A wireless sensor network architecture and performance optimization method and system based on a smart city

Abstract

The present application belongs to the technical field of wireless sensor network, and discloses a wireless sensor network architecture based on smart city and a performance optimization method. The present application adds a 6LoWPAN adaptation layer in the network infrastructure to realize the interconnection of heterogeneous nodes and IPv6 network; a multi-dimensional quantification grading model of heterogeneous nodes containing five indexes of computing capacity, residual energy, communication parameter and hardware specification is constructed to complete the pre-filtering of low-level nodes in the routing stage; non-uniform heterogeneous nodes of different scenes of smart city are deployed, and the transmission link is optimized by combining link failure prediction and topology self-healing strategy; the scene adaptive 6LoWPAN dynamic header compression algorithm and the dynamic data fusion mechanism based on node density and link signal-to-noise ratio are adopted to relieve the transmission pressure; finally, the network performance is comprehensively evaluated from multiple dimensions. The present application solves the problems of unprogrammed node deployment, poor heterogeneous adaptability and large data transmission volume, and shows good performance and applicability in different scenes of smart city.

Description

Technical Field

[0001] This invention belongs to the field of wireless sensor network technology, and in particular relates to a wireless sensor network architecture and performance optimization method based on smart cities. Background Technology

[0002] Wireless sensor networks (WSNs), as the core of the IoT sensing layer, are widely used in smart city infrastructure monitoring, environmental sensing, and traffic management. With the continuous growth in the number of IoT access nodes, IPv6-based low-power wireless personal area network (6LoWPAN) technology has become a key solution for large-scale interconnection of heterogeneous nodes. 6LoWPAN introduces an adaptation layer between the network layer and the data link layer to achieve header compression and fragmentation reassembly of IPv6 data packets, thereby supporting efficient communication between nodes on short-frame, long-link systems such as IEEE 802.15.4.

[0003] Currently, wireless sensor network architectures and performance optimization schemes typically employ a traditional five-layer network protocol stack structure: application layer, transport layer, network layer, data link layer, and physical layer. In practical implementations, routing protocols based on a single metric (such as minimum hop count, link quality, or node remaining energy) are often used for node connection paths, such as the RPL protocol. For data processing, static sampling intervals or simple data aggregation are commonly used to reduce redundancy. Regarding node deployment strategies, existing solutions often employ gridded uniform deployment or random distribution to adapt to general data acquisition needs.

[0004] However, the aforementioned solutions have significant shortcomings when facing the complex and diverse application scenarios of smart cities. First, at the architectural level, there is a lack of a dedicated cross-layer coordination mechanism between the traditional physical layer and the data link layer. The minimum length of an IPv6 packet is 1280 bytes, while the maximum payload of an IEEE 802.15.4 frame is only 127 bytes. When dealing with this "length mismatch," existing solutions mainly rely on a fixed fragmentation and reassembly process at the network layer, lacking a mechanism for dynamic header compression efficiency control for different service types. This results in low transmission efficiency and makes it difficult to form an efficient adaptation layer closed loop on low-bandwidth links. Second, in terms of node deployment strategies, traditional uniform or random deployment methods cannot simultaneously meet the needs of infrastructure construction for high coverage and environmental monitoring for low density and low interference, making it difficult to avoid network coverage gaps or communication conflicts in specific scenarios. At the same time, existing routing decisions do not perform multi-dimensional quantitative classification of heterogeneous nodes, resulting in insufficient accuracy in path selection.

[0005] Furthermore, existing routing algorithms based on a single path metric suffer from significant parameter coupling defects in their routing decision-making mechanisms. For example, simply selecting the node with the highest remaining energy may lead to poor link quality, while simply selecting the node with the best link quality may exacerbate node energy consumption due to frequent rerouting. Lacking a weighted scoring mechanism that integrates multiple dimensions such as coverage quality, transmission success rate, arrival reliability, node energy consumption, and end-to-end latency, existing solutions cannot effectively balance the differentiated requirements of various smart city services (such as real-time alarms and environmental monitoring) for path selection strategies. In addition, the network topology lacks link failure prediction and self-healing capabilities, resulting in insufficient network robustness and resilience in complex electromagnetic environments and node failure scenarios within smart cities.

[0006] Finally, in terms of data fusion and energy-saving optimization, existing solutions mostly adopt fixed data removal methods and lack refined management based on dynamic fusion thresholds. They cannot guide energy-saving decisions by quantitatively assessing changes in the number of data packets before and after fusion, leading to ineffective data transmission and energy waste. At the same time, overall network optimization lacks a closed-loop verification system with multi-dimensional performance indicators, making it difficult to achieve efficient technical effects of coordinated optimization of coverage, success rate, energy consumption, and latency. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides a wireless sensor network architecture and performance optimization method based on smart cities.

[0008] This invention is implemented as follows: a wireless sensor network architecture and performance optimization method based on smart cities. The wireless sensor network architecture and performance optimization method include: constructing a wireless sensor network architecture with an added 6LoWPAN adaptation layer; completing the interconnection of heterogeneous nodes with the IPv6 network based on 6LoWPAN technology; selecting front-end nodes based on the node heterogeneity quantization score; deploying non-uniform heterogeneous nodes for different smart city scenarios; calculating the weighted total score and connecting node paths using a topology self-healing mechanism; performing 6LoWPAN dynamic compression and dynamic data fusion; and evaluating network performance indicators from multiple dimensions.

[0009] Furthermore, the wireless sensor network architecture and performance optimization method include the following steps:

[0010] Step 1: Construct a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer;

[0011] Step 2: Connect heterogeneous nodes to the IPv6 network using 6LoWPAN technology;

[0012] Step 3: Select the preceding nodes based on the node heterogeneity quantization score;

[0013] Step 4: Deploy non-uniform heterogeneous nodes for different smart city scenarios;

[0014] Step 5: Calculate the weighted total score and combine it with the topological self-healing mechanism to connect the node paths;

[0015] Step six: Perform 6LoWPAN dynamic compression and dynamic data fusion;

[0016] Step 7: Evaluate network performance metrics from multiple dimensions.

[0017] Furthermore, the wireless sensor network architecture for constructing the newly added 6LoPWAN adaptation layer in step one includes:

[0018] A traditional five-layer network protocol stack framework for wireless sensor networks is constructed, and a 6LoWPAN adaptation layer is added between the physical layer and the data link layer to be responsible for IPv6 header compression, fragmentation and reassembly, and dynamic compression rate adjustment.

[0019] Furthermore, step two, which involves interconnecting heterogeneous nodes with the IPv6 network using 6LoWPAN technology, includes:

[0020] Compression efficiency for:

[0021] (1)

[0022] in, For compression efficiency, This is the length of the head after compression. This is the original IPv6 header length.

[0023] Dynamic compression ratio for:

[0024] (2)

[0025] in, For dynamic compression ratio, For the real-time requirements of the scenario, The remaining energy coefficient of the node. Based on the compression ratio.

[0026] The maximum payload length of an IEEE 802.15.4 frame is 127 bytes, while the minimum length of an IPv6 packet is 1280 bytes. By fragmenting the data into multiple smaller frames for transmission, and then reassembling them into a complete packet at the receiving end, the number of fragments, k, is obtained:

[0027] (3)

[0028] Where k is the number of fragments. This refers to the IPv6 packet length. This is the maximum payload length for an IEEE 802.15.4 frame.

[0029] Furthermore, step three, which involves selecting preceding nodes based on node heterogeneous quantization scores, includes:

[0030] Node total score for:

[0031] (4)

[0032] in, To quantify the total score for each node, These are quantitative scores for computing power, remaining energy, coverage radius, communication rate, and hardware specifications, respectively. These are the dimensional weights for computing power, remaining energy, coverage radius, communication rate, and hardware specifications, respectively.

[0033] Based on the total score, nodes are divided into three levels: high, medium, and low. Before making routing decisions, high-level nodes are prioritized to participate in path construction.

[0034] Furthermore, the deployment of non-uniform heterogeneous nodes for different smart city scenarios in step four includes:

[0035] Infrastructure monitoring scenarios employ a dense deployment approach; environmental monitoring scenarios employ a random and uniform deployment approach; and general equilibrium scenarios employ a centrally dense and peripherally sparse deployment approach.

[0036] Furthermore, step five, which involves calculating the weighted total score and connecting the node paths using the topology self-healing mechanism, includes:

[0037] The link failure prediction model is as follows:

[0038] (5)

[0039] in, This represents the probability of link failure. For the link signal-to-noise ratio, For the continuous packet loss rate of the node, For link lifetime, For signal-to-noise ratio weights, As a weight for packet loss rate, The weight for survival rounds.

[0040] When the failure probability exceeds a preset threshold, topology reconstruction is triggered, and the system automatically switches to the backup optimal path.

[0041] The weighted scoring model is as follows:

[0042] (6)

[0043] Where S is the total routing score. These are the normalized scores for coverage, transmission, arrival, energy, and delay, respectively. The scene weights are coverage, transmission, arrival, energy, and delay, respectively.

[0044] Furthermore, step six, which involves 6LoWPAN dynamic compression and dynamic data fusion, includes:

[0045] The dynamic fusion threshold K is:

[0046] (7)

[0047] Where K is the dynamic fusion threshold. For local node density, For the link signal-to-noise ratio, Based on the fusion threshold.

[0048] The number of data packets M generated after fusion is:

[0049] (4)

[0050] Where M is the number of data packets and N is the total number of original data packets.

[0051] Total energy savings in transmission for:

[0052] (5)

[0053] in, To save total transmission energy consumption, The energy consumption for transmitting a single data packet. This represents the reduction in the number of data packets transmitted after merging.

[0054] Furthermore, step seven, which involves a comprehensive evaluation of the network using multi-dimensional performance metrics, includes:

[0055] Evaluation metrics include network coverage, data transmission success rate, base station arrival rate, network energy consumption, and end-to-end latency. Simulations are used to verify the architecture's adaptability and effectiveness in three main scenarios: infrastructure detection, environmental monitoring, and general load balancing.

[0056] Another objective of this invention is to provide a wireless sensor network architecture and performance optimization system based on a smart city, wherein the wireless sensor network architecture and performance optimization system includes:

[0057] The architecture building module constructs a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer.

[0058] The heterogeneous access module uses 6LoWPAN technology to interconnect heterogeneous nodes with the IPv6 network.

[0059] The node quantization module filters out preceding nodes based on the heterogeneous quantization scores of the nodes.

[0060] The node deployment module deploys non-uniform heterogeneous nodes for different smart city scenarios.

[0061] The routing decision module calculates the weighted total score and combines it with the topology self-healing mechanism to connect the node paths;

[0062] The data fusion module performs 6LoWPAN dynamic compression and dynamic data fusion;

[0063] The performance evaluation module evaluates network performance metrics from multiple dimensions.

[0064] Another object of the present invention is to provide a computer device including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the wireless sensor network architecture and performance optimization method described above.

[0065] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by the processor, causes the processor to perform the wireless sensor network architecture and performance optimization method described above.

[0066] Another objective of this invention is to provide an information data processing terminal, which is used to implement the aforementioned wireless sensor network architecture and performance optimization method.

[0067] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0068] This invention addresses the technical shortcomings of WSN, such as disordered node deployment, poor heterogeneous adaptability, and large data transmission volume. It designs core technologies such as non-uniform heterogeneous node deployment, 6LoWPAN heterogeneous access and dynamic compression, multi-dimensional quantization of heterogeneous nodes, scenario-based weighted routing, topology self-healing, and dynamic data fusion, thus solving the comprehensive pain points of traditional WSN from the architectural level.

[0069] This invention divides current smart city scenarios into three typical scenarios: infrastructure detection, environmental monitoring, and general balancing. For each scenario, it plans a non-uniform heterogeneous node deployment standard. Through a multi-dimensional quantification model of heterogeneous nodes, attributes such as computing power, energy, and coverage radius are incorporated into the pre-screening process to improve the accuracy of routing decisions. Scenario-adaptive 6LoWPAN heterogeneous access and dynamic compression technology enable seamless interconnection between heterogeneous sensors and IPv6 networks, solving the challenge of heterogeneous device access. Based on different scenario weights, it helps links select the optimal path connection, balancing network coverage, transmission success rate, and base station arrival rate. The compression rate is adjusted in real time according to business needs to improve transmission efficiency. Dynamic compression technology further enhances the system's capabilities. The value data fusion algorithm dynamically adjusts the fusion threshold based on node density and signal-to-noise ratio to reduce redundant transmission; through the topology self-healing mechanism, it realizes link failure prediction and automatic reconstruction, thereby enhancing network resilience.

[0070] This invention adopts a wireless sensor network architecture and performance optimization method based on smart cities, which can comprehensively solve the problems of disordered deployment of traditional WSN nodes, poor heterogeneous adaptability and large data transmission volume. It provides theoretical guidance and technical support for the actual deployment of wireless sensor networks in smart cities and has important practical value.

[0071] The technical solution of this invention solves a technical problem that people have long desired to solve but have never been able to achieve: This invention solves four technical problems: First, it solves the technical defect that existing optimization schemes only improve a single problem and do not comprehensively optimize from the architectural level; second, it solves the technical problem of poor interoperability between heterogeneous sensor nodes and IPv6 networks; third, it solves the technical problem that traditional WSNs are difficult to adapt to the differentiated service needs of smart cities; and fourth, it solves the technical problem that routing decisions do not quantify the heterogeneous attributes of nodes and have low path selection accuracy. Attached Figure Description

[0072] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart of a wireless sensor network architecture and performance optimization method based on smart cities provided by an embodiment of the present invention.

[0074] Figure 2 This is a schematic diagram of the wireless sensor network architecture with a newly added 6LoPWAN adaptation layer provided in an embodiment of the present invention.

[0075] Figure 3This is a schematic diagram of the multi-dimensional quantization level distribution of heterogeneous nodes provided in an embodiment of the present invention.

[0076] Figure 4 This is a schematic diagram of 6LoPWAN heterogeneous access and data fusion provided in an embodiment of the present invention.

[0077] Figure 5 This is a schematic diagram of 6LoPWAN dynamic compression and dynamic data fusion for energy saving provided in an embodiment of the present invention.

[0078] Figure 6 This is a schematic diagram of heterogeneous node deployment in different smart city scenarios provided by embodiments of the present invention.

[0079] Figure 7 These are performance diagrams for three major scenarios provided in the embodiments of the present invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0081] To address the problems existing in the prior art, this invention provides a wireless sensor network architecture and performance optimization method based on smart cities. The invention will be described in detail below with reference to the accompanying drawings.

[0082] like Figure 1 As shown, the wireless sensor network architecture and performance optimization method based on smart cities provided in this embodiment of the invention includes the following steps:

[0083] S101, constructing a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer;

[0084] S102, which completes the interconnection of heterogeneous nodes with the IPv6 network based on 6LoWPAN technology;

[0085] S103, select the preceding nodes based on the node heterogeneity quantization score;

[0086] S104 is a non-uniform heterogeneous node deployed for different smart city scenarios.

[0087] S105, calculate the weighted total score and combine it with the topological self-healing mechanism to connect the node paths;

[0088] S106 performs 6LoWPAN dynamic compression and dynamic data fusion;

[0089] S107 evaluates network performance metrics from multiple dimensions.

[0090] 6LoWPAN, as an adaptation technology for IPv6 in low-power wireless networks, explicitly covers the transmission, header compression, fragmentation, and reassembly of IPv6 packets on low-power links in the standard documentation; RFC 6282 further specifies the header compression format of IPv6 packets in 6LoWPAN.

[0091] This wireless sensor network architecture and performance optimization method for smart cities addresses the multi-scenario sensor data acquisition needs of smart cities by setting up a 6LoPWAN adaptation layer in the wireless sensor network, enabling heterogeneous sensor nodes to have the data transmission foundation for interconnection with the IPv6 network. The system first constructs a wireless sensor network architecture including the 6LoPWAN adaptation layer. This layer, located between the physical layer and the data link layer, receives data transmission requests from low-power wireless sensor nodes. It performs IPv6 header compression and fragmentation reassembly on data packets that need to enter the IPv6 network, enabling restricted nodes to complete data interaction with the IPv6 network under low-bandwidth and low-power link conditions. The compression efficiency is determined by the ratio between the original IPv6 header length and the compressed header length; specifically, it is evaluated by subtracting the ratio of the compressed header length to the original IPv6 header length from 1, reflecting the degree to which compression reduces packet overhead.

[0092] During the network node organization phase, this method does not treat all nodes with the same capabilities. Instead, it selects leading nodes based on a node heterogeneity quantification score. The node heterogeneity quantification score is obtained by multiplying computing power score, remaining energy score, coverage radius score, communication rate score, and hardware specification score by their respective weights and then summing the results. Therefore, nodes with stronger computing power, higher remaining energy, larger coverage area, better communication rate, and higher hardware specifications receive higher priority in the comprehensive score and are selected as leading nodes to undertake more critical data access, forwarding, or aggregation functions. This scoring method ensures that the selection of leading nodes corresponds to the actual resource status and communication capabilities of the nodes, avoiding uneven node load caused by a single indicator.

[0093] During the node deployment phase, this method configures non-uniform heterogeneous nodes according to the sensing needs of different smart city application scenarios. Infrastructure detection scenarios employ a dense deployment approach to meet the high-frequency sensing and fine-grained monitoring needs of key areas; environmental monitoring scenarios use a randomized uniform deployment approach to maintain relatively balanced spatial coverage of monitoring nodes; and general balanced scenarios employ a center-dense, edge-loose deployment approach to ensure high data collection density in core areas while maintaining basic coverage in edge areas. These different deployment methods collectively serve the differentiated sensing needs of each scenario and provide a node distribution foundation for subsequent path connections and data fusion.

[0094] During the node path connection phase, this method calculates a weighted total score to select the connection path. The weighted total score is obtained by summing the coverage score, transmission score, arrival score, energy score, and delay score, each multiplied by its corresponding weight, thus ensuring that path selection is simultaneously constrained by coverage capability, transmission performance, reachability, energy state, and latency level. During network operation, the topology self-healing mechanism continuously calculates the link failure probability based on the link signal-to-noise ratio, continuous packet loss rate of nodes, and link lifetime. When the link failure probability exceeds a preset threshold, topology reconstruction is triggered, enabling node paths to reconnect according to changes in link quality.

[0095] During the data transmission and processing phase, the 6LoPWAN adaptation layer performs dynamic compression to reduce the transmission overhead of the IPv6 header in the wireless sensor network. Simultaneously, the system performs dynamic data fusion, aggregating data from different heterogeneous nodes. A continuous workflow chain, consisting of node selection, scenario-based deployment, weighted path connection, topology self-healing, dynamic compression, and dynamic data fusion, enables smart city wireless sensor networks to achieve IPv6 interconnection, stable path connections, and optimized data transmission performance under heterogeneous node conditions.

[0096] As a preferred embodiment, the wireless sensor network architecture and performance optimization method based on smart cities provided by this invention specifically includes the following steps:

[0097] Step 1: Construct a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer.

[0098] A traditional five-layer network protocol stack framework for wireless sensor networks is constructed, and a 6LoWPAN adaptation layer is added between the physical layer and the data link layer to be responsible for IPv6 header compression, fragmentation and reassembly, and dynamic compression rate adjustment.

[0099] Step 2: Connect heterogeneous nodes to the IPv6 network using 6LoWPAN technology.

[0100] Compression efficiency for:

[0101] (1)

[0102] in, For compression efficiency, This is the length of the head after compression. This is the original IPv6 header length.

[0103] Dynamic compression ratio for:

[0104] (2)

[0105] in, For dynamic compression ratio, For the real-time requirements of the scenario, The remaining energy coefficient of the node. Based on the compression ratio.

[0106] The maximum payload length of an IEEE 802.15.4 frame is 127 bytes, while the minimum length of an IPv6 packet is 1280 bytes. By fragmenting the data into multiple smaller frames for transmission, and then reassembling them into a complete packet at the receiving end, the number of fragments, k, is obtained:

[0107] (2)

[0108] Where k is the number of fragments. This refers to the IPv6 packet length. This is the maximum payload length for an IEEE 802.15.4 frame.

[0109] Step 3: Selecting preceding nodes based on node heterogeneity quantization scores includes:

[0110] Node total score for:

[0111] (4)

[0112] in, To quantify the total score for each node, These are quantitative scores for computing power, remaining energy, coverage radius, communication rate, and hardware specifications, respectively. These are the dimensional weights for computing power, remaining energy, coverage radius, communication rate, and hardware specifications, respectively.

[0113] Based on the total score, nodes are divided into three levels: high, medium, and low. Before making routing decisions, high-level nodes are prioritized to participate in path construction.

[0114] Step four involves deploying non-uniform heterogeneous nodes for different smart city scenarios, including:

[0115] Infrastructure monitoring scenarios employ a dense deployment approach; environmental monitoring scenarios employ a random and uniform deployment approach; and general equilibrium scenarios employ a centrally dense and peripherally sparse deployment approach.

[0116] Step 5, calculate the weighted total score and the node connection path based on the topology self-healing mechanism, including:

[0117] The link failure prediction model is as follows:

[0118] (5)

[0119] in, This represents the probability of link failure. For the link signal-to-noise ratio, For the continuous packet loss rate of the node, For link lifetime, For signal-to-noise ratio weights, As a weight for packet loss rate, The weight for survival rounds.

[0120] When the failure probability exceeds a preset threshold, topology reconstruction is triggered, and the system automatically switches to the backup optimal path.

[0121] The weighted scoring model is as follows:

[0122] (3)

[0123] Where S is the total routing score. These are the normalized scores for coverage, transmission, arrival, energy, and delay, respectively. The scene weights are coverage, transmission, arrival, energy, and delay, respectively.

[0124] Step six involves performing 6LoWPAN dynamic compression and dynamic data fusion, including:

[0125] The dynamic fusion threshold K is:

[0126] (7)

[0127] Where K is the dynamic fusion threshold. For local node density, For the link signal-to-noise ratio, Based on the fusion threshold.

[0128] The number of data packets M generated after fusion is:

[0129] (8)

[0130] Where M is the number of data packets and N is the total number of original data packets.

[0131] Total energy savings in transmission for:

[0132] (9)

[0133] in, To save total transmission energy consumption, The energy consumption for transmitting a single data packet. This represents the reduction in the number of data packets transmitted after merging.

[0134] Step 7: Evaluate network performance metrics from multiple dimensions.

[0135] Evaluation metrics include network coverage, data transmission success rate, base station arrival rate, network energy consumption, and end-to-end latency. Simulations are used to verify the architecture's adaptability and effectiveness in three main scenarios: infrastructure detection, environmental monitoring, and general load balancing.

[0136] To demonstrate the inventiveness and technical value of the technical solution of this invention, experimental simulation verification was conducted. During the simulation analysis, a wireless sensor network architecture was first constructed, with a research area of ​​500m × 500m, a sensor communication radius of 200~220m, and a transmission rate of 1.0~1.2Mbps. The node energy model encompassed three parts: transmission, reception, and idle listening, with idle listening energy consumption at 0.05J / round. A 6LoWPAN adaptation layer is added between the physical layer and the data link layer. Interoperability with IPv6 is achieved based on 6LoWPAN heterogeneous node technology. A scenario-adaptive dynamic compression scheme is adopted, with a basic compression coefficient of 0.95 for infrastructure monitoring scenarios, 0.75 for environmental monitoring scenarios, and 0.85 for general balanced scenarios. The dynamic compression rate is calculated in real time by combining scenario real-time requirement coefficients and node remaining energy coefficients. 6LoWPAN fragments are uniformly set to 32 bytes. Subsequently, a multi-dimensional quantitative level model for heterogeneous nodes is built. Based on indicators such as computing power, remaining energy, coverage radius, communication rate, and hardware specifications, a quantitative score is quantified and multiplied by the corresponding weights to obtain a total quantitative score. Nodes with a score higher than 80 are considered high-level nodes. Nodes with a score below 50 are classified as low-level nodes, with the rest being mid-level nodes. Non-uniform heterogeneous nodes are then deployed based on scenario requirements: dense deployment for infrastructure detection scenarios, random and uniform deployment across the entire area for environmental monitoring scenarios, and a centrally dense and loosely distributed deployment for general balanced scenarios. The total routing score is calculated based on the normalized scores of coverage, transmission, arrival, energy, and delay, combined with the corresponding scenario weights. For infrastructure detection scenarios, latency and success are each weighted at 0.4, arrival at 0.1, and coverage and energy at 0.05. For environmental monitoring scenarios, energy and coverage are each weighted at 0.4, latency at 0.1, and arrival and success at 0.05. For general balanced scenarios, all five indicators have a weight of 0.2. The link signal-to-noise ratio is also utilized. The probability of link failure is calculated using the global historical packet loss coefficient and the number of survival rounds. A failure threshold of 0.3 is set, with signal-to-noise ratio weight, packet loss rate weight, and survival round weight of 0.6, 0.2, and 0.2, respectively. When parameters exceed the limits, topology reconstruction is automatically triggered and a backup link is switched. Afterwards, scene-adaptive 6LoWPAN dynamic compression and dynamic data fusion processing are performed, with basic fusion thresholds for each scene. The values ​​are set to (6, 3, 5), corresponding to the number of infrastructure monitoring items, the number of environmental monitoring items, and the general equilibrium number, respectively, combined with the node density in the local area. Link signal-to-noise ratio The dynamic fusion threshold K is calculated in real time, with a single data fusion energy consumption of 0.3 joules and a single packet buffer energy consumption of 0.1 joules. The number of redundant packets is reduced by relying on dynamic aggregation rules. Finally, network performance is tested from multiple dimensions, including network coverage, transmission success rate, base station arrival rate, average energy consumption, and end-to-end latency. Simulation results show that: ① In three typical smart city application scenarios, the architecture proposed in this invention, based on heterogeneous node hierarchical pre-screening, scenario-based weighted routing and topology self-healing, adaptive dynamic compression, and dynamic fusion optimization strategies, can achieve targeted performance optimization according to the differentiated needs of different scenarios, meeting the core service quality requirements of each scenario. ② The 6LoWPAN heterogeneous access and scenario-adaptive dynamic compression technology achieve efficient interconnection between heterogeneous sensor nodes and the IPv6 network, with low encapsulation / decapsulation latency and high fragmentation / reassembly success rate. ③ Dynamic data fusion can adaptively match regional node density and link quality, effectively reducing overall network energy consumption. ④ Link failure prediction and topology self-healing mechanisms can quickly repair random faulty nodes and failed links, significantly improving network robustness and connectivity stability under complex electromagnetic and node failure environments.

[0137] Figure 2 This is a schematic diagram of a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer. It demonstrates the five-layer functionality of a wireless sensor network, as well as the newly inserted 6LoPWAN adaptation layer. The layers work together to optimize the entire process of data acquisition, fusion, forwarding, and management.

[0138] Figure 3 This diagram illustrates the multi-dimensional quantitative distribution of heterogeneous nodes. The nodes are divided into three levels—high, medium, and low—based on five dimensions and labeled accordingly, visually presenting the spatial distribution patterns of the three types of heterogeneous nodes.

[0139] Figure 4 This diagram illustrates heterogeneous access and data fusion in 6LoPWAN. The header compression mechanism significantly simplifies the IPv6 packet header and reduces packet processing time. The low latency of encapsulation and decapsulation demonstrates that sensor nodes can quickly access the IPv6 backbone network. Fragmentation reassembly success rates are all above 96.00%, with base stations rapidly reconstructing data based on precise fragment matching, significantly reducing the occurrence of fragment loss.

[0140] Figure 5 The diagrams show a comparison of 6LoPWAN dynamic compression and dynamic data fusion. In dynamic compression, the highest compression latency for the infrastructure monitoring scenario is 84.50ms, and the lowest is 31.90ms for the environmental monitoring scenario. In the energy consumption statistics of dynamic data fusion, the highest energy consumption for the environmental monitoring scenario is 100.0 joules, and the lowest is 38.4 joules during compression for the infrastructure monitoring scenario. These values ​​are consistent with theoretical patterns, verifying the energy-saving effects of adaptive compression and dynamic fusion.

[0141] Figure 6This diagram illustrates the heterogeneous node deployment across different smart city scenarios. The infrastructure monitoring scenario exhibits a dense node distribution. The environmental monitoring scenario shows a relatively random node distribution. The general balanced scenario employs a dense-at-the-center, loose-at-the-edge deployment approach. All three scenarios demonstrate a distribution strategy proposed in this invention.

[0142] Figure 7 Performance analysis was conducted for three main scenarios. Simulation analysis revealed significant differences in network performance indicators across different application scenarios, all highly adapted to the varying requirements of each scenario. In the environmental monitoring scenario, relying on coverage priority weighting and uniform node deployment across the entire domain, the network coverage reached a maximum of 91.11%, meeting its high coverage requirements, while maintaining an average remaining energy of 66.35 joules. In the infrastructure monitoring scenario, thanks to a strategy with a high latency weighting, the transmission success rate was optimal at 76.56%, with an average latency as low as 0.4961s, aligning with the scenario's requirements for low latency and high reliability transmission. In the communication balancing scenario, relevant performance indicators were averages, indicating overall balanced performance.

[0143] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0144] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A wireless sensor network architecture and performance optimization method for smart cities, characterized in that, include: Construct a wireless sensor network architecture with a newly added 6LoPWAN adaptation layer; Interconnection between heterogeneous nodes and IPv6 networks is achieved using 6LoWPAN technology; Selecting leading nodes based on node heterogeneity quantization scores; Deploy non-uniform heterogeneous nodes for different smart city scenarios; Calculate the weighted total score and connect the node paths using the topological self-healing mechanism; Perform 6LoWPAN dynamic compression and dynamic data fusion.

2. The method as described in claim 1, characterized in that, The wireless sensor network architecture for constructing the new 6LoPWAN adaptation layer includes: A new 6LoPWAN adaptation layer is added between the physical layer and the data link layer. This 6LoPWAN adaptation layer is used to perform IPv6 header compression and fragmentation reassembly functions.

3. The method as described in claim 1, characterized in that, It also includes performing IPv6 header compression through the 6LoPWAN adaptation layer, with a compression efficiency of 1 minus the ratio of the compressed header length to the original IPv6 header length.

4. The method as described in claim 1, characterized in that, The process of selecting front-end nodes based on node heterogeneity quantization scores includes: The node heterogeneous quantization score is the sum of the computing power score, remaining energy score, coverage radius score, communication rate score, and hardware specification score multiplied by their respective computing power weight, remaining energy weight, coverage radius weight, communication rate weight, and hardware specification weight.

5. The method as described in claim 1, characterized in that, The non-uniform heterogeneous nodes deployed for different smart city scenarios include: Infrastructure monitoring scenarios employ a dense deployment approach, environmental monitoring scenarios employ a random and uniform deployment approach, and general balanced scenarios employ a centrally dense and peripherally loose deployment approach.

6. The method as described in claim 1, characterized in that, The calculation of the weighted total score, combined with the topological self-healing mechanism, connects the node paths, including: The weighted total score is the sum of the coverage score, transmission score, arrival score, energy score, and delay score multiplied by their respective coverage weight, transmission weight, arrival weight, energy weight, and delay weight. The topology self-healing mechanism calculates the link failure probability by using the link signal-to-noise ratio, the continuous packet loss rate of nodes, and the link survival time. When the failure probability is higher than a preset threshold, topology reconstruction is triggered.

7. The method as described in claim 1, characterized in that, The 6LoWPAN dynamic compression and dynamic data fusion process includes: The 6LoWPAN dynamic compression is calculated in real time based on the scenario demand coefficient, the node remaining energy coefficient and the basic compression ratio. The dynamic data fusion is based on the local area node density, link signal-to-noise ratio and basic fusion threshold to solve the dynamic fusion threshold, determine the number of data packets after fusion, and calculate the total transmission energy saved.

8. A method for optimizing data transmission in wireless sensor networks based on the method described in any one of claims 1 to 7, characterized in that, include: IPv6 packets are fragmented and transmitted using the 6LoPWAN adaptation layer. The number of fragments is determined by rounding up the ratio of the IPv6 packet length to the maximum payload length of the IEEE 802.15.4 frame. Data fusion is performed on the transmitted data, and the number of data packets generated after fusion is the total number of original data packets divided by the fusion threshold. Calculate the total energy savings in transmission.

9. The method as described in claim 8, characterized in that, The number of fragments is determined by dividing the IPv6 packet length by the rounded-up value of the maximum payload length of the IEEE 802.15.4 frame, where the maximum payload length of the IEEE 802.15.4 frame is 127 bytes and the minimum length of the IPv6 packet is 1280 bytes.

10. The method as described in claim 8, characterized in that, The number of data packets generated after fusion is the total number of original data packets divided by the fusion threshold; the total energy saved is the difference between the total number of original data packets and the number of data packets generated after fusion, multiplied by the energy consumption of a single data packet. It also includes evaluating network performance indicators from multiple dimensions, including network coverage, data transmission success rate, base station arrival rate, network energy consumption, and end-to-end latency.

Images