Entanglement-Enabled Communication
Abstract: We introduce and analyse a multiple-access channel with two senders and one receiver, in the presence of i.i.d. noise coming from the environment. Partial side information about the environmental states allows the senders to modulate their signals accordingly. An adversarial jammer with its own access to information on environmental states and the modulation signals can jam a fraction of the transmissions. Our results show that for many choices of the system parameters, entanglement shared between the two senders allows them to communicate at non-zero rates with the receiver, while for the same parameters the system forbids any communication without entanglement-assistance, even if the senders have access to common randomness (local correlations). A simplified model displaying a similar behaviour but with a compound channel instead of a jammer is outlined to introduce basic aspects of the modeling. We complement these results by demonstrating that there even exist model parameters for which entanglement-assisted communication is no longer possible, but a hypothetical use of nonlocal no-signalling correlations between Alice and Bob could enable them to communicate to Charlie again.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Knowledge Gaps
Below is a single, consolidated list of knowledge gaps, limitations, and open questions that remain unresolved in the paper. Each item is framed to be concrete and actionable for future research.
- Precise capacity-region characterization: The paper proves existence of nontrivial entanglement-assisted rates for the MAVMACEI but does not derive the exact capacity region (e.g., sum rate and corner points) as a function of the jammer power constraint and the environmental distribution. Provide closed-form or computable multi-letter characterizations and tight inner/outer bounds.
- Tightness of thresholds: The classical zero-capacity threshold at and the entanglement-assisted threshold at are presented without a proof of optimality. Determine whether these thresholds are tight and whether improved entanglement strategies (e.g., higher-dimensional entanglement, other nonlocal games) can raise the entanglement-assisted threshold.
- Optimality of EPR-modulation scheme: The chosen measurement angles (, , , ) are motivated by CHSH but not proven optimal for the communication objective. Optimize measurement settings (and possibly state dimension) specifically for capacity under jamming, rather than CHSH winning probability.
- Generality beyond binary alphabets: The model is fully binary (inputs, outputs, states). Extend the construction and separation results to non-binary discrete alphabets and continuous channels (e.g., Gaussian MACs with power constraints), and assess whether the qualitative separations persist.
- Robustness to environmental models: Environmental states are i.i.d. and uniform. Analyze performance and thresholds under non-uniform, correlated, or memoryful environmental processes (e.g., Markovian ), and under noisy or delayed side-information at the senders.
- Jammer knowledge models: The jammer (James) observes and perfectly and is noncausal across the block subject to a per-block cost . Characterize capacities under:
- Causal jammers (state choices depending only on past).
- Partial/noisy jammer side-information (e.g., corrupted, partially hidden).
- Alternative cost functions (e.g., weighted costs, burst constraints).
- Symmetrizability conditions with environmental information: The zero-capacity claims rely implicitly on AVC symmetrizability. Provide necessary and sufficient symmetrizability conditions for AV(MA)CEI with partial environmental information and derive general tests that certify when classical capacity is zero versus positive.
- Role of common randomness with the receiver: The paper considers local correlations between the senders but not common randomness shared with the receiver, which in classical AVCs can break symmetrizability and yield positive random-code capacity. Analyze whether sender–receiver common randomness destroys the separation, and quantify the amount needed.
- Coding constructions: No explicit error-correcting code families are given for the entanglement-assisted setting. Develop constructive, efficiently decodable codes achieving the entanglement-assisted rate region and analyze their complexity.
- Finite-blocklength analysis: Only asymptotic capacity is considered. Provide finite-blocklength bounds (dispersion, error exponents) for both classical and entanglement-assisted MAVMACEI to understand practical performance and latency.
- Adaptive or multi-round modulation: The modulation is single-shot per use. Investigate whether adaptive measurement strategies (depending on past observations) or multi-round entanglement-assisted protocols further increase resilience to jamming.
- More-than-two users: The construction targets two senders. Extend the model and separation results to -user MACs and characterize how entanglement scales with the number of senders.
- Alternative nonlocal games: The analysis hinges on CHSH-type correlations. Explore whether other nonlocal games (e.g., magic square, KCBS, GYNI) or high-dimensional Bell inequalities yield stronger modulation advantages for communication under jamming.
- Non-signalling versus quantum correlations: The paper mentions parameter regimes where non-signalling correlations enable communication while entanglement does not, but does not give a full characterization or constructive proof. Specify the regimes, provide explicit constructions, and quantify the exact capacity gains under non-signalling assistance.
- Impact of partial sender cooperation: The model forbids any communication between Alice and Bob. Quantify how limited inter-sender classical communication (or side-channel constraints) affects the separation and capacity, both with and without entanglement.
- Decoder side-information: Charlie has no environmental side-information. Evaluate how providing Charlie with partial or noisy information about , , or affects the capacity region and the classical–quantum separation.
- Sensitivity to implementation imperfections: The entanglement-based modulation assumes ideal state preparation and measurement. Analyze robustness to entanglement fidelity loss, detector inefficiency, misalignment of measurement settings, and drift, and quantify the tolerable noise levels that preserve the separation.
- Resource accounting: The amount of entanglement per channel use (ebits consumed) and its distribution overhead are not quantified. Establish rate–resource trade-offs (capacity per ebit, entanglement cost per bit) and include constraints on entanglement generation/distribution.
- Model variations of the adder stage: The first stage uses a modulo-2 adder . Investigate alternative MAC combining rules (e.g., sum over reals with power constraints, OR/XOR with erasures) and whether the separation persists.
- Extension from compound to arbitrarily varying channels: The preliminary compound-channel intuition uses convexity to interchange min–max but does not fully specify conditions. Formalize the required convexity/compactness/continuity assumptions that validate the interchange and generalize to the AV(MA)CEI setting.
- Jamming strategy optimality: The paper assumes James flips only positions not already in error to maximize harm. Prove this strategy’s optimality for the given model or derive the truly optimal jammer policy against entanglement-modulated schemes.
- Interaction with receiver-side entanglement (category A+B): The paper distinguishes category B (entanglement among senders only). Study hybrid architectures where senders and receiver share entanglement and compare capacity gains against sender-only entanglement.
- Explicit rate quantification for the entanglement-assisted MAC through BSC composition: The paper informally relates the effective channel to a BSC with parameter but does not translate this into exact achievable MAC rates after the adder. Provide the full rate region based on mutual information expressions for the composed MAC.
- Generalization of environmental side-information granularity: In the specific MAVMACEI, senders know their local bits exactly ( for Alice, for Bob). Analyze intermediate regimes (noisy/partial/delayed side-information at senders) and quantify how entanglement-assisted gains degrade or persist.
- Practical distribution of entanglement in category B: The paper motivates category B by proximity of senders but does not address practical entanglement distribution networks, repeaters, or storage decoherence. Propose architectures and quantify feasibility under realistic constraints.
- Security and stealth aspects under adversarial jamming: While adversarial jamming is modeled, the framework does not address detection, estimation, or mitigation strategies beyond capacity. Explore jammer detection protocols compatible with entanglement-assisted modulation.
- Interplay with classical randomization and coding: Beyond local correlations, explore whether sophisticated classical randomized coding (e.g., stochastic encoders, mixed strategies) can partially close the gap and identify fundamental limits of purely classical schemes under this MAVMACEI.
- Extensions to interference/broadcast/relay networks: The paper focuses on MAC. Assess whether similar entanglement-enabled separations occur in interference channels, broadcast channels, or relay networks under environmental states and jamming.
Collections
Sign up for free to add this paper to one or more collections.