Fundamental Limits of Bistatic Integrated Sensing and Communications over Memoryless Relay Channels

This paper investigates the fundamental communication-sensing tradeoffs in bistatic integrated sensing and communications over memoryless relay channels by deriving an upper bound and a hybrid-partial-decode-and-compress-forward lower bound for the capacity-distortion function, demonstrating their optimality in specific cases and the benefits of integrated design.

The Gist

Imagine you are trying to send a secret letter to a friend across a noisy, foggy valley. You have a helper (a Relay) standing on a hill in the middle.

In the old way of doing things, you would either:

1. Send the letter: Focus all your energy on making sure the friend reads the message clearly.
2. Map the valley: Focus all your energy on shouting into the fog to hear the echo, so you can figure out how fast the wind is blowing or where the rocks are (Sensing).

Doing both at the same time usually means doing neither perfectly. This paper is about a new, smarter way to do both simultaneously using a "bistatic" setup (where the sender, the helper, and the listener are in different places).

Here is the breakdown of the paper's ideas using simple analogies:

1. The Setup: The "Three-Person Game"

• The Source (You): You have a message to send and you want to know about the environment (like wind speed).
• The Relay (The Helper): They stand in the middle. They hear you, but they also hear the environment.
• The Destination (Your Friend): They need to read your letter and figure out the wind speed based on what they hear from you and the helper.

The challenge is that the "air" (the channel) is full of static and random noise. The paper asks: What is the absolute best balance we can strike between sending a fast message and getting an accurate map of the environment?

2. The Core Problem: The Trade-off

Think of your bandwidth (the "pipe" for data) as a bucket of water.

• If you pour all the water into the Message, your friend gets the letter perfectly, but they know nothing about the wind.
• If you pour all the water into Sensing, they know the wind perfectly, but the letter is unreadable.
• The Goal: Find the "Goldilocks" zone where you get a good letter and a good map without wasting water.

3. The New Strategy: The "Hybrid" Approach

The authors propose a clever coding scheme called Hybrid Partial Decode-and-Compress-Forward. Let's translate that into a story:

• Partial Decode: The Helper doesn't try to read the whole secret letter. Instead, they just listen for the "big picture" parts (the common message) that are easy to understand.
• Compress-Forward: The Helper also listens to the noise and the echoes they hear. Instead of sending the raw, messy noise, they "compress" it into a tiny, efficient summary (like a compressed zip file) and send that to the friend.
• The Magic Trick: The friend receives the letter, the summary from the helper, and the direct signal from you. They mix all three together. Because they have the "summary" of the noise from the helper, they can cancel out the static in their own ears much better than before.

Why is this better?
In previous methods, the friend had to guess the noise while trying to read the letter. It was like trying to solve a puzzle while someone is shouting at you. This new method gives the friend a "cheat sheet" (the compressed noise data) from the helper, making the puzzle easier to solve.

4. The "Virtual" Helper

The paper also introduces a mathematical concept called an Auxiliary Random Variable.

• Analogy: Imagine the Helper has a "Super-Ear" that can hear everything perfectly if the Source and Helper were best friends working as a single team.
• The authors use this "Super-Ear" idea to calculate the Theoretical Limit. It's like calculating the maximum speed a car could go if it had a perfect engine, even if the current engine isn't quite there yet. This sets a "ceiling" on how good the system can possibly be.

5. The Results: When Does It Work Best?

The authors found that for three specific types of "valleys" (channel types), their new strategy hits the perfect ceiling. It turns out:

• If the Helper is far from the wind: The Helper just needs to send a summary of what they hear.
• If the Helper is close to the wind: The Helper can decode the message and send it along.
• The Hybrid Case: In the most complex scenarios, the "Hybrid" strategy (doing a bit of both) is the winner.

6. The Big Picture: Why Should We Care?

This isn't just about sending letters. This is about 6G networks and Self-Driving Cars.

• Self-Driving Cars: A car needs to talk to the traffic light (Communication) and sense if a pedestrian is stepping out (Sensing).
• The Benefit: Instead of having two separate systems (one for talking, one for radar), this research shows how to build one system that does both efficiently. It saves battery, saves spectrum (the airwaves), and makes the whole network smarter.

Summary

The paper proves that by having a middleman (Relay) help not just with the message, but also by "compressing" environmental data, we can get a much better balance between talking and listening. They found the mathematical "speed limit" for this technology and showed exactly how to drive at that speed in many real-world scenarios.

In one sentence: They figured out how to make a relay node act like a super-smart assistant that helps you send a message and map the surroundings at the same time, without either task getting in the other's way.

Technical Summary

Here is a detailed technical summary of the paper "Fundamental Limits of Bistatic Integrated Sensing and Communications over Memoryless Relay Channels."

1. Problem Statement

The paper investigates the fundamental performance limits of Bistatic Integrated Sensing and Communications (ISAC) systems operating over state-dependent discrete memoryless relay channels.

• System Model: A three-node network consisting of a Source, a Relay, and a Destination.
  • Bistatic Setup: The transmitter (Source) and the sensing estimator (Destination) are physically separated. The Relay assists in both communication and sensing.
  • Dual Functionality: The Source transmits a message $W$ to the Destination. Simultaneously, the Destination aims to estimate unknown channel state parameters $S_d$ (e.g., target velocity or reflection coefficients) correlated with the channel state $S$.
  • Relay Role: The Relay receives signals from the Source, processes them, and forwards information to the Destination. It aids in decoding the message and providing side information to improve state estimation.
• Objective: Characterize the Capacity-Distortion Function, denoted as $C(D)$. This function defines the maximum achievable communication rate $R$ for a given sensing distortion constraint $D$ (where lower $D$ implies higher sensing accuracy).

2. Methodology

The authors employ information-theoretic tools to derive bounds on the capacity-distortion tradeoff.

• Upper Bound (Converse):
  • Extends the classic Cut-Set Bound for relay channels to incorporate sensing constraints.
  • Introduces an auxiliary random variable $T$ to represent the "useful sensing information" extracted by the Relay under the assumption of full cooperation with the Source.
  • Derives a tradeoff constraint where resources used to convey sensing information ($T$) reduce the available rate for message transmission.
• Lower Bound (Achievability):
  • Proposes a novel Hybrid Partial-Decode-and-Compress-Forward (HPDCF) coding scheme.
  • Key Innovations in the Scheme:
    • Block Markov Encoding: Messages are split into common and private parts across blocks.
    • Simultaneous Decoding: Unlike previous schemes (e.g., Chong-Motani-Garg) that decode common messages and compressed information first, then private messages, this scheme enables the Destination to simultaneously decode the indices of the common message, the private message, and the compressed sensing information.
    • Benefit: This joint decoding treats the private message as useful observation rather than interference, improving the signal-to-noise ratio for retrieving compressed sensing data, thereby reducing distortion.
• Optimal Sensing Analysis:
  • Analyzes the minimum distortion ($D_{min}$) achievable when communication rate is zero (sensing-only mode).
  • Identifies specific channel classes where the upper and lower bounds coincide, yielding the exact optimal $C(D)$.

3. Key Contributions

1. General Bounds for Bistatic ISAC:

   • Derived a new Upper Bound on $C(D)$ by extending the cut-set bound with an auxiliary variable to model the relay's impact on sensing.
   • Established a Lower Bound using the HPDCF scheme, which strictly enlarges the achievable rate-distortion region compared to existing sequential decoding strategies.

2. Optimal Sensing Performance ($D_{min}$):

   • Characterized the minimum distortion $D_{min}$ for general relay channels.
   • Showed that $D_{min}$ is achieved when the Source sends pilot signals (or specific inputs) allowing the Relay to perfectly compress the received state information for the Destination.
   • Identified two specific channel classes ($C_1$ and $C_2$) where simpler strategies (deterministic signaling or relay estimation) achieve $D_{min}$.

3. Exact Capacity-Distortion Functions for Specific Channels:
   The paper proves that the upper and lower bounds coincide for three specific classes of relay channels, providing the exact optimal $C(D)$:

   • Class $C_3$ (Orthogonal Sender Components): Achieved via Partial-Decode-Forward. The relay decodes part of the message; sensing relies on the direct link and decoded signals.
   • Class $C_4$ (State-Dependent Cover-Kim Channels): Achieved via Compress-Forward. The relay compresses its received signal (which is a deterministic function of the source input) and forwards it.
   • Class $C_5$ (Two-Hop Relay Channels): Achieved via the Hybrid Partial-Decode-and-Compress-Forward strategy. The relay decodes partial messages and compresses state information.

4. Numerical Validation:

   • Provided numerical examples demonstrating that the integrated design (ISAC) significantly outperforms Time-Sharing (alternating between pure sensing and pure communication).
   • Showed that the proposed simultaneous decoding strategy strictly improves the achievable region over the standard Chong-Motani-Garg scheme.

4. Key Results

• Tradeoff Characterization: The paper successfully maps the Pareto frontier between communication rate and sensing distortion. It demonstrates that integrating sensing into the communication protocol allows for a better tradeoff than treating them as separate tasks.
• Decoding Strategy Superiority: The proposed simultaneous decoding of common messages, private messages, and compressed sensing data is proven to strictly enlarge the achievable rate-distortion region compared to sequential decoding. This is because it leverages the private message as side information to aid in the recovery of compressed sensing data.
• Optimality in Specific Scenarios:
  • For Class $C_3$, the relay does not need to compress state information; the destination estimates states based on the direct link and decoded signals.
  • For Class $C_4$, the relay acts purely as a compressor (Compress-Forward), and the destination estimates based on the source signal and the compressed relay signal.
  • For Class $C_5$, the hybrid strategy is necessary and optimal.
• Minimum Distortion: The paper establishes that even without communication, the system has a fundamental limit on sensing accuracy ($D_{min}$), which depends on the channel capacity between the Relay and Destination.

5. Significance and Impact

• Theoretical Foundation: This work provides the first rigorous information-theoretic characterization of bistatic ISAC over relay channels, filling a gap left by previous studies that focused mainly on monostatic scenarios or point-to-point links.
• System Design Guidance: The results offer concrete guidelines for future 6G network design. Specifically, they suggest that joint decoding strategies are superior for ISAC systems and that the choice of relaying strategy (Decode-Forward vs. Compress-Forward vs. Hybrid) depends heavily on the specific channel topology and state dependencies.
• Performance Gains: By quantifying the gains over time-sharing, the paper validates the efficiency of integrated designs, encouraging the development of hardware and protocols that support simultaneous sensing and communication.
• Future Directions: The authors highlight that extending these results to channels with memory (where past states influence current outputs) and finite blocklength regimes remains a challenging open problem, pointing toward the need for new analytical tools like neural estimation frameworks.

In summary, this paper establishes the fundamental limits of bistatic ISAC in relay networks, proving that a hybrid coding scheme with simultaneous decoding achieves optimal performance in specific scenarios and offers significant gains over traditional separated approaches.