5G工业互联网下的轻量级数据使用安全方案

Secure lightweight data using scheme in 5G industrial Internet systems

针对5G工业互联网环境下数据分析所面临的隐私和安全问题,提出了一种结合隐私计算技术与逻辑回归技术的轻量级解决方案.运用隐私计算技术来保护敏感数据,并通过数据加密和数字签名来确保数据传输的隐私性和完整性,以达到不暴露原始训练数据和模型参数的目标.采用轻量级的密码算法来实现隐私计算功能,以适应大量数据的训练和调参.结果表明,该方案模型的预测精确度高达98.62%,且具有较小的计算量和通信开销.数据使用者在无需获取原始数据的情况下,便可进行逻辑回归模型训练,从而实现数据的可用性和不可见性,同时还确保了模型参数的私密性,适用于5G工业互联网环境中的低功耗设备.

To address the privacy and security challenges of data analysis in the 5G industrial Internet environment,a lightweight solution based on privacy computing and logistic regression techniques is proposed.Privacy computing techniques are utilized to protect sensitive data.Data encryption and digital signatures are used to ensure the privacy and integrity of data transmission,achieving the goal of not disclosing original training data and model parameters.Lightweight cryptographic algorithms are employed to realize the function of privacy computing,so as to adapt to the training and tuning with a large amount of data.The results show that the prediction accuracy of the proposed model is 98.62%.It has small computational and communication costs.Logistic regression model training can be performed by data users without accessing the raw data.It can ensure both data availability and invisibility and prevent exposure of the model parameters,which is suitable for low-power devices in 5G industrial Internet environment.