Joint communication and sensing (JCAS) technology has been regarded as one of the innovations in the 6G network. With the channel modeling proposed by the 3rd Generation Partnership Project (3GPP) TR 38.901, this paper investigates the sensing capability using the millimeter-wave (mmWave) band with an orthogonal frequency division multiplexing (OFDM) waveform. Based on micro-Doppler (MD) analysis, we present two case studies, i.e., fan speed detection and human activity recognition, to demonstrate the target modeling with micro-motions, backscattering signal construction, and MD signature extraction using an OFDM waveform at 28 GHz. Simulated signatures demonstrate distinct fan rotation or human motion, and waveform parameters that affect the MD signature extraction are analyzed. Simulation results draw the validity of the proposed modeling and simulation methods, which also aim to facilitate the generation of data sets for various JCAS applications.
Taobao Search consists of two phases: the retrieval phase and the ranking phase. Given a user query, the retrieval phase returns a subset of candidate products for the following ranking phase. Recently, the paradigm of pre-training and fine-tuning has shown its potential in incorporating visual clues into retrieval tasks. In this paper, we focus on solving the problem of text-to-multimodal retrieval in Taobao Search. We consider that users' attention on titles or images varies on products. Hence, we propose a novel Modal Adaptation module for cross-modal fusion, which helps assigns appropriate weights on texts and images across products. Furthermore, in e-commerce search, user queries tend to be brief and thus lead to significant semantic imbalance between user queries and product titles. Therefore, we design a separate text encoder and a Keyword Enhancement mechanism to enrich the query representations and improve text-to-multimodal matching. To this end, we present a novel vision-language (V+L) pre-training methods to exploit the multimodal information of (user query, product title, product image). Extensive experiments demonstrate that our retrieval-specific pre-training model (referred to as MAKE) outperforms existing V+L pre-training methods on the text-to-multimodal retrieval task. MAKE has been deployed online and brings major improvements on the retrieval system of Taobao Search.
Locating mobile devices precisely in indoor scenarios is a challenging task because of the signal diffraction and reflection in complicated environments. One vital cause deteriorating the localization performance is the inevitable power dissipation along the propagation path of localization signals. In this paper, we propose a high-accuracy localization scheme based on the resonant beam system (RBS) and the binocular vision, i.e., binocular based resonant beam localization (BRBL). The BRBL system utilizes the energy-concentrated and self-aligned transmission of RBS to realize high-efficiency signal propagation and self-positioning for the target. The binocular method is combined with RBS to obtain the three-dimensional (3-D) coordinates of the target for the first time. To exhibit the localization mechanism, we first elaborate on the binocular localization model, including the resonant beam transmission analysis and the geometric derivation of the binocular method with RBS. Then, we establish the power model of RBS, and the signal and noise models of beam spot imaging, respectively, to analyse the performance of the BRBL system. Finally, the numerical results show an outstanding performance of centimeter level accuracy (i.e., $<5\mathrm{cm}$ in $0.4\mathrm{m}$ width and $0.4\mathrm{m}$ length effective range at $1\mathrm{m}$ vertical distance, $<13\mathrm{cm}$ in $0.6\mathrm{m}$ width and $0.6\mathrm{m}$ length effective range at $2\mathrm{m}$ vertical distance), which applies to indoor scenarios.
In large-scale e-commerce platforms like Taobao, it is a big challenge to retrieve products that satisfy users from billions of candidates. This has been a common concern of academia and industry. Recently, plenty of works in this domain have achieved significant improvements by enhancing embedding-based retrieval (EBR) methods, including the Multi-Grained Deep Semantic Product Retrieval (MGDSPR) model [16] in Taobao search engine. However, we find that MGDSPR still has problems of poor relevance and weak personalization compared to other retrieval methods in our online system, such as lexical matching and collaborative filtering. These problems promote us to further strengthen the capabilities of our EBR model in both relevance estimation and personalized retrieval. In this paper, we propose a novel Multi-Objective Personalized Product Retrieval (MOPPR) model with four hierarchical optimization objectives: relevance, exposure, click and purchase. We construct entire-space multi-positive samples to train MOPPR, rather than the single-positive samples for existing EBR models.We adopt a modified softmax loss for optimizing multiple objectives. Results of extensive offline and online experiments show that MOPPR outperforms the baseline MGDSPR on evaluation metrics of relevance estimation and personalized retrieval. MOPPR achieves 0.96% transaction and 1.29% GMV improvements in a 28-day online A/B test. Since the Double-11 shopping festival of 2021, MOPPR has been fully deployed in mobile Taobao search, replacing the previous MGDSPR. Finally, we discuss several advanced topics of our deeper explorations on multi-objective retrieval and ranking to contribute to the community.
Waterfall Recommender System (RS), a popular form of RS in mobile applications, is a stream of recommended items consisting of successive pages that can be browsed by scrolling. In waterfall RS, when a user finishes browsing a page, the edge (e.g., mobile phones) would send a request to the cloud server to get a new page of recommendations, known as the paging request mechanism. RSs typically put a large number of items into one page to reduce excessive resource consumption from numerous paging requests, which, however, would diminish the RSs' ability to timely renew the recommendations according to users' real-time interest and lead to a poor user experience. Intuitively, inserting additional requests inside pages to update the recommendations with a higher frequency can alleviate the problem. However, previous attempts, including only non-adaptive strategies (e.g., insert requests uniformly), would eventually lead to resource overconsumption. To this end, we envision a new learning task of edge intelligence named Intelligent Request Strategy Design (IRSD). It aims to improve the effectiveness of waterfall RSs by determining the appropriate occasions of request insertion based on users' real-time intention. Moreover, we propose a new paradigm of adaptive request insertion strategy named Uplift-based On-edge Smart Request Framework (AdaRequest). AdaRequest 1) captures the dynamic change of users' intentions by matching their real-time behaviors with their historical interests based on attention-based neural networks. 2) estimates the counterfactual uplift of user purchase brought by an inserted request based on causal inference. 3) determines the final request insertion strategy by maximizing the utility function under online resource constraints. We conduct extensive experiments on both offline dataset and online A/B test to verify the effectiveness of AdaRequest.
Modeling user's historical feedback is essential for Click-Through Rate Prediction in personalized search and recommendation. Existing methods usually only model users' positive feedback information such as click sequences which neglects the context information of the feedback. In this paper, we propose a new perspective for context-aware users' behavior modeling by including the whole page-wisely exposed products and the corresponding feedback as contextualized page-wise feedback sequence. The intra-page context information and inter-page interest evolution can be captured to learn more specific user preference. We design a novel neural ranking model RACP(i.e., Recurrent Attention over Contextualized Page sequence), which utilizes page-context aware attention to model the intra-page context. A recurrent attention process is used to model the cross-page interest convergence evolution as denoising the interest in the previous pages. Experiments on public and real-world industrial datasets verify our model's effectiveness.
Laser has unique advantages such as abundant spectrum resources and low propagation divergence in wireless charging and wireless communications, compared with radio frequency. Resonant beams, as a kind of intra-cavity laser beams, have been proposed as the carrier of wireless charging and communication, as it has unique features including high power, intrinsic safety, and self-aligned mobility. However, this system has problems such as intra-cavity echo interference and power fluctuation. To study the time-domain behavior of the resonant beam system, we create a simulation algorithm by discretizing the laser rate equations which model the dynamics of the excited atom density in the gain medium and the photon density in the cavity. The simulation results are in good agreement with theoretical calculation. We also propose a delay-divide demodulation method to address the echo interference issue, and use the simulation algorithm to verify its feasibility. The results show that the resonant beam charging and communication system with the proposed demodulator is feasible and performs well. The analysis in this work also helps researchers to deeply understand the behavior of the resonant beam system.
Short video has witnessed rapid growth in China and shows a promising market for promoting the sales of products in e-commerce platforms like Taobao. To ensure the freshness of the content, the platform needs to release a large number of new videos every day, which makes the conventional click-through rate (CTR) prediction model suffer from the severe item cold-start problem. In this paper, we propose GIFT, an efficient Graph-guIded Feature Transfer system, to fully take advantages of the rich information of warmed-up videos that related to the cold-start video. More specifically, we conduct feature transfer from warmed-up videos to those cold-start ones by involving the physical and semantic linkages into a heterogeneous graph. The former linkages consist of those explicit relationships (e.g., sharing the same category, under the same authorship etc.), while the latter measure the proximity of multimodal representations of two videos. In practice, the style, content, and even the recommendation pattern are pretty similar among those physically or semantically related videos. Besides, in order to provide the robust id representations and historical statistics obtained from warmed-up neighbors that cold-start videos covet most, we elaborately design the transfer function to make aware of different transferred features from different types of nodes and edges along the metapath on the graph. Extensive experiments on a large real-world dataset show that our GIFT system outperforms SOTA methods significantly and brings a 6.82% lift on click-through rate (CTR) in the homepage of Taobao App.
High-power and high-capacity simultaneous wireless information and power transfer (SWIPT) becomes more and more important with the development of Internet of Things technologies. Optical SWIPT, also known as simultaneous light information and power transfer (SLIPT), has unique advantages such as abundant spectrum resources and low propagation divergence, compared with RF technologies. However, optical SWIPT faces many challenges in beam steering and receiver positioning/tracking. Resonant beams generated by spatially separated laser resonators (SSLR) have many advantages, including high power, self-aligned mobility, and intrinsic safety. It has been proposed as the carrier of wireless charging and communication. Using resonant beams, mobile electronic devices can be remotely charged and supported with high-rate data transfer. In this paper, we propose a mobile optical SWIPT system based on asymmetric SSLR and present the system optimization procedure. We also determine the boundary of the achievable charging power and communication capacity, and discuss the trade-off between power transfer and information transfer. Numerical results show that the charging power of the optimized asymmetric system is much higher than that of the symmetric system in the previous work, and meanwhile, the channel capacity is kept almost unchanged.
Simultaneous wireless information and power transfer (SWIPT) has been envisioned as an enabling technology for future 6G by providing high-efficiency power transfer and high-rate data transmissions concurrently. In this paper, we propose a resonant beam charging and communication (RBCC) system utilizing the telescope internal modulator (TIM) and the semiconductor gain medium. TIM can concentrate the diverged beam into a small-size gain module, thus the propagation loss is reduced and the transmission efficiency is enhanced. Since the semiconductor gain medium has better energy absorption capacity compared with the traditional solid-state one, the overall energy conversion efficiency can be improved. We establish an analytical model of this RBCC system for SWIPT and evaluate its stability, output energy, and spectral efficiency. Numerical analysis shows that the proposed RBCC system can realize stable SWIPT over 10 meters, whose energy conversion efficiency is increased by 14 times compared with the traditional system using the solid-state gain medium without TIM, and the spectrum efficiency can be above 15 bit/s/Hz.