⚡ electrical engineering

Covert Routing with DSSS Signaling Against Cycle Detectors

This paper proposes a framework for covert multi-hop routing in wireless networks that utilizes DSSS signaling to optimize bandwidth, power, and spreading gain against cyclostationary detectors, revealing that end-to-end latency increases exponentially with covertness requirements and super-linearly with packet size.

Original authors: Swapnil Saha, Rahul Aggarwal, Fikadu Dagefu, Justin Kong, Jihun Choi, Brian Kim, Predrag Spasojevic

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Swapnil Saha, Rahul Aggarwal, Fikadu Dagefu, Justin Kong, Jihun Choi, Brian Kim, Predrag Spasojevic

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 note across a crowded schoolyard, but there's a strict hall monitor named Willie who is always watching. Willie doesn't just look for loud shouting; he has a special "super-sense" that can spot the hidden rhythm in a whisper. This paper is about how a group of friends (Alice, Bob, and their relay team) can use a clever trick called DSSS (Direct-Sequence Spread Spectrum) to pass that note without Willie catching them, while also making sure the note arrives fast enough.

The Secret Sauce: Spreading the Signal

Usually, if you whisper, you whisper quietly. But in this game, the friends use a technique that spreads their message out over a huge amount of space, like scattering a handful of glitter across the floor. To an untrained eye (or a simple detector), the glitter just looks like normal dust (noise). However, the glitter was thrown in a specific, repeating pattern.

Willie, the adversary, uses a "cycle detector." Think of this as a machine that knows exactly what that repeating pattern looks like. If the pattern is there, Willie knows a secret message is being sent. The paper shows that even with this super-sense, the friends can hide the message by carefully choosing how much "glitter" (bandwidth) to use, how hard to throw it (power), and how many friends to pass it through (hops).

The Two Big Goals: Hiding vs. Speed

The researchers set up two different challenges for the friends:

The "Hide at All Costs" Game: Here, the goal is to make sure Willie never finds the message. The friends will take a longer, winding path through the schoolyard, passing the note from person to person, even if it takes a long time. They might whisper so softly that the note barely moves, just to stay safe.
The "Speed Run" Game: Here, the goal is to get the note to Bob as fast as possible, but they still have to stay hidden. They have to find a balance: throw the note hard enough to reach the next friend quickly, but not so hard that Willie hears the whoosh.

What the Simulations Revealed

The authors didn't just guess; they ran detailed computer simulations in a realistic 3D environment (imagine a digital model of a school with 36 students and some concrete buildings). Here is what they found:

The "Exponential" Trap: If the friends demand extreme secrecy, the time it takes to send the note doesn't just go up a little bit—it explodes. However, the paper clarifies that this exponential growth only kicks in once the secrecy requirement crosses a specific threshold (when the detection error probability is 0.97 or higher). Below that level, the time increases more slowly (linearly or stays constant). It's like trying to sneak past a guard who is getting more and more suspicious; eventually, once you cross a certain line of suspicion, you have to hide in a closet for hours, and the time it takes to get to your destination skyrockets.
The "Super-Linear" Growth: The size of the message matters, too. If the note is huge (the paper tested a message size of 100 Mb), and the friends need to be very secretive, the time it takes to send it grows even faster than the size of the message itself.
The "Bottleneck" Effect: The paper found that the friend standing closest to Willie is the weak link. In their simulation, a specific node (Node 34) was the "bottleneck" that determined how safe the whole route was. If that one person got caught, the whole secret was out.

The Cycle Detector vs. The Energy Detector

Willie has two ways to catch them:

The Energy Detector: This is like a noise meter. It just listens for any sound louder than the background hum.
The Cycle Detector: This is the "super-sense" that looks for the specific rhythm of the message.

The paper suggests that which detector Willie uses changes the game. If the friends are only trying to be somewhat secretive, Willie's "super-sense" (the cycle detector) is actually harder to fool, forcing the friends to take longer, slower routes. But if the friends are trying to be extremely secretive, the cycle detector becomes less effective compared to the simple noise meter because it is more sensitive to noise. This degradation means the friends might actually be able to move faster than they thought under those very strict conditions.

The Bottom Line

The authors propose a new way to calculate the best path for these secret messages. They proved mathematically that finding the "safest" path is the same as finding the path with the best "signal-to-noise" advantage, which makes the computer math much easier to solve.

In their simulations, they found that for a message of 100 Mb, a route optimized for speed took 30 seconds to deliver. However, if they pushed for maximum secrecy, the time would balloon. The paper concludes that while multi-hop routing (passing the note along) is a great way to stay hidden, there is a heavy price to pay in time, especially when the message is big and the need for secrecy is high.

The paper doesn't claim to have "solved" the problem of perfect stealth forever; rather, it provides a framework and simulation results that show exactly how much time you lose when you try to be invisible to a very smart watcher. It's a map for the trade-off between speed and silence.

Technical Summary: Covert Routing with DSSS Signaling Against Cycle Detectors

Problem Statement
This paper addresses the challenge of establishing covert multi-hop communication in wireless networks where an adversary, Willie, employs a cyclostationary (cycle) detector to identify hidden transmissions. While Direct Sequence Spread Spectrum (DSSS) is a standard technique for Low Probability of Detection (LPD) communications by spreading signals over wide bandwidths, the statistical periodicity of repeating spreading codes creates cyclostationary features that cycle detectors can exploit.

The core problem involves routing a confidential message of $M$ bits from a source (Alice) to a destination (Bob) through a network of friendly relay nodes. The system must satisfy two competing objectives under Quality-of-Service (QoS) and link budget constraints:

Maximize End-to-End Covertness: Minimize the probability of detection by Willie across the entire route.
Minimize End-to-End Latency: Reduce the total time required to transmit the message.

A critical constraint is that the end-to-end covertness is determined by the "weakest link" (the hop with the highest detection probability), meaning a single vulnerable hop compromises the entire route.

Methodology
The authors propose a framework that jointly optimizes resource allocation (transmit power, bandwidth, and spreading gain) and route selection for each hop.

Signal Model: The system utilizes DSSS signaling. The transmitter adapts bandwidth ( $\Omega_v$ ), power ( $P_v$ ), and spreading gain ( $\eta_v$ ) based on dynamic channel conditions and adversary proximity.
Adversary Model: Willie employs a cycle detector that estimates the Degree of Cyclostationarity (DCS) of the received signal. The detection error probability (DEP) at Willie is modeled as inversely proportional to the Signal-to-Noise Ratio (SNR) at Willie and the message length ( $M$ ).
Optimization Formulation:
- Covertness Maximization: The problem is formulated to maximize the minimum DEP across all hops in a route. The authors prove an equivalence between maximizing the minimum DEP and maximizing the minimum detection SNR gain ( $\theta_v$ ) along the path. This equivalence allows the use of efficient widest-path algorithms (a variant of Dijkstra's algorithm) rather than computationally intensive Monte Carlo simulations for every route evaluation.
- Latency Minimization: The problem minimizes the sum of latencies ( $\lambda_v = M/D_v$ ) subject to a minimum DEP constraint per hop. This is solved using standard shortest-path algorithms.
Single-Hop Optimization: For both objectives, the paper derives optimal parameters. For covertness maximization, the optimal strategy involves using maximum bandwidth and minimum required data rate to maximize processing gain. For latency minimization, the strategy involves maximizing transmit power (within covertness limits) to increase data rates.

Key Contributions

Novel Routing Algorithms: The paper introduces two routing algorithms specifically designed for covert multi-hop networks against cycle detectors: one for covertness maximization and one for latency minimization, both guaranteeing QoS.
Equivalence Theorem: The authors demonstrate that the covertness-maximization problem (based on DEP) is mathematically equivalent to a problem based on detection SNR gain. This theoretical insight enables the use of efficient graph-theoretic algorithms for route computation.
3D Environment Validation: The proposed framework is validated through numerical simulations in a realistic 3D environment (250m $\times$ 250m $\times$ 9.5m) with concrete buildings, using ray-tracing to model channel gains.

Numerical Results
Simulations using a 36-node network topology yield the following insights:

Latency vs. Covertness: End-to-end latency increases exponentially as the covertness requirement becomes more stringent (specifically when the required DEP exceeds $\approx 0.97$ ). This is driven by the need for more hops and higher processing gains near the adversary.
Latency vs. Message Size ( $M$ ): The relationship between latency and packet size is super-linear under relaxed covertness requirements and exponential under stringent requirements.
Detector Comparison: The paper compares cycle detectors against energy detectors. It finds that the adversary's choice of detector significantly impacts the achievable latency:
- Under low covertness requirements, the cycle detector imposes higher latency penalties than the energy detector.
- Under stringent covertness requirements, the energy detector becomes more effective at detecting transmissions, imposing higher latency penalties than the cycle detector.
- The transition point where the cycle detector becomes less effective than the energy detector occurs at a DEP requirement of approximately $10^{-1}$ .

Significance
The paper claims to provide the first study of multi-hop routing covertness specifically against adversaries employing cycle detectors. By establishing the equivalence between covertness metrics and SNR gain, the work enables computationally efficient resource allocation and routing design for covert networks. The results offer critical insights for network designers regarding the trade-offs between latency, message size, and the level of protection required against advanced detection adversaries. The findings highlight that as covertness demands increase, the network topology must adapt significantly (more hops, lower power), leading to non-linear increases in communication delay.