Description
Date depot: 20 septembre 2024
Titre: Uplink Radiated Energy Prediction of Cellular Networks in Complex Dynamic Environment
Directeur de thèse: Farid NAIT-ABDESSELAM (LIPADE)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: Artificial Intelligence (AI) has emerged as a powerful tool for predicting the performance of wireless networks. Through the utilization of AI algorithms, wireless network operators can analyze complex data patterns, optimize network parameters, and forecast network behavior with greater accuracy andefficiency. Among the performance metric, the radio frequency (RF) electromagnetic field (EMF) exposure is raising more and more public concerns. Assessments for compliance with RF-EMF regulations are carried out on radio equipment prior to market placement. This ensures adherence to exposure limits outlined by organizations like the International Commission on Non-Ionizing Radiation Protection (ICNIRP). RF EMF exposure limits are usuallydesigned to be averaged over a specific time frame, making the UE time-averaged transmit power more pertinent than its instantaneous counterpart. In terms of the RF sources, there are downlink (DL) and uplink (UL) sources. Common DL sources include cellular base stations, WiFi access points. The assessments of RF EMF exposure has been widely explored, such as. For uplink (UL) sources, user equipment (UE) is of great interest due to its extensive daily usage by humans. Presently, popular measurement methods include network tool-based approaches and the use of external add- on sensors. However, these measurement- based methods require substantial investment and adaptation to accommodate various mobile devices and systems. And due to the large uncertainty caused by the dynamic complex environment, the propagation of EM waves in the measurement data exhibits significant fluctuation due to the reflection, refraction, and scattering in the environment. The second challenge in the UL exposure assessments comes from the different UL power control algorithms in different generation of wireless technologies, e.g., 3G, 4G, and 5G. To be more specific, 4G LTE power control algorithms are relatively static and rely on pre- defined parameters to adjust transmit power based on the channel conditions and network requirements. 5G NR employs more advanced and adaptive power control mechanisms, lever- aging techniques such as beamforming, massive MIMO (Multiple-Input Multiple-Output), and dynamic scheduling. Last but not least, in the literature of UL RF-EMF exposure assessments, people are in- terested in the power density (or electric field) to quantify the RF-EMF exposure level [8]. However, it is not complete to assess exposure without considering the amount of data transmitted, which is throughput.
Résumé dans une autre langue: Artificial Intelligence (AI) has emerged as a powerful tool for predicting the performance of wireless networks. Through the utilization of AI algorithms, wireless network operators can analyze complex data patterns, optimize network parameters, and forecast network behavior with greater accuracy andefficiency. Among the performance metric, the radio frequency (RF) electromagnetic field (EMF) exposure is raising more and more public concerns. Assessments for compliance with RF-EMF regulations are carried out on radio equipment prior to market placement. This ensures adherence to exposure limits outlined by organizations like the International Commission on Non-Ionizing Radiation Protection (ICNIRP). RF EMF exposure limits are usuallydesigned to be averaged over a specific time frame, making the UE time-averaged transmit power more pertinent than its instantaneous counterpart. In terms of the RF sources, there are downlink (DL) and uplink (UL) sources. Common DL sources include cellular base stations, WiFi access points. The assessments of RF EMF exposure has been widely explored, such as. For uplink (UL) sources, user equipment (UE) is of great interest due to its extensive daily usage by humans. Presently, popular measurement methods include network tool-based approaches and the use of external add- on sensors. However, these measurement- based methods require substantial investment and adaptation to accommodate various mobile devices and systems. And due to the large uncertainty caused by the dynamic complex environment, the propagation of EM waves in the measurement data exhibits significant fluctuation due to the reflection, refraction, and scattering in the environment. The second challenge in the UL exposure assessments comes from the different UL power control algorithms in different generation of wireless technologies, e.g., 3G, 4G, and 5G. To be more specific, 4G LTE power control algorithms are relatively static and rely on pre- defined parameters to adjust transmit power based on the channel conditions and network requirements. 5G NR employs more advanced and adaptive power control mechanisms, lever- aging techniques such as beamforming, massive MIMO (Multiple-Input Multiple-Output), and dynamic scheduling. Last but not least, in the literature of UL RF-EMF exposure assessments, people are in- terested in the power density (or electric field) to quantify the RF-EMF exposure level [8]. However, it is not complete to assess exposure without considering the amount of data transmitted, which is throughput.
Doctorant.e: Sun Qunfei
Titre: Uplink Radiated Energy Prediction of Cellular Networks in Complex Dynamic Environment
Directeur de thèse: Farid NAIT-ABDESSELAM (LIPADE)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Systèmes et réseaux
Resumé: Artificial Intelligence (AI) has emerged as a powerful tool for predicting the performance of wireless networks. Through the utilization of AI algorithms, wireless network operators can analyze complex data patterns, optimize network parameters, and forecast network behavior with greater accuracy andefficiency. Among the performance metric, the radio frequency (RF) electromagnetic field (EMF) exposure is raising more and more public concerns. Assessments for compliance with RF-EMF regulations are carried out on radio equipment prior to market placement. This ensures adherence to exposure limits outlined by organizations like the International Commission on Non-Ionizing Radiation Protection (ICNIRP). RF EMF exposure limits are usuallydesigned to be averaged over a specific time frame, making the UE time-averaged transmit power more pertinent than its instantaneous counterpart. In terms of the RF sources, there are downlink (DL) and uplink (UL) sources. Common DL sources include cellular base stations, WiFi access points. The assessments of RF EMF exposure has been widely explored, such as. For uplink (UL) sources, user equipment (UE) is of great interest due to its extensive daily usage by humans. Presently, popular measurement methods include network tool-based approaches and the use of external add- on sensors. However, these measurement- based methods require substantial investment and adaptation to accommodate various mobile devices and systems. And due to the large uncertainty caused by the dynamic complex environment, the propagation of EM waves in the measurement data exhibits significant fluctuation due to the reflection, refraction, and scattering in the environment. The second challenge in the UL exposure assessments comes from the different UL power control algorithms in different generation of wireless technologies, e.g., 3G, 4G, and 5G. To be more specific, 4G LTE power control algorithms are relatively static and rely on pre- defined parameters to adjust transmit power based on the channel conditions and network requirements. 5G NR employs more advanced and adaptive power control mechanisms, lever- aging techniques such as beamforming, massive MIMO (Multiple-Input Multiple-Output), and dynamic scheduling. Last but not least, in the literature of UL RF-EMF exposure assessments, people are in- terested in the power density (or electric field) to quantify the RF-EMF exposure level [8]. However, it is not complete to assess exposure without considering the amount of data transmitted, which is throughput.
Résumé dans une autre langue: Artificial Intelligence (AI) has emerged as a powerful tool for predicting the performance of wireless networks. Through the utilization of AI algorithms, wireless network operators can analyze complex data patterns, optimize network parameters, and forecast network behavior with greater accuracy andefficiency. Among the performance metric, the radio frequency (RF) electromagnetic field (EMF) exposure is raising more and more public concerns. Assessments for compliance with RF-EMF regulations are carried out on radio equipment prior to market placement. This ensures adherence to exposure limits outlined by organizations like the International Commission on Non-Ionizing Radiation Protection (ICNIRP). RF EMF exposure limits are usuallydesigned to be averaged over a specific time frame, making the UE time-averaged transmit power more pertinent than its instantaneous counterpart. In terms of the RF sources, there are downlink (DL) and uplink (UL) sources. Common DL sources include cellular base stations, WiFi access points. The assessments of RF EMF exposure has been widely explored, such as. For uplink (UL) sources, user equipment (UE) is of great interest due to its extensive daily usage by humans. Presently, popular measurement methods include network tool-based approaches and the use of external add- on sensors. However, these measurement- based methods require substantial investment and adaptation to accommodate various mobile devices and systems. And due to the large uncertainty caused by the dynamic complex environment, the propagation of EM waves in the measurement data exhibits significant fluctuation due to the reflection, refraction, and scattering in the environment. The second challenge in the UL exposure assessments comes from the different UL power control algorithms in different generation of wireless technologies, e.g., 3G, 4G, and 5G. To be more specific, 4G LTE power control algorithms are relatively static and rely on pre- defined parameters to adjust transmit power based on the channel conditions and network requirements. 5G NR employs more advanced and adaptive power control mechanisms, lever- aging techniques such as beamforming, massive MIMO (Multiple-Input Multiple-Output), and dynamic scheduling. Last but not least, in the literature of UL RF-EMF exposure assessments, people are in- terested in the power density (or electric field) to quantify the RF-EMF exposure level [8]. However, it is not complete to assess exposure without considering the amount of data transmitted, which is throughput.
Doctorant.e: Sun Qunfei