In today's digitally connected world, keyboards remain the primary interface for inputting sensitive information, making them a persistent target for eavesdropping attacks. While prior keystroke inference techniques have exploited side-channel signals such as acoustics and vibrations, they typically rely on conspicuous, short-range sensors and require victim-specific data for model training, limiting their practicality, scalability, and stealth. In this paper, we present RadKey, an RF backscatter system for covert, long-range, through-wall keystroke eavesdropping. RadKey comprises two components: a compact batteryless backscatter tag and an RF reader. The tag captures keystroke-induced vibrations and acoustic signals, modulating them onto the frequency shift of its backscattered RF signal using two magnetically-coupled LC resonators. This design also enables spectral separation between the excitation and backscatter signals, mitigating self-interference for the RF reader and thus extending eavesdropping range. The RF reader demodulates the backscattered RF signal to infer typed content. It employs a dedicated signal processing pipeline that extracts user- and keyboard-independent keystroke features across time and frequency domains, enabling strong generalizability. To further enhance adaptability, RadKey integrates an LLM for online adaptation, leveraging LLM outputs as pseudo ground-truth labels to refine the classifier during runtime. We have built a prototype of the full RadKey system and evaluated it through extensive over-the-air experiments. Results show that RadKey achieves accurate and robust keystroke inference across diverse users in real-world settings. A demo video is available at: https://radkey-submission.github.io/RadKey/

The Open Radio Access Network (O-RAN) represents a significant advancement in cellular networks, promoting openness, intelligence, and flexibility in 5G deployment. However, designing a scheduler for 5G O-RAN presents significant challenges due to its unique network architecture, large scheduling space, and stringent timing requirements of various control loops. Existing model-based and model-free schedulers both have inherent drawbacks that hinder their performance and adoption in 5G O-RAN. Model-based schedulers struggle because accurately modeling wireless system is often impossible, and they usually suffer from high complexity due to the NP-hard problem structure and large scheduling space. On the other hand, model-free schedulers often face convergence issue under a large scheduling space, and they usually cannot provide performance guarantees or even satisfy constraints. In this paper, we present MOGUL—a MOdel-GUided Learning approach that retains the strengths of both model-based and model-free methods while avoiding their pitfalls. MOGUL employs a model-based optimization problem to derive a reduced yet promising scheduling space, which is then used as the action space for model-free online Deep Multi-agent Reinforcement Learning (DMARL) to determine the final scheduling decision. Moreover, MOGUL is specifically tailored to the O-RAN architecture, allowing seamless integration into various control loops while meeting their stringent timing requirements. Experimental results demonstrate that MOGUL outperforms both state-of-the-art model-based and model-free algorithms.

Electric vehicles (EVs) have become one of the promising solutions to the ever-evolving environmental and energy crisis. The key to the wide adoption of EVs is a pervasive charging infrastructure, composed of both the private/home chargers and the public/commercial charging stations. The security of EV charging, however, has not been thoroughly investigated. This paper investigates the communication mechanisms between the chargers and EVs, and exposes the lack of protection on the authenticity in the SAE J1772 charging control protocol. To showcase our discoveries, we propose a new class of attacks, ChargeX, which aims to manipulate the charging states or charging rates of EV chargers with the goal of disrupting the charging schedules, causing denial of service (DoS), or degrading the battery performance. ChargeX inserts a hardware attack circuit to strategically modify the charging control signals. We design and implement multiple attack systems, and evaluate the attacks on a public charging station and two home chargers using a simulated vehicle load in the lab environment. Extensive experiments on different types of chargers demonstrate the effectiveness and generalization of ChargeX. Specifically, we demonstrate that ChargeX can force a Tesla’s charging state to switch from “stand by” to “charging”, potentially leading to overcharging. Additionally, ChargeX can transition any charging state to an error state, effectively launching a DoS attack on Tesla. If deployed, ChargeX may significantly demolish people’s trust in the EV charging infrastructure.

In this paper, GeoFL develops a hierarchical federated learning (FL) framework to address the unique challenges in large-scale geo-distributed scenarios. The key idea is to deploy multiple aggregators to geo-distributed clients and aggregate the local model and the global model efficiently and effectively. By assigning each aggregator as a relay layer, GeoFL can elaborately aggregate the geo-distributed clients and systematically determine when to upload the model to the central server based on bandwidth to efficiently update the global model under inadequate and heterogeneous WAN bandwidth constraints. GeoFL designs three key components to optimize the inefficient model aggregation and cope with the non-importance model updates. It further addresses the statistical heterogeneity across geo-distributed aggregators by considering the clients’ graph relationship, delivering an end-to-end clien-taggregator- server architecture for large-scale clients. Compared with existing works, our results on large-scale real-life datasets show that GeoFL speeds up the training process by 1.4×–8× and reduces 6%–80% unnecessary communication rounds between the aggregator and the central server.

Open-Radio Access Network (O-RAN) has become an important paradigm for 5G and beyond radio access networks. This paper presents an xApp calledxSlicefor the Near-Real-Time (Near-RT) RAN Intelligent Controller (RIC) of 5G O-RANs.xSliceis an online learning algorithm that adaptively adjusts MAC-layer resource allocation in response to dynamic network states, including time-varying wireless channel conditions, user mobility, traffic fluctuations, and changes in user demand. To address these network dynamics, we first formulate the Quality-of-Service (QoS) optimization problem as a regret minimization problem by quantifying the QoS demands of all traffic sessions through weighting their throughput, latency, and reliability. We then develop a deep reinforcement learning (DRL) framework that utilizes an actor-critic model to combine the advantages of both value-based and policy-based updating methods. A graph convolutional network (GCN) is incorporated as a component of the DRL framework for graph embedding of RAN data, enablingxSliceto handle a dynamic number of traffic sessions. We have implementedxSliceon an O-RAN testbed with 10 smartphones and conducted extensive experiments to evaluate its performance in realistic scenarios. Experimental results show thatxSlicecan reduce performance regret by 67% compared to the state-of-the-art solutions. Source code is available athttps://github.com/xslice-5G/code

Low Probability of Intercept (LPI) wireless communication is a critical aspect of modern wireless technology. Despite various techniques developed for LPI wireless communication, most of them require some form of pre-existing knowledge (e.g., encryption keys) or specific information (e.g., eavesdropper location and channel details). In this paper, we propose a novel physical-layer precoding technique called spatial pilot perturbation (SPP) to achieve efficient LPI wireless communications. Unlike existing methods, SPP operates without the need for pre-shared information between the two communication devices, nor any knowledge about potential eavesdroppers. It remains transparent to users and thus backward-compatible with off-the-shelf 5G/WiFi user devices. The core idea of SPP is to use different precoders for pilot and data symbols in a signal frame at the physical layer. Through a systematic precoder design, the pilot and data symbols will experience identical compound channels upon arrival at intended receivers, but experience different compound channels when intercepted by eavesdroppers. Consequently, the intended receivers can demodulate the signal frame, while eavesdroppers cannot. We have implemented SPP on 5G and WiFi testbeds and evaluated its performance through over-the-air experiments. Extensive experimental results show that SPP achieves an eavesdropping rate of ≤0.2% for 5G and ≤0.9% for WiFi, both at the cost of less than 18% throughput.

Eye motion tracking plays a vital role in many applications such as human-computer interaction (HCI), virtual reality, and disease detection.Camera-based eye tracking, albeit accurate and easy to use, may raise privacy concerns and appear to be unreliable in poor lighting conditions.In this paper, we present RadEye, a radar system capable of detecting fine-grained human eye motions from a distance.RadEye is realized through an integrated hardware and software design.It customizes a sub-6GHz FMCW radar so as to detect millimeter-level eye movement while extending its detection range using low frequency.It further employs a deep neural network (DNN) to refine the detection accuracy through camera-guided supervisory training.We have built a prototype of RadEye.Extensive experimental results show that it achieves 90% accuracy when detecting human eye rotation directions (up, down, left, and right) in various scenarios.

Sub-6GHz radio sensing offers several compelling advantages, such as resilience to poor lighting conditions, privacy preservation, and the ability to see through walls. However, in indoor environments, the sub-6GHz ISM spectrum is heavily occupied by WiFi devices, leaving little available spectrum for sensing purposes. In this paper, we introduce SiWiS, a new approach to integrate radio sensing capabilities into individual WiFi devices for fine-grained human activity detection. SiWiS comprises two main components: (i) a new hardware component that can be easily installed on an off-the-shelf WiFi device, and (ii) a dual-branch deep neural network (DNN) optimized for concurrent human mask segmentation and pose estimation. We have built a prototype of SiWiS and installed it on a commercial WiFi router for evaluation. Extensive experimental results demonstrate a significant performance improvement over WiFi channel state information (CSI) based sensing methods. More importantly, zero-shot experiments confirm that SiWiS can be directly transferred to unseen real-world environments.

Beam selection is a fundamental problem in millimeter-wave (mmWave) communication systems. Yet, most existing beam selection techniques focus on the exploitation of spatial channel features to reduce their airtime overhead in stationary mmWave networks. In this article, we exploit the temporal correlation of wireless channels to facilitate beam selection in mobile mmWave networks. Specifically, we present a temporal beam prediction (TBP) scheme for a mobile mmWave device to predict its future beam direction based on its history beam selection profile. TBP has two challenges in its design: 1) nonuniform history data samples due to the bursty nature of data traffic and 2) nonsmooth beam angles over time due to the multipath effect of channels and the imperfect radiation pattern of phased-array antennas. TBP addresses these two challenges by employing a new mobility-aware LSTM model that takes data timestamp for its training, together with an adversarial learning model to exploit user-independent features for beam steering. We have evaluated TBP through over-the-air (OTA) experiments on a 60-GHz mmWave testbed. Experimental results show that the average prediction error of TBP is less than 7° and that TBP improves the throughput by 60% in representative mmWave networks.

The use of mobile phones while driving is a major source of distraction for vehicle drivers and has resulted in a large number of car accidents. While surveillance cameras can be used to detect the violation of phone use, they do not work well in some scenarios (e.g., darkness and blockage) and may raise privacy concerns. In this paper, we present PhoLoc, a roadside device to detect the violation of phone use in personal vehicles using the cellular signals emitted by cellphones. PhoLoc is equipped with two sensors: a multi-antenna radio receiver and a low-cost lidar. It jointly processes the multimodal data from the two sensors to estimate the relative location of a phone in a vehicle. The enabler of PhoLoc is a new near-field localization scheme, which is capable of estimating the location of a moving phone at a specific time moment by overhearing its cellular signals. We have built a prototype of PhoLoc and evaluated its performance in realistic scenarios. Experimental results show that PhoLoc achieves 4.2% false positive rate and 13.8% false negative rate in the detection of phone call violation.

We study the data packet transmission problem (mmDPT) in dense cell-free millimeter wave (mmWave) networks, i.e., users sending data packet requests to access points (APs) via uplinks and APs transmitting requested data packets to users via downlinks. Our objective is to minimize the average delay in the system due to APs’ limited service capacity and unreliable wireless channels between APs and users. This problem can be formulated as a restless multi-armed bandits problem with fairness constraint (RMAB-F). Since finding the optimal policy forRMAB-Fis intractable, existing learning algorithms are computationally expensive and not suitable for practical dynamic dense mmWave networks. In this paper, we propose a structured reinforcement learning (RL) solution formmDPTby exploiting the inherent structure encoded inRMAB-F. To achieve this, we first design a low-complexity and provably asymptotically optimal index policy forRMAB-F. Then, we leverage this structure information to develop a structured RL algorithm calledmmDPT-TS, which provably achieves an$\tilde {\mathcal {O}}(\sqrt {T})$Bayesian regret. More importantly,mmDPT-TSis computation-efficient and thus amenable to practical implementation, as it fully exploits the structure of index policy for making decisions. Extensive emulation based on data collected in realistic mmWave networks demonstrate significant gains ofmmDPT-TSover existing approaches.

Facial expression recognition has enormous potential for downstream applications by revealing users’ emotional status when interacting with digital content. Previous studies consider using cameras or wearable sensors for expression recognition. However, these approaches bring considerable privacy concerns or extra device burdens. Moreover, the recognition performance of camera-based methods deteriorates when users are wearing masks. In this paper, we propose FacER, an active acoustic facial expression recognition system. As a software solution on a smartphone, FacER avoids the extra costs of external microphone arrays. Facial expression features are extracted by modeling the echoes of emitted near-ultrasound signals between the earpiece speaker and the 3D facial contour. Besides isolating a range of background noises, FacER is designed to identify different expressions from various users with a limited set of training data. To achieve this, we propose a contrastive external attention-based model to learn consistent expression features across different users. Extensive experiments with 20 volunteers with or without masks show that FacER can recognize 6 common facial expressions with more than 85% accuracy, outperforming the state-of-the-art acoustic sensing approach by 10% in various real-life scenarios. FacER provides a more robust solution for recognizing facial expressions in a convenient and usable manner.

UHF RFID tags have been widely used for contactless inventory and tracking applications. One fundamental problem with RFID readers is their limited tag reading rate. Existing RFID readers (e.g., Impinj Speedway) can read about 35 tags per second in a read zone, which is far from enough for many applications. In this paper, we present the first-of-its-kind RFID reader (mReader), which borrows the idea of multi-user MIMO (MU-MIMO) from cellular networks to enable concurrent multi-tag reading in passive RFID systems. mReader is equipped with multiple antennas for implicit beamforming in downlink transmissions. It is enabled by three key techniques: uplink collision recovery, transition-based channel estimation, and zero-overhead channel calibration. In addition, mReader employs a Q-value adaptation algorithm for medium access control to maximize its tag reading rate. We have built a prototype of mReader on USRP X310 and demonstrated for the first time that a two-antenna reader can read two commercial off-the-shelf (COTS) tags simultaneously. Numerical results further show that mReader can improve the tag reading rate by 45% compared to existing RFID readers.

Leaf wetness detection is one of the key technologies for preventing plant diseases in agriculture. In this poster, we propose mmLeaf, leveraging a commercial off-the-shelf millimeter-wave (mmWave) radar to detect actual leaf wetness in diverse environments and lighting conditions. mmLeaf captures mmWave signals reflected by monitored leaves with a two-dimensional (2D) scanning system. Then, we use a multiple-input multiple-output (MIMO) array and synthetic aperture radar (SAR) to reconstruct the signal distribution of different planes of the leaves. A deep learning model takes the fused signal distribution as inputs to classify the leaf wetness. We implement mmLeaf using a frequency-modulated continuous-wave (FMCW) radar and evaluate its performance with a potted plant indoors. By exploring the use of mmWave signals, mmLeaf delivers an end-to-end detection framework that achieves up to 90% accuracy in classifying leaf wetness under different distances.

Rate adaptation is a key mechanism in current IEEE 802.11 networks and next-generation cellular systems. Observing that the operating time scale of rate adaptation is usually much smaller than the user association and scheduling, we study a joint design of wireless user association and scheduling and rate adaptation with different time scales to maximize cumulative system throughput while guaranteeing desired fairness among users. We develop a maximum-weight type user association and scheduling algorithm that combines the virtual queues (tracking the scheduling debt for each user to ensure the desired fairness guarantee) and Upper Confidence Bound (UCB) estimates in its weight measure; each selected user then adopts the UCB algorithm to perform rate adaptation in a smaller time scale. We show that our proposed algorithm yields a cumulative regret growing with the square root of the time horizon up to a logarithmic factor, and achieves zero cumulative fairness violation after a certain number of time frames. We demonstrate the efficiency of the proposed algorithm via simulations using synthetic and realistic data traces.

Degree-of-freedom (DoF)-based models have been proven to be highly successful in modeling and analysis of MIMO systems. Among existing DoF-based models, the number of DoFs used for interference cancellation (IC) is solely based on the number of interfering data streams. However, from both experimental and simulation results, we find that signal strengths of an interference link vary significantly in different directions in the eigenspace. In this paper, we exploit the difference in interference signal strengths in the eigenspace and perform IC with DoFs only on those directions with strong signals. To differentiate interference signal strengths on an interference link, we introduce a novel concept called “effective rank threshold.” Based on this threshold, DoFs are consumed only to cancel strong interferences in the eigenspace while weak interferences are treated as noise in throughput calculation. To better understand the benefits of this approach, we study a fundamental trade-off between network throughput and effective rank threshold for an MU-MIMO network. Our simulation results show that network throughput under optimal rank threshold is significantly higher than that under existing DoF IC models. To ensure the new DoF IC model is feasible at PHY layer, we propose an algorithm to set the weights for all nodes that can offer our desired DoF allocation.

Federated Learning (FL) is a promising technique to enhance the safety and efficiency of intelligent transportation systems. While FL has been extensively studied, the communication and networking challenges related to the operations of FL in dynamic yet dense vehicular networks remain under-explored. Limited storage and communication capacities of individual vehicles throttle the timely training of an FL model in distributed vehicular networks. In this paper, we present a communication framework for FL (CF4FL) in transportation systems. CF4FL aims to accelerate the convergence of FL training process through the innovation of two complementary networking components: (i) a deadline-driven vehicle scheduler (DDVS), and (ii) a concurrent vehicle polling scheme (CVPS). DDVS identifies a subset of vehicles for local model training in each iteration of FL, with the aim of minimizing data loss while respecting the deadline constraints derived from vehicles’ storage, computation, and energy budgets. CVPS takes advantage of multiple antennas on an edge server to enable concurrent local model transmissions in dynamic vehicular networks, thereby reducing the airtime overhead of each FL iteration. We have evaluated CF4FL through a blend of experimentation and simulation. Trace-driven simulation shows that, compared to existing scheduling and transmission schemes, CF4FL reduces the convergence time of FL training by 39%.

LoRa has emerged as a key wireless communication technology for a gateway to provide geographically-distributed IoT devices with low-rate, long-range connections. In this paper, we present MaLoRaGW, the first-of-its-kind Multi-antenna LoRa GateWay that enables multi-user MIMO (MU-MIMO) LoRa communications in both uplink and downlink. MaLoRaGW was inspired by the success of MU-MIMO in cellular and Wi-Fi networks. The key component of MaLoRaGW is a joint baseband PHY design for uplink packet detection and downlink beamforming. Its innovation lies in three modules: spatial signal projection, accurate channel estimation, and implicit beamforming, all of which reside only in a LoRa gateway and require no modification on LoRa client devices. We have built a prototype of two-antenna MaLoRaGW on a USRP device and extensively evaluated its performance with commercial LoRa dongles in three scenarios: lab, office building, and university campus. Our experimental results show that, compared to the state-of-the-art, the two-antenna MaLoRaGW increases uplink throughput by 10% and downlink throughput by 95%.

Smart home Internet of Things (IoT) has a vibrant market with a wide range of appliances and sensors, spanning across smart home, smart city, and smart factory. However, the security and privacy of these IoT systems have raised serious concerns. Currently, most IoT devices rely on electromagnetic wave-based radio frequency (RF) for communication. Yet, RF has several inherent limitations, such as shortage of spectrum, susceptible to interference, and vulnerable to eavesdropping or jamming attacks. This article presents URadio, a wideband ultrasonic communication system. By leveraging recent advances in reduced Graphene Oxide (rGO), we design a new type of electrostatic ultrasonic transducer, which can achieve more than$6\times $bandwidth than commercial ultrasonic transducers. With this new transducer, we design an OFDM communication system to maximize its data rate for smart home applications. We build a prototype of URadio on a wireless testbed and evaluate its performance in several real-world environments. Our experiments show that URadio can reach up to 360 kb/s data rate at a distance of 81 cm or 20 Kb/s data rate at a distance of 20 m, which supports a variety of smart home applications. We further showcase URadio’s resilience against eavesdropping and jamming attacks, as well as demonstrate its capability of securely localizing objects in an indoor environment.

Wireless Internet of Things (IoT) applications have penetrated every aspect of our society and become increasingly important in smart homes, smart cities, and smart hospitals. However, many WiFi-based IoT devices (e.g., light switches, door/window open alert sensors, and Google Home) do not have input interfaces such as keypad or touchscreen due to their limits in physical size, power consumption, and/or manufacturing cost, making it inconvenient and onerous for end users to authenticate those IoT devices for wireless Internet access. In this article, we present AuthIoT, a learning-based authentication scheme for wireless IoT devices without input interfaces. The key component of AuthIoT is a channel state information (CSI)-based character classification algorithm for a WiFi access point (AP), which recognizes the passcode from an IoT device when an end user holds it in hand and writes the passcode over the air. AuthIoT has two salient features: 1) it is transferable for cross-environment applications and 2) it works in more realistic scenarios where AP is equipped with nonlinear antenna array. We have built a prototype of AuthIoT and evaluated its performance on two testbeds: 1) Intel 5300 WiFi card with three linear antennas and 2) USRP N310 with four nonlinear (square-shaped) antennas. The experimental results show that AuthIoT achieves 84% and 83% recognition accuracy on the two testbeds.

Recently, networks operate at frequencies over 28 GHz (mmWave) have emerged as a viable solution for 5G mobile networks to provide Gbps data rate. Due to the high directivity and attenuation of mmWave signals, mmWave communication links are highly vulnerable to the frequent mmWave channel blockages, which can trigger excessive handovers. Thanks to its ability to enrich the scattering environment and create reflective signal multipaths, Reconfigurable Intelligent Surface (RIS) has great potential to counter the blockage effect and thus greatly reduce the number of unnecessary handovers. However, this potential has not been well explored. In this paper, we propose a RIS-assisted handover scheme by leveraging deep reinforcement learning (DRL). Under various channel blockage conditions, the DRL agent manages to reduce the cumulative handover overhead by jointly adjusting beamformers and RIS phase shifts. Compared with the existing schemes without considering RIS, the RIS-assisted handover scheme significantly reduces the number of handovers and achieves higher spectrum efficiency. Besides, to alleviate the impact from the limited observations of the fast fading channels, we propose a lightweight algorithm to sense the blockage status and such sensing results can be utilized to improve the performance of model training. Numerical results show that DRL agent is able to further improve the performance when integrated with the blockage status sensing algorithm.

Autonomous vehicles (AVs), equipped with numerous sensors such as camera, LiDAR, radar, and ultrasonic sensor, are revolutionizing the transportation industry. These sensors are expected to sense reliable information from a physical environment, facilitating the critical decision-making process of the AVs. Ultrasonic sensors, which detect obstacles in a short distance, play an important role in assisted parking and blind spot detection events. However, due to their weak security level, ultrasonic sensors are particularly vulnerable to signal injection attacks, when the attackers inject malicious acoustic signals to create fake obstacles and intentionally mislead the vehicles to make wrong decisions with disastrous aftermath. In this paper, we systematically analyze the attack model of signal injection attacks toward moving vehicles. By considering the potential threats, we propose SoundFence, a physical-layer defense system which leverages the sensors' signal processing capability without requiring any additional equipment. SoundFence verifies the benign measurement results and detects signal injection attacks by analyzing sensor readings and the physical-layer signatures of ultrasonic signals. Our experiment with commercial sensors shows that SoundFence detects most (more than 95%) of the abnormal sensor readings with very few false alarms, and it can also accurately distinguish the real echo from injected signals to identify injection attacks.

Wireless local area networks (WLANs) are a key component of the telecommunications infrastructure in our society. While many solutions have been produced to improve their downlink throughput, the techniques for enhancing their uplink throughput remain limited. The stagnation can be attributed to the lack of fine-grained inter-node synchronization due to the hardware limitation of most devices. In this paper, we present an uplink distributed multiple-input-and-multiple-output scheme (termed UD-MIMO) for WLANs to enable concurrent uplink transmission in the absence of fine-grained inter-node synchronization. The enabling technique behind UD-MIMO is a practical solution to decoding uplink packets from asynchronous users. UD-MIMO makes it possible for WLANs to significantly improve their uplink throughput while not requiring tight internode synchronization. We have built a prototype of UD-MIMO on a wireless testbed and demonstrate its compatibility with commercial off-the-shelf Atheros 802.11 client devices (with modified Linux driver). Our experimental results show that, for a WLAN with 8 APs in a conference room, UD-MIMO offers 3.4× throughput compared to interference-avoidance approach.

Current data-driven intelligent transportation systems are mainly reliant on IEEE 802.11p to collect and exchange information. Despite promising performance of IEEE 802.11p in providing low-latency communications, it is still vulnerable to jamming attacks due to the lack of a PHY-layer countermeasure technique in practice. In this paper, we propose JammingBird, a novel receiver design that tolerates strong constant jamming attacks. The enablers of JammingBird are two MIMO-based techniques: Jamming-resistant synchronizer and jamming suppressor. Collectively, these two new modules are able to detect, synchronize, and recover desired signals under jamming attacks, regardless of the PHY-layer technology employed by the jammers. We have implemented JammingBird on a vehicular testbed and conducted extensive experiments to evaluate its performance in three common vehicular scenarios: Parking lots (0~15 mph), local traffic areas (25~45 mph), and highways (60~70 mph). In our experiments, while the jamming attacks degrade the throughput of conventional 802.11p-based receivers by 86.7%, JammingBird maintains 83.0% of the throughput on average. Experimental results also show that JammingBird tolerates the jamming signals with 25 dB stronger power than the desired signals.

ZigBee is a wireless communication technology that has been widely used to provide low-bandwidth wireless services for Internet-of-Things applications, such as building automation, medical data collection, and industrial equipment control. As ZigBee operates in the industrial, scientific and medical radio frequency bands, it may suffer from unintentional interference from coexisting radio devices (e.g., WiFi and Bluetooth) and/or radio jamming attacks from malicious devices. Although many results have been produced to enhance ZigBee security, there is no technique that can secure ZigBee against jamming attack. In this article, we propose a new ZigBee receiver by leveraging MIMO technology, which is capable of decoding its desired signal in the presence of constant jamming attack. The enabler is a learning-based jamming mitigation method, which can mitigate the unknown interference using an optimized neural network. We have built a prototype of our proposed ZigBee receiver on a wireless testbed. Experimental results show that it is capable of decoding its packets in the face of 20-dB stronger jamming. The proposed ZigBee receiver offers an average of 26.7-dB jamming mitigation capability compared to off-the-shelf ZigBee receivers.

In cellular networks, multiuser multiple-input multiple-output (MU-MIMO) is a key technology and has already been deployed in many real systems. Recently, device-to-device (D2D) communication has emerged as another promising technology as it offers several advantages, such as traffic offloading, low-latency transmissions, and enhanced spectral efficiency. Although there are many results of these two technologies, most of them are limited to their respective domains and there is a lack of practical design to combine both technologies for cellular networks. In this article, we present DM-COM, a practical scheme for enabling the coexistence of D2D and MU-MIMO subsystems in cellular networks. The enabler of DM-COM is a new approach for managing the mutual interference between the two subsystems, which does not require channel state information and is, therefore, amenable to practical implementation. We have built a prototype of DM-COM on a wireless testbed and evaluated its performance in a real-world wireless environment. Our experimental results show that, using DM-COM in a small cellular network, D2D users achieve 1.9 bit/s/Hz spectral efficiency, while MU-MIMO users have less than 8% throughput degradation compared to the case without D2D users.

MU-MIMO and OFDMA are two key techniques in IEEE 802.11ax standard. Although these two techniques have been intensively studied in cellular networks, their joint optimization in Wi-Fi networks has been rarely explored as OFDMA was introduced to Wi-Fi networks for the first time in 802.11ax. The marriage of these two techniques in Wi-Fi networks creates both opportunities and challenges in the practical design of MAC-layer protocols and algorithms to optimize airtime overhead, spectral efficiency, and computational complexity. In this paper, we present DeepMux, a deep-learning-based MU-MIMO-OFDMA transmission scheme for 802.11ax networks. DeepMux mainly comprises two components: deep-learning-based channel sounding (DLCS) and deep-learning-based resource allocation (DLRA), both of which reside in access points (APs) and impose no computational/communication burden on Wi-Fi clients. DLCS reduces the airtime overhead of 802.11 protocols by leveraging the deep neural networks (DNNs). It uses uplink channels to train the DNNs for downlink channels, making the training process easy to implement. DLRA employs a DNN to solve the mixed-integer resource allocation problem, enabling an AP to obtain a near-optimal solution in polynomial time. We have built a wireless testbed to examine the performance of DeepMux in real-world environments. Our experimental results show that DeepMux reduces the sounding overhead by 62.0% ~ 90.5% and increases the network throughput by 26.3% ~ 43.6%.

Timely vehicle-to-vehicle (V2V) communication is a key component of intelligent transportation systems to improve driving safety and efficiency. Although many results have been produced for vehicular networks, most of them focused on improving vehicular communication capacity and reliability. Very limited progress has been made so far in the design of practical V2V communication schemes for large-scale vehicular networks. In this paper, we present VehCom, a fully distributed message broadcast scheme for V2V communication networks. VehCom offers a delay guarantee for each vehicle's message broadcast while minimizing the packet loss rate. The enabler of VehCom is an asynchronous packet reception technique, which leverages a vehicle's multiple antennas to decode asynchronous collided packets from its neighboring vehicles. We have implemented the asynchronous packet reception technique on a vehicular wireless testbed, and examined the performance of VehCom in a large-scale vehicular network where i) each vehicle is equipped with four antennas, ii) each vehicle has 240 vehicles in its communication range, and iii) each vehicle broadcasts a 624-bit packet over 10 MHz spectrum in every 100 ms (guaranteed delay). Our experimental and analytical results show that the packet loss rate is less than 3.9% on parking lots, less than 4.1% on local roads, and less than 6.2% on highways.

Multi-user MIMO (MU-MIMO) is a key technology for current and next-generation wireless local area networks (WLANs). While it has widely been deployed in WLANs, its potential is not fully exploited in real-world systems. This can be attributed to the large airtime overhead induced by channel acquisition in existing MU-MIMO protocols, which significantly compromises the throughput gain of MU-MIMO. In this paper, we present LB-SciFi, a learning-based channel feedback framework for MU-MIMO in WLANs. LB-SciFi takes advantage of recent advances in deep neural network autoencoder (DNN-AE) to compress channel state information (CSI) in 802.11 protocols, thereby conserving airtime and improving spectral efficiency. The key component of LB-SciFi is an online DNN-AE training scheme, which allows an AP to train DNN-AEs by leveraging the side information of existing 802.11 protocols. With this training scheme, DNN-AEs are capable of significantly lowering the airtime overhead for MU-MIMO while preserving its backward compatibility with incumbent Wi-Fi client devices. We have implemented LB-SciFi on a wireless testbed and evaluated its performance in indoor wireless environments. Experimental results show that LB-SciFi offers an average of 73% airtime overhead reduction and increases network throughput by 69% on average when compared to 802.11 feedback protocols.

Co-channel interference is a fundamental issue in wireless local area networks (WLANs). Although many results have been developed to handle co-channel interference for concurrent transmission, most of them require network-wide fine-grained synchronization and data sharing among access points (APs). Such luxuries, however, are not affordable in many WLANs due to their hardware limitation and data privacy concern. In this paper, we present TCCI, a co-channel interference management scheme to enable concurrent transmission in WLANs. TCCI requires neither network-wide fine-grained synchronization nor inter-network data sharing, and therefore is amenable to real-world implementation. The enabler of TCCI is a new detection and beamforming method for an AP, which is capable of taming unknown interference by leveraging its multiple antennas. We have built a prototype of TCCI on a wireless testbed and demonstrated its compatibility with commercial Atheros 802.11 devices. Our experimental results show that TCCI allows co-located APs to serve their users simultaneously and achieves up to 113% throughput gain compared to existing interference-avoidance protocol.

As most Internet-of-Things (IoT) devices are powered by small-sized batteries and expected to operate for many years without battery replacement, energy-efficient wireless IoT communication has been considered as a crucial component of the future network infrastructure. In this article, we propose a practical design (termed WiFi-IoT) to add energy-efficient IoT communication capability into WLANs. WiFi-IoT features two innovative techniques: 1) an asymmetric physical (PHY) design and 2) a transparent coexistence scheme. The asymmetric PHY allows an access point (AP) to communicate with multiple IoT devices at a much low sampling rate (250 ksps), thereby significantly reducing the power consumption for the IoT devices. The transparent coexistence scheme enables a multiantenna AP to serve Wi-Fi and IoT devices simultaneously, leading to an efficient utilization of spectrum. We have built a prototype of WiFi-IoT on a USRP2 wireless testbed and evaluated its performance in real-world wireless environments. The experimental results show that a two-antenna AP can simultaneously serve one broadband Wi-Fi device and 24 narrowband IoT devices on the same spectrum.

Non-orthogonal multiple access (NOMA) has emerged as a new multiple access paradigm for wireless networks. Although many results have been produced for NOMA, most of them are limited to theoretical exploration and performance analysis in cellular networks. Very limited progress has been made so far in the design of practical NOMA schemes for wireless local area networks (WLANs). In this paper, we propose a practical downlink NOMA scheme for WLANs and evaluate its performance in real-world wireless environments. Our NOMA scheme has three key components: precoder design, user grouping, and successive interference cancellation (SIC). On the transmitter side, we first formulate the precoding design problem as an optimization problem and then devise an efficient algorithm to construct precoders for downlink NOMA transmissions. We further propose a lightweight user grouping algorithm to ensure the success of SIC at the receivers. On the receiver side, we propose a new SIC method to decode the desired signal in the presence of strong interference. In contrast to existing SIC methods, our SIC method does not require channel estimation to decode the signals, thereby improving its resilience to interference. We have built a prototype of the proposed NOMA scheme on a wireless testbed. Experimental results show that, compared to orthogonal multiple access (OMA), the proposed NOMA scheme can significantly improve the weak user's date rate (93% on average) and considerably improve WLAN's weighted sum rate (36% on average).

Spectrum shortage is a fundamental problem in wireless networks, and this problem becomes increasingly acute with the rapid proliferation of wireless devices. To address this issue, spectrum sharing in the context of cognitive radio networks (CRNs) has been regarded as a promising solution. Although there is a large body of work on spectrum sharing in the literature, most existing work is limited to theoretical exploration and the progress in practical solution design remains scarce. In this paper, we propose a practical scheme to enable transparent spectrum sharing for a small CRN by leveraging recent advances in multiple-input multiple-output (MIMO) technology. The key components of our scheme are two MIMO-based interference management techniques: blind beamforming (BBF) and blind interference cancellation (BIC). These two techniques enable secondary users to mitigate cross-network interference in the absence of inter-network coordination, fine-grained synchronization, and mutual knowledge. We have built a prototype of our scheme on a wireless testbed and demonstrated its compatibility with commercial Wi-Fi devices (primary users). Experimental results show that, for a secondary device with two/three antennas, BBF and BIC achieve an average of 25 dB and 33 dB interference cancellation capabilities in real-world wireless environments, respectively.

While Narrow-Band Internet of Things (NB-IoT) has been standardized by 3GPP to provide wireless Internet access for IoT devices, this service is expected to come with a monthly fee (e.g., $1 or $2 per month per device). As the number of IoT devices tends to be large, the service charge will impose a considerable financial burden on the end users. In this paper, we propose an Energy-Efficient IoT (EE-IoT) communication scheme by taking advantage of the existing WiFi infrastructure that is widely available in home, office, campus, and city environments. EE-IoT will not only avoid monthly service charge for the end users but also maintain a low power consumption for IoT devices. The key component of EE-IoT is an asymmetric physical (PHY) design, which enables an OFDM-based broadband AP to communicate with multiple QAM-based narrowband IoT devices at a low sampling rate (250 ksps) in both uplink and downlink. The trick in our design is that, instead of using the same carrier frequency as the AP, each IoT device tunes its carrier frequency to a particular subcarrier of the AP's OFDM signal, making it possible to encode/decode the data on that subcarrier at a low sampling rate. Based on this new PHY, we propose a MAC protocol to enable EE-IoT in WLANs. We have built a prototype of EE-IoT on a USRP2 wireless testbed and evaluated its performance in an office building environment. Experimental results show that an AP can serve 24 IoT devices simultaneously and each IoT device can achieve more than 187 kbps in the downlink and more than 125 kbps in the uplink.

As the proliferation of mobile devices has led to an ever-growing demand for wireless Internet services, the spectrum shortage issue becomes increasingly severe and spectrum sharing is regarded as a promising approach to addressing the spectrum shortage issue. In this paper, we propose a practical underlay spectrum sharing scheme for cognitive radio networks (CRNs) where the primary users are oblivious to the secondary users. The key components of our scheme are two MIMO-based interference cancellation (IC) techniques to handle cross-network interference on the secondary network side. The first one is a blind beamforming technique for secondary transmitters. This IC technique allows a secondary transmitter to nullify its generated interference for primary users without requiring channel state information (CSI). The second one is a blind interference cancellation (BIC) technique for secondary receivers. This IC technique enables a secondary receiver to decode its desired signal in the presence of strong unknown interference from primary transmitters. Based on these two MIMO-based IC techniques, we develop a MAC protocol for the secondary network to enable underlay spectrum sharing in CRNs. We have implemented the proposed underlay spectrum sharing scheme on a GNURadio-USRP2 wireless testbed. Experimental results show that the secondary users can achieve an average of 1 bit/s/Hz spectrum efficiency without degrading the performance of the primary users in a real-world office building environment.

In wireless LANs (WLANs), network-wide time and frequency synchronization among user devices is widely regarded as a necessity for uplink MU-MIMO. Therefore, to enable uplink MU-MIMO in 802.11ax, dedicated MAC protocols (e.g., trigger frame and timing advance mechanism) have been proposed to synchronize user devices in the time and frequency domains. Such MAC protocols increase not only network complexity but also communication overhead. In this paper, we show that the time and frequency synchronization among user devices is not a necessity for uplink MU-MIMO. We propose a practical uplink MU-MIMO solution which does not require time and frequency alignments of the signals from user devices. The key component in our solution is a new PHY design for AP's receiver, which can decode the concurrent signals from multiple asynchronous user devices. We have built a prototype of our uplink MU-MIMO solution on USPR2-GNURadio testbed. Experimental results show that, using the new PHY, an M-antenna AP can successfully decode the concurrent signals from M asynchronous user devices (2 ≤ M ≤ 4).

In this paper, we study the optimal transmitter beamforming design in a coordinated multicell multicast network under the wiretap channel model, where each cell serves a group of users and a single-antenna eavesdropper tries to overhear the confidential information. We study the problem of finding the optimal beamforming strategy at each base station to minimize the total transmit power of the multicell network while guaranteeing secrecy and QoS requirements of each user. The resulting non-convex optimization problem is solved through a sequence of convex semi-definite relaxations. The effectiveness of the proposed algorithm is validated via numerical simulations.

Interference neutralization (IN) is regarded as a promising interference management techniques for multi-hop wireless networks. Yet most existing results of IN are limited to two-hop networks such as the relay-aided cellular network. Little progress has been made so far in the exploration of IN in generic multi-hop (more than two hops) networks. This paper aims to bridge this gap by developing an optimization framework for IN in a generic multi-hop network with the objective of maximizing the end-to-end throughput of multiple coexisting communication sessions. We first derive a mathematical model for IN in a special one-hop network to characterize the capability of IN, and then generalize this model to a multi-hop network. Based on the IN model, we develop a cross-layer optimization framework for a multi-hop network with the objective of fully translating the benefits of IN to the end-to-end throughput of the multi-hop sessions. To evaluate the performance of IN in multi-hop networks, we compare its performance against the case where IN is not employed. Simulation results show that the use of IN can significantly (more than 50%) increase the session throughput and, more notably, the throughput gain of IN increases with the node density and traffic intensity in the network.

Underwater acoustic (UWA) networks are a key form of communications for human exploration and activities in the oceanographic space of the earth. A fundamental issue of UWA communications is large propagation delays due to water medium, which has posed a grand challenge in UWA network protocol design. Conventional wisdom of addressing this issue is to live with this disadvantage by inserting a guard interval to introduce immunity to propagation delays. Recent advances in interference alignment (IA) open up a new direction to address this issue and promise a great potential to improve network throughput by exploiting large propagation delays. In this paper, we investigate propagation delay-based IA (PD-IA) in multi-hop UWA networks. We first develop a set of simple constraints to characterize PD-IA feasible region at the physical layer. Based on the set of PD-IA constraints, we develop a distributed PD-IA scheduling algorithm to greedily maximize interference overlapping possibilities in a multi-hop UWA network. Simulation results show that the proposed PD-IA algorithm yields higher throughput than an idealized benchmark algorithm without propagation delays, indicating that large propagation delays are not adversarial but beneficial for network throughput performance.

Interference alignment (IA) is widely regarded as a promising interference management technique in wireless networks. Despite its rapid advances in cellular networks, most results of IA are limited to information-theoretic exploration or physical-layer signal design. Little progress has been made so far to advance IA in cellular networks from a networking perspective. In this paper, we aim to fill this gap by studying IA in large-scale cellular networks. For the uplink, we propose an OFDM-based IA scheme and prove its feasibility at the physical layer by showing that all data streams in the IA scheme can be transported free of interference. Based on the IA scheme, we develop a cross-layer IA optimization framework that can fully translate the benefits of IA to throughput gain in cellular networks. Furthermore, we show that the IA optimization problem in the downlink can be solved in the exactly same way as that in the uplink. Simulation results show that our OFDM-based IA scheme can significantly increase the user throughput and the throughput gain increases with user density in the network.

There have been some rapid advances on the design of full duplex (FD) transceivers in recent years. Although the benefits of FD have been studied for single-hop wireless communications, its potential on throughput performance in a multi-hop wireless network remains unclear. As for multi-hop networks, a fundamental problem is to compute the achievable end-to-end throughput for one or multiple communication sessions. The goal of this paper is to offer some fundamental understanding on end-to-end throughput performance limits of FD in a multi-hop wireless network. We show that through a rigorous mathematical formulation, we can cast the multi-hop throughput performance problem into a formal optimization problem. Through numerical results, we show that in many cases, the end-to-end session throughput in a FD network can exceed 2x of that in a half duplex (HD) network. Our finding can be explained by the much larger design space for scheduling that is offered by removing HD constraints in throughput maximization problem. The results in this paper offer some new understandings on the potential benefits of FD for end-to-end session throughput in a multi-hop wireless network.

Interference neutralization (IN) is an interference management technique that allows simultaneous transmission of multiple links by nullifying their mutual interference in the air via cooperation among the transmitters. Although IN has been studied from information theoretic perspective, its potential for a general multi-hop wireless network has not been explored. The goal of this paper is to understand IN in a multi-hop wireless network from networking perspective. We first establish an IN reference model. Based on this reference model, we develop a set of feasibility constraints for a subset of links to be active simultaneously. By identifying each eligible neutralization node (called neut), we study IN in a general multi-hop network and develop a set of necessary constraints to characterize neut selection, IN, and scheduling. These constraints allow us to study the performance of multi-hop networks without the need of getting involved into onerous signal design issues at the physical layer. Finally, we apply our IN model and constraints to study a throughput maximization problem and show that the use of IN can generally increase network throughput. In particular, throughput gain is most significant when the node density increases.

Jamming attack is a serious threat to the wireless communications. Reactive jamming maximizes the attack efficiency by jamming only when the targets are communicating, which can be readily implemented using software-defined radios. In this paper, we explore the use of the multi-input multi-output (MIMO) technology to achieve jamming resilient orthogonal frequency-division multiplexing (OFDM) communication. In particular, MIMO interference cancellation treats jamming signals as noise and strategically cancels them out, while transmit precoding adjusts the signal directions to optimize the decoding performance. We first investigate the reactive jamming strategies and their impacts on the MIMO-OFDM receivers. We then present a MIMO-based anti-jamming scheme that exploits MIMO interference cancellation and transmit precoding technologies to turn a jammed non-connectivity scenario into an operational network. We implement our jamming resilient communication scheme using software-defined radios. Our testbed evaluation shows the destructive power of reactive jamming attack, and also validates the efficacy and efficiency of our defense mechanisms in the presence of numerous types of reactive jammers with different jamming signal powers.

Interference alignment (IA) is a powerful technique to handle interference in wireless networks. Since its inception, IA has become a central research theme in the wireless communications community. Due to its intrinsic nature of being a physical layer technique, IA has been mainly studied for point-to-point or single-hop scenario. There is a lack of research of IA from a networking perspective in the context of multi-hop wireless networks. The goal of this paper is to make such an advance by bringing IA technique to multi-hop MIMO networks. We develop an IA model consisting of a set of constraints at a transmitter and a receiver that can be used to determine IA for a subset of interfering streams. We further prove the feasibility of this IA model by showing that a DoF vector can be supported free of interference at the physical layer as long as it satisfies the constraints in our IA model. Based on the proposed IA model, we develop an IA design space for a multi-hop MIMO network. To study how IA performs in a multi-hop MIMO network, we compare the performance of a network throughput optimization problem based on our developed IA design space against the same problem when IA is not employed. Simulation results show that the use of IA can significantly decrease the DoF consumption for IC, thereby improving network throughput.

Recently, a new MIMO degree-of-freedom (DoF) model was proposed to allocate DoF resources for spatial multiplexing (SM) and interference cancellation (IC) in a multi-hop network. Although this DoF model promises many benefits, it hinges upon a global node ordering to keep track of IC responsibilities among all the nodes. An open question about this model is whether its global ordering property can be achieved among the nodes in the network through distributed operations. In this paper, we explore this question by studying DoF scheduling in a multi-hop MIMO network, with the objective of maximizing the minimum throughput among a set of sessions. We propose an efficient DoF scheduling algorithm to solve it and show that our algorithm only requires local operations. We prove that the resulting DoF scheduling solution is globally feasible and show that there exists a corresponding feasible global node ordering for IC, albeit such global ordering is implicit. Simulation results show that the solution values obtained by our algorithm are relatively close to the upper bound values computed by CPLEX solver, thereby indicating that our algorithm is highly competitive.

Cyclostationary analysis is a powerful tool to study the Signal-Selective Direction Estimation (SSDE) problem as different types of wireless signals have different cyclostationary patterns. Generally speaking, each type of wireless signal has unique cyclic frequencies with the frequency-selective property, which distinguishes itself from other types of signals. The cyclostationary property of a signal may be induced by its modulation method, its carrier frequency, and/or its frame structure. In this paper, we study the cyclostationary property of IEEE 802.11 (WiFi) signals induced by their underlying OFDM frame structure, which includes pilots, cyclic prefix (CP), and preambles. We first analyze the pilot-induced, CP-induced, and preamble-induced cyclostationary properties of WiFi signals, respectively. We then derive their spectral correlation function (SCF) and investigate their applicability to solving the SSDE problem. Simulation results show that the pilot-induced cyclostationary property of WiFi signals is a promising feature that can be used to solve the SSDE problem.

Reactive jamming is considered the most powerful jamming attack as the attack efficiency is maximized while the risk of being detected is minimized. Currently, there are no effective anti-jamming solutions to secure OFDM wireless communications under reactive jamming attack. On the other hand, MIMO has emerged as a technology of great research interest in recent years mostly due to its capacity gain. In this paper, we explore the use of MIMO technology for jamming resilient OFDM communication, especially its capability to communicate against the powerful reactive jammer. We first investigate the jamming strategies and their impacts on the OFDM-MIMO receivers. We then present a MIMO-based anti-jamming scheme that exploits interference cancellation and transmit precoding capabilities of MIMO technology to turn a jammed non-connectivity scenario into an operational network. Our testbed evaluation shows the destructive power of reactive jamming attack, and also validates the efficacy and efficiency of our defense mechanisms.

Recent advances in information theory (IT) have shown great promises of interference alignment (IA) for cellular networks. However, due to a number of assumptions, these IT results cannot be directly applied to address practical problems. The goal of this paper is to fill in this gap by studying IA for cellular networks with more practical settings. We propose an IA scheme that includes constraints at each user and each base station (BS) for the uplink communication of a cellular network. We prove the feasibility of the IA scheme by constructing the encoding and decoding vectors for each data stream so that it can be transported free of interference. Based on this IA scheme, we study an uplink user throughput maximization problem and show the throughput improvement of the IA scheme over two other schemes.

Degree-of-Freedom (DoF)-based model is a simple yet powerful tool to analyze MIMO's spatial multiplexing (SM) and interference cancellation (IC) capabilities in a multi-hop network. Recently, a new DoF model was proposed and was shown to achieve the same rate region as the matrix-based model (under SM and IC). The essence of this new DoF model is a novel node ordering concept, which eliminates potential duplication of DoF allocation for IC. In this paper, we investigate DoF scheduling for a multi-hop MIMO network based on this new DoF model. Specifically, we study how to perform DoF allocation among the nodes for SM and IC so as to maximize the minimum rate among a set of sessions. We formulate this problem as a mixed integer linear programming (MILP) and develop an efficient DoF scheduling algorithm to solve it. We show that our algorithm is amenable to local implementation and has polynomial time complexity. More importantly, it guarantees the feasibility of final solution (upon algorithm termination), despite that node ordering establishment and adjustment are performed locally. Simulation results show that our algorithm can offer a result that is close to an upper bound found by CPLEX solver, thus showing that the result found by our algorithm is highly competitive.

Interference alignment (IA) is a major advance in information theory. Despite its rapid advance in the information theory community, most results on IA remain point-to-point or single-hop and there is a lack of advance of IA in the context of multi-hop wireless networks. The goal of this paper is to make a concrete step toward advancing IA technique in multi-hop MIMO networks. We present an IA model consisting of a set of constraints at a transmitter and a receiver that can be used to determine a subset of interfering streams for IA. Based on this IA model, we develop an IA optimization framework for a multihop MIMO network. For performance evaluation, we compare the performance of a network throughput optimization problem under our proposed IA framework and the same problem when IA is not employed. Simulation results show that the use of IA can significantly decrease the DoF consumption for IC, thereby improving network throughput.

In this paper, we investigate the linear processing at either base station or relay station in cellular network with a non-regenerative MIMO relay assisting. The purpose of linear processing is to equalize the two-hop fading channel in order to support multiuser transmission in downlink and uplink. Under the transmit power constraint at each station, the closed-form expressions of the suboptimal MMSE processing matrix in downlink and the optimal MMSE processing matrix in uplink are derived when the linear processing is performed at either base station or relay station. Simulation results demonstrate that the proposed MMSE strategies significantly outperform the corresponding ZF strategies in terms of BER.