Koichiro Kitagawa

A handover optimization algorithm with mobility robustness for LTE systems

A large number of cells will be deployed to provide high speed services in any places using the Long-Term Evolution (LTE) system. The management of such a large number of cells increases the operating expenditure (OPEX). Self-organizing network (SON) attracts attention as an effective way to reduce OPEX. This paper proposes a self-optimization algorithm for handover (HO) parameters. In conventional studies, the HO optimization techniques are discussed in models with stationary mobility of user equipment (UE). On the other hand, the key feature of the proposed algorithm is the mobility robustness, which means that the HO performance is robust against the change in UE mobility. In order to realize the mobility robustness, the proposed algorithm adaptively adjusts the HO parameters considering cause of HO failures, which changes in UE mobility. We examine the performance of the proposed algorithm through the computer simulations and confirm the mobility robustness. The simulation study demonstrates the following; The HO parameters are initially set to the optimum values for UEs with 3 km/h based on the random walk. Then, when the mobility changes from 3 km/h random walk to 300 km/h linear motion, the HO failure rates increases to 19%. The proposed algorithm reduces such increasing HO failure rates less than 0.2%.

Performance evaluation of handover in LTE-Advanced systems with pico cell Range Expansion

Cell Range Expansion (CRE) combined with Inter-Cell Interference Coordination (ICIC) in Heterogeneous Networks (HetNet) is a promising technique to enhance the utilization of the resource in pico cells. CRE increases the number of the UE sets connected to pico cells (pico UEs) by changing the trigger timing of the handover (HO). However, inappropriate trigger timing of HO increases HO failures and inefficient HOs, thereby degrading HO performance. HO performance directly affects the continuity of the communication service. Hence, not only the utilization of the resource in pico cells but HO performance should be taken into consideration when CRE is adopted. In this paper, we evaluate the effect of CRE on HO performance through system level simulations, and clarify the application range of CRE. As an example, we examine the CRE bias values which keep the HO failure rate to under 1% and the Ping-Pong HO rate to under about 1%. It is shown that the applicable CRE bias values range from 0 dB to 6 dB, which is smaller than the applicable CRE bias values reported by the conventional studies which range from 0 dB to 9 dB. We also show that the CRE bias values over 6 dB result in high HO failure rates over 5%. Therefore, CRE should be adopted with consideration of HO performance in order to keep good HO performance.

Experimental Results between Non-Linear and Linear Precoding Using Multiuser MIMO Testbed

This paper presents the development of an 8-by-8 real-time multiuser-multiple input multiple output (MU-MIMO) testbed and experimental results of linear and non-linear precoding subject to limited channel state information (CSI) feedback. Linear precoding techniques such as minimum mean square error precoding have already been tested and the related results are available in the literature. However, they are not capacity-achieving. Non-linear precoding techniques such as vector perturbation and Tomlinson-Harashima precoding, on the other hand, achieve in theory the capacity of MU-MIMO, but such claims have not been verified in practice to date. One reason is that theoretical non-linear precoding works rely on the assumption of full and perfect CSI at the transmitter as non-linear precoding techniques are sensitive to CSI error. The merit of this work is that it experimentally confirms (through hardware implementation and pre-field test experimentation) the feasibility of non-linear precoding subject to limited CSI feedback. Experimental results reveal that the non-linear precoding achieves higher throughput performance than the linear precoding scheme by about 35% under realistic CSI feedback.

Development of multiuser MIMO testbed adopting Tomlinson-Harashima Precoding and limited CSI feedback

This paper presents the development of an 8-by-2 real-time multiuser-multiple input multiple output (MU-MIMO) testbed adopting Tomlinson-Harashima Precoding (THP) and limited channel state information (CSI) feedback. In this paper, we present an overview of the MU-MIMO testbed and show the implementation validity of the testbed. The experimental results show that the testbed achieves up to 2.4 dB degradation of BLER compared to the floating point computer simulation results.

Indoor experiment of 8-by-2 multiuser MIMO transmission using Tomlinson-Harashima-Precoding subject to limited CSI feedback

This paper presents indoor experimental results of 8-by-2 multiuser MIMO (MU-MIMO) transmission using Tomlinson-Harashima-Precoding (THP) subject to limited Channel State Information (CSI) feedback. The channel correlation in indoor environment tends to be high because user equipments for MU-MIMO are situated close to each other due to the limited space. In such environment, MU-MIMO using linear precoding degrades the spectrum efficiency. The merit of this work is to experimentally confirm the feasibility of THP, which is robust to highly correlated channel, subject to limited CSI feedback. Experimental results reveal that the THP does not degrade under the highly correlated environment and the peak spectrum efficiency reaches 17 bit/sec/Hz.