Machine learning based channel modeling for molecular MIMO communications
18th IEEE International Workshop on Signal Processing Advances in Wireless Communications, SPAWC 2017, Sapporo, Japonya, 3 - 06 Temmuz 2017, cilt.2017-July, ss.1-5, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Cilt numarası: 2017-July
- Doi Numarası: 10.1109/spawc.2017.8227765
- Basıldığı Şehir: Sapporo
- Basıldığı Ülke: Japonya
- Sayfa Sayıları: ss.1-5
- Anahtar Kelimeler: Channel model, Molecular communication, Molecular MIMO, Random movement
- Boğaziçi Üniversitesi Adresli: Evet
Özet
In diffusion-based molecular communication, information particles locomote via a diffusion process, characterized by random movement and heavy tail distribution for the random arrival time. As a result, the molecular communication shows lower transmission rates than the traditional communication. To compensate for such low rates, researchers have recently proposed the molecular multiple-input multiple-output (MIMO) technique. Although channel models exist for single-input single-output (SISO) systems for some simple environments, extending the results to multiple molecular emitters complicates the modeling process. In this paper, we introduce a novel machine learning technique for modeling the molecular MIMO channel and confirm the effectiveness via extensive numerical studies.