Cryptocurrency introduces usability challenges by requiring users to manage signing keys. Popular signing key management services (e.g., custodial wallets), however, either introduce a trusted party or burden users with managing signing key shares, posing the same usability challenges. TEE (Trusted Execution Environment) is a promising technology to avoid both, but practical implementations of TEEs suffer from various side-channel attacks that have proven hard to eliminate. This paper explores a new approach to side-channel mitigation through economic incentives for TEE-based cryptocurrency wallet solutions. By taking the cost and profit of side-channel attacks into consideration, we designed a Stick-and-Carrot-based cryptocurrency wallet, CrudiTEE, that leverages penalties (the stick) and rewards (the carrot) to disincentivize attackers from exfiltrating signing keys in the first place. We model the attacker’s behavior using a Markov Decision Process (MDP) to evaluate the effectiveness of the bounty and enable the service provider to adjust the parameters of the bounty’s reward function accordingly.

Payment Channel Network (PCN) is proposed as a promising layer-two solution to tackle the scalability problem of current blockchain systems, which allows the two transacting parties to perform off-chain transactions through their established payment channel. For the transacting parties who are not directly connected, PCN allows them to route the transaction through some intermediate nodes with sufficient balance. Designing an efficient routing protocol is one of the most important and challenging problems in improving the performance of PCN. To tackle this challenge, we propose Real-Time Recursive Routing (RTRR), an efficient routing algorithm that can achieve a short routing time with strong privacy protection and high flexibility in the dynamic scenario. In addition, we investigate the bidding process in RTRR and derive the equilibrium strategy, which implies that the proposed protocol prefers to route the transaction through the nodes with a higher success rate, contributing to a better performance. Both the theoretical analyses and the empirical experiment results demonstrate the high efficiency of RTRR.

Loop filters are used in video coding to remove artifacts or improve performance. Recent advances in deploying convolutional neural network (CNN) to replace traditional loop filters show large gains but with problems for practical application. First, different model is used for frames encoded with different quantization parameter (QP), respectively. It is expensive for hardware. Second, float points operation in CNN leads to inconsistency between encoding and decoding across different platforms. Third, redundancy within CNN model consumes precious computational resources. This paper proposes a CNN as the loop filter for intra frames and proposes a scheme to solve the above problems. It aims to design a single CNN model with low redundancy to adapt to decoded frames with different qualities and ensure consistency. To adapt to reconstructions with different qualities, both reconstruction and QP are taken as inputs. After training, the obtained model is compressed to reduce redundancy. To ensure consistency, dynamic fixed points (DFP) are adopted in testing CNN. Parameters in the compressed model are first quantized to DFP and then used for inference of CNN. Outputs of each layer in CNN are computed by DFP operations. Experimental results on JEM 7.0 report 3.14%,5.21 %, 6.28% BD-rate savings for luma and two chroma components with all intra configuration when replacing all traditional filters.