Keyword-based matching

Proxy-Free Privacy-Preserving Task Matching with Efficient Revocation in Crowdsourcing 1

这篇文章提出了一个无需代理重加密的基于关键字的任务匹配模型，支持用户撤销。作者公布代码。

基本模型是经典的基于可搜索加密的关键字匹配模型，之后加入了 Server-Local Revocation (SLR)，实现用户撤销的思想很简单：服务器维护一个撤销用户列表 Revocation List (RL)，当有用户离开系统时，将该用户的相关信息存入RL实现撤销，在每次执行匹配算法时，服务器会先验证改用户是否在RL在，如果在RL中，则此次搜索终止，如不在RL中，说明此次搜索合法，进行匹配操作。

之后为了减少RL的存储开销，作者提出了Global Revocation (GR)，为密钥、密文、搜索陷门加入了版本号（version number），系统周期性地执行$ReKeyGen$算法，清空RL，重新生成公钥、主私钥和密文更新密钥（用来更新之前版本的密文），并为worker分发新的密钥。

优点：撤销机制简单

缺点：每次匹配均需进行两次pairing操作以检查用户权限；在 GR 中，密钥机构需要周期性地上线执行$ReKeyGen$算法，即使是在RL为空时？（作者没有考虑这一点）。

思考：学习到了撤销机制的实现方法：维护撤销用户列表和版本号。

📝 Note: 写文章思路：先写出一个基础方案，之后进行扩展，本文在基础匹配方案上扩展了用户撤销机制。先写SLR，然后针对 SLR 中RL的存储开销问题，提出GR，让系统定期清空RL，引入版本号。

ABE

Attribute-Based Encryption with Publicly Verifiable Outsourced Decryption

Transform算法输出的不是ElGamal形式的密文，而是把中间计算过程中的配对值输出。 验证者通过 sampling technique 随机选择pairing results进行验证。 Achieving public verification by combining sampling technique and game theory.

Multi-Keyword Searchable and Verifiable Attribute-Based Encryption Over Cloud Data

只支持合取的关键字查询；需要公布一个关键字集合之间的映射；使用签名验证密文的complete； 是否可以做多关键字的检索 without bilinear pairing ?

TrustAccess: A Trustworthy Secure Ciphertext-Policy and Attribute Hiding Access Control Scheme based on Blockchain

对向量使用ElGamal算法加密，利用同态性质验证向量内积为0； 属性向量$v$中有有一维是1，那么$h^r$就暴露了，别的$v$是否就能求出？ Access policy使用polynomial编码 -> hidden policy

Registered Attribute-Based Encryption (RABE)

size of CRS quadratic in L time of registration linear in L—-slow size of mpk, hdk and time of enc, dec polylogarithmically with L

Main Contribution: resolve leakage of master secret key (msk) problem.

limitations:

  • Exist priori bound L on the number of users in the system
  • Trusted output CRS

Task Assignment

Bid-based

Personalized Location Privacy Trading in Double Auction for Mobile Crowdsensing

保护用户位置隐私

Truthful Online Double Auctions for Mobile Crowdsourcing: An On-Demand Service Strategy

不保护bid

Truthful online double auctions for dynamic mobile crowdsourcing

不保护bid

Edge Resource Prediction and Auction for Distributed Spatial Crowdsourcing With Differential Privacy

采用非对称加密保护bid

Location-based

Towards preserving worker location privacy in spatial crowdsourcing 2015

加法同态，在位置密文上进行运算比较来分配

PriRadar: A Privacy-Preserving Framework for Spatial Crowdsourcing, 2020

利用属性加密，将位置经纬度设置为密文访问策略。且保护了任务内容

Protecting Location Privacy of Users Based on Trajectory Obfuscation in Mobile Crowdsensing 2022

差分隐私

RoPriv: Road Network-aware Privacy-preserving Framework in Spatial Crowdsourcing 2023

差分隐私

iTAM: Bilateral Privacy-Preserving Task Assignment for Mobile Crowdsensing 2021

minimize travel distance; Bilateral privacy includes task requirements and worker profiles; Paillier同态

Privacy-Preserving Online Task Assignment in Spatial Crowdsourcing: A Graph-based Approach 2022

差分隐私混淆位置

@INPROCEEDINGS{9796827,
 author={Wang, Hengzhi and Wang, En and Yang, Yongjian and Wu, Jie and Dressler, Falko},
 booktitle={IEEE INFOCOM 2022 - IEEE Conference on Computer Communications}, 
 title={Privacy-Preserving Online Task Assignment in Spatial Crowdsourcing: A Graph-based Approach}, 
 year={2022},
 volume={},
 number={},
 pages={570-579},
 doi={10.1109/INFOCOM48880.2022.9796827}}

Keyword-based

Secure Task Recommendation in Crowdsourcing 2016

利用可搜索加密以及代理重加密

Privacy-Preserving Task Recommendation Services for Crowdsourcing 2018

SSE 代理重加密 多关键字，使用多项式表示任务需求和工人兴趣

Anonymous Privacy-Preserving Task Matching in Crowdsourcing 2018

SSE 代理加密，提供用户匿名性且可实现追踪和撤销

Privacy-Preserving Task Matching with Threshold Similarity Search via Vehicular Crowdsourcing 2021

SSE proxy re-encryption. 先基于关键字检索，然后根据位置筛选， 同时保护了关键字和位置

PPTA: Privacy-Preserving Task Assignment Based on Inner-Product Functional Encryption in SAM 2023 Xu

同时保护关键字和位置

PPTA: A location privacy-preserving and flexible task assignment service for spatial crowdsourcing 2023 Zhou

仅位置，使用两方的加法秘密共享,将工人位置和任务位置分别分割成两个秘密共享出去，但是安全性较弱，要求两个crowdsourcing server provider不合谋

Blockchain-based Crowdsourcing

Differentially Private Crowdsourcing With the Public and Private Blockchain

identity and location

TrustCom:

2020

Efficient Revocable Attribute-Based Encryption with Hidden Policies

  • outsourced decryption, hidden policies and revocation
  • support revocation: - update attribute group key and re-encrypt all ciphertext
  • security proof

2021

White-Box Traceable Attribute-Based Encryption with Hidden Policies and Outsourced Decryption

  • same authors for above paper
  • add Algorithm $Trace, Revoke$ compare with Efficient Revocable Attribute-Based Encryption with Hidden Policies
  • no security proof

A Privacy-Enhanced Mobile Crowdsensing Strategy for Blockchain Empowered Internet of Medical Things

This scheme classifies the uses by spectral clustering based on the social network. Therefore, the worker is assigned a specific task.

Publicly Verifiable Inner Product Evaluation over Outsourced Data Streams under Multiple Keys

生成Proof需要向量明文，不可行

SanIdea: Exploiting Secure Blockchain-Based Access Control via Sanitizable Encryption

TIFS'24

Sanitize 后的密文没有用到，解密时直接忽略了前面Sanitize加的随机数，什么操作？Sanitize了个什么东西？

将Authority的secret keys存储在区块链上？– Introduction

  1. Shu, Jiangang, et al. “Proxy-free privacy-preserving task matching with efficient revocation in crowdsourcing.” IEEE Transactions on Dependable and Secure Computing 18.1 (2018): 117-130. ↩︎