Review for XZY20

[XZY20] - InPPTD

InPPTD: An Lightweight Incentive-based Privacy-Preserving Truth Discovery for Crowd Sensing Systems

(IEEE Internet of Things Journal)

"... we propose an incentive-based privacy-preserving truth discovery framework, named InPPTD."

---

Introduction

privacy concerns:

• worker's sensed data
• worker's reliability (i.e., weight)

other issues:

• workers may be lazy and selfish
• workers may maliciously manipulate

two categories:

• single-server scheme
• two-server scheme

single-server scheme:

• achieve the privacy preservation for worker's sensed data and weight
• each worker has to perform costly operation
• workers need to keep online

two-server scheme:

• no costly operation for workers
• cannot guarantee strong privacy protection

research gap:

The existing PPTD schemes, whether with a single server or two servers, have not yet considered the issue of incentives, ...

---

Related Works

CRH¹, optimized CRH², TruthFinder³: do not consider participants' privacy

single-server scheme:

• Miao et al.⁴, threshold Paillier cryptosystem
• Zheng et al.⁵, lightweight homomorphic encryption, on-line, weight tamper
• Xu et al.⁶, lightweight homomorphic encryption, on-line, weight tamper
• Li et al.⁷⁸, local differential privacy

two-server scheme:

• Zheng et al.⁹, noise perturbation
• Miao et al.¹⁰, L-PPTD
• Tang et al.¹¹, GC
• Zheng et al.¹²

requirements:

• high efficiency
• workers' failure resistance
• data and weight privacy preservation
• resistance to the modification attack from malicious workers
• reduction of lazy workers

---

System Model

entities:

• Service Provider (SP)
• Cloud Provider (CP)
• Workers

three phases:

• Report Phase
• Iteration Phase
• Rewards Phase

Report Phase:

• worker reports his perturbed sensed data to CP
• random number is uploaded to SP

For SP: $PK=(N,g), SK=(\lamda, \mu)$

---

Iteration Phase:

• CP computes the secure distance function, sens $\prod_{k=1}^K C_k$ and $C_k^{2^{a_k}}$ to SP
• SP decrypts, and obtains $w_k-a_k$
• SP sends $E(w_k-a_k)$ and $E(\sum_{k=1}^K(w_k-a_k)\cdot r_{m,k})$ to CP
• CP computes $E(\sum_{k=1}^K w_kx_m,k)$ and $E(\sum_{k=1}^K w_k)$, and sends to SP
• SP decrypts, and computes ground truth $t_m$

For sensed data $x_{m,k}$, worker chooses two random numbers $r_{m,k},r'_{m,k}$:

$$\~{x_{m,k}} = x_{m,k} - r_{m,k}$$ $$\~{x_{m,k}^2} = x_{m,k}^2 - r'_{m,k}$$

to SP: random values to CP: perturbed data

For SP:

$$E(r_{m,k}) = g^{r_{m,k}} \cdot h^r$$ $$E(r'_{m,k}) = g^{r'_{m,k}} \cdot h^r'$$

to CP: encrypted random values

For CP:

$$E(x_{m,k}) = E(\~{x}_{m,k}) \cdot E(r_{m,k})$$ $$E(x^2_{m,k}) = E(\~{x}^2_{m,k}) \cdot \cdot E(r'_{m,k})$$

Secure Weight Estimation:

For CP:

computes the secure distance function $E((x_{m,k} - t_m)^2)$

$$E((x_{m,k} - t_m)^2) = E(x^2_{m,k}) \cdot E(t^2_m) \cdot E(x_{m,k})^{-2t_m}$$

aggregates the distance of objects and workers:

$$C_k = \prod_{m=1}^M E((x_{m,k} - t_m)^2)$$ $$C = \prod_{k=1}^K C_k$$

chooses a random number $a_k$, computes $C'_{k} = (C_k)^{2^{a_k}}$

to SP: $C'_k$ and $C$

SP:

decrypts and computes $w_k - a_k$

Truth Estimation:

For SP:

computes $E(\sum_{k=1}^K(w_k - a_k)\cdot r_{m,k})$

to CP: $E(w_k - a_k)$

For CP:

computes:

$$E(\sum_{k=1}^K w_k) = \prod_{k=1}^K E(w_k - a_k)\cdot E(a_k)$$

and $E(\sum_{k=1}^K w_kx_{m,k})$

to SP: $E(\sum_{k=1}^K w_kx_{m,k})$, $E(\sum_{k=1}^K w_k)$

For SP:

$t_m = \sum_{k=1}^K w_kx_{m,k} / \sum_{k=1}^K w_k$

---

Rewards Phase:

• each worker redeem his rewards from SP

For CP:

aggregates all random numbers $S^i = \sum_{k=1}^K a_k$

For SP:

computes the total weight $W^i = \sum_{k=1}^K (w_k - a_k) + S^i$

$$ps_{k}^i = \frac{w_k - a_k}{W^i} \cdot P^i$$

For CP:

$$pc_k^i = \frac{a_k}{W^i} \cdot P^i$$

---

1. Q. Li, Y. Li, J. Gao, B. Zhao, W. Fan, and J. Han, “Resolving conflicts in heterogeneous data by truth discovery and source reliability estimation,” in Proceedings of 2014 ACM SIGMOD international conference on Management of data (SIGMOD/PODS). ACM, 2014, pp. 1187–1198. ↩

2. Y. Li, Q. Li, J. Gao, L. Su, B. Zhao, W. Fan, and J. Han, “Conflicts to harmony: A framework for resolving conflicts in heterogeneous data by truth discovery,” IEEE Transactions on Knowledge and Data Engineering, vol. 28, no. 8, pp. 1986–1999, 2016. ↩

3. X. Yin, J. Han, and S. Y. Philip, “Truth discovery with multiple conflicting information providers on the web,” IEEE Transactions on Knowledge and Data Engineering, vol. 20, no. 6, pp. 796–808, 2008. ↩

4. C. Miao, W. Jiang, L. Su, Y. Li, S. Guo, Z. Qin, H. Xiao, J. Gao, and K. Ren, “Cloud-enabled privacy-preserving truth discovery in crowd sensing systems,” in Proceedings of the 13th ACM Conference on Embedded Networked Sensor Systems (SenSys). ACM, 2015, pp. 183–196. ↩

5. Y. Zheng, H. Duan, X. Yuan, and C. Wang, “Privacy-aware and efficient mobile crowdsensing with truth discovery,” IEEE Transactions on Dependable and Secure Computing, vol. 17, no. 1, pp. 121–133, 2020. ↩

6. G. Xu, H. Li, C. Tan, D. Liu, Y. Dai, and K. Yang, “Achieving efficient and privacy-preserving truth discovery in crowd sensing systems,” Computers & Security, vol. 69, pp. 114–126, 2017. ↩

7. Y. Li, C. Miao, L. Su, J. Gao, Q. Li, B. Ding, Z. Qin, and K. Ren, “An efficient two-layer mechanism for privacy-preserving truth discovery,” in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018, pp. 1705–1714. ↩

8. Y. Li, H. Xiao, Z. Qin, C. Miao, L. Su, J. Gao, K. Ren, and B. Ding, “Towards differentially private truth discovery for crowd sensing systems,” arXiv preprint arXiv:1810.04760, 2018. ↩

9. Y. Zheng, H. Duan, X. Yuan, and C. Wang, “Privacy-aware and efficient mobile crowdsensing with truth discovery,” IEEE Transactions on Dependable and Secure Computing, vol. 17, no. 1, pp. 121–133, 2020. ↩

10. C. Miao, L. Su, W. Jiang, Y. Li, and M. Tian, “A lightweight privacy-preserving truth discovery framework for mobile crowd sensing systems,” in Proceedings of 2017 IEEE Conference on Computer Communications (Infocom). IEEE, 2017, pp. 1–9. ↩

11. X. Tang, C. Wang, X. Yuan, and Q. Wang, “Non-interactive privacy-preserving truth discovery in crowd sensing application,” in Proceedings of 2018 IEEE Conference on Computer Communications (INFOCOM). IEEE, 2018, pp. 1–9. ↩

12. Y. Zheng, H. Duan, and C. Wang, “Learning the truth privately and confidently: Encrypted confidence-aware truth discovery in mobile crowdsensing,” IEEE Transactions on Information Forensics and Security, vol. 13, no. 10, pp. 2475–2489, 2018. ↩