Integrasi AI dan IoT untuk Personalisasi Pengalaman Jaringan Multimedia.pptx

Integrasi AI dan IoT untuk Personalisasi Pengalaman Jaringan Multimedia" adalah konsep pemanfaatan kecerdasan buatan (Artificial Intelligence/AI) dan Internet of Things (IoT) dalam membangun ekosistem multimedia yang mampu menyesuaikan layanan, konten, serta kualitas jaringan sesuai kebutuhan dan preferensi pengguna secara real-time.

1. Integrasi AI dan IoT untuk Personalisasi Pengalaman Jaringan Multimedia Dr. Nur Alamsyah, S.T., M.Kom Scopus ID: 57209454458 LinkedIn : dr-nur-alamsyah-289948206 Scopus ID: 57209454458

2. Latar Belakang 70% traffic internet global berasal dari video streaming. Indonesia: 111,43 juta pengguna video streaming. Tantangan : latency, rebuffering, Quality of Experience ( QoE ) tidak konsisten . Sumber : Arslan, M. Y., et al. (2024). Towards AI-assisted sustainable adaptive video streaming. arXiv . https://arxiv.org/abs/2406.02302 Mardian, R., et al. (2022). User experience of 5G video services in Indonesia. Information, 13 (3), 155. https://doi.org/10.3390/info13030155

3. Peran AI dalam Multimedia Deep Reinforcement Learning (DRL) digunakan untuk meningkatkan Adaptive Bitrate (ABR) dalam Dynamic Adaptive Streaming over HTTP (DASH) . Model Gated Recurrent Unit (GRU) untuk prediksi bandwidth. Dampak : QoE meningkat signifikan , rebuffering berkurang . Sumber : Souane, A., et al. (2023). Deep reinforcement learning-based approach for video streaming DASH. Applied Sciences, 13(21), 11697. https://doi.org/10.3390/app132111697 Tran, Q. V., et al. (2024). Joint wireless resource allocation and bitrate adaptation with DRL. Digital Communications and Networks. https://doi.org/10.1016/j.dcan.2024.07.002

4. Peran IoT dalam Multimedia IoT sebagai penyedia data real-time (sensor jaringan , perangkat pengguna ). Studi di Indonesia: pengalaman pengguna 4G memengaruhi persepsi layanan video 5G. Sumber : Mardian, R., et al. (2022). User experience of 5G video services in Indonesia. Information, 13(3), 155. https://doi.org/10.3390/info13030155

5. Konsep Personalisasi Pengalaman Multimedia Personalisasi berbasis preferensi , perangkat , dan kondisi jaringan . Adaptive streaming berbasis profil pengguna . Tujuan: menjaga konsistensi QoE . Sumber : Liotou , E., et al. (2023). Cache-enabled adaptive video streaming: A QoE -based evaluation study. IEEE Access, 11, 120345–120359. https://doi.org/10.1109/ACCESS.2023.3296542 Pratama, D., et al. (2023). Adaptive video streaming: Navigating challenges and embracing personalization. ResearchGate Preprint.

6. Integrasi AI dan IoT IoT → data → AI → keputusan real-time. Arsitektur : IoT devices → Edge/Cloud → AI engine → QoE . Contoh : Federated Deep Reinforcement Learning-based Adaptive Bitrate (FDRLABR) untuk berbagai perangkat IoT. Sumber : Vo, T. D., et al. (2023). Federated deep reinforcement learning-based bitrate adaptation. arXiv . https://arxiv.org/abs/2306.15860

7. Contoh Adaptive Streaming Proactive Energy-Aware Adaptive Streaming meningkatkan QoE hingga 44.8% dalam kondisi daya rendah . Model GRU untuk prediksi bandwidth → mengurangi rebuffering. Sumber : Meng, T., et al. (2021). Proactive energy-aware adaptive video streaming on mobile devices. USENIX Annual Technical Conference (ATC). https://www.usenix.org/conference/atc21/presentation/meng

8. Personalisasi dalam Video Conference AI: noise cancellation, background blur. IoT: sensor kamera & mic untuk adaptasi kualitas video. Algoritma DeepBuffer (buffer-aware ABR) mengurangi stall & waste data. Sumber : Huang, T. Y., et al. (2023). DeepBuffer : Buffer-aware adaptive bitrate algorithm. IEEE INFOCOM 2023. https://godka.github.io/deepbuffer_infocom_23.pdf

9. AR/VR dan Metaverse Virtual Reality (VR) dan Augmented Reality (AR) membutuhkan IoT devices (headset, motion sensor). AI: real-time rendering & adaptive viewport streaming. Studi: AI untuk prediksi viewport di 360° video streaming. Sumber : Zhang, Y., & Guo, J. (2024). A deep reinforcement learning approach to balance viewport prediction and video transmission in 360° video streaming. IJCAI 2024. https://www.ijcai.org/proceedings/2024/680

10. Keuntungan Integrasi AI–IoT QoE meningkat dengan Proactive ABR . Efisiensi bandwidth & energi meningkat . Contoh : AI-aided joint Bitrate and Resource allocation ( iABR ) . Sumber : Meng, T., et al. (2021). Proactive energy-aware adaptive video streaming on mobile devices. USENIX ATC. Cao, H., et al. (2019). iABR : AI-aided joint bitrate and resource allocation. UCL Discovery. https://discovery.ucl.ac.uk/id/eprint/10100605

11. Tantangan dan Isu Privasi data IoT. Kompleksitas arsitektur edge & cloud. Biaya implementasi tinggi . Studi di Indonesia: banyak pengguna masih bertahan di 4G meskipun menantikan 5G. Sumber : Mardian, R., et al. (2022). User experience of 5G video services in Indonesia. Information, 13(3), 155. Arslan, M. Y., et al. (2024). Towards AI-assisted sustainable adaptive video streaming. arXiv .

12. Kesimpulan & Arah Masa Depan AI + IoT = personalisasi multimedia adaptif , efisien , berkelanjutan . Arah masa depan : Sixth Generation (6G) mobile network , holographic communication, metaverse. Sumber : Arslan, M. Y., et al. (2024). Towards AI-assisted sustainable adaptive video streaming. arXiv . Zhang, Y., & Guo, J. (2024). A deep reinforcement learning approach to balance viewport prediction and video transmission. IJCAI.