skip to main content


Title: Reinforcement Learning Toward Decision-Making for Multiple Trusted-Third-Parties in PUF-Cash
Electronic money is the digital representation of physical banknotes enabling offline and online payments. An electronic e-Cash scheme, termed PUF- Cash was proposed in prior work. PUF-Cash preserves user anonymity by leveraging the random and unique statistical properties of physically unclonable functions (PUFs). PUF-Cash is extended meaningfully in this work by the introduction of multiple trusted third parties (TTPs) for token blinding and a fractional scheme to diversify and mask Alice's spending habits from the Bank. A reinforcement learning (RL) framework based on stochastic learning automata (SLA) is proposed to efficiently select a subset of TTPs as well as the fractional amounts for blinding per TTP, based on the set of available TTPs, the computational load per TTP and network conditions. An experimental model was constructed in MATLAB with multiple TTPs to verify the learning framework. Results indicate that the RL approach guarantees fast convergence to an efficient selection of TTPs and allocation of fractional amounts in terms of perceived reward for the end-users.  more » « less
Award ID(s):
1849739
NSF-PAR ID:
10228419
Author(s) / Creator(s):
; ; ;
Date Published:
Journal Name:
IEEE 6th World Forum on Internet of Things (WF-IoT)
Page Range / eLocation ID:
1 to 6
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Electronic money or e-Cash is becoming increasingly popular as the preferred strategy for making purchases, both on- and off-line. Several unique attributes of e-Cash are appealing to customers, including the convenience of always having "cash-on-hand" without the need to periodically visit the ATM, the ability to perform peer-to-peer transactions without an intermediary, and the peace of mind associated in conducting those transactions privately. Equally important is that paper money provides customers with an anonymous method of payment, which is highly valued by many individuals. Although anonymity is implicit with fiat money, it is a difficult property to preserve within e-Cash schemes. In this paper, we investigate several artificial intelligence (AI) approaches for improving performance and privacy within a previously proposed e-Cash scheme called PUF-Cash. PUF-Cash utilizes physical unclonable functions (PUFs) for authentication and encryption operations between Alice, the Bank and multiple trusted third parties (mTTPs). The AI methods select a subset of the TTPs and distribute withdrawal amounts to maximize the performance and privacy associated with Alice's e-Cash tokens. Simulation results show the effectiveness of the various AI approaches using a large test-bed architecture. 
    more » « less
  2. null (Ed.)
    Electronic money (e-money or e-Cash) is the digital representation of physical banknotes augmented by added use cases of online and remote payments. This paper presents a novel, anonymous e-money transaction protocol, built based on physical unclonable functions (PUFs), titled PUF-Cash. PUF-Cash preserves user anonymity while enabling both offline and online transaction capability. The PUF’s privacy-preserving property is leveraged to create blinded tokens for transaction anonymity while its hardware-based challenge–response pair authentication scheme provides a secure solution that is impervious to typical protocol attacks. The scheme is inspired from Chaum’s Digicash work in the 1980s and subsequent improvements. Unlike Chaum’s scheme, which relies on Rivest, Shamir and Adlemans’s (RSA’s) multiplicative homomorphic property to provide anonymity, the anonymity scheme proposed in this paper leverages the random and unique statistical properties of synthesized integrated circuits. PUF-Cash is implemented and demonstrated using a set of Xilinx Zynq Field Programmable Gate Arrays (FPGAs). Experimental results suggest that the hardware footprint of the solution is small, and the transaction rate is suitable for large-scale applications. An in-depth security analysis suggests that the solution possesses excellent statistical qualities in the generated authentication and encryption keys, and it is robust against a variety of attack vectors including model-building, impersonation, and side-channel variants. 
    more » « less
  3. Electronic money (e‐money or e‐Cash) is the digital representation of physical banknotes augmented by added use cases of online and remote payments. This paper presents a novel, anonymous e‐money transaction protocol, built based on physical unclonable functions (PUFs), titled PUF‐Cash. PUF‐Cash preserves user anonymity while enabling both offline and online transaction capability. The PUF’s privacy‐preserving property is leveraged to create blinded tokens for transaction anonymity while its hardware‐based challenge–response pair authentication scheme provides a secure solution that is impervious to typical protocol attacks. The scheme is inspired from Chaum’s Digicash work in the 1980s and subsequent improvements. Unlike Chaum’s scheme, which relies on Rivest, Shamir and Adlemans’s (RSA’s) multiplicative homomorphic property to provide anonymity, the anonymity scheme proposed in this paper leverages the random and unique statistical properties of synthesized integrated circuits. PUF‐Cash is implemented and demonstrated using a set of Xilinx Zynq Field Programmable Gate Arrays (FPGAs). Experimental results suggest that the hardware footprint of the solution is small, and the transaction rate is suitable for large‐scale applications. An in‐depth security analysis suggests that the solution possesses excellent statistical qualities in the generated authentication and encryption keys, and it is robust against a variety of attack vectors including model‐building, impersonation, and side‐ channel variants. 
    more » « less
  4. Pedagogical planners can provide adaptive support to students in narrative-centered learning environments by dynamically scaffolding student learning and tailoring problem scenarios. Reinforcement learning (RL) is frequently used for pedagogical planning in narrative-centered learning environments. However, RL-based pedagogical planning raises significant challenges due to the scarcity of data for training RL policies. Most prior work has relied on limited-size datasets and offline RL techniques for policy learning. Unfortunately, offline RL techniques do not support on-demand exploration and evaluation, which can adversely impact the quality of induced policies. To address the limitation of data scarcity and offline RL, we propose INSIGHT, an online RL framework for training data-driven pedagogical policies that optimize student learning in narrative-centered learning environments. The INSIGHT framework consists of three components: a narrative-centered learning environment simulator, a simulated student agent, and an RL-based pedagogical planner agent, which uses a reward metric that is associated with effective student learning processes. The framework enables the generation of synthetic data for on-demand exploration and evaluation of RL-based pedagogical planning. We have implemented INSIGHT with OpenAI Gym for a narrative-centered learning environment testbed with rule-based simulated student agents and a deep Q-learning-based pedagogical planner. Our results show that online deep RL algorithms can induce near-optimal pedagogical policies in the INSIGHT framework, while offline deep RL algorithms only find suboptimal policies even with large amounts of data.

     
    more » « less
  5. Physical unclonable functions (PUF) in silicon exploit die-to-die manufacturing variations during fabrication for uniquely identifying each die. Since it is practically a hard problem to recreate exact silicon features across dies, a PUF-based authentication system is robust, secure and cost-effective, as long as bias removal and error correction are taken into account. In this work, we utilize the effects of inherent process variation on analog and radio-frequency (RF) properties of multiple wireless transmitters (Tx) in a sensor network, and detect the features at the receiver (Rx) using a deep neural network based framework. The proposed mechanism/ framework, called RF-PUF, harnesses already-existing RF communication hardware and does not require any additional PUF-generation circuitry in the Tx for practical implementation. Simulation results indicate that the RF-PUF framework can distinguish up to 10000 transmitters (with standard foundry defined variations for a 65 nm process, leading to non-idealities such as LO offset and I-Q imbalance) under varying channel conditions, with a probability of false detection <10^-3 
    more » « less