Use of Bayesian Changepoint Detection for Spectrum Sensing in Mobile Cognitive Radio

Chede, Akpaki Steaven V.; Rottenberg, François; Dossou, Michel; Louveaux, Jérôme

doi:10.1109/vtc2021-spring51267.2021.9448935

Cited by 6 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One potential disadvantage of the approach presented in [21] is that the integration of CSS and the removal of the cyclic prefix (CP) may introduce additional complexity and computational overhead to the system. The authors of [22] explored the use of Bayesian changepoint detection method and incorporates knowledge about the environment and user mobility parameters to improve spectrum occupancy detection in a mobile cognitive radio scenario. The drawback of this method is that it exhibits reduced performance at higher SNR, indicating a potential limitation in scenarios with stronger signals or lower noise levels.…”

Section: Literature Reviewmentioning

confidence: 99%

Optimizing Spectrum Sensing in Cognitive Radio Networks Using Bayesian-Optimized Random Forest Classifier

2023

IJIES

View full text Add to dashboard Cite

Achieving high spectral efficiency is a fundamental objective in cognitive radio (CR) networks, where fast and precise spectrum sensing plays a vital role. The cognitive radios rely on spectrum sensing to detect available frequencies, identify idle channels, and utilize them for transmitting data. A new approach to spectrum sensing (SS) is presented in this study, using a machine learning-based model. The model leverages different techniques for feature extraction and separates redundant and non-redundant features to enhance detection performance and reduce training overhead. A hybrid approach is introduced, combining energy, MME, and cyclostationary features extracted from sample data with NCA. The selected features are then trained with a Bayesian-optimized random forest classifier (RF) as a machine learning model. The proposed algorithms are evaluated in terms of average training time, classification speed, and accuracy. The experimental results demonstrate that the proposed bayesian-optimized random forest classifier (RF) achieved a probability of detection of 0.94.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Optimizing Spectrum Sensing in Cognitive Radio Networks Using Bayesian-Optimized Random Forest Classifier

2023

IJIES

View full text Add to dashboard Cite

show abstract

“…In view of advantages of the quick detection theory, the CSS problem related to hypothesis test (ie, changepoint detection, sequential test, and Bayesian detection) is gained more in‐depth insight on the performance or efficiency in cognitive UAV or radio networks, that is, .References 4‐20. These related works have varying degrees of advantages or disadvantages.…”

Section: Introductionmentioning

confidence: 99%

“…Morere et al formulated a Bayesian optimisation for continuous UAV trajectories within a partially observable Markov decision process framework in Reference 6 which is solved by Monte‐Carlo tree search and a reward function balancing exploration and exploitation. In Reference 7, a low‐complexity likelihood ratio test based on Shiryaev algorithm for spectrum sensing was introduced at low SNR by Chede et al Apparently, References 4‐7 took quickest detection for spectrum sensing into consideration for cognitive UAV or radio networks, they usually focused on the spectrum sensing performance improvement while ignore sensing efficiency (it is very important to consider spectrum sensing efficiency for battery limited UAV networks), especially the tradeoff problem between the spectrum sensing performance and efficiency or the optimal sensing strategy.…”

Section: Introductionmentioning

confidence: 99%

Bayesian quickest detection‐based spectrum sensing for cognitive UAV networks

Liang

2023

Trans Emerging Tel Tech

View full text Add to dashboard Cite

With the gradual popularization of unmanned aerial vehicles (UAVs), spectrum resources are increasingly scarce. At this time, cognitive radio (CR) technology came into being, which can provide available spectrum resources for UAVs. However, different from the conventional CR networks, the UAV's location flexibility significantly reduces the spectrum sensing performance and increases the detection delay. Therefore, we propose a Bayesian quickest detection-based spectrum sensing for cognitive UAV networks in this paper. To this aim, we establish a multislot spectrum sensing model in a cognitive UAV network to realize cooperative spectrum sensing (CSS) within the single UAV and avoid disadvantages of cooperation among multiple UAVs. On the basis of this, we make use of Bayesian quickest detection to implement multislot spectrum sensing and evaluate its false alarm probability and detection delay. Further, the tradeoff problem between the false alarm and the detection delay is developed and converted into an optimization stopping time problem. Subsequently, the stopping time problem is theoretically proved and solved. Finally, compared to Bayesian detection, numerical simulation results show that the superiority of Bayesian quickest detection with respect to the false alarm probability and the detection delay is evident, which proves the correctness and effectiveness of the proposed tradeoff problem.

show abstract

“…X. Ma et al formulated the Bayesian two‐stage sequential change diagnosis problem in Reference 19 to obtain the optimal rule. A. S. V. Chede proposed a low‐complexity Shiryaev algorithm for mobile CR to implement Bayesian changepoint detection in Reference 20. The aforementioned works, basically used a series of CUSUM algorithms and quickest detection theory to address CSS problem, however, such algorithms pay less attention to the impact of specific sensing parameters or environment on CSS, especially when the UAVs may follow various flight trajectories in a mobile or multi‐channel scenario, which is often the case in CUAVNs.…”

Section: Introductionmentioning

confidence: 99%

Optimal sequential decision rule for cooperative spectrum sensing of CUAVNs

Bao

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Cognitive radio (CR) technology is beneficial for unmanned aerial vehicles (UAVs) to access the spectrum resources needed for communication. Then, in order to integrate CR technology into UAVs without causing excessive interference to the primary users (PUs), cooperative spectrum sensing (CSS) is used as a key function of CR technology to detect the PUs for the UAV's available spectrum. However, the flexible locations of UAVs may significantly degrade the performance and efficiency of CSS within them. Hence, on the basis of a cognitive UAV networks (CUAVN) framework, we propose an internal cooperation paradigm to achieve the cooperative gain of CSS within an UAV. Furthermore, we convert the decision rule as an optimal stopping problem to optimize the decision rule for CSS. The optimization problem is carried out by the error probability regarding the PU presence of the phenomenon and the stopping time required to reach a global decision about the PU status. By optimizing the error probability respectively under the condition of a fixed sensing time, the optimization problem is transformed into an optimal stopping problem and solved by Markov stopping theory. Following this, we further determine a pair of thresholds in the optimal sequential decision rule from the minimal cost function. At last, numerical simulation results verify the correctness and effectiveness of the optimal sequential detection rule, and corroborate that in contrast to the external cooperative paradigm, the superiority of the internal cooperative paradigm under various network environments and parameters, in terms of the CSS performance and the stopping time.

show abstract

Use of Bayesian Changepoint Detection for Spectrum Sensing in Mobile Cognitive Radio

Cited by 6 publications

References 13 publications

Optimizing Spectrum Sensing in Cognitive Radio Networks Using Bayesian-Optimized Random Forest Classifier

Optimizing Spectrum Sensing in Cognitive Radio Networks Using Bayesian-Optimized Random Forest Classifier

Bayesian quickest detection‐based spectrum sensing for cognitive UAV networks

Optimal sequential decision rule for cooperative spectrum sensing of CUAVNs

Contact Info

Product

Resources

About