Review

Info

Enabling Privacy-Assured Fog-Based Data Aggregation in E-Healthcare Systems

(IEEE Transactions on Industrial Informatics'21)

"In this article, we design an improved symmetric homomorphic cryptosystem and a fog-based communication architecture to support delay- or time-sensitive monitoring and other-related applications."

---

Medical Monitoring Systems

• 属于 e-healthcare system 中的一部分
• 通过类似 wearable body area network (WBAN) 的环境进行无线通信
• 可被植入，或穿戴在病人身上，或安装在医疗仪器中
• 会不断地周期地发送信息 (continuously and periodically)

为什么需要 fog-based

• 设备使用 ZigBee 或蓝牙等近距离通信协议
• 如果传输到 Medical Cloud Server (MCS)，会因为大量的信息内容和通信延迟而花费太长的时间

使用雾节点需要注意的是，因为雾节点处于不安全的环境中，所以传输的数据都需要进行加密。

both data-in-transit and data-at-rest at the fog servers (FSs) should remain encrypted.

这篇文章的贡献：

• design a new symmetric homomorphic encryption-based data aggregation scheme for e-healthcare systems
• use the proposed scheme in a fog-based architecture to support a time- or delay-sesitive medical monitoring system

Related Work

How to execute data aggregation in privacy-preservign manners:

• Paillier encryption technique
• support fine-grained demands in smart grid
• dynamic data aggregation framework for smart grid
• MapReduce framework
• several aggregated statistics

All above schemes are based on cloud server.

How to solve the problem of network latency:

• two-layer encryption

How to preserve the integrity of the data:

• ElGamal cryptosystem
• emergency-response approach
• P2DA by using Boneh-Goh-Nissim encryption system
• aggregate multidimensional data, batch verification

None of above can satisfy all of the security requirements, and several schemes are too time-consuming to be practical.

Problem Statement

Four types of entities:

• medical workers
• MCS
• FS
• WBAN

One MCS connects to $m$ FSs, each FS connects to $n$ WBANs, each WBAN connects to $l$ various medical sensors in real-time, denoted by $\{md_1, md_2, \ldots, md_l\}$.

Threat Moedl:

• Fully Trustworthy: MCS, FSs

Preliminaries

Symmetric Homomorphic Cryptosystem (SHE):

• $KeyGen(\lambda)\to (s,v,u,d)$
• $Enc(K, m, r)\to c$: $c=s^d(rv + m) \mbox{ mod } u$
• $Dec(K, c)\to m$: $m = (cs^{-d} \mbox{ mod } v) \mbox{ mod } v$

同态加法：

$c_1 + c_2 = s^d ((r_1 + r_2)v + (m_1 + m_2)) \mbox{ if } d_1 = d_2 = d$

同态标量乘法：

$c_1 \times m_2 = s^{d_1} ((r_1 m_2)v + m_1 m_2)$

The proposed scheme

初始化：

• TTP 选取安全参数 $k$，生成 $(q, P, \mathbb{G}_1, \mathbb{G}_2, e)$，以及哈希函数 $H(\cdot): \{0,1\}^* \to \mathbb{G}_1$
• TTP 公开参数
• $FS_i$ 选取 $x_i\in \mathbb{Z}_q^*$ 作为私钥，$Y_i = x_i P$ 作为公钥
• TTP 将 $K_{i,j} = (s_{i,j}, v_{i,j}, u_{i,j}, d_{i,j})$ 分配给 $WBAN_{i,j}$，每一个 $FS_i$ 产生用户的私钥 $K_i = \{K_{i,1}, K_{i,2},\ldots, K_{i,n}\}$