Kaibin Huang

Semantic communication (SemCom) has emerged as a promising paradigm for achieving unprecedented communication efficiency in sixth-generation (6G) networks by leveraging artificial intelligence (AI) to extract and transmit the underlying meanings of source data. However, deploying SemCom over digital systems presents new challenges, particularly in ensuring robustness against transmission errors that may distort semantically critical content. To address this issue, this paper proposes a novel framework, termed generative feature imputing, which comprises three key techniques. First, we introduce a spatial-error-concentration packetization strategy that spatially concentrates feature distortions by encoding feature elements based on their channel mappings—a property crucial for both the effectiveness and reduced complexity of the subsequent techniques. Second, building on this strategy, we propose a generative feature imputing method that utilizes a diffusion model to efficiently reconstruct missing features caused by packet losses. Finally, we develop a semantic-aware power allocation scheme that enables unequal error protection by allocating transmission power according to the semantic importance of each packet. Experimental results demonstrate that the proposed framework outperforms conventional approaches, such as Deep Joint Source-Channel Coding (DJSCC) and JPEG2000, under block fading conditions, achieving higher semantic accuracy and lower Learned Perceptual Image Patch Similarity (LPIPS) scores.

The development of sixth-generation (6G) mobile networks imposes unprecedented latency and reliability demands on multiple-input multiple-output (MIMO) communication systems, a key enabler of high-speed radio access. Recently, deep unfolding-based detectors, which map iterative algorithms onto neural network architectures, have emerged as a promising approach, combining the strengths of model-driven and data-driven methods to achieve high detection accuracy with relatively low complexity. However, algorithmic innovation alone is insufficient; software-hardware co-design is essential to meet the extreme latency requirements of 6G (i.e., 0.1 milliseconds). This motivates us to propose leveraging in-memory computing, which is an analog computing technology that integrates memory and computation within memristor circuits, to perform the intensive matrix-vector multiplication (MVM) operations inherent in deep MIMO detection at the nanosecond scale. Specifically, we introduce a novel architecture, called the deep in-memory MIMO (IM-MIMO) detector, characterized by two key features. First, each of its cascaded computational blocks is decomposed into channel-dependent and channel-independent neural network modules. Such a design minimizes the latency of memristor reprogramming in response to channel variations, which significantly exceeds computation time. Second, we develop a customized detector-training method that exploits prior knowledge of memristor-value statistics to enhance robustness against programming noise. Furthermore, we conduct a comprehensive analysis of the IM-MIMO detector's performance, evaluating detection accuracy, processing latency, and hardware complexity. Our study quantifies detection error as a function of various factors, including channel noise, memristor programming noise, and neural network size. The insights gained from these tradeoffs provide valuable guidance for practical system implementation.

A distinctive function of sixth-generation (6G) networks is the integration of distributed sensing and edge artificial intelligence (AI) to enable intelligent perception of the physical world. This resultant platform, termed integrated sensing and edge AI (ISEA), is envisioned to enable a broad spectrum of Internet-of-Things (IoT) applications, including remote surgery, autonomous driving, and holographic telepresence. Recently, the communication bottleneck confronting the implementation of an ISEA system is overcome by the development of over-the-air computing (AirComp) techniques, which facilitate simultaneous access through over-the-air data feature fusion. Despite its advantages, AirComp with uncoded transmission remains vulnerable to interference. To tackle this challenge, we propose AirBreath sensing, a spectrum-efficient framework that cascades feature compression and spread spectrum to mitigate interference without bandwidth expansion. This work reveals a fundamental tradeoff between these two operations under a fixed bandwidth constraint: increasing the compression ratio may reduce sensing accuracy but allows for more aggressive interference suppression via spread spectrum, and vice versa. This tradeoff is regulated by a key variable called breathing depth, defined as the feature subspace dimension that matches the processing gain in spread spectrum. To optimally control the breathing depth, we mathematically characterize and optimize this aforementioned tradeoff by designing a tractable surrogate for sensing accuracy, measured by classification discriminant gain (DG). Experimental results on real datasets demonstrate that AirBreath sensing effectively mitigates interference in ISEA systems, and the proposed control algorithm achieves near-optimal performance as benchmarked with a brute-force search.

Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm of >edge model caching. In this paper, we develop a novel model placement framework, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To tackle this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with a (1 − ϵ) /2-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models.

The Wizard-of-Oz (WoZ) method has long been a core prototyping technique in Human-Computer Interaction (HCI), in which users interact with systems that seem autonomous but are actually controlled by hidden human operators. Advances in interactive technologies have expanded the landscape of future system behaviors, broadening both where and how WoZ is used. However, as more envisioned behaviors become technically feasible, the distinction between engineering a system and simulating an interaction becomes blurred, making it essential to clarify when and why to employ wizarding. This paper presents the first systematic review of WoZ in HCI, drawing on 194 papers from SIGCHI venues to identify ten application domains, five wizard control types, eight motivations, and five categories of concerns. Building on these findings, we propose a reciprocal evolution framework that interprets how technology and wizarding shape each other, and derive guidelines for the rigorous application of WoZ. We further illustrate the framework through emerging prototyping practices with Large Language Models (LLMs).

Federated fine-tuning attempts to fine-tune a pre-trained model with private data from distributed clients by exchanging models rather than data under the orchestration of a parameter server (PS). To overcome the bottleneck imposed by the growing communication and memory overhead for clients in such systems due to ever-increasing model sizes, we propose FeedSign, a federated fine-tuning algorithm in which the upload and download payloads are exactly 1 bit per aggregation step. Meanwhile, the memory overhead is reduced to the minimum required for inference. The reduction is achieved by using zeroth-order (ZO) optimizers on large models and shared pseudo-random number generators (PRNGs) across devices to represent gradient estimates as seed-sign pairs. We conduct theoretical analysis on FeedSign and show that it converges at an exponential rate under the widely used Polyak-Łojaciewicz assumption. Moreover, FeedSign is robust against data heterogeneity and Byzantine attacks due to the 1-bit vague updates. We extensively tested FeedSign on models across different structures and sizes (11M to 13B). We found that the proposed method performs better or closely matches its ZO and FO counterparts, depending on the scenario, despite having an order-of-magnitude lower communication overhead.

Rydberg Atomic REceivers (RAREs) have demonstrated remarkable capabilities for radio-frequency signal measurement, enabling advanced quantum wireless sensing. Existing RARE-based sensing systems popularly adopt the heterodyne detection methodology, which requires an additional reference source to serve as an atomic mixer. However, this approach entails a bulky transceiver architecture and is limited in the supportable sensing bandwidth. To address these limitations, we propose a self-heterodyne sensing paradigm where the transmitter’s self-interference naturally provides the reference signal. We demonstrate that a self-heterodyne RARE functions as an atomic autocorrelator, eliminating the need for external reference sources while supporting substantially wider bandwidth than conventional heterodyne methods. Next, a two-stage algorithm is devised to perform target ranging in self-heterodyne RARE systems. This algorithm is shown to closely approach the Cramér-Rao lower bound. Furthermore, we introduce the power-trajectory (P-trajectory) design for RAREs, which maximizes the sensing sensitivity through time-varying transmission power control. An internal noise (ITN)-limited P-trajectory is developed to capture the profile of the asymptotically optimal time-varying power in the presence of ITN only. This design is then extended to the practical P-trajectory by incorporating both the ITN and external noise. Numerical results validate that the proposed selfheterodyne sensing can achieve a ~ 100 MHz-level bandwidth with high sensitivity, substantially surpassing existing heterodyne counterparts and paving the way for high-resolution quantum wireless sensing.