🔢 mathematics

Rate Loss in Quantum Channels with Classical State and Applications for Quantum Broadcast Channels

This paper demonstrates that the absence of channel state information at the receiver causes a strict rate loss in non-commutative classical-quantum channels and proves that coset-code-based strategies can strictly outperform conventional unstructured IID codes for 3-user CQ broadcast channels to mitigate this penalty.

Original authors: Igor Bernard, Arun Padakandla

Published 2026-06-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Igor Bernard, Arun Padakandla

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to send a secret message through a noisy, magical tunnel. This tunnel is a Quantum Channel. In this world, the "noise" isn't just static; it's a shifting state of the tunnel itself, like the walls changing color or texture randomly.

Here is the problem the paper solves: Who knows what?

The Sender (Tx): Knows exactly what the tunnel looks like right now. They can adjust their message to fit the current shape of the tunnel.
The Receiver (Rx): Is blindfolded. They don't know what the tunnel looks like. They just see the message arrive and have to guess what it meant.

The Core Discovery: The "Rate Loss" Penalty

The paper proves that if the Receiver is blindfolded to the tunnel's changing state, they suffer a "Rate Loss."

The Analogy:
Imagine you are sending a package through a delivery service.

Scenario A (Both know): You know the truck is small today, so you pack a small box. The driver knows the truck is small, so they know exactly how to stack it. You can send a lot of packages efficiently.
Scenario B (Only you know): You know the truck is small, so you pack a small box. But the driver (Receiver) doesn't know the truck is small. They are used to seeing big trucks. When they see your small box, they have to guess: "Is this a tiny box for a small truck? Or is it a tiny box for a big truck?" Because they are guessing, they have to be more careful, and they can't accept as many packages per hour.

The paper proves that for certain types of "quantum" tunnels (specifically those where the states don't play nicely together, called non-commutative), this penalty is strict. You simply cannot send as much data as you could if the receiver had a peek at the tunnel's state.

The Broadcast Problem: The 3-Way Split

The authors then take this idea to a Broadcast Channel, where one sender tries to talk to three different receivers at the same time.

The Analogy:
Imagine a radio station (Sender) broadcasting to three listeners (Rx 1, Rx 2, Rx 3).

The station sends one signal that everyone hears.
Rx 2 and Rx 3 just want to hear their own specific songs.
Rx 1 wants to hear a specific song, but to them, the songs meant for Rx 2 and Rx 3 sound like interference (noise).

In a normal (classical) world, you can often treat this interference as random noise and just ignore it. But in this quantum world, the interference is tricky.

The "Bivariate" Puzzle:
For Rx 1, the interference isn't just one song; it's a mixture of the songs meant for Rx 2 and Rx 3.

If Rx 2 and Rx 3 use standard, random coding (like throwing random letters into a bag), the mixture of their signals creates a huge, messy cloud of noise. Rx 1 can't make sense of it.
Because Rx 1 can't decode the interference, the Sender has to "pre-code" (prepare) for a massive amount of noise, which drastically slows down the speed for Rx 1. This is the Rate Loss in action again.

The Solution: The "Coset Code" Strategy

The paper proposes a smarter way to code the messages, using something called Coset Codes.

The Analogy:
Instead of throwing random letters into a bag, imagine the Sender gives Rx 2 and Rx 3 a special, structured set of instructions.

Think of it like a dance routine. Rx 2 and Rx 3 aren't dancing randomly; they are dancing in a specific pattern (a "coset").
Even though Rx 1 doesn't know the individual steps of Rx 2 or Rx 3, the sum of their movements creates a very simple, predictable pattern.
Rx 1 can easily decode this "sum pattern" (the interference) because it's structured. Once Rx 1 figures out the pattern of the interference, they can subtract it out and hear their own message clearly.

The Big Win

The paper proves that for a specific type of 3-user quantum channel:

Old Way (Unstructured Codes): If everyone uses random, unstructured codes, Rx 1 gets stuck. They can't decode the interference, so their speed is capped at a lower limit.
New Way (Coset Codes): If the Sender uses these structured codes, Rx 1 can decode the interference pattern. This allows Rx 1 to achieve a higher speed that was previously impossible.

Summary

The paper shows that in the quantum world, if the receiver doesn't know the channel's state, you lose speed. However, by using structured, algebraic coding strategies (like coset codes) instead of random ones, you can turn that "noise" (interference from other users) into a predictable pattern. This allows the receiver to decode the interference, remove it, and achieve a faster communication rate than anyone thought possible with standard random methods.

In short: When the receiver is blind to the channel's state, randomness hurts you. But if you organize the interference like a structured dance, you can teach the receiver to dance along, turning noise into a signal.

Technical Summary: Rate Loss in Quantum Channels with Classical State and Applications for Quantum Broadcast Channels

Problem Statement
This paper addresses the problem of rate loss in classical-quantum (CQ) communication systems. Rate loss is defined as the strict penalty in achievable communication rates incurred when the receiver (Rx) lacks Channel State Information (CSI) that is available to the transmitter (Tx). While rate loss has been extensively characterized in classical network information theory (e.g., in source coding with side information and channel coding with states), its existence and characterization in truly non-commutative quantum channels have remained unaddressed.

The authors investigate two primary scenarios:

Point-to-Point (PTP) CQ Channels: Specifically, CQ channels with classical state information at the transmitter (QSTx), where the state is known causally or non-causally at the Tx but is unknown at the Rx.
CQ Broadcast Channels (CQBC): Specifically, 3-user broadcast channels where the lack of state information at the receiver leads to interference that cannot be efficiently mitigated by conventional coding strategies.

Methodology
The authors employ a combination of analytical information-theoretic proofs and structural comparisons of coding strategies:

Construction of a Noisy BB84-like Channel: To demonstrate rate loss in the non-commutative regime, the authors construct a specific QSTx model based on the BB84 protocol. The channel involves binary input ( $X$ ) and state ( $S$ ) alphabets, with output states living in a qubit space ( $\mathbb{C}^2$ ). The output states are noisy mixtures of orthogonal and non-orthogonal basis states (e.g., $|0\rangle, |1\rangle, |+\rangle, |-\rangle$ ).
Analytical Capacity Characterization:
- They derive the exact single-letter capacity for the causal case and a single-letter inner bound for the non-causal case where the Rx is oblivious to the state.
- They utilize convexity arguments involving the Holevo information and binary entropy functions to prove that the capacity with Rx state information strictly exceeds the capacity without it for non-commutative channels.
- Crucially, the proof avoids reliance on cardinality bounds for auxiliary random variables, instead using convexity of the function $g_\delta(t)$ .
Structured vs. Unstructured Coding Comparison:
- The authors analyze a 3-user CQBC where the interference at one receiver is a bivariate function of the codewords intended for the other two receivers.
- They compare conventional unstructured i.i.d. random codes against structured coset-code-based strategies.
- The analysis focuses on the ability of the receiver to decode the "interference" (the sum of other users' codewords). In the quantum setting, the authors show that unstructured codes fail to compress the entropy of the bivariate interference effectively, whereas coset codes exploit algebraic closure properties to decode the sum directly.

Key Contributions and Results

Proof of Rate Loss in Non-Commutative Channels (Propositions 1 & 2):
- The authors analytically prove that for the constructed noisy BB84-like QSTx, the achievable rate when the Rx is unaware of the state is strictly less than the rate when the Rx knows the state, provided the state probability $q$ is strictly between 0 and 1.
- They show that the single-letter non-causal inner bound coincides with the causal capacity, yet both are strictly smaller than the capacity with full state information at the Rx. This establishes the first rigorous characterization of rate loss in a non-commutative quantum channel.
Sub-optimality of Unstructured Codes in 3-User CQBCs (Theorem 2 & Proposition 3):
- The paper demonstrates that for specific 3-user CQBCs, coset-code-based strategies strictly outperform unstructured i.i.d. code-based strategies.
- In the proposed 3-user example, Receiver 1 faces interference $W = X_2 \oplus X_3$ $W = X_{2} \oplus X_{3}$ .
  - Unstructured Strategy: To achieve the state-informed rate, Receiver 1 must decode $X_2$ and $X_3$ separately. The normalized entropy of the sum of two independent unstructured codes is high ( $\min\{1, 2 - 2h_2(\epsilon)\}$ ), creating a bottleneck that prevents achieving the optimal rate triple.
  - Structured (Coset) Strategy: By using nested coset codes, the sum $X_2 \oplus X_3$ forms a coset of a linear code with lower entropy ( $1 - h_2(\epsilon)$ ). Receiver 1 can decode the sum directly without decoding the individual components, effectively treating the interference as a state it can "cancel" or decode, thereby achieving a rate triple that is unachievable by the unstructured subclass.
Explicit Parameter Regime:
- The authors provide a concrete set of parameters ( $\tau=0.05, \delta_1=0, \delta_2=0.158, \delta_3=0.159$ ) where the corner point of the achievable rate region is attainable via structured coding but lies outside the closure of the reduced unstructured inner bound.

Significance and Claims
The paper claims to fill a critical gap in quantum network information theory by:

Establishing the existence of rate loss in non-commutative quantum channels, a phenomenon previously unproven in this context.
Demonstrating the necessity of structured coding (specifically coset codes) for optimal performance in multi-user quantum broadcast channels. The results show that the "flip side" of auxiliary random variables in quantum settings—often leading to intractable optimization problems—can be overcome by structured strategies that exploit algebraic properties to mitigate interference.
Providing a structural characterization of optimal coding strategies, suggesting that for non-commutative channels, the ability to decode interference components (via structured codes) is essential to avoid the strict rate penalties associated with missing state information at the receiver.

The authors maintain a modest tone, noting that their findings are specific to the identified non-commutative channels and that the proof relies on a specific reduction of the general inner bound to a "reduced unstructured subclass" to make the comparison explicit. They do not claim these results apply universally to all quantum channels but rather highlight a specific class where the gap between structured and unstructured strategies is provable and significant.