ABOUT
Pekka Kyösti received the M.Sc. degree in mathematics and D.Sc. (Hons.) in Telecommunications from the University of Oulu, Finland, in 2000 and 2018, respectively. From 1998 to 2002, he was with Nokia Networks. From 2002 to 2016, he was with Elektrobit Ltd/Anite Ltd. He was moved to Keysight Technologies Finland Oy along the acquisition in 2016. Since 2002, he has been involved in radio channel measurements, estimation and modelling. He has contributed to numerous widely adopted channel models like, e.g., WINNER, IMT-Advanced, and METIS, as well as to development and standardization of radiated testing in fading radio conditions. Currently he is involved in activities aiming at 5G and beyond 5G systems at Keysight Technologies Finland Oy and at Centre for Wireless Communications (CWC), University of Oulu. His research interests are radio channel characterization and modelling, emulation of multi-dimensional fading radio channel, and over-the-air testing of radio devices.
Tutorial: Radio Channel Modelling for Evaluation of 5G Systems
Abstract: In this tutorial we identify channel model requirements and present a channel model for the fifth generation air interface evaluations. The described model covers frequency bands from typical cellular frequencies up to millimeter waves and a variety of different environments, with emphasis on the urban outdoor. The model enables assessment of single radio links with, e.g., the massive multiple-input-multiple-output (MIMO) and very large antenna arrays, device-to-device links with both link ends moving, up to system level evaluations with a multitude of different types of transceivers. In addition to the overview, some selected model features are described in more detail. Also, a few exemplary model outputs are depicted and discussed. A comparison to corresponding geometry-based stochastic model is performed in urban outdoor environment with the second moment distributions of propagation parameters and with the multiuser (MU) MIMO sum rate capacity. The simulations indicate substantial differences in MU-MIMO performances between the models.
In the second part of this tutorial we discuss and pick high-lights of the recent “5G” channel model specified by 3GPP in the technical report TR 38.901. The specification contains four different modelling approaches/principles: 1) the base-line model is a geometry based stochastic channel model (GSCM) mainly targeted for system level simulations. 2) A reduced variability version of the base line model is the clustered delay line (CDL) approach with five tabulated parameter sets, three for non line of sight (NLOS) and two for line of sight (LOS) conditions. 3) A further reduced category is the tapped delay line models, that are obtained from CDL tables by removing angular and polarimetric parameters and substituting them by amplitude distributions, Doppler spectra, and MIMO correlation matrices. 4) Finally, the specification contains also an alternative hybrid model concept, which combines deterministic, i.e. ray tracing, and stochastic modelling. The technical report does not specify a single unambiguous channel model for performance simulations/emulations. Instead it provides a selection of mathematical procedures and parameterizations for generating realizations of radio channel impulse responses and transmission losses.
Keynote: Putting Channel Models in Practice – Radiated Testing of 5G Devices in Fading Propagation Conditions
Abstract: Radiated testing of massive multiple-input-multiple-output (MIMO) devices in fading radio channel conditions is expected to be essential in development of the fifth generation (5G) base stations (BS) and user equipment (UE) operating at or close to the millimetre wave (mm-wave) frequencies. In this talk we define concepts of fading radio channel emulation and over-the-air (OTA) testing. We introduce main categories of OTA fading emulation setups and focus in more detail to the multi-probe anechoic chamber based (MPAC) system, originally designed for 4G UE testing but upgraded to 5G use as well. We briefly discuss design criteria of MPAC setups and methods for mapping radio channel models onto the probe configuration. The important design parameters for the setup are its dimension, dictated mainly by the measurement distance (the range length), the configuration of switchable probes including their number and locations, and the number of active probes used for the emulation. Parameter values have been determined by simulations using novel metrics. Some of the metrics are also introduced on high level.