Scalability is a critical challenge for blockchain-based cryptocurrencies and has become a significant bottleneck in large-scale applications such as retail. Payment channel networks (PCNs) are among the most promising solutions to this challenge. However, nodes in real-world PCNs exhibit heterogeneous behaviors and capabilities, causing existing schemes to perform poorly when deployed in practical environments. In this paper, we first demonstrate, through a measurement study, the existence of node heterogeneity in real-world PCNs; and then propose RD-PCN, a role-differentiated PCN that accommodates such heterogeneity by explicitly separating node roles and responsibilities. We address two key challenges in realizing RD-PCN. To improve channel fund utilization, RD-PCN introduces multi-party payment channels (MPCs) to selectively connect nodes at the network edge. We accordingly design a PCN-compatible MPC, termed PC-MPC, representing the first practical solution for deploying MPCs in PCNs. To securely outsource routing tasks to resource-rich nodes, we design a privacy-preserving and fair routing scheme based on trusted execution environments and the proposed topology synchronization mechanism. We prove the security of PC-MPC and the routing scheme under the universally composable framework. Experiments show that RD-PCN improves the payment success ratio by at least 18.9% compared to representative schemes, while increasing fund utilization by 60%.

The Link Flooding Attack (LFA) poses a significant threat to the network as a novel form of indirect, distributed denial-of-service attack. There is an urgent need for a defense capable of eliminating LFA. LFA operates as a closed-loop, continuously monitoring target links and adjusting traffic to adapt to network changes, thereby sustaining the attack. This implies that the effective defense against LFA should be feedback-control-based, enabling adaptive responses to evolving attack strategies. However, existing works for eliminating LFA operate using an open-loop way, which lacks feedback throughout the defense process, preventing them from efficiently detecting and blocking attack flows. In this paper, we adopt the concept of feedback control to propose Loop-Filter, which first filters part of the traffic on the attacked link and then refines the detected attack flows based on feedback from the link and flows to optimize traffic filtering. This process is repeated iteratively until the attacked link returns to normal. To implement the concept, we propose a blockchain-based collaboration defense architecture, along with economic incentives, to promote cooperation between domains. Meanwhile, we leverage reinforcement learning to learn how to rapidly and accurately adjust the detection results of flows based on the feedback. We also design a lossless compression method of filtering rules and dynamically deploy them to drop attack traffic maximally. Finally, we validate the effectiveness and efficiency of Loop-Filter under various traffic conditions.

Payment channel networks (PCNs) are emerged as the most widely deployed solution to mitigate the scalability problem of permissionless cryptocurrencies, allowing vast payments to be carried out off-chain. Routing, which finds feasible paths between the senders and receivers, is critical for PCNs. However, existing solutions either fail to achieve high scalability that can maintain low storage/computation/network communication overhead, or they are susceptible to privacy disclosure. In this paper, we propose SorTEE, a service-oriented routing solution for PCNs, which adopts a set of service nodes to alleviate the per-user burden of routing and achieves more comprehensive privacy guarantees than the state-of-the-art by leveraging trusted execution environments (TEEs). SorTEE demands users communicate with the TEE by the secure channel to protect the privacy of transaction value. Then, an oblivious path mechanism is designed to construct redundant paths with the pseudo senders and receivers generated by TEEs to confuse its untrusted controller. Further, we report a novel attack that allows malicious service nodes to drop the valid paths for profit, and design a feedback mechanism to relieve it. Moreover, SorTEE hides the identities of the senders/receivers for the intermediate nodes of payment paths by introducing a novel identity information transfer scheme called encrypted identity chain. Based on security analysis and performance evaluation, our results demonstrate that SorTEE is able to achieve sufficient privacy-preserving payment and low per-user overhead.

Underwater communication is extremely challenging for small underwater robots which typically have stringent power and size constraints. In our previous work, we developed an artificial electrocommunication system which could be an alternative for the communication of small underwater robots. This paper further presents a new electrocommunication system that utilizes Binary Frequency Shift Keying (2FSK) modulation and deep-learning-based demodulation for underwater robots. We first derive an underwater electrocommunication model that covers both the near-field area and a large transition area outside of the near-field area. 2FSK modulation is adopted to improve the anti-interference ability of the electric signal. A deep learning algorithm is used to demodulate the electric signal by the receiver. Simulations and experiments show that with the same testing condition, the new communication system outperforms the previous system in both the communication distance and the data transmitting rate. In specific, the newly developed communication system achieves stable communication within the distance of 10 m at a data transfer rate of 5 Kbps with a power consumption of less than 0.1 W. The substantial increase in communication distance further improves the possibility of electrocommunication in underwater robotics.