Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2607.14013

Square-Root Law for Covert Communication with Warden-Favorable Side Information

Hossein Ahmadi, Christian Deppe, Boulat A. Bash, Eduard Axel Jorswieck

Published Jul 15, 2026• 0 citations
arXivHTMLPDF

Abstract

Covert communication enables Alice to transmit to Bob while making the transmission difficult for Willie to detect. We study a scalar Gaussian covert-overlay model in which Alice's low-power covert signal is superimposed on an aggregate public component generated by Alice or other trackable sources. Willie is given all physically obtainable side information, including protocol details, timing, pilots, channel estimates, and calibration information, and subtracts his best estimate of the public component before testing. Covertness is imposed on the resulting residual through a relative-entropy constraint with budget $δ$ conditioned on Willie's side information. In the stationary case, the residual under no covert transmission has variance $σ_0^2=σ_W^2+σ_e^2$, where $σ_W^2$ is Willie's receiver-noise variance and $σ_e^2$ is the irreducible cancellation error. Over $n$ channel uses, the maximal reliably transmissible covert payload is $R_C^\star\sqrt{n}(1+o(1))$ bits, where $R_C^\star=\frac{σ_0^2}{σ_B^2\ln 2}\sqrtδ$, and $σ_B^2$ is Bob's receiver-noise variance. Thus, the square-root-law (SRL) constant is governed by the variance at Willie's actual detector input, not by receiver noise alone. Low-power Gaussian signaling achieves this constant, and a matching converse establishes first-order optimality within the conditioned additive Gaussian innovation model. For known time-varying conditioned residual variances, we also derive the first-order allocation, which assigns more covert power to larger residual variances. The results require a Gaussian post-cancellation null residual with known conditioned variance; non-Gaussian residuals and fixed non-vanishing variance uncertainty are outside the scope of this paper.

DOI
10.48550/arxiv.2607.14013

Related papers

On the Limits of Covert Backscatter Communication Over Undecodable Ambient Signals

Weiyu Chen, Haiyang Ding, Shilian Wang +2 more

This paper investigates the limit covert rate of the backscatter communication systems based on undecodable ambient radio frequency (RF) signals over additive white Gaussian noise channel. In the considered network, source node Alice expects to transmit information to receiver Bob without being detected by warden Willie, by backscattering undecodable ambient RF signals. In this case, it is challenging for Bob to recover the information from Alice, since the undecodable signals result in both unknown additive and multiplicative interference. A subtle communication strategy is proposed to alleviate the influences of the interference, by which we prove that no matter Willie can or cannot decode the information from the RF source and then eliminate its interference, the limit covert rate is subject to the well-known square root law (i.e.,O(√n) bits pernchannel uses). This interesting result means that even if the RF source is a cooperative node of Willie (e.g., a jammer deployed by Willie to restrict Alice from active transmission), Alice can still covertly and reliably transmit information to Bob by backscattering the ambient RF signals (e.g., the malicious jamming signals).

IEEE Transactions on Information Forensics and Security•Jan 1, 2023•9 citations

Covert Communication in the Presence of an Uninformed Jammer

Tamara V. Sobers, Boulat A. Bash, Saikat Guha +2 more

Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to O(√n) bits in n channel uses. However, this assumes that Willie has an accurate statistical characterization of the channel. When Willie has uncertainty about such and his receiver is limited to a threshold test on the received power, Alice can transmit covertly with a power that does not decrease with n, thus conveying O(n) bits covertly and reliably in n uses of an AWGN channel. Here, we consider covert communication of O(n) bits in n channel uses while generalizing the environment and removing any restrictions on Willie's receiver. We assume that an uninformed “jammer” is present to help Alice, and we consider AWGN and block fading channels. In some scenarios, Willie's optimal detector is a threshold test on the received power. When the channel between the jammer and Willie has multiple fading blocks per codeword, a threshold test on the received power is not optimal. However, we establish that Alice can remain covert with a transmit power that does not decrease with n even when Willie employs an optimal detector.

IEEE Transactions on Wireless Communications•Jun 30, 2017•451 citations

Covert Communication in the Presence of an Uninformed Jammer

Tamara V. Sobers, Boulat A. Bash, Saikat Guha +2 more

Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to $\mathcal{O}(\sqrt{n})$ bits in $n$ channel uses. However, this assumes Willie has an accurate statistical characterization of the channel. When Willie has uncertainty about such and his receiver is limited to a threshold test on the received power, Alice can transmit covertly with a power that does not decrease with $n$, thus conveying $\mathcal{O}(n)$ bits covertly and reliably in $n$ uses of an AWGN channel. Here, we consider covert communication of $\mathcal{O}(n)$ bits in $n$ channel uses while generalizing the environment and removing any restrictions on Willie's receiver. We assume an uninformed "jammer" is present to help Alice, and we consider AWGN and block fading channels. In some scenarios, Willie's optimal detector is a threshold test on the received power. When the channel between the jammer and Willie has multiple fading blocks per codeword, a threshold test on the received power is not optimal. However, we establish that Alice can remain covert with a transmit power that does not decrease with $n$ even when Willie employs an optimal detector.

arXiv:1608.00698•Aug 2, 2016•2 citations

Covert communication with the help of an uninformed jammer achieves positive rate

Tamara V. Sobers, Boulat A. Bash, Dennis Goeckel +2 more

Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to O(√n) bits in n channel uses. However, this result is predicated on Willie having an accurate statistical characterization of the background noise. When Willie has uncertainty about his noise characteristics and his receiver is limited to a power detector, O(n) bits can be covertly delivered. Here, we establish covert communication of O(n) bits in n channel uses while: (i) generalizing the environment; and (ii) removing any restrictions on Willie's receiver. We consider the case where an additional node, called the "jammer", is present to help Alice and Bob communicate covertly. This jammer is termed "uninformed", as it does not know the content or timing of Alice's transmission. We consider both AWGN channels and a scenario with a block faded jammer with a single fade per codeword. In these scenarios, we are able to establish the optimality of the power detector for Willie, from which the achievability of reliable and covert communication at a positive rate follows.

Asilomar Conference on Signals, Systems and Computers (ACSSC)•Nov 1, 2015•36 citations

Covert Communication Gains From Adversary’s Ignorance of Transmission Time

Boulat A. Bash, Dennis Goeckel, Don Towsley

The recent square root law (SRL) for covert communication demonstrates that Alice can reliably transmit O(√n) bits to Bob in n uses of an additive white Gaussian noise (AWGN) channel while keeping ineffective any detector employed by the adversary; conversely, exceeding this limit either results in detection by the adversary with high probability or nonzero decoding error probability at Bob. This SRL is under the assumption that the adversary knows when Alice transmits (if she transmits); however, in many operational scenarios, he does not know this. Hence, here, we study the impact of the adversary's ignorance of the time of the communication attempt. We employ a slotted AWGN channel model with T(n) slots each containing n symbol periods, where Alice may use a single slot out of T(n). Provided that Alice's slot selection is secret, the adversary needs to monitor all T(n) slots for possible transmission. We show that this allows Alice to reliably transmit O(min{(n log T(n))1/2, n}) bits to Bob (but no more) while keeping the adversary's detector ineffective. To achieve this gain over SRL, Bob does not have to know the time of transmission provided T(n)cTn, cT= O(1).

arXiv:1403.1013•IEEE Transactions on Wireless Communications•Sep 29, 2016•152 citations

Robust Power Allocation in Covert Communication: Imperfect CDI

Moslem Forouzesh, Paeiz Azmi, Nader Mokari +1 more

The study of the fundamental limits of covert communications, where a transmitter Alice wants to send information to a desired recipient Bob without detection of that transmission by an attentive and capable warden Willie, has emerged recently as a topic of great research interest. Previous work has assumed that the distribution of the multipath fading on the channel, which we term the channel distribution information (CDI), is known to Alice, hence facilitating her characterization of Willie's capabilities to detect the signal. However, in practice, Willie tends to be passive and the environment heterogeneous, implying a lack of signaling interchange between the transmitter and Willie makes it difficult if not impossible for Alice to estimate the CDI exactly and provide covertness guarantees. In this paper, we address this issue by developing covert communication schemes for various assumptions on Alice's imperfect CDI: 1) when the transmitter knows the channel distribution is within some distance of a nominal distribution; 2) when only the mean and variance of the channel distribution are available; 3) when Alice knows the channel distribution is complex Gaussian but the variance is unknown. In each case, we formulate new optimization problems to find the power allocations that maximize covert rate subject to a covertness requirement under uncertain CDI. Moreover, Willie may be able to estimate CDI exactly or be faced with an error in estimation, and we investigate two these issues. Numerical results are presented to compare the proposed schemes from various aspects, for attaining desirable covert system performance.

arXiv:1901.04914•IEEE Transactions on Vehicular Technology•Apr 29, 2021•25 citations
Covert communication enables Alice to transmit to Bob while making the transmission difficult for Willie to detect. We study a scalar Gaussian covert-overlay model in which Alice's low-power covert signal is superimposed on an aggregate public component generated by Alice or other trackable sources. Willie is given all physically obtainable side information, including protocol details, timing, pilots, channel estimates, and calibration information, and subtracts his best estimate of the public component before testing. Covertness is imposed on the resulting residual through a relative-entropy constraint with budget δδδ conditioned on Willie's side information. In the stationary case, the residual under no covert transmission has variance σ02=σW2+σe2σ_0^2=σ_W^2+σ_e^2σ02​=σW2​+σe2​, where σW2σ_W^2σW2​ is Willie's receiver-noise variance and σe2σ_e^2σe2​ is the irreducible cancellation error. Over nnn channel uses, the maximal reliably transmissible covert payload is RC⋆n(1+o(1))R_C^\star\sqrt{n}(1+o(1))RC⋆​n​(1+o(1)) bits, where RC⋆=σ02σB2ln⁡2δR_C^\star=\frac{σ_0^2}{σ_B^2\ln 2}\sqrtδRC⋆​=σB2​ln2σ02​​δ​, and σB2σ_B^2σB2​ is Bob's receiver-noise variance. Thus, the square-root-law (SRL) constant is governed by the variance at Willie's actual detector input, not by receiver noise alone. Low-power Gaussian signaling achieves this constant, and a matching converse establishes first-order optimality within the conditioned additive Gaussian innovation model. For known time-varying conditioned residual variances, we also derive the first-order allocation, which assigns more covert power to larger residual variances. The results require a Gaussian post-cancellation null residual with known conditioned variance; non-Gaussian residuals and fixed non-vanishing variance uncertainty are outside the scope of this paper.
This paper investigates the limit covert rate of the backscatter communication systems based on undecodable ambient radio frequency (RF) signals over additive white Gaussian noise channel. In the considered network, source node Alice expects to transmit information to receiver Bob without being detected by warden Willie, by backscattering undecodable ambient RF signals. In this case, it is challenging for Bob to recover the information from Alice, since the undecodable signals result in both unknown additive and multiplicative interference. A subtle communication strategy is proposed to alleviate the influences of the interference, by which we prove that no matter Willie can or cannot decode the information from the RF source and then eliminate its interference, the limit covert rate is subject to the well-known square root law (i.e.,O(√n) bits pernchannel uses). This interesting result means that even if the RF source is a cooperative node of Willie (e.g., a jammer deployed by Willie to restrict Alice from active transmission), Alice can still covertly and reliably transmit information to Bob by backscattering the ambient RF signals (e.g., the malicious jamming signals).
Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to O(√n) bits in n channel uses. However, this assumes that Willie has an accurate statistical characterization of the channel. When Willie has uncertainty about such and his receiver is limited to a threshold test on the received power, Alice can transmit covertly with a power that does not decrease with n, thus conveying O(n) bits covertly and reliably in n uses of an AWGN channel. Here, we consider covert communication of O(n) bits in n channel uses while generalizing the environment and removing any restrictions on Willie's receiver. We assume that an uninformed “jammer” is present to help Alice, and we consider AWGN and block fading channels. In some scenarios, Willie's optimal detector is a threshold test on the received power. When the channel between the jammer and Willie has multiple fading blocks per codeword, a threshold test on the received power is not optimal. However, we establish that Alice can remain covert with a transmit power that does not decrease with n even when Willie employs an optimal detector.
Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to O(n)\mathcal{O}(\sqrt{n})O(n​) bits in nnn channel uses. However, this assumes Willie has an accurate statistical characterization of the channel. When Willie has uncertainty about such and his receiver is limited to a threshold test on the received power, Alice can transmit covertly with a power that does not decrease with nnn, thus conveying O(n)\mathcal{O}(n)O(n) bits covertly and reliably in nnn uses of an AWGN channel. Here, we consider covert communication of O(n)\mathcal{O}(n)O(n) bits in nnn channel uses while generalizing the environment and removing any restrictions on Willie's receiver. We assume an uninformed "jammer" is present to help Alice, and we consider AWGN and block fading channels. In some scenarios, Willie's optimal detector is a threshold test on the received power. When the channel between the jammer and Willie has multiple fading blocks per codeword, a threshold test on the received power is not optimal. However, we establish that Alice can remain covert with a transmit power that does not decrease with nnn even when Willie employs an optimal detector.
Recent work has established that when transmitter Alice wishes to communicate reliably to recipient Bob without detection by warden Willie, with additive white Gaussian noise (AWGN) channels between all parties, communication is limited to O(√n) bits in n channel uses. However, this result is predicated on Willie having an accurate statistical characterization of the background noise. When Willie has uncertainty about his noise characteristics and his receiver is limited to a power detector, O(n) bits can be covertly delivered. Here, we establish covert communication of O(n) bits in n channel uses while: (i) generalizing the environment; and (ii) removing any restrictions on Willie's receiver. We consider the case where an additional node, called the "jammer", is present to help Alice and Bob communicate covertly. This jammer is termed "uninformed", as it does not know the content or timing of Alice's transmission. We consider both AWGN channels and a scenario with a block faded jammer with a single fade per codeword. In these scenarios, we are able to establish the optimality of the power detector for Willie, from which the achievability of reliable and covert communication at a positive rate follows.
The recent square root law (SRL) for covert communication demonstrates that Alice can reliably transmit O(√n) bits to Bob in n uses of an additive white Gaussian noise (AWGN) channel while keeping ineffective any detector employed by the adversary; conversely, exceeding this limit either results in detection by the adversary with high probability or nonzero decoding error probability at Bob. This SRL is under the assumption that the adversary knows when Alice transmits (if she transmits); however, in many operational scenarios, he does not know this. Hence, here, we study the impact of the adversary's ignorance of the time of the communication attempt. We employ a slotted AWGN channel model with T(n) slots each containing n symbol periods, where Alice may use a single slot out of T(n). Provided that Alice's slot selection is secret, the adversary needs to monitor all T(n) slots for possible transmission. We show that this allows Alice to reliably transmit O(min{(n log T(n))1/2, n}) bits to Bob (but no more) while keeping the adversary's detector ineffective. To achieve this gain over SRL, Bob does not have to know the time of transmission provided T(n)cTn, cT= O(1).
The study of the fundamental limits of covert communications, where a transmitter Alice wants to send information to a desired recipient Bob without detection of that transmission by an attentive and capable warden Willie, has emerged recently as a topic of great research interest. Previous work has assumed that the distribution of the multipath fading on the channel, which we term the channel distribution information (CDI), is known to Alice, hence facilitating her characterization of Willie's capabilities to detect the signal. However, in practice, Willie tends to be passive and the environment heterogeneous, implying a lack of signaling interchange between the transmitter and Willie makes it difficult if not impossible for Alice to estimate the CDI exactly and provide covertness guarantees. In this paper, we address this issue by developing covert communication schemes for various assumptions on Alice's imperfect CDI: 1) when the transmitter knows the channel distribution is within some distance of a nominal distribution; 2) when only the mean and variance of the channel distribution are available; 3) when Alice knows the channel distribution is complex Gaussian but the variance is unknown. In each case, we formulate new optimization problems to find the power allocations that maximize covert rate subject to a covertness requirement under uncertain CDI. Moreover, Willie may be able to estimate CDI exactly or be faced with an error in estimation, and we investigate two these issues. Numerical results are presented to compare the proposed schemes from various aspects, for attaining desirable covert system performance.