論文

In this paper, two-stage BP detection is proposed for overloaded MIMO. The proposal combines BP with the MMSE pre-cancellation algorithm followed by normal BP detection. In overloaded MIMO systems, the loops in a factor graph degrade the demodulation performance of BP detection. MMSE pre-cancellation reduces the number of connections or coefficient values in the factor graph which improves the convergence characteristics of posteriori probabilities. Numerical results obtained through computer simulation show that the BERs of the proposed two-stage BP detection outperforms the conventional BP with MMSE pre-cancellation in a low bit energy range when the MMSE block size is four and the number of MMSE blocks is one. When the pre-cancellation is applied for complexity reduction, the proposed scheme reduces multiplication operations and summation operations by the same factor of 0.7 though the amount of the performance improvement to the conventional scheme is limited.

東海道新幹線車両の屋根上に設置されている架線電圧検知・無線通信共用アンテナは，60 Hzの架線電圧を検知する機能と，VHF帯の無線通信を行う機能の双方を有している．この共用アンテナは，車両の屋根から突起しているため騒音源の一つであるにもかかわらず，新幹線の開業以来その形状に変化がなかった．本アンテナの低騒音化のためには，特に，架線電圧を検知する検知棒の短縮が必要である．そこで，本論文では，折り返しモノポール構造の採用と並列共振回路の付加により，検知棒を短縮し小形化したアンテナを提案する．検知棒を短縮すると架線電圧検知性能が劣化するが，騒音を低減するアンテナレドーム形状内でアンテナ基板面積を可能な限り大きくすることで架線電圧検知性能の劣化を実用上問題ない範囲に収めた．本アンテナの有効性を計算と測定により実証するとともに，積雪・降雨の架線電圧検知性能への影響を実験と計算により確認する．更に，東海道新幹線区間全線の走行試験により性能を評価した結果を示す．

Index modulation (IM) is receiving attention because of its high energy efficiency. In precoding index modulation (PIM), some of the data bits are used for the modulation of symbols and the rest are used for the selection of precoding coefficients. In conventional PIM, the precoding matrices are orthogonal and unitary. In the proposed PIM, the number of the columns of the precoding matrix is expanded more than that of the rows. Because of the expanded precoding matrices, the number of data bits used for the selection of precoding coefficients is increased. As a result, a code rate can be reduced compared to that of the conventional PIM and the number of candidate constellation points for demodulation can be decreased as compared to that of a multiple-input multiple-output MIMO system under the same throughput. Numerical results obtained through computer simulation show that the proposed PIM with QPSK symbols improves the performance by about 2.5 dB at a bit error rate of 10-3 as compared to overloaded MIMO with 16QAM symbols for two transmit antennas and one receive antenna. It also achieves about 3.5 dB better performance than the conventional PIM under the same antenna condition. Furthermore, the optimum number of index modulation bits is found by the simulation for the proposed PIM. In addition, the PIM scheme reduces demodulation complexity by a factor of 32 as compared to that of the MIMO under specific modulation parameters.

As one of the key technologies in the fifth-generation mobile communication system, non-orthogonal multiple access (NOMA) has been investigated. In NOMA, multiple terminals are assigned the same frequency resources by a scheduler on the basis of the difference in propagation losses between a base station and user terminals. Each terminal cancels the signals for the other terminals and extracts its desired signal. On the other hand, the application of joint maximum likelihood (ML) detection to overloaded signals has also been investigated, and joint ML detection can be applied to a NOMA downlink. In this paper, the effect of joint ML detection in a heterogeneous NOMA network is presented. The numerical results obtained through system-level simulation show that joint ML detection in a heterogeneous NOMA downlink can effectively offload mobile traffic from a macro base station to a pico base station. It is shown that a heterogeneous NOMA network with joint ML detection improves the throughput performance by 0.2 bit/user/subcarrier as compared to that without joint ML detection at a cumulative probability of 0.5. The system throughput is also increased about twofold with joint ML detection.

Massive multiple-input multiple-output (MIMO) realizes simultaneous transmission to a large number of mobile stations (MSs) and improves frequency utilization efficiency. It is drawing attention as the key technology of the fifth-generation (5G) mobile communication systems. The 5G system is going to be implemented in a high frequency band and massive MIMO beamforming (BF) is applied to compensate propagation loss. In the conventional BF scheme, a transmit beam is selected based on the power of received signals over subcarriers. The signal on a different subcarrier is transmitted with a different directivity. To improve the accuracy of beam selection, this paper proposes a transmit beam selection scheme for massive MIMO. The proposed scheme calculates the expected responses of the signals over the subcarriers based on the relative directivity between a base station (BS) and a MS. The MS calculates the correlation between the received signals and each of the expected response sequences. It then selects the beam with the highest correlation value. It is shown in this paper that the proposed scheme can improve the average signal-to-noise ratio of a received signal by about 1.5dB as compared with that of the power based search scheme. It is also shown that the proposed scheme with limited response coefficients can reduce the computational complexity by a factor of 1/100 while it still increases the average SNR by about 1.0dB.

Low-resolution analog-to-digital converters (ADCs) may need to be implemented in a full-digital massive multi-user MIMO system for low-power consumption. In this paper, the performance with a proposed ADC design for low-resolution ADCs to suppress the amount of degradation due to the quantization process in the full-digital massive multi-user MIMO system is evaluated. In the proposed ADC design, quantization range limit is specified based on the mean amplitude of a received signal and the input amplitude is bounded to eliminate thermal noise. Numerical results show that, through clipping the thermal noise by setting the appropriate quantization range limit, the full-digital massive MIMO with low-resolution ADCs (even 1-bit ADCs) realizes better system throughput than that with infinite-resolution ADCs. It is also shown that the throughput with low-resolution ADCs is equivalent to that with infinite-resolution ADCs when the number of users is less than three (the number of streams is less than six).

In this paper, mixed Gibbs sampling multiple-input multiple-output (MIMO) detection with forcible search is proposed. In conventional Gibbs sampling MIMO detection, the problem of stalling occurs under high signal-to-noise ratios (SNRs) which degrades the detection performance. Mixed Gibbs sampling (MGS) is one solution to this problem. In MGS, random sampling is carried out with a constant probability regardless of whether a current search falls into a local minimum. In the proposed scheme, combined with MGS, multiple candidate symbols are forcibly changed when the search is captured by a local minimum. The search restarts away from the local minimum which effectively enlarges the search area in the solution space. Numerical results obtained through computer simulation show that the proposed scheme achieves better performance in a large scale MIMO system with QPSK signals.

In this paper, mixed Gibbs sampling multiple- input multiple-output (MIMO) detection with forcible search is proposed. In conventional Gibbs sampling MIMO detection, the problem of stalling occurs under high signal-to-noise ratios (SNRs) which degrades the detection performance. Mixed GS (MGS) is one solution to this problem. In the MGS, random sampling is carried out with a constant probability without judging if a current search is at a local minimum. In the proposed scheme, combined with MGS, multiple candidate symbols are forcibly changed when the search is captured by a local minimum. The search restarts away from the local minimum which effectively enlarges the search area in a solution space. Numerical results obtained through computer simulation show that the proposed scheme achieves better performance in a large scale MIMO system with QPSK signals.

In this paper, the latency of non-orthogonal multiple access (NOMA) grant-free (GF) transmission in uplink is evaluated by link and system level simulations. In order to cancel interference among NOMA signals, multiuser detection (MUD) consisting of minimum mean square error demodulation followed by successive interference cancellation is applied. Retransmission happens only when a packet error occurs after serial interference cancellation (SIC). Therefore, the latency is reduced as compared to GF transmission without MUD. Numerical results obtained through system level simulation show that the application of MUD in uplink GF transmission improves the latency and increases the successful transmission rate without retransmission by a factor of about 3.0.

In this paper, a metric function for Gibbs sampling multiple-input multiple-output (MIMO) detection is proposed. In conventional Gibbs sampling MIMO detection, an exponential function is used in the calculation of a metric for the selection of candidate symbols. However, the exponential function can be implemented by a look-up table and may require a large amount of memory. This paper proposes a metric function based on a simple fraction. The proposed metric substitutes the exponential function though it increases the number of multiplication operations. It is shown by numerical results obtained through computer simulation that the proposed metric function improves the performance under a high signal-to-noise ratio condition in a large scale MIMO system since its curve is close to that of the exponential function when an input metric distance approaches to zero.

In this paper, the application of minimum mean square error (MMSE) pre-cancellation prior to belief propagation (BP) is proposed as a detection scheme for overloaded multiple-input multiple-output (MIMO) systems. In overloaded MIMO systems, the loops in the factor graph degrade the demodulation performance of BP. Therefore, the proposed scheme applies MMSE pre-cancellation prior to BP and reduces the number of loops. Furthermore, it is applied to the selected transmit and receive nodes so that the condition number of an inverse matrix in the MMSE weight matrix is minimized to suppress the residual interference and the noise after MMSE pre-cancellation. Numerical results obtained through computer simulation show that the proposed scheme achieves better bit error rate (BER) performance than BP without MMSE pre-cancellation. The proposed scheme improves the BER performance by 2.9-5.6 dB at a BER of 5.0 × 10− compared with conventional BP. Numerical results also show that MMSE pre-cancellation reduces the complexity of BP by a factor of 896 in terms of the number of multiplication operations. 3

In this paper, block parallel Gibbs sampling (BPGS) multiple-input multiple-output (MIMO) detection is proposed. In a conventional Gibbs sampling scheme, MIMO detection is carried out sequentially symbol-by-symbol. The proposed scheme divides a symbol vector to blocks and updates candidate transmit symbols in parallel in a block so that the total number of iterations in a unit period increases. Furthermore, maximum ratio combining (MRC) is adopted to BPGS to improve accuracy in this paper. Numerical results obtained through computer simulations show that bit error rate performance under high bit-energy-to-noise-spectrum-density conditions improves with the proposed scheme. It is also shown that the block size of three achieves the best performance when number of antennas is 16×16 and the number of iterations is 50.

This research presents the performance of Gibbs sampling iterative MIMO detection and decoding scheme with maximum ratio combining (MRC).

This paper presents a multiuser (MU)-multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM) scheme that realizes signal separation among mobile terminals by spreading codes on top of analog beamforming with analog-digital hybrid massive MIMO. Multiuser MIMO scheme suffers from interference among terminals caused by inaccurate information of channel responses. The channel state information (CSI) of the terminals are outdated owing to a Doppler shift if there is a time lag between channel estimation and CSI feedback. Thus, this paper proposes an MU-MIMO-OFDM scheme with analog-digital hybrid massive MIMO and orthogonal spreading codes that separates signals for mobile terminals. Numerical results obtained through computer simulation show that on WINNER II Umi NLOS channel the proposed scheme improves bit error rate (BER) performance by 1.0dB at a bit error rate of 10 in comparison with the conventional MU-MIMO. -3

Reducing the noise from the sound of train travel is an important issue in high-speed railways such as the Shinkansen. There has been no change in the shape of the antenna installed on the roof on the leading car of the Tokaido Shinkansen since the line started service, even though this is a source of the noise. Here, by changing (miniaturizing) the shape of this antenna from the standpoint of noise countermeasures, we developed an antenna that satisfied both the overhead line voltage detection performance and wireless detection performance, analyzed its performance by examination in factory and running tests, and confirmed that it can be provided for practical use. In this paper, we report a summary of the antenna, the results of the performance evaluation under environments of rain and accumulated snow, and the results of performance evaluation by running tests.

In this paper, a precoder design algorithm using Gibbs sampling for hybrid multiple-input multiple-output (MIMO) systems is proposed. The hybrid MIMO systems are expected to be implemented in high frequency bands. To reduce power consumption and increase the bandwidth of a transmit signal, the number of phase shifters and the number of quatntization levels in digital-analog (D/A) converters have to be limited. A baseband precoder needs to take the effect of quantization into account. Owing to the quantization, it is difficult for a conventional precoder design algorithm with orthogonal matching pursuit (OMP) to reach the optimal solution. On the other hand, the proposed precoder design algorithm using Gibbs sampling is robust to quantization errors. Numerical results obtained through computer simulation show that the proposed algorithm achieves spectral efficiency of 50 bit/s/Hz at a signal-to-noise ration of 20 dB for a quantization level of 24 on an independent Rayleigh fading channel.

In this paper, a precoder design algorithm using spatial signal synthesis with selected multiple antenna subsets for hybrid multiple-input multiple-out-put (MIMO) systems is proposed. The hybrid MIMO systems are expected to be implemented in high frequency bands. To reduce power consumption and increase the bandwidth of a transmit signal, the resolution of phase shifters and the number of quantization levels in digital-analog (D/A) converters have to be limited. A baseband precoder needs to take the effect of quantization into account. On the other hand, when the number of transmit antennas is larger than that of receive antennas, spatial signal synthesis can be realized. The selection of antenna sets also needs to take the effect of quantization into account. In this paper, a transmit antenna selection scheme that improves the spectral efficiency by spatial signal synthesis with multiple antenna subsets is proposed. Numerical results obtained through computer simulation show that the proposed scheme with two antenna subsets of four transmit antennas and two receive antennas improves the spectrum efficiency by about 1.0 bit/s/Hz when the number of transmit and receive are 4.

The sum rate performance of quantized precoding using Gibbs sampling is evaluated in a massive multiple-input multiple-output (MIMO) system in this paper. The massive MIMO is a promising technique to improve spectrum efficiency and energy efficiency. In a full digital massive MIMO system, however, the resolution of digital-to-analogue converters (DACs) in transmit antenna branches have to be low enough because of their power consumption. Therefore, quantized precoding or precoding with the low resolution DACs is investigated. A conventional optimization criterion minimizes errors between a desired received signal and a designed signal. However, the system sum rate may decrease as it increases transmit power. This paper proposes two optimization criteria that take not only the errors but also the transmit power into account to select the candidate quantized transmit signal in order to maximize the sum rate. Moreover, Gibbs sampling is applied to obtain the suboptimal solution of the optimization problem. Numerical results obtained through computer simulation show that the proposed sum rate based criterion for the optimization achieves the approximately 1.3 times higher sum rate than the conventional criterion on a Rician fading channel. On the other hand, the proposed symbol-power-to-mean-square-error ratio (SMSER) based criterion shows faster optimization convergence.

Throughput analysis of the usage of joint maximum-likelihood detection (MLD) in a distributed antenna system (DAS) with multiple mobile terminal scheduling is evaluated in this paper. In DAS, mobile terminals are closer to desired antennas which improves system throughput and increases frequency utilization efficiency. However, since there are more antennas in a cell, interference can occur between transmitted signals from the antennas when multiple mobile terminals are scheduled. Therefore, we propose a novel mobile terminal scheduling method and the use of joint MLD to reduce the effects of interference. A system level simulation shows that the usage of joint MLD in densely packed DAS provides better system throughput.