-
Automatic Data Visualization Generation from Chinese Natural Language Questions
Authors:
Yan Ge,
Victor Junqiu Wei,
Yuanfeng Song,
Jason Chen Zhang,
Raymond Chi-Wing Wong
Abstract:
Data visualization has emerged as an effective tool for getting insights from massive datasets. Due to the hardness of manipulating the programming languages of data visualization, automatic data visualization generation from natural languages (Text-to-Vis) is becoming increasingly popular. Despite the plethora of research effort on the English Text-to-Vis, studies have yet to be conducted on data…
▽ More
Data visualization has emerged as an effective tool for getting insights from massive datasets. Due to the hardness of manipulating the programming languages of data visualization, automatic data visualization generation from natural languages (Text-to-Vis) is becoming increasingly popular. Despite the plethora of research effort on the English Text-to-Vis, studies have yet to be conducted on data visualization generation from questions in Chinese. Motivated by this, we propose a Chinese Text-to-Vis dataset in the paper and demonstrate our first attempt to tackle this problem. Our model integrates multilingual BERT as the encoder, boosts the cross-lingual ability, and infuses the $n$-gram information into our word representation learning. Our experimental results show that our dataset is challenging and deserves further research.
△ Less
Submitted 14 September, 2023;
originally announced September 2023.
-
Beam Management with Orientation and RSRP using Deep Learning for Beyond 5G Systems
Authors:
Khuong N. Nguyen,
Anum Ali,
Jianhua Mo,
Boon Loong Ng,
Vutha Va,
Jianzhong Charlie Zhang
Abstract:
Beam management (BM), i.e., the process of finding and maintaining a suitable transmit and receive beam pair, can be challenging, particularly in highly dynamic scenarios. Side-information, e.g., orientation, from on-board sensors can assist the user equipment (UE) BM. In this work, we use the orientation information coming from the inertial measurement unit (IMU) for effective BM. We use a data-d…
▽ More
Beam management (BM), i.e., the process of finding and maintaining a suitable transmit and receive beam pair, can be challenging, particularly in highly dynamic scenarios. Side-information, e.g., orientation, from on-board sensors can assist the user equipment (UE) BM. In this work, we use the orientation information coming from the inertial measurement unit (IMU) for effective BM. We use a data-driven strategy that fuses the reference signal received power (RSRP) with orientation information using a recurrent neural network (RNN). Simulation results show that the proposed strategy performs much better than the conventional BM and an orientation-assisted BM strategy that utilizes particle filter in another study. Specifically, the proposed data-driven strategy improves the beam-prediction accuracy up to 34% and increases mean RSRP by up to 4.2 dB when the UE orientation changes quickly.
△ Less
Submitted 4 February, 2022;
originally announced February 2022.
-
Sub-Chain Beam for mmWave Devices: A Trade-off between Power Saving and Beam Correspondence
Authors:
Jianhua Mo,
Daehee Park,
Boon Loong Ng,
Vutha Va,
Anum Ali,
Chonghwa Seo,
Jianzhong Charlie Zhang
Abstract:
Beam correspondence, or downlink-uplink (DL-UL) beam reciprocity, refers to the assumption that the best beams in the DL are also the best beams in the UL. This is an important assumption that allows the existing beam management framework in 5G to rely heavily on DL beam swee** and avoid UL beam swee**: UL beams are inferred from the measurements of the DL reference signals. Beam correspondenc…
▽ More
Beam correspondence, or downlink-uplink (DL-UL) beam reciprocity, refers to the assumption that the best beams in the DL are also the best beams in the UL. This is an important assumption that allows the existing beam management framework in 5G to rely heavily on DL beam swee** and avoid UL beam swee**: UL beams are inferred from the measurements of the DL reference signals. Beam correspondence holds when the radio configurations are symmetric in the DL and UL. However, as mmWave technology matures, the DL and the UL face different constraints often breaking the beam correspondence. For example, power constraints may require a UE to activate only a portion of its antenna array for UL transmission, while still activating the full array for DL reception. Meanwhile, if the UL beam with sub-array, named as sub-chain beam in this paper, has a similar radiation pattern as the DL beam, the beam correspondence can still hold. This paper proposes methods for sub-chain beam codebook design to achieve a trade-off between the power saving and beam correspondence.
△ Less
Submitted 22 December, 2021;
originally announced December 2021.
-
Beam Codebook Design for 5G mmWave Terminals
Authors:
Jianhua Mo,
Boon Loong Ng,
Sanghyun Chang,
Pengda Huang,
Mandar Kulkarni,
Ahmad AlAmmouri,
Jianzhong Charlie Zhang,
Jeongheum Lee,
Won-Joon Choi
Abstract:
A beam codebook of 5G millimeter wave (mmWave) for data communication consists of multiple high-peak-gain beams to compensate the high pathloss at the mmWave bands. These beams also have to point to different angular directions, such that by performing beam searching over the codebook, a good mmWave signal coverage over the full sphere around the terminal (spherical coverage) can be achieved. A mo…
▽ More
A beam codebook of 5G millimeter wave (mmWave) for data communication consists of multiple high-peak-gain beams to compensate the high pathloss at the mmWave bands. These beams also have to point to different angular directions, such that by performing beam searching over the codebook, a good mmWave signal coverage over the full sphere around the terminal (spherical coverage) can be achieved. A model-based beam codebook design that assumes ideal omni-directional antenna pattern, and neglects the impact of terminal housing around the antenna, does not work well because the radiation pattern of a practical mmWave antenna combined with the impact of terminal housing is highly irregular. In this paper, we propose a novel and efficient data-driven method to generate a beam codebook to boost the spherical coverage of mmWave terminals. The method takes as inputs the measured or simulated electric field response data of each antenna and provides the codebook according to the requirements on the codebook size, spherical coverage, etc. The method can be applied in a straightforward manner to different antenna type, antenna array configuration, placement and terminal housing design. Our simulation results show that the proposed method generates a codebook better than the benchmark and 802.15.3c codebooks in terms of the spherical coverage.
△ Less
Submitted 2 August, 2019;
originally announced August 2019.
-
Grip-Aware Analog mmWave Beam Codebook Adaptation for 5G Mobile Handsets
Authors:
Ahmad AlAmmouri,
Jianhua Mo,
Boon Loong Ng,
Jianzhong Charlie Zhang,
Jeffrey G. Andrews
Abstract:
This paper studies the effect of the user hand grip on the design of beamforming codebooks for 5G millimeter-wave (mmWave) mobile handsets. The high-frequency structure simulator (HFSS) is used to characterize the radiation fields for fourteen possible handgrip profiles based on experiments we conducted. The loss from hand blockage on the antenna gains can be up to 20-25 dB, which implies that the…
▽ More
This paper studies the effect of the user hand grip on the design of beamforming codebooks for 5G millimeter-wave (mmWave) mobile handsets. The high-frequency structure simulator (HFSS) is used to characterize the radiation fields for fourteen possible handgrip profiles based on experiments we conducted. The loss from hand blockage on the antenna gains can be up to 20-25 dB, which implies that the possible hand grip profiles need to be taken into account while designing beam codebooks. Specifically, we consider three different codebook adaption schemes: a grip-aware scheme, where perfect knowledge of the hand grip is available; a semi-aware scheme, where just the application (voice call, messaging, etc.) and the orientation of the mobile handset is known; and a grip-agnostic scheme, where the codebook ignores hand blockage. Our results show that the ideal grip-aware scheme can provide more than 50% gain in terms of the spherical coverage over the agnostic scheme, depending on the grip and orientation. Encouragingly, the more practical semi-aware scheme we propose provides performance approaching the fully grip-aware scheme. Overall, we demonstrate that 5G mmWave handsets are different from pre-5G handsets: the user grip needs to be explicitly factored into the codebook design.
△ Less
Submitted 2 August, 2019;
originally announced August 2019.
-
A Hybrid Instance-based Transfer Learning Method
Authors:
Azin Asgarian,
Parinaz Sobhani,
Ji Chao Zhang,
Madalin Mihailescu,
Ariel Sibilia,
Ahmed Bilal Ashraf,
Babak Taati
Abstract:
In recent years, supervised machine learning models have demonstrated tremendous success in a variety of application domains. Despite the promising results, these successful models are data hungry and their performance relies heavily on the size of training data. However, in many healthcare applications it is difficult to collect sufficiently large training datasets. Transfer learning can help ove…
▽ More
In recent years, supervised machine learning models have demonstrated tremendous success in a variety of application domains. Despite the promising results, these successful models are data hungry and their performance relies heavily on the size of training data. However, in many healthcare applications it is difficult to collect sufficiently large training datasets. Transfer learning can help overcome this issue by transferring the knowledge from readily available datasets (source) to a new dataset (target). In this work, we propose a hybrid instance-based transfer learning method that outperforms a set of baselines including state-of-the-art instance-based transfer learning approaches. Our method uses a probabilistic weighting strategy to fuse information from the source domain to the model learned in the target domain. Our method is generic, applicable to multiple source domains, and robust with respect to negative transfer. We demonstrate the effectiveness of our approach through extensive experiments for two different applications.
△ Less
Submitted 3 December, 2018;
originally announced December 2018.
-
Boosting Model Performance through Differentially Private Model Aggregation
Authors:
Sophia Collet,
Robert Dadashi,
Zahi N. Karam,
Chang Liu,
Parinaz Sobhani,
Yevgeniy Vahlis,
Ji Chao Zhang
Abstract:
A key factor in develo** high performing machine learning models is the availability of sufficiently large datasets. This work is motivated by applications arising in Software as a Service (SaaS) companies where there exist numerous similar yet disjoint datasets from multiple client companies. To overcome the challenges of insufficient data without explicitly aggregating the clients' datasets du…
▽ More
A key factor in develo** high performing machine learning models is the availability of sufficiently large datasets. This work is motivated by applications arising in Software as a Service (SaaS) companies where there exist numerous similar yet disjoint datasets from multiple client companies. To overcome the challenges of insufficient data without explicitly aggregating the clients' datasets due to privacy concerns, one solution is to collect more data for each individual client, another is to privately aggregate information from models trained on each client's data. In this work, two approaches for private model aggregation are proposed that enable the transfer of knowledge from existing models trained on other companies' datasets to a new company with limited labeled data while protecting each client company's underlying individual sensitive information. The two proposed approaches are based on state-of-the-art private learning algorithms: Differentially Private Permutation-based Stochastic Gradient Descent and Approximate Minima Perturbation. We empirically show that by leveraging differentially private techniques, we can enable private model aggregation and augment data utility while providing provable mathematical guarantees on privacy. The proposed methods thus provide significant business value for SaaS companies and their clients, specifically as a solution for the cold-start problem.
△ Less
Submitted 4 December, 2018; v1 submitted 12 November, 2018;
originally announced November 2018.
-
Directional Cell Search Delay Analysis for Cellular Networks with Static Users
Authors:
Yingzhe Li,
Francois Baccelli,
Jeffrey G. Andrews,
Jianzhong Charlie Zhang
Abstract:
Cell search is the process for a user to detect its neighboring base stations (BSs) and make a cell selection decision. Due to the importance of beamforming gain in millimeter wave (mmWave) and massive MIMO cellular networks, the directional cell search delay performance is investigated. A cellular network with fixed BS and user locations is considered, so that strong temporal correlations exist f…
▽ More
Cell search is the process for a user to detect its neighboring base stations (BSs) and make a cell selection decision. Due to the importance of beamforming gain in millimeter wave (mmWave) and massive MIMO cellular networks, the directional cell search delay performance is investigated. A cellular network with fixed BS and user locations is considered, so that strong temporal correlations exist for the SINR experienced at each BS and user. For Poisson cellular networks with Rayleigh fading channels, a closed-form expression for the spatially averaged mean cell search delay of all users is derived. This mean cell search delay for a noise-limited network (e.g., mmWave network) is proved to be infinite whenever the non-line-of-sight (NLOS) path loss exponent is larger than 2. For interference-limited networks, a phase transition for the mean cell search delay is shown to exist in terms of the number of BS antennas/beams $M$: the mean cell search delay is infinite when $M$ is smaller than a threshold and finite otherwise. Beam-swee** is also demonstrated to be effective in decreasing the cell search delay, especially for the cell edge users.
△ Less
Submitted 3 September, 2017;
originally announced September 2017.
-
Modeling and Analyzing the Coexistence of Wi-Fi and LTE in Unlicensed Spectrum
Authors:
Yingzhe Li,
Francois Baccelli,
Jeffrey G. Andrews,
Thomas D. Novlan,
Jianzhong Charlie Zhang
Abstract:
We leverage stochastic geometry to characterize key performance metrics for neighboring Wi-Fi and LTE networks in unlicensed spectrum. Our analysis focuses on a single unlicensed frequency band, where the locations for the Wi-Fi access points (APs) and LTE eNodeBs (eNBs) are modeled as two independent homogeneous Poisson point processes. Three LTE coexistence mechanisms are investigated: (1) LTE w…
▽ More
We leverage stochastic geometry to characterize key performance metrics for neighboring Wi-Fi and LTE networks in unlicensed spectrum. Our analysis focuses on a single unlicensed frequency band, where the locations for the Wi-Fi access points (APs) and LTE eNodeBs (eNBs) are modeled as two independent homogeneous Poisson point processes. Three LTE coexistence mechanisms are investigated: (1) LTE with continuous transmission and no protocol modifications; (2) LTE with discontinuous transmission; and (3) LTE with listen-before-talk (LBT) and random back-off (BO). For each scenario, we have derived the medium access probability (MAP), the signal-to-interference-plus-noise ratio (SINR) coverage probability, the density of successful transmissions (DST), and the rate coverage probability for both Wi-Fi and LTE. Compared to the baseline scenario where one Wi-Fi network coexists with an additional Wi-Fi network, our results show that Wi-Fi performance is severely degraded when LTE transmits continuously. However, LTE is able to improve the DST and rate coverage probability of Wi-Fi while maintaining acceptable data rate performance when it adopts one or more of the following coexistence features: a shorter transmission duty cycle, lower channel access priority, or more sensitive clear channel assessment (CCA) thresholds.
△ Less
Submitted 19 February, 2016; v1 submitted 5 October, 2015;
originally announced October 2015.
-
What Will 5G Be?
Authors:
Jeffrey G. Andrews,
Stefano Buzzi,
Wan Choi,
Stephen Hanly,
Angel Lozano,
Anthony C. K. Soong,
Jianzhong Charlie Zhang
Abstract:
What will 5G be? What it will not be is an incremental advance on 4G. The previous four generations of cellular technology have each been a major paradigm shift that has broken backwards compatibility. And indeed, 5G will need to be a paradigm shift that includes very high carrier frequencies with massive bandwidths, extreme base station and device densities and unprecedented numbers of antennas.…
▽ More
What will 5G be? What it will not be is an incremental advance on 4G. The previous four generations of cellular technology have each been a major paradigm shift that has broken backwards compatibility. And indeed, 5G will need to be a paradigm shift that includes very high carrier frequencies with massive bandwidths, extreme base station and device densities and unprecedented numbers of antennas. But unlike the previous four generations, it will also be highly integrative: tying any new 5G air interface and spectrum together with LTE and WiFi to provide universal high-rate coverage and a seamless user experience. To support this, the core network will also have to reach unprecedented levels of flexibility and intelligence, spectrum regulation will need to be rethought and improved, and energy and cost efficiencies will become even more critical considerations. This paper discusses all of these topics, identifying key challenges for future research and preliminary 5G standardization activities, while providing a comprehensive overview of the current literature, and in particular of the papers appearing in this special issue.
△ Less
Submitted 12 May, 2014;
originally announced May 2014.
