Self-adaptive compression cooperation method and system for animal husbandry data
Through a four-level collaborative architecture and intelligent processing, the system solves the problems of real-time compression latency and security vulnerabilities of edge devices under high-concurrency data streams, achieving efficient and secure data processing and environmental monitoring, reducing maintenance costs, and improving system adaptability and data sharing efficiency.
Patent Information
- Application Number
- CN202511172098.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-21
- Publication Date
- 2025-11-04
AI Technical Summary
Edge devices have limited computing power, high latency in real-time compression under high-concurrency data streams, significant risk of critical data loss, poor environmental adaptability, high equipment failure rate, numerous security vulnerabilities, low efficiency in cross-farm data sharing, high maintenance costs, and a lack of predictive capabilities, resulting in data processing delays and economic losses.
It adopts a four-level collaborative architecture of terminal perception layer, edge processing layer, disaster recovery storage layer and application decision layer, and combines IoT devices, embedded AI nodes, blockchain distributed storage and federated learning framework to achieve adaptive data compression and real-time processing. It ensures data security through smart contracts and encryption technology, and optimizes data stream processing by using sliding time windows and lightweight models.
It reduces data distortion rate, improves system environmental adaptability and security, reduces the impact of equipment failure, enhances data sharing efficiency and predictive capabilities, reduces maintenance costs, and ensures the real-time performance and security of data processing.
Smart Images
Figure CN120896960A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the intersection of edge intelligence and agricultural informatization, specifically providing a method and system for adaptive compression and collaboration of livestock data. Background Technology
[0002] Edge streaming processing and adaptive compression system: A system that combines edge computing, real-time data processing, and intelligent compression technologies. Its core goal is to make data processing more efficient and faster-responding, while reducing the consumption of network bandwidth and storage resources. It's an intelligent system that "works near the data source and adjusts the data size as it goes."
[0003] Decentralized disaster recovery and collaborative architecture: a fault-resistant system structure that relies on multiple equal nodes to back up each other and cooperate, so that even if some nodes fail, the entire system will not be affected.
[0004] Data analysis and decision-making systems: These systems process and analyze collected data to identify patterns. These patterns are then used to help individuals and systems make decisions that are more aligned with reality.
[0005] Environmental monitoring and automated control: An integrated system that monitors environmental parameters in real time and automatically adjusts relevant equipment according to preset rules or intelligent algorithms to maintain the environment in an ideal state, allowing the environment to "regulate itself".
[0006] The above has the following disadvantages:
[0007] (1) Hardware resource bottleneck: Edge devices (such as sensors) have limited computing power, and real-time compression latency is high under high concurrency data flow, which significantly increases the risk of loss of critical data.
[0008] (2) Poor environmental adaptability: In high temperature / high humidity pasture environments, the equipment failure rate is high, frequently causing data acquisition interruptions and increasing compression distortion rate.
[0009] (3) Security vulnerabilities: Edge nodes (such as feeders) have weak physical protection, and penetration tests show that hackers have a very high success rate in intrusion, resulting in extremely high risks of data leakage and compliance.
[0010] (4) Low collaboration efficiency: Cross-ranch data sharing relies on the central platform for relay, and the daily processing time for collaboration requests is long, which delays joint prevention and control of diseases and amplifies economic losses.
[0011] (5) High maintenance costs: Remote ranches have long equipment failure repair cycles, high annual maintenance costs, and increased operating costs.
[0012] (6) Lack of predictive ability: based on threshold warnings rather than predictive regulation (such as warming preparation before a cold wave), the ability to buffer environmental fluctuations is weak, and livestock stress mortality increases. Summary of the Invention
[0013] This invention addresses the shortcomings of the prior art by providing a highly practical adaptive compression and collaborative method for livestock data.
[0014] A further technical objective of this invention is to provide a rationally designed, safe, and applicable adaptive compression and collaborative system for livestock data.
[0015] The technical solution adopted by this invention to solve its technical problem is:
[0016] The adaptive compression and collaborative method for livestock data employs a four-level collaborative approach: terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer.
[0017] The terminal sensing layer consists of IoT devices deployed in the farm to collect raw data streams;
[0018] The edge processing layer enables embedded AI nodes to process data locally, performing streaming compression and preprocessing.
[0019] The disaster recovery storage layer is based on a distributed network built on blockchain, storing encrypted shards on edge devices and backup nodes;
[0020] The application decision layer drives cross-field data collaboration through a learning engine and outputs control instructions to the execution terminal.
[0021] Furthermore, livestock data is collected in real time through IoT devices in the edge processing layer, and the data stream is segmented by time window at the edge node, and the compression algorithm is automatically switched according to the data fluctuation threshold.
[0022] Furthermore, in the disaster recovery storage layer, compressed data is fragmented, encrypted, and distributed across at least three edge device nodes, with node status monitored via smart contracts.
[0023] Furthermore, in the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.
[0024] The livestock data adaptive compression and collaborative system adopts a four-level collaborative approach: terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer.
[0025] The terminal sensing layer consists of IoT devices deployed in the farm to collect raw data streams;
[0026] The edge processing layer enables embedded AI nodes to process data locally, performing streaming compression and preprocessing.
[0027] The disaster recovery storage layer is based on a distributed network built on blockchain, storing encrypted shards on edge devices and backup nodes;
[0028] The application decision layer drives cross-field data collaboration through a learning engine and outputs control instructions to the execution terminal.
[0029] Furthermore, livestock data is collected in real time through IoT devices in the edge processing layer, and the data stream is segmented by time window at the edge node, and the compression algorithm is automatically switched according to the data fluctuation threshold.
[0030] Furthermore, in the disaster recovery storage layer, compressed data is fragmented, encrypted, and distributed across at least three edge device nodes, with node status monitored via smart contracts.
[0031] Furthermore, in the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.
[0032] Compared with existing technologies, the adaptive compression and collaborative method and system for livestock data of the present invention have the following outstanding advantages:
[0033] This invention receives sensor data streams in real time via the MQTT protocol and incorporates a priority queue (disease data > environmental data). A sliding time window divides the data into blocks, and data correlation analysis is performed within each window (e.g., coupled calculation of temperature and humidity). A CRC checksum and keyframe reconstruction algorithm ensure reduced distortion. A lightweight MobileNet model predicts data value, discarding invalid data directly.
[0034] Kalman filtering dynamically corrects temperature and humidity data; master node election is based on the improved Raft protocol; device fingerprint (MAC address) + asymmetric encryption (ECDSA algorithm); Merkle root hash is generated and uploaded to the chain every 10 minutes; cross-validation by adjacent sensors eliminates single-point false alarms; soil moisture sensor + evaporation model → dynamically adjusts irrigation amount. Attached Figure Description
[0035] To more clearly illustrate the technical solutions in the embodiments of the present 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0036] Figure 1 This is a schematic diagram of the architecture of an adaptive compression and collaborative method for livestock data. Detailed Implementation
[0037] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0038] The following is a preferred embodiment:
[0039] like Figure 1 As shown, the adaptive compression and collaborative method for livestock data in this embodiment adopts a four-level collaboration of terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer.
[0040] The terminal sensing layer consists of IoT devices deployed in the farm, which collect raw data streams at a frequency of 50-100ms / time.
[0041] The edge processing layer processes data locally for embedded AI nodes (such as NVIDIA Jetson Nano), performing streaming compression and preprocessing.
[0042] The disaster recovery storage layer is built on a distributed network based on blockchain, and the encrypted shards are stored on edge devices (feeders, environmental controllers) and backup nodes;
[0043] The application decision layer drives cross-field data collaboration through a learning engine and outputs control commands to the execution terminal (automatic ventilation / feeding machine).
[0044] In the edge processing layer, livestock data is collected in real time through IoT devices, and the data stream is divided into time windows at the edge nodes. The compression algorithm is automatically switched according to the data fluctuation threshold.
[0045] In the disaster recovery storage layer, compressed data is sharded, encrypted, and distributed across at least three edge device nodes. The node status is monitored through smart contracts. The disaster recovery storage uses sharding technology, which significantly reduces data recovery time in case of failure.
[0046] In the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.
[0047] Based on the above method, the livestock data adaptive compression collaborative system in this embodiment adopts a four-level collaborative approach: terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer.
[0048] The terminal sensing layer consists of IoT devices deployed in the farm to collect raw data streams;
[0049] The edge processing layer processes data locally for embedded AI nodes, performing streaming compression and preprocessing;
[0050] The disaster recovery storage layer is built on a distributed network based on blockchain, storing encrypted shards on edge devices and backup nodes;
[0051] The application decision layer drives cross-field data collaboration through a learning engine and outputs control instructions to the execution terminal.
[0052] In the edge processing layer, livestock data is collected in real time through IoT devices, and the data stream is divided into time windows at the edge nodes. The compression algorithm is automatically switched according to the data fluctuation threshold.
[0053] In the disaster recovery storage layer, compressed data is fragmented, encrypted, and distributed across at least three edge device nodes, with node status monitored via smart contracts.
[0054] In the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.
[0055] The above-described specific embodiments are merely specific examples of the present invention. The patent protection scope of the present invention includes, but is not limited to, the above-described specific embodiments. Any technical solution that conforms to the above-described specific embodiments of the present invention and any appropriate changes or substitutions made by those skilled in the art should fall within the patent protection scope of the present invention.
[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. An adaptive compression and collaborative method for livestock data, characterized in that, It adopts a four-level collaborative approach: terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer. The terminal sensing layer consists of IoT devices deployed in the farm to collect raw data streams; The edge processing layer enables embedded AI nodes to process data locally, performing streaming compression and preprocessing. The disaster recovery storage layer is based on a distributed network built on blockchain, storing encrypted shards on edge devices and backup nodes; The application decision layer drives cross-field data collaboration through a learning engine and outputs control instructions to the execution terminal.
2. The adaptive compression and collaborative method for livestock data according to claim 1, characterized in that, In the edge processing layer, livestock data is collected in real time through IoT devices. Data streams are segmented by time windows at edge nodes, and compression algorithms are automatically switched based on data fluctuation thresholds.
3. The adaptive compression and collaborative method for livestock data according to claim 2, characterized in that, In the disaster recovery storage layer, compressed data is fragmented, encrypted, and distributed across at least three edge device nodes, with node status monitored via smart contracts.
4. The adaptive compression and collaborative method for livestock data according to claim 3, characterized in that, In the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.
5. An adaptive compression and collaborative system for livestock data, characterized in that, It adopts a four-level collaborative approach: terminal perception layer, edge processing layer, disaster recovery storage layer, and application decision layer. The terminal sensing layer consists of IoT devices deployed in the farm to collect raw data streams; The edge processing layer enables embedded AI nodes to process data locally, performing streaming compression and preprocessing. The disaster recovery storage layer is based on a distributed network built on blockchain, storing encrypted shards on edge devices and backup nodes; The application decision layer drives cross-field data collaboration through a learning engine and outputs control instructions to the execution terminal.
6. The livestock data adaptive compression and collaborative system according to claim 5, characterized in that, In the edge processing layer, livestock data is collected in real time through IoT devices. Data streams are segmented by time windows at edge nodes, and compression algorithms are automatically switched based on data fluctuation thresholds.
7. The livestock data adaptive compression and collaborative system according to claim 6, characterized in that, In the disaster recovery storage layer, compressed data is fragmented, encrypted, and distributed across at least three edge device nodes, with node status monitored via smart contracts.
8. The livestock data adaptive compression and collaborative system according to claim 7, characterized in that, In the application decision layer, cross-field encrypted parameters are aggregated based on the federated learning framework to generate control instructions and send them to the execution terminal.