Coladangelo and Colleagues Demonstrate 1.1n Output Length for Single-Copy Secure Pseudorandom States

Researchers Boyang Chen at Tsinghua University, in collaboration with the University of Washington and UC Santa Barbara, have identified a fundamental limit to extending quantum pseudorandom states, mirroring classical pseudorandom generators but with crucial distinctions. The work demonstrates the first definitive separation between single-copy secure pseudorandom states ($\mathsf{1PRS}$) of varying output lengths, proving the existence of such states with a limited stretch, specifically up to 1.1 times the input length, and simultaneously showing that states with sharply greater expansion, of Ω(n 2+ε ), do not exist. This finding, achieved through a new bounding technique within the Common Haar Random State model, strongly advances understanding of the inherent constraints in generating truly scalable quantum randomness and has implications for the development of secure cryptographic protocols.

Quantum pseudorandom states exhibit a definitive limit to length expansion

Single-copy secure pseudorandom states ($\mathsf{1PRS}$) cannot be stretched to arbitrary polynomial lengths, establishing a limit previously assumed absent in quantum systems, according to work from Tsinghua University and the University of Washington. The concept of pseudorandomness is central to modern cryptography, allowing for the generation of sequences that appear random despite being produced by deterministic algorithms. Classical pseudorandom generators (PRGs) can, in theory, produce outputs of any desired length, given sufficient computational resources. However, the quantum realm introduces unique challenges and opportunities. A modest expansion of the input is represented by the existence of $\mathsf{1PRS}$ with an output length of 1.1n, where ‘n’ represents the key size, while the creation of $\mathsf{1PRS}$ with output lengths of Ω(n 2 +ε) is impossible under the constructed quantum oracle. This finding represents the first definitive separation of $\mathsf{1PRS}$ based on output length, challenging the long-held belief that quantum pseudorandom states could be extended to any desired polynomial length, much like their classical counterparts. The separation is achieved via a ‘black-box’ approach, meaning the result holds regardless of the specific internal workings of the pseudorandom state generator, focusing solely on input-output behaviour.

States increasing input length to 1.1n have been proven to exist, demonstrating a degree of stretch is achievable within the quantum framework. However, generating $\mathsf{1PRS}$ with an output length of at least n 2 +ε, a significantly larger expansion, is impossible, with ε representing an arbitrarily small positive number. This research utilised the Common Haar Random State model, a method for generating quantum states based on random unitary transformations, alongside a new technique to measure the resources needed by any $\mathsf{1PRS}$ generator within this model. The Common Haar Random State model is frequently employed in theoretical cryptography due to its relative simplicity and mathematical tractability. Further analysis revealed that the attack’s success hinges on bounding the effective number of these resource states, thereby limiting the potential for larger expansions. Specifically, the researchers demonstrate that any attempt to achieve an expansion of Ω(n 2 +ε) requires an impractically large number of quantum states, rendering it infeasible. This bounding technique represents a novel contribution to the field, providing a powerful tool for analysing the limitations of quantum pseudorandomness.

Limitations on expanding quantum states impact secure communication protocols

The findings challenge assumptions about scaling quantum cryptography, a field reliant on generating unpredictable sequences of quantum states for secure communication. Quantum Key Distribution (QKD), for example, relies on the principles of quantum mechanics to ensure secure key exchange, but often requires substantial random number generation. Quantum systems face inherent limitations, stemming from the no-cloning theorem and the fragility of quantum states, while classical pseudorandom generators can theoretically produce outputs of any desired length. This distinction is particularly acute when considering the Common Haar Random State model, raising concerns about its suitability for all applications. The inability to arbitrarily expand quantum pseudorandom states necessitates a careful consideration of the trade-offs between security and efficiency in quantum cryptographic systems.

Vital insight into the practical boundaries of quantum cryptography systems is provided by the fact that quantum random sequences cannot be expanded limitlessly. Outputs 1.1 times the input length can be created, but substantially larger expansions are not possible. This has direct implications for applications requiring long random sequences, such as generating one-time pads or masking data. A distinction has been made between methods of generating quantum randomness, highlighting the need for careful consideration when designing systems requiring lengthy outputs. Establishing a clear boundary on how much quantum pseudorandomness can be expanded represents a strong advance in cryptography, with implications extending to the design of secure communication protocols. The resource constraints identified necessitate a re-evaluation of existing cryptographic schemes and the development of new approaches that account for these limitations. Future research may focus on exploring alternative models for generating quantum pseudorandom states or developing hybrid approaches that combine classical and quantum techniques to overcome these limitations. The work also underscores the importance of rigorously analysing the security of quantum cryptographic protocols in light of these fundamental constraints, ensuring that deployed systems remain robust against potential attacks.

The research demonstrated a fundamental limit to expanding quantum pseudorandom states, showing outputs can be generated up to 1.1 times the input length but not to lengths that grow much faster. This matters because quantum systems, unlike classical ones, cannot create arbitrarily long random sequences due to inherent physical constraints. The findings establish a clear boundary for quantum randomness expansion, informing the design of secure communication protocols and cryptographic systems. Researchers suggest further work may explore alternative methods for generating these states or combining quantum and classical techniques.