Blockchain Sharding Load Balancing via CNN-LSTM Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blockchain sharding methods for the Internet of Vehicles face challenges such as poor load balancing, neglect of cross-shard transactions, and incomplete reputation value models, leading to inefficient resource utilization and security concerns.
Innovation Solution
A blockchain sharding method based on a CNN-LSTM prediction model is introduced, which allocates vehicle nodes into different shards, uses a DBFT protocol for consensus, and employs a reputation value model to optimize shard load balancing and reduce cross-shard transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing load balancing strategies are used that optimize based on current node status, then resource allocation can be simplified, but load balancing effect deteriorates due to lack of predictability for dynamic changes
Solution Approach 1:
The patent applies preliminary action by using the CNN-LSTM prediction model to forecast future transaction volumes and node statuses before actual load balancing decisions are made. This allows the system to proactively redistribute transactions to prevent future imbalances rather than reactively responding to current conditions, thereby improving load balancing effectiveness while maintaining reasonable complexity.
2Reliability
If all node accounts are considered during load balancing optimization, then comprehensive optimization can be achieved, but network resource pressure increases
Solution Approach 1:
The patent applies segmentation by dividing the blockchain network into multiple shards, allowing load balancing optimization to be performed independently within each shard rather than across the entire network. This segmentation reduces the computational resources required for optimization while maintaining comprehensive coverage through distributed optimization across multiple partitions.
Solution Approach 2:
The patent applies partial action by focusing load balancing optimization only on hot accounts (accounts with high transaction volumes) rather than all node accounts. The prediction model identifies and prioritizes these critical accounts for optimization, achieving effective load balancing with reduced network resource consumption by ignoring low-impact accounts.
3Reliability
If traditional reputation value models with low tolerance are used, then data security can be enforced, but data sharing volume decreases due to limited practicality
Solution Approach 1:
The patent applies dynamics by implementing a dynamic reputation value model that adapts to changing node behaviors and transaction patterns over time. Rather than using static tolerance thresholds, the system continuously updates reputation scores based on observed behavior, allowing flexible enforcement of security policies that can accommodate legitimate high-volume data sharing while maintaining protection against malicious actors.
Data Source
AI summary
Disclosed are a blockchain sharding method and system based on a convolutional neural network-long short-term memory (CNN-LSTM) prediction model. The method includes: using a double-layer chain system architecture, namely a vehicle layer and an edge layer; and using a delegated byzantine fault tolerance (DBFT) as a consensus protocol of a network at the vehicle layer and using proof of authority (POA) as a consensus protocol at the edge layer. Load balancing of the blockchain sharding system is optimized through a CNN-LSTM prediction model. A transaction status of the entire sharding network in a next period is predicted. Hot accounts are effectively allocated through a designed account allocation algorithm. The blockchain sharding method can maintain load balancing of shards, reduce cross-shard transactions, and implement effective utilization of resources. Malicious behavior of nodes is restricted. Security of the network is ensured.