Search

The ferroelectric field-effect transistor (FeFET) is one of the most promising candidates for emerging nonvolatile memory devices owing to its low write energy and high $I_{\mathrm {ON}}/I_{\mathrm {OFF}}$ ratio. For FeFET applications as nonvolatile memory devices, 1FeFET, 1T-1FeFET, 2T-1FeFET, and 3T-1FeFET cells have been proposed. The 1FeFET cell exhibits the highest density but suffers from write disturbance. Although the 1T-1FeFET and 2T-1FeFET cells resolve the write disturbance, they use a write scheme with a negative write voltage ( $V_{\mathrm {W}}$ ), which requires voltage swings of many control signals, leading to a significantly high write energy consumption. The 3T-1FeFET cell uses a write scheme without a negative $V_{\mathrm {W}}$ ; however, it exhibits the largest area overhead. Although the 1T-1FeFET cell resolves the write disturbance with a small area overhead; however, it exhibits high write energy consumption because of the use of a negative $V_{\mathrm {W}}$ . In this paper, to significantly reduce the write energy consumption, we propose a less control signal swing (LCSS) write scheme without using a negative $V_{\mathrm {W}}$ . Simulation results indicate that the worst, average, and best cases of the proposed LCSS write scheme can achieve 35%, 66%, and 96% lower write energy consumption, respectively, than the write scheme with a negative $V_{\mathrm {W}}$ in the 1T-1FeFET cell. We also identify the available sensing schemes for each FeFET cell in the read operation according to the FeFET threshold voltage distribution.

Recently, nonvolatile systems with nonvolatile flip-flops (NVFFs) have gained prominence for their energy efficiency in energy-harvesting devices and battery-operated Internet of Things applications. They are normally-off instantly-on, and thus, can save energy effectively owing to their zero standby power consumption. An NVFF stores the computing state in nonvolatile memories (NVMs) when the power is off. A ferroelectric field-effect transistor (FeFET) is one of the most promising NVMs owing to its high $\text{I}_{\mathrm {on}}/\text{I}_{\mathrm {off}}$ ratio and low write power. Three FeFET-based NVFFs (previous FeFET-out NVFF-1/-2 and FeFET-in NVFF) were recently proposed to improve the area, power, and speed; however, they still have their own problems. Previous FeFET-out NVFF-1 has large area overhead and previous FeFET-out NVFF-2 does not properly perform restore operation. Previous FeFET-in NVFF has a long clock-to-Q delay and high operating energy. This paper introduces two novel FeFET-based NVFFs (proposed FeFET-out and -in NVFFs). Proposed FeFET-out NVFF reduces the large area overhead of previous FeFET-out NVFF-1 and corrects the malfunction in the restore operation of previous FeFET-out NVFF-2. Proposed FeFET-in NVFF achieves a better clock-to-Q delay, operating energy, and area than the previous FeFET-in NVFF. Monte Carlo simulations based on an industry-compatible 10-nm FinFET model are performed for a comparative analysis. Proposed FeFET-out NVFF achieves 17.6% smaller area with slightly higher (6.3%) operating energy and only 0.8% slower clock-to-Q delay than previous FeFET-out NVFF-1. Proposed FeFET-in NVFF achieves 18.9% shorter clock-to-Q and 3.0% smaller operating energy with 8.7% smaller area than the previous FeFET-in NVFF.

Developing an accurate tourism forecasting model is essential for making desirable policy decisions for tourism management. Early studies on tourism management focus on discovering external factors related to tourism demand. Recent studies utilize deep learning in demand forecasting along with these external factors. They mainly use recursive neural network models such as LSTM and RNN for their frameworks. However, these models are not suitable for use in forecasting tourism demand. This is because tourism demand is strongly affected by changes in various external factors, and recursive neural network models have limitations in handling these multivariate inputs. We propose a multi-head attention CNN model (MHAC) for addressing these limitations. The MHAC uses 1D-convolutional neural network to analyze temporal patterns and the attention mechanism to reflect correlations between input variables. This model makes it possible to extract spatiotemporal characteristics from time-series data of various variables. We apply our forecasting framework to predict inbound tourist changes in South Korea by considering external factors such as politics, disease, season, and attraction of Korean culture. The performance results of extensive experiments show that our method outperforms other deep-learning-based prediction frameworks in South Korea tourism forecasting., Accepted to IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2021)

In this paper, we consider multi-user signal transmission by means of zero-forcing (ZF) beamforming with an aid of user grouping in spatially correlated massive antenna environments. For the user grouping, we exploit transmission performance loss (TPL) due to nulling out inter-user interference. We partition users into a number of groups so that users in each group experience low TPL. We transmit user signal in each group by means of ZF beamforming, while allocating transmission resource to user groups in an orthogonal manner. Finally, we verify the performance of the proposed scheme by computer simulation.

Conventional channel estimation schemes may require for the use of orthogonal pilot signals for each antenna in frequency-division duplex (FDD) wireless systems. It is a very challenging issue to get channel information with an affordable signaling overhead in massive multi-input multi-output (m-MIMO) FDD transmission environments. The signaling overhead for the channel estimation can be reduced by exploiting the downlink channel correlation matrix (DCCM) of m-MIMO channel. With the use of DCCM, the channel information can be estimated using a reduced number of pilot signals. However, the accuracy of DCCM may seriously affect the estimation performance. In this paper, we consider the estimation of DCCM in m-MIMO transmission environments. Since the distance between a base station and a user is much larger than that between antennas, a set of antenna pairs of equal distance may experience similar spatial correlation due to the far-field effect. We consider the estimation of DCCM by averaging the correlation of these antenna pairs without reception of a large number of pilot signals, significantly reducing the pilot signaling overhead. We also consider the improvement of estimated DCCM by dropping out inaccurate correlation terms. Finally, we verify the proposed scheme by computer simulation.