Digital › Security2 min read

Shandong University's BEATS verifies batches of ECDSA signatures with under 1 GB of memory

Shandong University researchers proposed BEATS, a batch ECDSA verification scheme built on IVC. On a standard VM the verifier ran 48-240 times faster than direct SNARK constructions.

Shandong University's BEATS verifies batches of ECDSA signatures with under 1 GB of memory
Photo: devopstom, Wikimedia Commons, CC BY-SA 2.0 · License: CC BY-SA 2.0

On 5 November 2026, EurekAlert published a release about BEATS (Batch ECDSA Transaction verification Scheme), developed by a research team at Shandong University. The scheme applies memory-friendly Incremental Verifiable Computation (IVC) to batch verification of ECDSA signatures. According to the authors, it supports an arbitrary number of signatures with memory use under 1 GB. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗ The work was published in Frontiers of Computer Science.

The signature verification bottleneck

As blockchain transaction volumes grow, signature verification has become a limit on scalability, the authors say. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗ Digital signatures ensure that only authorized transactions are recorded, but standard ECDSA verification is computationally expensive when scaled to millions of transactions. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗

The authors name two problems in existing batch verification methods: high memory consumption and loss of efficiency as batch size grows. Such methods often require specialized, high-performance hardware, which limits participation in decentralized networks. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗

How BEATS works and what was measured

BEATS uses IVC to break the verification of a large batch of signatures into small incremental steps that are processed recursively. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗ This allows proofs to be generated in parallel while data is still being received from the network, which hides the computational latency. The authors say this makes the framework suitable for low-latency applications. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗

The evaluation was run on a standard VM. Compared with direct SNARK constructions such as SpartanBP, BEATS sped up the prover by 3-7 times and the verifier by 48-240 times, according to the authors. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗

On very large batches, the verifier speedup reached up to 174 times over the baseline. Peak memory usage stayed below 1 GB. The authors conclude that nodes with modest hardware could therefore take part in high-throughput blockchain networks. eurekalert.orgEurekAlerteurekalert.orgOpen source ↗

The figures come from the authors' own evaluation as reported in the release; the material does not describe independent verification or deployment in a live network.

Sources · 1

Found a mistake?
  1. ↩
Add RealTech as a preferred source in Google

Read also

Up next

The next story opens automatically

More
Follow us

Type at least two letters — results appear instantly

↑↓to choose↵to openescto close