Coexistence through adaptive sensing and Markov chains

Tallon, Justin; Blumm, Christian; Puschmann, André; Paisana, Francisco; Belt, Jonathan van de; Francesco, Paolo Di; Kaminski, Nicholas J.; Ahmadi, Hamed

doi:10.1109/dyspan.2015.7343844

Cited by 4 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main challenges in realizing such a system include limitations in observing the channel in a half duplex radio receiver, modeling limitations of the non-stationary PU transmissions [7] and the inherent nondeterminism of channel occupancy. The solutions presented during the challenge tried to tackle some of these challenges, for example predicting the PU channel occupancy pattern using a hidden Markov model (HMM) [3]. However, models based on such dynamic Bayesian Networks (DBNs) like Kalman filters and HMMs may be sub-optimal mainly due to relatively simple state transition structures or internal state space structure.…”

Section: B Suggestions For Future Workmentioning

confidence: 99%

“…A joint team from CONNECT/CTVR, Trinity College Dublin and Technische Universität Ilmenau (TUI) devised a system that makes use of a state machine consisting of sensing, learning, decision making and transmit/receive [3]. A four channel frequency domain energy detection is used during the sensing state.…”

Section: B Team 2: Cnct/tuimentioning

confidence: 99%

“…The DARPA spectrum challenge 3 , a prominent one among them, put state of the art (SoA) cognitive algorithms to the test by designing competitive and cooperative competitions. This challenge gave insight to the adaptability of the SoA radio algorithms in using the spectrum aggressively or sharing it among other users based on the environment.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

IEEE 5G Spectrum Sharing Challenge: A Practical Evaluation of Learning and Feedback

et al. 2016

View full text Add to dashboard Cite

Abstract-We present the results, experiences and takeaways from comparing a diverse set of dynamic spectrum access methods during the IEEE DySPAN 2015 spectrum challenge. Five solutions for coexistence with a given wireless link were implemented and tested in an unknown environment during the conference in Stockholm. The challenge was framed broadly, enabling participants to use their own hardware, antennas, physical layer or medium access control solutions to compete in a unified setup. Each solution was ran two times and ranked using a single metric. Between the two runs the teams were allowed to improve their solution. The metric considered wanted throughput and unwanted interference. In addition to the metric, all solutions were evaluated by a jury. In this paper, we give a detailed overview of the challenge, how we organized it, the participating teams and finally the winners. We conclude with some takeaways on dynamic spectrum access.

show abstract

Section: B Suggestions For Future Workmentioning

confidence: 99%

Section: B Team 2: Cnct/tuimentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

IEEE 5G Spectrum Sharing Challenge: A Practical Evaluation of Learning and Feedback

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Our results in comparison with neural networks and conventional hidden Markov models showed that the proposed method performs equally with those, while being much less computationally intensive. A version of the proposed learning technique has been implemented in [10] on Ettus Research 1 Universal Software Radio Platforms (USRPs).…”

mentioning

confidence: 99%

Deep Learning Meets Cognitive Radio: Predicting Future Steps

Shenfield

Khan

Ahmadi

2020

2020 IEEE 91st Vehicular Technology Conference (VTC2020-Spring)

Self Cite

View full text Add to dashboard Cite

Learning the channel occupancy patterns to reuse the underutilised spectrum frequencies without interfering with the incumbent is a promising approach to overcome the spectrum limitations. In this work we proposed a Deep Learning (DL) approach to learn the channel occupancy model and predict its availability in the next time slots. Our results show that the proposed DL approach outperforms existing works by 5%. We also show that our proposed DL approach predicts the availability of channels accurately for more than one time slot.

show abstract