• Thursday, 15 June, 9:00-10:30, Room Theatre Piki Hall
• Session Chair: Jorma Olavi Lilleberg (University of Oulu, Finland)

09:00 Multi-cell Interference Coordination for Multigroup Multicast Transmission

Oskari Tervo and Harri Pennanen (University of Oulu, Finland); Symeon Chatzinotas (University of Luxembourg, Luxembourg); Björn Ottersten (University of Luxembourg, Luxembourg); Markku Juntti (University of Oulu, Finland)

Multicasting has become a particularly important technique in the context of cache-enabled cloud radio access networks proposed for 5G systems, where it can be used to transmit common information to multiple users to improve both spectral and energy efficiency. For the efficient spectrum utilization, the future communications are based on aggressive frequency reuse, where the required data rates can be achieved with multiple-input multiple-output precoding techniques. This approach, however, calls for advanced interference coordination techniques. This paper summarizes some of the core approaches proposed in the literature and discusses the main future challenges.

09:18 Analysis of Out-Of-Band Interference from Saturated Power Amplifiers in Massive MIMO

Steve Blandino (Imec, Belgium); Claude Desset and Andre Bourdoux (IMEC, Belgium); Liesbet Van der Perre (KUL, Belgium); Sofie Pollin (KU Leuven, Belgium)

Nonlinear power amplifiers distort the transmitted signal and out-of-band (OOB) radiation becomes a source of interference for users operating in adjacent channels. This paper studies the effect of Massive Multiple Input Multiple Output (MIMO) system on the OOB interference. Massive MIMO relies on channel based precoding which ensures the signal is added constructively at the receiver user equipment. However, the effect of the precoding on the OOB radiation should be investigated given that a random user could experience an increased interference. Assuming a third order polynomial model to describe the behavior of the power amplifier, we model the received power spectrum density. Mathematical analysis and system level simulation confirm that OOB does not recombine constructively avoiding large interference. Massive MIMO allows to increase the in-band received power for the target users without increasing the interference in adjacent bands. In other words, less stringent requirements are demanded for power amplifier design, confirming that simpler hardware with respect to conventional transmission scheme is sufficient. We show that in Massive MIMO, even power amplifiers with efficiency of 65.4% satisfy 3GPP OOB constraints.

09:36 Transparent Spectral Confinement Approach for 5G

Jamal Bazzi and Katsutoshi Kusume (DOCOMO Euro-Labs, Germany); Petra Weitkemper (UniBw Munich, Germany); Kazuaki Takeda (NTT DOCOMO, Inc., Japan); Anass Benjebbour (NTT DOCOMO, INC., Japan)

This paper proposes a transparent spectral confinement approach for OFDM to enable multiplexing of multiple services with diverse requirements in one system band. Besides mobile broadband services, new service types like machine type and ultra-reliable low latency communications foreseen for future 5G systems set new requirements for the chosen waveform to support asynchronous access and multiplexing different numerologies. That is not best handled by OFDM as it is. Thus, various spectral confinement techniques have been proposed in the literature, which, however, require specific processing at both the transmitter and receiver. This tight link would increase signaling overheads to agree on both sides to apply certain respective processing. The transparent approach proposed in this paper decouples the tight link and thus keeps the system simple and robust. We show by means of numerical evaluations that OFDM with spectral confinement techniques like windowing or filtering at the transmitter, but without respective receiver processing, outperforms the conventional OFDM.

09:54 Ultra Reliable Communication via CC-HARQ in Finite Block-Length

Endrit Dosti, Mohammad Shehab and Hirley Alves (University of Oulu, Finland); Matti Latva-aho (UoOulu, Finland)

In this paper, we suggest a power allocation strategy for the Chase Combining Hybrid Automatic Repeat Request (CC-HARQ) protocol with ultra-reliability constraints. The proposed optimal power allocation scheme, would allow us to reach any outage probability target in the finite block-length regime. We cast an optimization problem as minimization of the average transmitted power under a given outage probability and maximum transmit power constraint. To solve the problem and attain the closed form solution, we utilize the Karush-Kuhn-Tucker (KKT) conditions. We show that in the finite block-length regime the transmitted power is highly dependent on the number of channel uses. However, as the block size increases, the transmitted power becomes constant. Furthermore, we show that by using the proposed power allocation scheme, we can achieve very large average and sum power gains when compared to the one shot transmission.

10:12 Optimized Fast Convolution Based Filtered-OFDM Processing for 5G

Juha Yli-Kaakinen, Toni A Levanen, Markku K. Renfors and Mikko Valkama (Tampere University of Technology, Finland)

This paper investigates the application of flexible fast-convolution (FC) filtering scheme for multiplexing orthogonal frequency-division multiplexing (OFDM) physical resource blocks (PRBs) in a spectrally well-localized manner. This scheme is able to suppress interference leakage between adjacent PRBs, thus supporting independent waveform parametrization and numerologies for different PRBs, as well as asynchronous multiuser operation. These are considered as important features in the 5G waveform development. This contribution focuses on optimizing FC based OFDM transmultiplexers such that the in-band interference is minimized subject to the given out-of-band emission constraint. The performance of the optimized designs is demonstrated using resource block groups (RBGs) of different sizes and with various design parameters. The proposed scheme has great flexibility in tuning the filtering bandwidths dynamically according the resource allocation to different users with different requirements regarding the OFDM waveform numerology. Also the computational complexity is competitive with existing time-domain filtering approaches and becomes superior when the number of filtering bands is increased.