Abstract
We propose two positioning based intra-frequency handover methods assisted by radio maps. The objective is to minimize the number of handovers in a use-case consisting of a robot equipped with a positioning device, moving along predetermined routes, in two indoor scenarios. The proposed methods predict the next position of the robot. The position is input to functions, constructed by symbolic regression from field measurements, that describe the radio maps for each cell, mapping positions to received powers. The function returning the highest power determines the serving cell. With one of the proposed methods, which implements a neural network, we manage to significantly reduce the number of handovers performed along the routes of the robot. The reduction in handovers allows to save resources, and scale up the use of simultaneous transmissions, which are needed to minimize handover setup times, making possible ultra reliable communications in cellular networks.
Original language | English |
---|---|
Title of host publication | 11th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops, ICUMT 2019 |
Publisher | IEEE Institute of Electrical and Electronic Engineers |
ISBN (Electronic) | 978-1-7281-5764-1, 978-1-7281-5763-4 |
ISBN (Print) | 978-1-7281-5765-8 |
DOIs | |
Publication status | Published - Oct 2019 |
MoE publication type | A4 Article in a conference publication |
Event | 11th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops, ICUMT 2019 - Dublin, Ireland Duration: 28 Oct 2019 → 30 Oct 2019 |
Conference
Conference | 11th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops, ICUMT 2019 |
---|---|
Country/Territory | Ireland |
City | Dublin |
Period | 28/10/19 → 30/10/19 |
Keywords
- Deep learning
- intra-frequency handover
- neural networks
- positioning
- radio maps
- ultra reliable communications