Horcrux: Synthesize, Split, Shift and Stay Alive; Preventing Channel Depletion via Universal and Enhanced Multi-hop Payments

Anqi Tian

Network and Distributed System Security (NDSS) Symposium 2025 · Day 3 · Blockchain Security 2

Overview

This article delves into "Horcrux," a groundbreaking protocol presented at the NDSS Symposium, designed to fundamentally address the persistent problem of fund depletion in payment channel networks (PCNs). Presented by Anqi Tian, this work offers a novel approach to enhance the scalability and reliability of blockchain-based payment systems, particularly those relying on off-chain channels like the Lightning Network. The core innovation of Horcrux lies in its ability to synthesize, split, and shift funds within multi-hop payments, ensuring channels remain balanced and operational, thereby preventing the significant performance degradation and service interruptions caused by depleted channels.

Watch on YouTube · Slides

Key moments

  1. 0:00 Introduction and the channel depletion problem
  2. 3:20 Limitations of current channel rebalancing solutions
  3. 4:45 Horcrux's three-party virtual channel construction
  4. 6:25 Novel incentive fee model for secure updates
  5. 7:00 Time slicing to remove intermediary online assumption
  6. 7:50 Overview of Horcrux's final construction and phases
  7. 9:40 Performance evaluation: high success rate, zero depletion

Horcrux: Synthesize, Split, Shift and Stay Alive; Preventing Channel Depletion via Universal and Enhanced Multi-hop Payments

Speakers: Anqi Tian

Conference: NDSS Symposium

YouTube: https://www.youtube.com/watch?v=aov2dvkkMIY

Overview

This article delves into "Horcrux," a groundbreaking protocol presented at the NDSS Symposium, designed to fundamentally address the persistent problem of fund depletion in payment channel networks (PCNs). Presented by Anqi Tian, this work offers a novel approach to enhance the scalability and reliability of blockchain-based payment systems, particularly those relying on off-chain channels like the Lightning Network. The core innovation of Horcrux lies in its ability to synthesize, split, and shift funds within multi-hop payments, ensuring channels remain balanced and operational, thereby preventing the significant performance degradation and service interruptions caused by depleted channels.

The talk highlights how existing rebalancing solutions for PCNs often fall short due to limitations such as reliance on costly on-chain transactions, extra trust assumptions, or restricted rebalancing capabilities. Horcrux introduces concepts like flow neutrality and multi-party virtual channels (VCs), coupled with an incentive fee model and time slicing, to create a universally applicable and trustless solution. This not only enhances the payment capabilities and privacy of PCNs but also significantly reduces operational costs and improves the overall success rate of multi-hop payments, marking a significant advancement in off-chain scalability solutions.

The importance of Horcrux cannot be overstated in the context of increasing blockchain adoption. As centralized payment systems continue to process vast numbers of transactions, blockchain networks struggle with inherent scalability limitations. PCNs offer a promising avenue for off-chain scaling, but their widespread utility is hampered by the fragility introduced by fund depletion. Horcrux provides a robust, efficient, and secure mechanism to overcome this critical hurdle, paving the way for more stable, efficient, and widely adopted decentralized payment systems.

Background

▶ Watch: Introduction and the channel depletion problem (0:00)

The rapid growth of blockchain technology has brought with it the challenge of scalability. While blockchain-based payment systems offer decentralization and security, they inherently struggle to match the transaction throughput of traditional centralized systems. This bottleneck necessitates innovative scaling solutions, among which payment channels have emerged as a leading contender. A payment channel allows two users to conduct an arbitrary number of off-chain transactions after establishing a single on-chain channel, requiring only two on-chain transactions (one to open, one to close). Extending this concept, payment channel networks (PCNs), exemplified by the popular Lightning Network, enable any two users to transact with each other through a chain of existing channels, without the need to establish a direct channel between them.

Despite their promise, PCNs face a critical challenge known as fund depletion. This occurs when a channel's liquidity is significantly skewed in one direction due to the repeated use of the same payment path for multi-hop payments (MHPs). Once a channel becomes depleted, it can only facilitate unidirectional payments, severely limiting its utility and leading to an imbalanced PCN. This imbalance dramatically shortens the lifespan of involved channels and can cause payment failures across the network.

Prior attempts to address fund depletion have varied in their approaches and effectiveness. Early methods like splicing and loop required an on-chain transaction to replenish funds, which is costly and only provides a one-time refunding. More sophisticated rebalancing techniques emerged, broadly classified into cycle-based rebalancing and non-cycle-based rebalancing.

Cycle-based rebalancing, as seen in works like Revive (CCS 2017), aims to rebalance channels by routing a payment through a cyclical path. This method utilizes coins from adjacent channels without incurring on-chain costs. However, cycle-based solutions suffer from significant drawbacks: they are heavily reliant on the existence of suitable cycle paths, offer limited rebalancing amounts (e.g., if a channel has only one coin, only one coin can be rebalanced), and often introduce extra trust assumptions, such as requiring an honest leader in Revive.

To overcome the path restrictions of cycle-based approaches, non-cycle-based solutions were developed. An example is SADUF (NDSS 2022), which enables coin transfers between adjacent channels through an on-chain binding. SADUF boasts no path restrictions and supports multiple rebalancing operations. Yet, it too has limitations: it typically supports only single-node rebalancing, requires the use of smart contracts, and still incurs some on-chain costs for the binding mechanism.

In summary, the existing landscape of rebalancing solutions for PCNs presented three major limitations:

  1. Inadequate focus on fundamental depletion: Most solutions merely address the symptoms of depletion in individual channels rather than providing a universal, preventative solution.
  2. Reliance on scripting languages and smart contracts: This introduces complexity, potential vulnerabilities, and platform-specific dependencies.
  3. Extra trust assumptions: Many methods require an honest leader or continuous online presence, undermining the decentralized and trustless ethos of blockchain.

The overarching goal for further research, and the problem Horcrux aims to solve, was to fundamentally address channel depletion while removing these restrictive assumptions, including the need for on-chain interactions, smart contracts, and constant online presence.

Key Findings

▶ Watch: Horcrux's three-party virtual channel construction (4:45)

Horcrux presents several pivotal findings and contributions that fundamentally redefine how fund depletion in payment channel networks can be prevented. The core innovations center around achieving flow neutrality through novel cryptographic and protocol designs, effectively removing the limitations of prior rebalancing solutions.

The primary discovery is a method to fundamentally address channel depletion by ensuring intermediaries in multi-hop payments allocate funds to offset balance shifts. This concept, termed flow neutrality, keeps involved channels from becoming imbalanced and significantly amplifies their payment capabilities by leveraging balances across multiple channels. Unlike previous reactive rebalancing methods, Horcrux is designed to prevent imbalance proactively.

Another key finding is the successful construction of a multi-party virtual channel (VC) that circumvents the high costs and privacy concerns associated with direct fund transfers between channels. This VC design allows for multiple rebalancing operations while keeping payment details invisible to intermediaries, thus enhancing privacy. Crucially, this construction also enables the offloading of VC closure from the underlying channels, making it resilient to domino attacks—a vulnerability identified in prior work like Donor (NSS 2023), where the closure of one channel could trigger a cascade of closures.

Horcrux also successfully addresses the critical challenge of removing extra trust assumptions inherent in previous solutions. Specifically, it eliminates the need for an online assumption for all parties and the reliance on smart contracts. This is achieved through an innovative incentive fee model based on time and the technique of time slicing, which together ensure the secure and honest updating of channel states even when intermediaries are offline for significant periods.

The practical implementation and evaluation of Horcrux reveal its superior performance compared to existing solutions. Experiments conducted on real Lightning Network datasets demonstrate significantly higher success rates for payments (80% short-term, 90% long-term for Horcrux vs. 60% for SADUF), drastically reduced channel depletion (nearly zero for Horcrux vs. thousands for SADUF and Revive), and substantially lower financial costs (less than $1 total for Horcrux, significantly cheaper than SADUF). These empirical results confirm Horcrux's effectiveness and efficiency, validating its promise as a universal and enhanced multi-hop payment protocol.

Technical Deep Dive

▶ Watch: Novel incentive fee model for secure updates (6:25)

Horcrux's technical ingenuity lies in its multi-layered approach to achieving flow neutrality and robust multi-party virtual channels without burdensome trust assumptions. The protocol's design can be broken down into several interconnected components.

At its core, Horcrux introduces flow neutrality, a concept where intermediaries actively manage their local channel balances to offset the balance shifts caused by Multi-Hop Payments (MHPs). Instead of simply forwarding funds, intermediaries "allocate funds to offset balance shifts," effectively preventing the imbalance that leads to depletion. This proactive approach ensures that the channels involved in an MHP remain balanced, thereby extending their lifespan and enhancing their capacity to facilitate future payments. This is achieved by leveraging balances from two channels, essentially amplifying their combined payment capabilities.

To facilitate this flow neutrality efficiently and privately, Horcrux constructs a three-party virtual channel (VC). This VC allows for the transfer of funds between channels without the need for expensive on-chain transactions or exposing payment details. For instance, if user I needs to facilitate a payment between user A and user B, user I locks a certain amount of coins (e.g., five coins) for the payment within the VC. When the VC is closed, user I reallocates these five coins back to their respective two channels, ensuring that the MHP proceeds without introducing the typical balance shifts. This mechanism supports multiple rebalancing operations within the VC and crucially keeps the actual payment amounts invisible to the intermediaries, significantly enhancing privacy. Furthermore, this VC construction is designed to offload the closure of the VC from its underlying channels. This is a critical security enhancement, making Horcrux resistant to domino attacks, where the failure or closure of one channel could cascade and force the closure of others, as demonstrated by Lucas Armier et al. in their NSS 2023 work, Donor.

While the three-party VC is powerful, the initial iteration still required the intermediary (user I in the example) to remain online for security. To address this, Horcrux extends to a multi-party VC construction that removes the online assumption. The primary challenge here is that payments within the VC are invisible to intermediaries. An adversary could exploit this by tricking intermediaries into updating channels with outdated VC states, which end-users cannot supervise directly. To counteract this, Horcrux introduces a novel incentive fee model based on time. Each VC state is assigned a timestamp. When end-users update VC states and corresponding timestamps during the close phase, intermediaries are incentivized to always choose the latest state. This is because the fee structure is time-based, meaning intermediaries earn more for providing service over a longer duration. By always selecting the most recent state, intermediaries maximize their fees, and as long as at least one honest end-user is present, the VC will be closed securely with the correct, latest state.

To further remove the continuous online assumption for intermediaries, Horcrux employs time slicing. This technique divides the entire lifespan of the VC into small, discrete "slices." Intermediaries are only required to come online during these specific time slices to check for messages or close requests. If they receive a close request and VC states, they stay online to process them; otherwise, they can go idle securely. The talk uses an example where, according to specific parameter settings, intermediaries might only need to be online for a total of one hour across the entire VC lifespan, significantly reducing their operational burden and enabling greater participation.

The complete Horcrux protocol involves three main phases:

  1. Setup Phase: Each intermediary locks a certain amount of coins into a temporary account. This amount is determined by the maximum potential payment between the end-users. For example, if the maximum payment is four coins, four coins might be locked.
  2. Update Phase: During the designated time slices, end-users cooperate to update the VC states and their corresponding timestamps. Intermediaries go online during these slices to check for messages and updates, remaining offline and secure during the interval times.
  3. Close Phase: When end-users decide to close the VC (e.g., user A pays user B two coins), they send the final VC states to the intermediaries. The intermediaries then split any necessary "extra" coins to their right channel to facilitate the final payment between the end-users. Crucially, multi-channel updates are performed consistently across all involved channels. After successfully updating their channels and ensuring flow neutrality, intermediaries collect their fees, and their overall channel balances remain unchanged, demonstrating the effectiveness of the flow-neutral rebalancing.

This detailed construction ensures that Horcrux is an efficient, multi-party VC protocol that fundamentally addresses fund depletion, achieves high universality, and operates with minimal trust assumptions.

Demo / Proof of Concept

▶ Watch: Overview of Horcrux's final construction and phases (7:50)

The talk provided a clear demonstration of Horcrux's practical viability and superior performance through its implementation and rigorous evaluation. The researchers implemented the Horcrux protocol to assess its time and financial costs, as well as its effectiveness in preventing channel depletion compared to existing solutions.

The financial cost analysis revealed that Horcrux is exceptionally economical. The total cost for using Horcrux was measured to be less than $1, making it significantly cheaper than alternative protocols like SADUF. This low cost is a critical advantage, as high transaction or rebalancing fees can deter widespread adoption of off-chain solutions.

Beyond cost, a key part of the demonstration involved an experiment conducted on a real Lightning Network (LN) dataset to evaluate Horcrux's performance under realistic conditions. The results were compelling:

  • Short-term Performance: Horcrux achieved an impressive 80% success rate for multi-hop payments. In stark contrast, SADUF, a prominent non-cycle-based rebalancing solution, only managed a 60% success rate. More critically, while SADUF and cycle-based solutions like Revive resulted in "thousands of depleted channels" over the short term, Horcrux maintained nearly zero channel depletion. This highlights Horcrux's proactive prevention of balance skew.
  • Long-term Performance: The advantages of Horcrux became even more pronounced over extended periods. The success rate for Horcrux improved to 90% in the long term, representing an 18% improvement over its short-term performance. Crucially, even in the long run, Horcrux continued to exhibit nearly zero channel depletion. This sustained high performance and stability are directly attributed to Horcrux's ability to mitigate balance skills introduced by multi-hop payments.

The experimental evaluation unequivocally demonstrated that Horcrux outperforms current solutions across critical metrics: it achieves a higher payment success rate, requires less initial deposit (due to efficient fund utilization), and virtually eliminates channel depletion. This empirical evidence validates Horcrux as a robust and efficient protocol for enhancing the stability and reliability of payment channel networks.

Defensive Implications

▶ Watch: Performance evaluation: high success rate, zero depletion (9:40)

Horcrux offers significant defensive implications for operators, users, and developers within payment channel networks, particularly those based on technologies like the Lightning Network. Its fundamental approach to preventing channel depletion transforms the operational landscape, leading to a more stable, efficient, and resilient off-chain payment ecosystem.

For network operators and liquidity providers, Horcrux provides a powerful tool to maintain the health and profitability of their channels. By adopting Horcrux, operators can dramatically reduce the incidence of channel depletion, which currently necessitates costly and often inefficient rebalancing operations. The flow neutrality mechanism means that channels remain balanced even after numerous multi-hop payments, reducing the need for manual intervention or reliance on external rebalancing services. This translates to lower operational overhead, more predictable liquidity management, and ultimately, a higher return on investment for locked funds. The resistance to domino attacks also means increased network stability, as the failure of one channel is less likely to trigger a cascading effect across the network.

End-users benefit from a significantly improved payment experience. The higher payment success rates (80-90%) demonstrated by Horcrux mean fewer failed transactions and a more reliable service. This directly addresses a major pain point in current PCNs where payments can frequently fail due to insufficient liquidity along a path. Furthermore, the enhanced privacy features, where payment details within virtual channels are invisible to intermediaries, bolster user confidence and security. The removal of the online assumption for intermediaries means that network participants can operate with greater flexibility without needing to be constantly connected, improving the overall user experience.

For developers and protocol designers, Horcrux presents a robust framework for building the next generation of scalable blockchain applications. The principles of multi-party virtual channels, incentive fee models based on time, and time slicing offer reusable design patterns for creating trustless and efficient off-chain protocols. The fact that Horcrux operates without requiring smart contracts or continuous online presence for all parties simplifies development and deployment, making it more accessible and less prone to platform-specific vulnerabilities. Integrating Horcrux-like mechanisms into future PCN implementations could lead to more robust, scalable, and user-friendly decentralized payment systems.

In essence, Horcrux shifts the defensive paradigm from reactive damage control (rebalancing depleted channels) to proactive prevention. It provides the architectural foundation for PCNs to "stay alive" and thrive, fostering greater adoption and utility of off-chain scaling solutions by addressing their most critical vulnerability.

Key Takeaways

  • Fundamental Depletion Prevention: Horcrux fundamentally addresses channel fund depletion in PCNs by implementing flow neutrality, ensuring intermediaries offset balance shifts during multi-hop payments, proactively preventing channel imbalance.
  • Trustless Multi-Party Virtual Channels: The protocol introduces an efficient and private multi-party virtual channel (VC) construction that allows for multiple rebalancing operations without exposing payment details to intermediaries, enhancing privacy.
  • Elimination of Trust Assumptions: Horcrux removes critical trust assumptions, including the need for continuous online presence and reliance on smart contracts, through an incentive fee model based on time and time slicing.
  • Enhanced Network Resilience: The VC design is resistant to domino attacks, where the closure of one channel could lead to cascading failures, thereby significantly improving the overall stability and resilience of PCNs.
  • Superior Performance and Cost-Effectiveness: Empirical evaluations show Horcrux achieves significantly higher payment success rates (80-90%), nearly zero channel depletion, and substantially lower operational costs (less than $1 total) compared to existing solutions like SADUF.
  • Universal Applicability: Horcrux offers a universal solution for enhancing the scalability and reliability of blockchain-based payment systems, providing a robust framework for future off-chain scaling efforts.

About the Speaker(s)

Anqi Tian is the presenter of the Horcrux protocol at the NDSS Symposium. While specific title and company details were not provided in the transcript or metadata, the context of presenting a detailed technical paper at a prestigious security conference like NDSS indicates that Anqi Tian is a researcher actively involved in cutting-edge work in blockchain security and scalability. The presentation details a "joint work," suggesting collaboration within a research institution or academic setting focused on offering fundamental solutions to complex problems in decentralized systems.

Reviews

Dr. Zero (Offensive Security Researcher) — SOLID

Horcrux is legitimate academic cryptography/systems work on a real and underappreciated problem in payment channel networks. The flow-neutrality framing, time-sliced online requirements, and domino-attack resistance are genuine technical contributions. But this is a conference proceedings presentation, not a security research talk in the traditional offensive/defensive sense — and the write-up reads like an AI-generated summary of the paper rather than a firsthand account of the talk.

Heather Calloway (CISO) — PASS

Rigorous cryptographic research on payment channel network rebalancing — this is outside my lane entirely. No governance angle, no institutional risk exposure, no defender or operator relevance for enterprise security programs.

→ Top-rated talks at Network and Distributed System Security (NDSS) Symposium 2025

All talks from Network and Distributed System Security (NDSS) Symposium 2025