device generates independent control signaling for each integrated sensing node in the network based on the solved optimal parameter set, or generates broadcast control signaling containing the configurations of all nodes. Its content encapsulates resource allocation strategies according to predefined protocol fields. The control signaling can adopt the physical layer downlink control information format or the higher-layer radio resource control signaling format; this embodiment does not impose any restrictions.

[0120] For example, for each integrated sensing node k, the control signaling includes at least the following information fields: Node identifiers, such as node identification (ID) or physical address; The total transmit power of this node, or the power of each link; The subcodeword length (integer, in sign count) of each neighbor link of this node. The encoded data rate that each neighboring link of this node should use.

[0121] In addition, optional fields such as timestamp, signaling validity period, checksum or cyclic redundancy check code can be included to ensure the reliability of signaling transmission.

[0122] Furthermore, control signals are sent to each integrated sensing node via a wireless communication network, so that each integrated sensing node can adjust the sub-codeword segmentation method, transmission power, and encoding data rate of the broadcast encoder according to the control signals.

[0123] The broadcast encoder is located within the broadcast coding module of each integrated sensing node. Its main function is to divide the total message bits to be transmitted into blocks based on the sub-codeword length and coding data rate carried in the control signaling, encode each block into different orthogonal sub-codewords, and output the result to the transceiver antenna measurement module after power scaling according to a specified transmit power. The sub-codeword segmentation method refers to how the broadcast encoder allocates codewords of a fixed total length (e.g., N=100 symbols) to different neighboring links. Specifically, it represents the sub-codeword length allocated to each link. The transmit power is the total power allocated to each integrated sensing node for its total transmitted signal, or further refined into the transmit power allocated to the sub-codewords of each link. In this embodiment, the control signaling can carry the total node power, and the node can then automatically allocate link power according to the sub-codeword length ratio or a preset rule; alternatively, it can directly carry the power value for each link. The coding data rate is the number of information bits carried per symbol on each link (unit: bits / symbol), determined by the modulation and coding scheme. The control signaling can indicate a modulation and coding index or directly provide the data rate value.

[0124] In practical applications, the fusion center distributes encapsulated control signals to each integrated sensing node via its wireless communication interface, such as a 4G / 5G module, a WiFi module, or a dedicated UAV communication link. The distribution can be unicast or broadcast. Each integrated sensing node receives the control signals through its transceiver antenna measurement module and then transmits them to its internal broadcast encoding module. The broadcast encoding module parses the signals, extracting the node's transmit power, subcodeword length, and encoded data rate. Subsequently, each integrated sensing node adjusts its operating parameters according to the signals, such as adjusting the subcodeword segmentation method, transmit power, and encoded data rate. After completing parameter adjustments, each integrated sensing node enters the next communication measurement cycle. The broadcast encoding module generates a broadcast signal according to the new subcodeword segmentation method, transmit power, and data rate, which is then transmitted by the transceiver antenna measurement module. Meanwhile, the decoding estimation module also slices, decodes, and estimates channel parameters of the received signal based on the same subcodeword configuration (the subcodeword positions of each link are known), thereby achieving the desired integrated sensing performance of the fusion center.

[0125] The method provided in this invention encapsulates the resource allocation strategy into standardized control signals and sends them to each integrated sensing node. This enables each integrated sensing node to accurately adjust the sub-codeword segmentation method, transmission power, and encoding data rate of its broadcast encoder according to the optimal calculation results of the fusion center. As a result, the joint optimization strategy is actually executed at the physical layer, realizing the reliable transformation of theoretically optimal resource allocation into engineering-executable control instructions.

[0126] Figure 5 This is a comparative schematic diagram of the network-level communication awareness performance curves provided by the present invention, such as... Figure 5 As shown, the horizontal axis (X-axis) represents the lower bound (LB) of the relative square position, in square meters (m). 2The vertical axis (Y-axis) reflects the error level of location awareness (the larger the value, the lower the location awareness accuracy). The vertical axis represents the network communication transmission capacity (CT), measured in bits / ch.use (the number of bits used per channel), measuring the transmission capacity of the communication system (the larger the value, the higher the communication capacity). The three curves in the figure correspond to different communication scheduling strategies and power-codeword length allocation methods, combined with the total network power constraint (solid line: 40 dB; dashed line: 80 dB). The red dotted curve represents weighted communication scheduling + node transmission power and sub-codeword length allocation based on an average strategy (solid line: 40 dB; dashed line: 80 dB). The green cross curve represents greedy communication scheduling + node transmission power and sub-codeword length allocation based on an average strategy (solid line: 40 dB; dashed line: 80 dB). The blue triangular curve represents greedy communication scheduling + node transmission power and sub-codeword length allocation based on projective gradient descent and discrete search (solid line: 40 dB; dashed line: 80 dB) (the allocation method of this invention). The figure clearly illustrates the impact of power constraints, communication scheduling strategies, and power-codeword allocation methods on the performance of the "communication-sensing integrated" system through multi-dimensional comparisons on the horizontal axis: increasing power constraints can improve capacity; greedy scheduling is superior to weighted scheduling; advanced power-codeword allocation methods (projective gradient descent + discrete search) can significantly enhance the performance of greedy scheduling; as the position error bound increases, the overall communication capacity increases (reduced sensing accuracy may alleviate communication resource competition). Figure 5 As shown, in both low-power scenarios (dashed lines) and high-power scenarios (solid lines), the network performance of the resource allocation method provided by this invention is superior to that of conventional resource allocation methods, verifying the effectiveness of the resource allocation method for the integrated sensor network provided by this invention.

[0127] Figure 6 This is a schematic diagram of the resource allocation system of the integrated sensor network provided by the present invention, as shown below. Figure 6 As shown, the resource allocation system 600 of the integrated sensor network includes the following modules: Multiple integrated sensing nodes 610, each of the integrated sensing nodes 610 includes a broadcast coding module and a decoding estimation module; the broadcast coding module is used to divide the total codeword to be transmitted into multiple orthogonal sub-codewords for broadcasting to different neighboring nodes; the decoding estimation module is used to decode the received sub-codewords and use the decoded sub-codewords as a known auxiliary sequence for channel parameter estimation and extraction; A converged central computing device 620 is connected to each of the integrated sensing nodes via a wireless communication network. It is used to collect channel estimation parameters reported by each integrated sensing node and execute the resource allocation method of the integrated sensing network described in any one of the claims to control the allocation of communication and sensing resources for each integrated sensing node.

[0128] Specifically, such as Figure 6 As shown, the resource allocation system 600 of the sensor network includes multiple sensor nodes 610 and a fusion central computing device 620. The multiple sensor nodes 610 include sensor node 1, sensor node 2, sensor node 3, ..., sensor node n. Figure 7 This is a schematic diagram of the integrated sensing node provided by the present invention, as shown below. Figure 7 As shown, each integrated sensing node includes a broadcast encoding module and a decoding estimation module.

[0129] In practical applications, the channel estimation parameters reported by the integrated sensing node are implemented in the following way: the broadcast coding module is used to divide the total codeword to be transmitted into multiple orthogonal sub-codewords for broadcasting to different neighboring nodes; the decoding estimation module is used to decode the received sub-codewords and use the decoded sub-codewords as a known auxiliary sequence to estimate and extract the channel parameters, thereby obtaining the channel estimation parameters reported by the integrated sensing node.

[0130] The fusion center computing device 620 is connected to each of the integrated sensing nodes via a wireless communication network, and is used to collect the channel estimation parameters reported by each of the integrated sensing nodes, and execute the resource allocation method of the integrated sensing network described in any one of the claims, so as to control the communication and sensing resource allocation of each of the integrated sensing nodes.

[0131] It should be noted that the resource allocation method and technical effects of the integrated sensor network executed by the fusion central computing device 620 are consistent with those of the method embodiment, and will not be repeated here.

[0132] The system provided in this embodiment of the invention includes multiple integrated sensing nodes 610 and a fusion center computing device 620. Through codeword segmentation and data-assisted sensing structure on the node side, and a limited code length joint optimization method on the fusion center side, it achieves efficient unified scheduling of communication and sensing resources in an anchorless cooperative network, improves the overall network performance, and has good practicality and promotion value.

[0133] According to the resource allocation system 600 of the integrated sensing network provided by the present invention, each of the integrated sensing nodes 610 further includes a transceiver antenna measurement module; The transceiver antenna measurement module is used to transmit wireless signals according to the communication measurement instructions issued by the broadcast coding module, and to receive wireless signals from different neighboring nodes, and to transmit the received signals to the decoding estimation module.

[0134] Specifically, see Figure 7 As shown, Figure 7This is a structural schematic diagram of the integrated sensing node provided by the present invention. Each integrated sensing node also includes a transceiver antenna measurement module. The core function of the transceiver antenna measurement module is the transmission and reception of physical layer wireless signals and the interaction of commands and data with other modules. Specifically: ① Transmit function: Receive communication measurement commands from the broadcast coding module, and transmit broadcast signals to neighboring nodes according to the parameters such as node transmit power and subcodeword length specified in the command.

[0135] ② Receiving function: Receives wireless signals transmitted from other integrated sensing nodes and transmits the received raw signals to the decoding and estimation module for subsequent decoding and channel parameter estimation.

[0136] ③ Interactive function: It can communicate bidirectionally with the broadcast encoding module, decoding estimation module and fusion center through serial port and other interfaces to complete the issuance of measurement commands and the reporting of measurement data.

[0137] In short, this module is the only physical layer front-end in the integrated sensing node that directly interacts with the wireless channel. It is responsible for signal transmission and reception and serves as a bridge connecting baseband processing and radio frequency antenna.

[0138] The system provided in this embodiment of the invention includes a transceiver antenna measurement module in the integrated sensing node 610. By setting the transceiver antenna measurement module, the integrated sensing node 610 realizes the decoupling of command transmission and signal transmission and reception between the broadcast coding module and the physical antenna, ensuring the reliable transmission of orthogonal subcodewords and the complete transmission of received signals to the decoding estimation module.

[0139] The resource allocation device for the integrated sensor network provided by the present invention will be described below. The resource allocation device for the integrated sensor network described below can be referred to in correspondence with the resource allocation method for the integrated sensor network described above.

[0140] Figure 8 This is a schematic diagram of the resource allocation device for the integrated sensor network provided by the present invention, as shown below. Figure 8 As shown, the resource allocation device 800 for the integrated sensor network is applied to the convergence center computing equipment; the resource allocation device 800 for the integrated sensor network includes the following modules: The acquisition module 810 is used to acquire the channel estimation parameters reported by each integrated sensing node; the channel estimation parameters are obtained by each integrated sensing node by decoding the received broadcast signal and using the decoded subcodeword as an auxiliary sequence for channel estimation; The resource allocation module 820 is used to allocate communication data rates to each link based on the channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, using a greedy communication scheduling strategy, allocating the transmit power of each of the integrated sensing nodes using a projection gradient descent strategy, and determining the sub-codeword length of each link using an integer discrete search strategy, thereby obtaining a resource allocation strategy; the resource allocation strategy includes optimal communication data rate, optimal transmit power, and optimal sub-codeword length; The distribution module 830 is used to distribute the resource allocation strategy to each of the integrated sensing nodes in order to control the communication and sensing resource allocation of each of the integrated sensing nodes.

[0141] According to the resource allocation device 800 for a sensor-integrated network provided by the present invention, the channel estimation parameters are determined through the following steps, including: Each of the aforementioned integrated sensing nodes is controlled to split the total codeword to be transmitted into multiple orthogonal sub-codewords according to the link. Each sub-codeword corresponds to a neighbor node, and is broadcast to different neighbor nodes respectively. Each of the integrated sensing nodes is controlled to extract and decode the sub-codewords allocated to each of the integrated sensing nodes from the received broadcast signal. The decoded sub-codewords corresponding to each integrated sensing node are used as the known auxiliary sequence for channel parameter estimation to obtain the channel estimation parameters of each integrated sensing node.

[0142] According to the present invention, a resource allocation device 800 for a sensor-integrated network is provided, wherein the resource allocation module 820 is specifically used for: Fix the transmit power and subcodeword length of the current node in each of the aforementioned integrated sensing nodes, and initialize the communication data rate of each link to zero; Calculate the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate, and determine the marginal impact value as the loss function value of the link; the loss function value of the link characterizes the degree of impact of increasing the coding rate of the link on network positioning performance; Traverse all links in the network and mark the links whose communication load has reached its limit as fully loaded links; Select the link with the smallest marginal impact value from the links that have never been marked as fully loaded, allocate the communication load increment to the link with the smallest marginal impact value first, and update the communication data rate corresponding to the link with the smallest marginal impact value. Repeat the above allocation process until the total transmission capacity of all links meets the preset network communication constraints or reaches the maximum number of iterations, and obtain the communication data rate corresponding to each link.

[0143] According to the present invention, a resource allocation device 800 for a sensor-integrated network is provided, wherein the resource allocation module 820 is further configured to: The loss function value of the link is calculated based on the derivative of the relative squared position error bound with respect to the mean square error of channel estimation, the derivative of the mean square error of channel estimation with respect to the data rate, and the derivative of the data rate with respect to the transmission capacity. The relative squared position error bound is calculated based on the mean square error of channel estimation for each link. The mean square error of channel estimation is related to the communication data rate, node transmit power, and subcodeword length of the link.

[0144] According to the present invention, a resource allocation device 800 for a sensor-integrated network is provided, wherein the resource allocation module 820 is further configured to: Fix the current communication data rate of each link, and initialize the node transmit power and subcodeword length; Determine the gradient direction of the loss function of the network under the node's transmit power, and search the iteration step size by backtracking straight line method to initially update the node transmit power allocation; The initially updated node transmit power allocation is projected onto the feasible region that satisfies the total power constraint of the entire network and the minimum energy requirement of each link to obtain the optimized node transmit power allocation. With the optimized node transmit power fixed, and aiming to maximize the transmission capacity contributed by each broadcast node, the optimal subcodeword length for each link is searched using the discrete grid descent method; the optimal subcodeword length for each link corresponds to the subcodeword length of each link. Repeat the above steps until the improvement value of the positioning error is less than the threshold or the maximum number of iterations is reached, to obtain the transmit power allocated to each of the integrated sensing nodes and the subcodeword length of each of the links.

[0145] According to the resource allocation device 800 of the integrated sensing network provided by the present invention, the sending module 830 is specifically used for: The resource allocation strategy is encapsulated as control signaling; The control signaling is transmitted to each of the integrated sensing nodes via a wireless communication network, so that each integrated sensing node can adjust the sub-codeword segmentation method, transmission power, and encoding data rate of the broadcast encoder according to the control signaling.

[0146] Figure 9 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 9As shown, the electronic device may include: a processor 910, a communications interface 920, a memory 930, and a communication bus 940, wherein the processor 910, the communications interface 920, and the memory 930 communicate with each other via the communication bus 940. The processor 910 can call logical instructions in the memory 930 to execute a resource allocation method for a converged computing center, which is applied to a converged computing center. The method includes: The channel estimation parameters reported by each integrated sensing node are obtained; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; Based on the aforementioned channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each of the aforementioned integrated sensing nodes, and an integer discrete search strategy is used to determine the sub-codeword length of each of the aforementioned links, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; The resource allocation strategy is distributed to each of the integrated sensing nodes to control the allocation of communication and sensing resources for each integrated sensing node.

[0147] Furthermore, the logical instructions in the aforementioned memory 930 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0148] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the resource allocation method for the integrated sensing network provided by the above methods, the method being applied to a convergence center computing device; the method includes: The channel estimation parameters reported by each integrated sensing node are obtained; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; Based on the aforementioned channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each of the aforementioned integrated sensing nodes, and an integer discrete search strategy is used to determine the sub-codeword length of each of the aforementioned links, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; The resource allocation strategy is distributed to each of the integrated sensing nodes to control the allocation of communication and sensing resources for each integrated sensing node.

[0149] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the resource allocation method for the integrated sensing network provided by the above methods, the method being applied to a convergence center computing device; the method includes: The channel estimation parameters reported by each integrated sensing node are obtained; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; Based on the aforementioned channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each of the aforementioned integrated sensing nodes, and an integer discrete search strategy is used to determine the sub-codeword length of each of the aforementioned links, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; The resource allocation strategy is distributed to each of the integrated sensing nodes to control the allocation of communication and sensing resources for each integrated sensing node.

[0150] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0151] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A resource allocation method for a sensor-integrated network, characterized in that, Applied to a converged computing center; the method includes: The channel estimation parameters reported by each integrated sensing node are obtained; the channel estimation parameters are obtained by each integrated sensing node by decoding the received sub-codewords and using the decoded sub-codewords as a known auxiliary sequence to estimate the channel parameters; Based on the aforementioned channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, a greedy communication scheduling strategy is used to allocate communication data rates to each link, a strategy based on projection gradient descent is used to allocate the transmit power of each of the aforementioned integrated sensing nodes, and an integer discrete search strategy is used to determine the sub-codeword length of each of the aforementioned links, thus obtaining a resource allocation strategy; the resource allocation strategy includes the optimal communication data rate, the optimal transmit power, and the optimal sub-codeword length; The resource allocation strategy is distributed to each of the integrated sensing nodes to control the allocation of communication and sensing resources for each integrated sensing node.

2. The resource allocation method for a sensor-integrated network according to claim 1, characterized in that, The channel estimation parameters are determined through the following steps: Each of the aforementioned integrated sensing nodes is controlled to split the total codeword to be transmitted into multiple orthogonal sub-codewords according to the link. Each sub-codeword corresponds to a neighbor node, and is broadcast to different neighbor nodes respectively. Each of the integrated sensing nodes is controlled to extract and decode the sub-codewords allocated to each of the integrated sensing nodes from the received broadcast signal. The decoded sub-codewords corresponding to each integrated sensing node are used as the known auxiliary sequence for channel parameter estimation to obtain the channel estimation parameters of each integrated sensing node.

3. The resource allocation method for a sensor-integrated network according to claim 1, characterized in that, The method of allocating communication data rates to each link using a greedy communication scheduling strategy includes: Fix the transmit power and subcodeword length of the current node in each of the aforementioned integrated sensing nodes, and initialize the communication data rate of each link to zero; Calculate the marginal impact of increasing the communication load per unit on network positioning performance for each link at the current communication data rate, and determine the marginal impact value as the loss function value of the link; the loss function value of the link characterizes the degree of impact of increasing the coding rate of the link on network positioning performance; Traverse all links in the network and mark links whose communication load has reached its limit as fully loaded links; Select the link with the smallest marginal impact value from the links that have never been marked as fully loaded, allocate the communication load increment to the link with the smallest marginal impact value first, and update the communication data rate corresponding to the link with the smallest marginal impact value. Repeat the above allocation process until the total transmission capacity of all links meets the preset network communication constraints or reaches the maximum number of iterations, and obtain the communication data rate corresponding to each link.

4. The resource allocation method for a sensor-integrated network according to claim 3, characterized in that, The calculation of the marginal impact of each link increasing the communication load by one unit on network positioning performance at the current communication data rate includes: The loss function value of the link is calculated based on the derivative of the relative squared position error bound with respect to the mean square error of channel estimation, the derivative of the mean square error of channel estimation with respect to the data rate, and the derivative of the data rate with respect to the transmission capacity. The relative squared position error bound is calculated based on the mean square error of channel estimation for each link. The mean square error of channel estimation is related to the communication data rate, node transmit power, and subcodeword length of the link.

5. The resource allocation method for a sensor-integrated network according to claim 1, characterized in that, The method of allocating the transmit power of each of the integrated sensing nodes using a strategy based on projected gradient descent, and determining the sub-codeword length of each link using an integer discrete search strategy, includes: Fix the current communication data rate of each link, and initialize the node transmit power and subcodeword length; Determine the gradient direction of the loss function of the network under the node's transmit power, and search the iteration step size by backtracking straight line method to initially update the node transmit power allocation; The initially updated node transmit power allocation is projected onto the feasible region that satisfies the total power constraint of the entire network and the minimum energy requirement of each link to obtain the optimized node transmit power allocation. With the optimized node transmit power fixed, and aiming to maximize the transmission capacity contributed by each broadcast node, the optimal subcodeword length for each link is searched using the discrete grid descent method. Repeat the above steps until the improvement value of the positioning error is less than the threshold or the maximum number of iterations is reached, to obtain the transmit power allocated to each of the integrated sensing nodes and the subcodeword length of each of the links.

6. The resource allocation method for a sensor-integrated network according to claim 1, characterized in that, The step of distributing the resource allocation strategy to each of the integrated sensing nodes to control the communication and sensing resource allocation of each integrated sensing node includes: The resource allocation strategy is encapsulated as control signaling; The control signaling is transmitted to each of the integrated sensing nodes via a wireless communication network, so that each integrated sensing node can adjust the sub-codeword segmentation method, transmission power, and encoding data rate of the broadcast encoder according to the control signaling.

7. A resource allocation system for a sensor-integrated network, characterized in that, include: Multiple integrated sensing nodes, each of which includes a broadcast encoding module and a decoding estimation module; The broadcast encoding module is used to divide the total codeword to be sent into multiple orthogonal sub-codewords for broadcasting to different neighboring nodes; The decoding estimation module is used to decode the received sub-codewords and use the decoded sub-codewords as a known auxiliary sequence to estimate and extract channel parameters; A converged central computing device, which is connected to each of the integrated sensing nodes via a wireless communication network, is used to collect channel estimation parameters reported by each of the integrated sensing nodes and execute the method as described in any one of claims 1 to 6 to control the allocation of communication and sensing resources for each of the integrated sensing nodes.

8. The resource allocation system for the integrated sensor network according to claim 7, characterized in that... Each of the aforementioned integrated sensing nodes also includes a transceiver antenna measurement module; The transceiver antenna measurement module is used to transmit wireless signals according to the communication measurement instructions issued by the broadcast coding module, and to receive wireless signals from different neighboring nodes, and to transmit the received signals to the decoding estimation module.

9. A resource allocation device for a sensor-integrated network, characterized in that, Applied to a converged computing center; the device includes: The acquisition module is used to acquire the channel estimation parameters reported by each integrated sensing node; the channel estimation parameters are obtained by each integrated sensing node by decoding the received broadcast signal and using the decoded subcodewords as an auxiliary sequence for channel estimation; The resource allocation module is used to allocate communication data rates to each link based on the channel estimation parameters, with the goal of minimizing the relative squared position error bound, under finite code length transmission conditions, using a greedy communication scheduling strategy, allocating the transmit power of each of the integrated sensing nodes using a strategy based on projection gradient descent, and determining the sub-codeword length of each link using an integer discrete search strategy, thereby obtaining a resource allocation strategy; the resource allocation strategy includes optimal communication data rate, optimal transmit power, and optimal sub-codeword length; The distribution module is used to distribute the resource allocation strategy to each of the integrated sensing nodes in order to control the communication and sensing resource allocation of each of the integrated sensing nodes.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the resource allocation method of the integrated sensor network as described in any one of claims 1 to 6.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the resource allocation method of the integrated sensor network as described in any one of claims 1 to 6.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the resource allocation method of the integrated sensor network as described in any one of claims 1 to 6.

Images