With the advancements of generative artificial intelligence (GenAI) models, their capabilities are expanding significantly beyond content generation and the models are increasingly being used across diverse applications. Particularly, GenAI shows great potential in addressing challenges in the electric vehicle (EV) ecosystem ranging from charging management to cyber-attack prevention. In this article, we specifically consider Internet of Electric Vehicles (IoEV) and we categorize GenAI for IoEV into four different layers, namely, EV’s battery layer, individual EV layer, smart grid layer, and security layer. We introduce various GenAI techniques used in each layer of IoEV applications. Subsequently, public datasets available for training the GenAI models are summarized. Finally, we provide recommendations for future directions. This survey not only categorizes the applications of GenAI in IoEV across different layers but also serves as a valuable resource for researchers and practitioners by highlighting the design and implementation challenges within each layer. Furthermore, it provides a roadmap for future research directions, enabling the development of more robust and efficient IoEV systems through the integration of advanced GenAI techniques.

Mobile crowdsensing (MCS) is a novel sensing paradigm by utilizing mobile users (MUs) to collect data from environment. Considering the finite sensing and computing resources of MUs, it is crucial to inspire MUs to take part in crowdsensing willingly. In this study, a multiagent-deep-reinforcement-learning (DRL)-based incentive mechanism is investigated to tackle the joint data sensing and computing issues. Specifically, due to the heterogeneity of sensing tasks, multiple MCS platforms (MCPs) motivate MUs to participate in different tasks. The interaction between MCPs and MUs is modeled as a multileader-multifollower Stackelberg game with Stackelberg equilibrium proved by derivation. Moreover, the Stackelberg game is transformed as a Markov decision process (MDP) to deal with a multiagent DRL method without any prior knowledge. Due to the continuous high-dimensional action space of multiple MCPs and MUs, a multiagent double actors deep deterministic policy gradient (MA-DADDPG) algorithm is proposed to obtain the optimal sensing data size, computing resource, and incentive payment policies. Extensive simulation results illustrate the effectiveness of the proposed crowdsensing incentive mechanism.

Channel knowledge map (CKM) has been envisioned as a promising technology to achieve environment-aware communications for future sixth-generation (6G) wireless networks. The angle of departure (AoD) and the angle of arrival (AoA) are crucial parameters of CKM widely used for location-specific applications. Conventional stochastic methods of char-acterization fail to correlate the AoD and AoA with location-specific transmission environments. In this paper, we propose to utilize the CKM technology to characterize the location-specific AoD and AoA, by constructing the AoD and AoA map based on sparse measurement data. We further leverage the data-driven deep machine learning (DML) technique to obtain the AoD and AoA for locations without measurements. Simulation results show that the proposed method can construct both the AoD and the AoA maps with high fidelity to the true map. Moreover, Method II which predicts AoDs and AoAs sequentially achieves better performance than the independent prediction in Method I.

Mobile crowdsourcing network is a promising paradigm to leverage mobile users (MUs) to perform large-scale sensing task. Due to limited sensing-computation resource and data security risk, it is necessary to design an efficient incentive mechanism to motivate MUs to complete crowdsourcing task. In this paper, a deep reinforcement learning (DRL)-based contract incentive mechanism is proposed by jointly considering participation contribution, sensing task, and computation resource of MUs. Specifically, considering the heterogeneous willingness of the MUs, we formulate a three-dimensional sensing-computation-reward contract incentive mechanism to obtain the maximum utility of mobile crowdsourcing platform. Moreover, based on the individual rationality and incentive compatibility constraints, we derive the optimal contract under the partial information asymmetry scenario. In the case of the complete information asymmetry scenario, we formulate the contract incentive issue as an Markov decision process. Considering the infinite and continuous action and state spaces, we develop the deep deterministic method to obtain the efficient sensing task, computation resource, and incentive reward policy. Finally, we conduct numerical simulation to demonstrate the feasibility of our DRL-based contract crowdsourcing incentive mechanism.

Vehicular edge computing network emerges as a key technique to offload vehicles’ tasks to the nearby roadside unit (RSU). Considering that the RSU may not always meet computation requirements of task vehicles (TVs), utilizing idle computing resources of the surrounding resource vehicles (RVs) becomes a feasible solution to enhance the TVs’ experience. Due to the self-interested property and limited resource of RVs, an efficient incentive mechanism should be designed to encourage RVs to participate in task offloading and resource cooperation. In this work, a deep reinforcement learning-assisted contract incentive mechanism is investigated by considering TVs’ task offloading requirements, RVs’ computation resources, and the RSU’s transmission time incentive. To truthfully reveal TV-RV task-resource coordination types, a contract is designed with computation task-transmission time contract items. The joint task offloading and resource cooperation optimization issue is formulated to maximize the RSU’s utility with incentive compatible (IR), individual rationality (IC), and task offloading constraints. The optimal transmission time strategy is first derived from IR and IC constraints. To obtain the task offloading and task data size strategies, a Markov decision process is formulated. A multiagent parametrized deep Q-network scheme is developed to handle the discrete-continuous hybrid action space problem. Numerical simulations show the feasibility and effectiveness of our proposed incentive method to solve the joint task offloading and resource cooperation problem.

Symbiotic radio (SR) is a promising solution to achieve high spectrum- and energy-efficiency due to its spectrum sharing and low-power consumption properties, in which the secondary system achieves data transmissions by backscattering the signal originating from the primary system. In this paper, we are interested in the pilot design and signal detection when the primary transmission adopts orthogonal frequency division multiplexing (OFDM). In particular, to preserve the channel orthogonality among the OFDM sub-carriers, each secondary symbol is designed to span an entire OFDM symbol. The comb-type pilot structure is employed by the primary transmission, while the preamble pilot structure is used by the secondary transmission. With the designed pilot structures, the primary signal can be detected via the conventional methods by treating the secondary signal as a part of the composite channel, i.e., the effective channel of the primary transmission. Furthermore, the secondary signal can be extracted from the estimated composite channel with the help of the detected primary signal. The bit error rate (BER) performance with both perfect and estimated CSI, the diversity orders of the primary and secondary transmissions, and the sensitivity to symbol synchronization error are analyzed. Simulation results show that the performance of the primary transmission is enhanced thanks to the backscatter link established by the secondary transmission. More importantly, even without the direct link, the primary and secondary transmissions can be supported via only the backscatter link.

In this paper, we investigate reconfigurable intelligent surface (RIS)-enabled full-duplex backscatter communications in multi-user symbiotic radio (SR) systems. Specifically, an RIS is integrated with a primary transmission in which the primary transmitter (PT) broadcasts common messages to multiple primary receivers (PRs) and the RIS. Thanks to the full-duplex backscatter communication nature, the RIS absorbs part of the PT signals to decode the PT messages and, meanwhile, reflects the remaining part to convey its own messages for the PRs. By doing so, the RIS and the PRs can receive common messages, like the pairing messages, from the PT, and simultaneously establish information links between them without requiring additional spectrum and radio resources. However, it brings a challenging task on how to properly design the reflection matrix to resolve the conflicts between the signal absorption and reflection in the RIS-enabled full-duplex backscatter communication. Towards this end, we formulate an optimization problem that aims to jointly optimize the PT transmit beamforming vector and the RIS reflection matrix subject to the transmission rate constraints for the PT and the RIS. The problem is solved by the proposed alternating optimization (AO) method combined with difference-of-convex (DC) algorithm. Simulation results are presented to show that the RIS-enabled full-duplex backscatter communication can efficiently save the transmit power as compared to its half-duplex counterpart.

In reconfigurable intelligent surface (RIS)-assisted symbiotic radio (SR), the RIS acts as a secondary transmitter by modulating its information bits over the incident primary signal and simultaneously assists the primary transmission, then a cooperative receiver is used to jointly decode the primary and secondary signals. Most existing works of SR focus on using RIS to enhance the reflecting link while ignoring the ambiguity problem for the joint detection caused by the multiplication relationship of the primary and secondary signals. Particularly, in case of a blocked direct link, joint detection will suffer from severe performance loss due to the ambiguity, when using the conventional on-off keying and binary phase shift keying modulation schemes for RIS. To address this issue, we propose a novel modulation scheme for RIS-assisted SR that divides the phase-shift matrix into two components: the symbol-invariant and symbol-varying components, which are used to assist the primary transmission and carry the secondary signal, respectively. To design these two components, we focus on the detection of the composite signal formed by the primary and secondary signals, through which a problem of minimizing the bit error rate (BER) of the composite signal is formulated to improve both the BER performance of the primary and secondary ones. By solving the problem, we derive the closed-form solution of the optimal symbol-invariant and symbol-varying components, which is related to the channel strength ratio of the direct link to the reflecting link. Moreover, theoretical BER performance is analyzed. Finally, simulation results show the superiority of the proposed modulation scheme over its conventional counterpart.

This paper studies the reconfigurable intelligent surface (RIS)-assisted symbiotic radio (SR) system, where an RIS acts as a secondary transmitter to transmit its information by leveraging the primary signal as its RF carrier and simultaneously assists the primary transmission. Conventionally, all reflecting elements of the RIS are used to transmit the secondary signal, which, however, would limit its capability for assisting the primary transmission. To address this issue, we propose a novel RIS partitioning scheme, where the RIS is partitioned into two sub-surfaces, one to assist the primary transmission and the other to transmit the secondary signal. Naturally, there exists a fundamental tradeoff between the assistance and transmission capabilities of RIS regarding the surface partitioning strategy. Considering the coupling effect between the primary and secondary transmissions, we focus on the detection of the composite signal formed by the primary and secondary ones, based on which we propose a novel two-step detector. Then, we formulate the assistance-transmission tradeoff problem to minimize the bit error rate (BER) of the composite signal by jointly optimizing the surface partitioning strategy and the phase shifts of the two sub-surfaces, such that the overall BER of RIS-assisted SR is minimized. By solving this problem, we show that the optimized surface partitioning strategy depends on the channel strength ratio of the direct link to the reflected link. Moreover, performance analysis shows that when the direct link is blocked, exchanging the number of reflecting elements used for assistance and transmission can still achieve almost the same BER of the composite signal thanks to the coupling effect. Finally, extensive simulations show that our proposed RIS partitioning scheme outperforms the conventional schemes which use all reflecting elements for either assistance or transmission.

Radio environment map (REM) is a database repository widely adopted for applications such as spectrum sensing, interference management and network planning for wireless networks. However, conventional REM construction requires measurement-capable devices (MCDs) to upload location-based measurements, which exposes the privacy of users. In this paper, we propose a new method to construct the REM utilizing federated learning (FL) to preserve user privacy. In this method, multiple mobile MCDs collect the signal strength measurements in parallel and store the measurements locally. FL is then used to train a shared deep machine learning (DML) model for multiple MCDs collectively. In addition, we pre-process the location information and apply power adjustment to the measurements, either to increase the dynamic range of the input data to the FL or decrease the dynamic range of the measurements. The performance of the proposed FL-based method is evaluated in terms of the colormaps, heatmaps and cumulative distribution function (CDF) of estimation errors. The colormaps show that the power adjustment greatly improves the performance of the FL-based REM. The heatmaps show that smaller estimation errors can be achieved with more participating MCDs. The simulation results also show that the proposed FL-based method achieves better performance than the distance-based method and the nearest-neighbor-based (NN-based) method, in terms of estimation errors.

Reconfigurable intelligent surface (RIS) has recently emerged as a promising technology for wireless communications, which intelligently controls the phase shift of each unit cell to form desired beams. Most prior works on RIS consider time-division duplexing (TDD) systems, in which the same phase shifts can be applied to both uplink and downlink due to the channel reciprocity. However, for frequency-division duplexing (FDD) systems, using the same phase shifts will result in beam misalignment, thereby leading to performance degradation. To address this issue, in this article, we study the practical RIS design and beamforming optimization for FDD systems. By representing the phase shifts of RIS with the equivalent circuit model which includes the resistance, inductances, and tunable capacitance, we propose a methodology to design the circuit parameters (i.e., inductances and capacitance) to meet the desired reflection requirements (i.e., phase tuning range, reflectivity, and zero phase slope) of both the uplink and downlink transmissions in FDD systems. Given the designed inductances, a practical binary RIS reflection model corresponding to two reflection states is then proposed. Furthermore, based on the proposed reflection model, a problem is formulated to jointly optimize the active and passive beamforming such that the minimum array response gain of the uplink and the downlink is maximized. An efficient iterative algorithm is proposed to obtain a suboptimal solution. Simulation results show that our proposed RIS design outperforms those benchmarks which design the circuits by only optimizing either uplink or downlink.

In this paper, we investigate a novel uplink reconfigurable intelligent surface (RIS) based multi-user multi-input multi-output symbiotic radio system. It indicates that each RIS, as an Internet-of-Things (IoT) device, enhances the primary transmission from a nearby user to the base station (BS), and simultaneously transmits its own information to the BS by backscattering modulation. By embedding environmental sensors on the RISs, the proposed system enables the IoT transmission of locally collected environmental data to the BS while assisting the primary communications from the users to the BS. We consider both the case of perfect and imperfect channel state information (CSI), and design the active beamforming at the BS and the passive beamforming at the RISs jointly to maximize the weighted sum-rate of both the primary and IoT transmissions. For the perfect CSI case, we propose an algorithm based on the block coordinate descent (BCD) method to solve the problem. We also propose another algorithm with a similar framework to reduce the computational complexity. For the imperfect CSI case, an algorithm based on BCD and the online successive convex approximation technique is proposed. Simulation results show that the proposed system achieves significant performance gain over a number of baseline schemes for both the perfect and imperfect CSI cases. Furthermore, when the channel estimation error is small, the performance loss due to imperfect CSI is insignificant.

Timely identification of outliers occurring in key performance indicators (KPIs) of mobile cellular networks is crucial for prompt action to unexpected events. The KPIs of cellular networks typically exhibit seasonality and trend effects and these make the detection of outliers challenging. In this paper, an online unsupervised deep machine learning (DML) algorithm for outlier detection for time series with seasonality and trend is proposed. The proposed algorithm utilizes a neural network based on a temporal convolution network (TCN) and an autoencoder (AE) to reconstruct a time series that captures the normal behavior of the input data and then the reconstruction errors between the input and reconstructed time series at the output of the TCN-AE network are used for outlier detection. To train the TCN-AE network to learn the normal behavior of the input time series, we propose a two-step training process to overcome the presence of outliers in the training data. In addition, a novel loss function specially designed to address the seasonality and trend effects of the time series is proposed. Furthermore, a pre-processing technique is used to combat the adverse trend effect which might cause false alarms in outlier detection. The performance of the proposed TCN-AE-based outlier detection algorithm is evaluated using synthetic time series, Yahoo Webscope dataset and real time series of KPIs of mobile cellular networks. The results show that the proposed TCN-AE-based outlier detection algorithm achieves better detection accuracy in terms of F-score than other DML-based algorithms such as long short-term memory (LSTM)-AE-based algorithm and the convolutional neural network (CNN) based algorithm.

Small cell networks (SCNs) have emerged as a promising solution to meet the demand for increasing data traffic for the sixth generation and beyond wireless networks. However, power consumption and two-tier interference issues are two bottlenecks that hinder further development. This paper proposes a novel reconfigurable intelligent surface (RIS)-based SCN in which an RIS serves multiple micro users as a small cell base station while assisting the macro user’s transmission. Compared to the conventional SCNs, the RIS-based SCN can achieve significant power reduction. Meanwhile, the reflected signal can be regarded as a multipath component instead of interference to the macro user. We propose two transmission schemes and formulate the design of the phase shift matrix at the RIS and the beamforming vector at the macro base station as an optimization problem. The alternating optimization algorithm is developed to optimize the phase shift matrix and the beamforming vector to minimize the total power consumption under the user rate and phase shift constraints. Simulation results show that the total power consumption can be reduced significantly by deploying the RIS in the SCN when the number of reflective elements is sufficiently large.

Mobile edge computing can effectively reduce service latency and improve service quality by offloading computation-intensive tasks to the edges of wireless networks. Due to the characteristic of flexible deployment, wide coverage and reliable wireless communication, unmanned aerial vehicles (UAVs) have been employed as assisted edge clouds (ECs) for large-scale sparely-distributed user equipment. Considering the limited computation and energy capacities of UAVs, a collaborative mobile edge computing system with multiple UAVs and multiple ECs is investigated in this paper. The task offloading issue is addressed to minimize the sum of execution delays and energy consumptions by jointly designing the trajectories, computation task allocation, and communication resource management of UAVs. Moreover, to solve the above non-convex optimization problem, a Markov decision process is formulated for the multi-UAV assisted mobile edge computing system. To obtain the joint strategy of trajectory design, task allocation, and power management, a cooperative multi-agent deep reinforcement learning framework is investigated. Considering the high-dimensional continuous action space, the twin delayed deep deterministic policy gradient algorithm is exploited. The evaluation results demonstrate that our multi-UAV multi-EC task offloading method can achieve better performance compared with the other optimization approaches.

Small cell network (SCN) is a promising solution to meet the demand for increasing data traffic for the sixth generation and beyond wireless networks. However, the power consumption and two-tier interference issues are two bottlenecks that hinder its further development. In this paper, we propose a novel intelligent reflecting communication (IRC) system in which a reconfigurable intelligent surface (RIS) is used to serve multiple micro users in an SCN while assisting the transmission from a macro base station (MBS) to a macro user. Compared to the conventional SCN, the RIS can achieve significant power reduction as it transmits the information by passively reflecting the incident signals. In addition, the reflected signal can be regarded as a multipath component instead of an interference to the macro user. We are interested in minimizing the total power consumption by jointly designing the phase shift matrix at the RIS and the beamforming vector at the MBS under the user rate constraints and the practical phase shift constraints. The solution is obtained by alternating optimization to iteratively solve two subproblems, one to optimize the phase shift matrix, and the other to optimize the beamforming vector. A TDMA transmission scheme is also proposed as an alternative to serve multiple users. Simulation results demonstrate that the total power consumption can be reduced significantly by deploying the RIS in the SCN when the number of reflecting elements is sufficiently large.

In this paper, we investigate a reconfigurable in-telligent surface (RIS)-based uplink massive multi-input multi-output symbiotic radio system, where each RIS, as an IoT device, enhances the primary transmission from a nearby user to the base station (BS) and simultaneously transmits its own infor-mation to the BS by backscattering modulation. By embedding environmental sensors on the RISs, the proposed system enables the IoT transmission of locally collected environmental data to the BS while assisting the primary transmission. Assuming imperfect channel state information (CSI), we jointly design the active beamforming at the BS and the passive beamforming at the RISs to maximize the weighted sum-rate of both the primary and IoT transmissions. An algorithm based on the block coordinate descent method is proposed to solve it. Simulation results show that the proposed system achieves significant performance gain compared with different baseline schemes. Besides, when the channel estimation error is small, the performance loss due to imperfect CSI is insignificant.

To support massive access for future wireless communications, we propose a novel intelligent reflecting surface (IRS) enhanced downlink multi-user multi-input single-output (MU-MISO) symbiotic radio (SR) system, where each IRS, acting as a reflecting Internet-of-Things (IoT) device, transmits its message to a nearby primary receiver (PR) by reflecting the RF signals from the primary transmitter (PT), and simultaneously enhances the transmission from the PT to the associated PR. Thus, each PR jointly decodes its own message as well as the one from the corresponding IRS. We are interested in maximizing the weighted sum-rate of both primary and IoT transmissions by jointly designing the active transmit beamforming at PT and the passive beamforming at each IRS, subject to the maximum transmit power constraint at PT. Besides, as the passive elements at IRS can only reflect the incident signal with discrete phase shifts in practice, the discrete reflection coefficient (RC) constraint is further considered at the IRSs. Due to the non-convexity of the formulated problems, we solve them with fractional programming (FP) technique and alternating optimization (AO) method. Simulation results have verified the effectiveness of the proposed algorithms compared to different benchmark schemes.

The Internet-of-Things (IoT) paradigm holds the promise to revolutionize the way we live and work through connecting various machine-type communication terminals. In this work, we investigate a symbiotic network from a cross-layer perspective, where passive IoT devices coexist in symbiosis with an ambient network that uses carrier sense multiple access with collision notifications (CSMA/CNs) MAC protocol. In the ambient network, each full-duplex mobile user (MU) aims to transmit its own packets to the common access point (AP) while receiving the signal backscattered from its associated backscatter device (BD). Considering the imperfectness of carrier sensing of the MU, two key parameters, i.e., the probability of detection and the probability of false alarm, are quantified. Then we derive the PHY-layer outage probabilities and analyze the corresponding diversity orders for both the CSMA/CN and the BD system. By incorporating the outage probabilities and the carrier sensing metrics into the MAC-layer analysis, the cross-layer outage capacities of the CSMA/CN and the BD system are derived. Simulation results demonstrate that the system performance can be improved by appropriately setting the PHY-layer parameters, such as the BD reflection coefficient α and the number of samples for carrier sensing K, as well as the MAC-layer parameters, such as the sensing duration and the initial contention window. With the BD reflection coefficient being 0.05, the outage capacity of the overall system has improved by 41.85% compared with carrier sense multiple access with collision avoidance protocol.

To support massive access for future wireless networks, we propose a novel reconfigurable intelligent surface (RIS)-enhanced downlink multi-user multi-input single-output (MU-MISO) symbiotic radio (SR) system. In the proposed system, each RIS not only enhances the primary transmission from the primary transmitter (PT) to the associated primary receiver (PR) nearby, but also acts as an Internet-of-Things (IoT) device to enable IoT transmissions to the same PR. Therefore, each PR needs to jointly decode the information from both the PT and its corresponding RIS. We are interested in maximizing the weighted sum-rate of both primary and IoT transmissions by jointly designing the active transmit beamforming at PT and the passive beamforming at each RIS under the maximum transmit power constraint at the PT and various constraints on the reflection coefficients (RCs), which include the ideal, continuous-phase and the discrete-phase cases. The formulated problem is non-convex, which cannot be solved directly. Thus, fractional programming (FP) method and alternating optimization (AO) technique are adopted to tackle the problem. In particular, three low-complexity algorithms are proposed to trade off between computational complexity and convergence rate. Compared to different benchmark schemes, simulation results demonstrate that with the aid of the RISs, the PRs can benefit from the enhanced primary transmission from the PT, and receive information from the associated RISs via IoT transmission.

Reconfigurable Intelligent Surface (RIS) is a promising solution to reconfigure the wireless environment in a controllable way. To compensate for the double-fading attenuation in the RIS-aided link, a large number of passive reflecting elements (REs) are conventionally deployed at the RIS, resulting in large surface size and considerable circuit power consumption. In this paper, we propose a new type of RIS, called active RIS, where each RE is assisted by active loads (negative resistance), that reflect and amplify the incident signal instead of only reflecting it with the adjustable phase shift as in the case of a passive RIS. Therefore, for a given power budget at the RIS, a strengthened RIS-aided link can be achieved by increasing the number of active REs as well as amplifying the incident signal. We consider the use of an active RIS to a single input multiple output (SIMO) system. However, it would unintentionally amplify the RIS-correlated noise, and thus the proposed system has to balance the conflict between the received signal power maximization and the RIS-correlated noise minimization at the receiver. To achieve this goal, it has to optimize the reflecting coefficient matrix at the RIS and the receive beamforming at the receiver. An alternating optimization algorithm is proposed to solve the problem. Specifically, the receive beamforming is obtained with a closed-form solution based on linear minimum-mean-square-error (MMSE) criterion, while the reflecting coefficient matrix is obtained by solving a series of sequential convex approximation (SCA) problems. Simulation results show that the proposed active RIS-aided system could achieve better performance over the conventional passive RIS-aided system with the same power budget.

Conventionally, a substantial number of reflecting elements (REs) is deployed at the intelligent reflecting surface (IRS) to mitigate the effect of the double-fading attenuation in the IRS-aided link, leading to a large surface size and considerable power consumption. In this paper, a new type of IRS, called active IRS, is proposed to solve this challenge by allowing each RE to amplify the incident signal with the assistance of the active loads (negative resistances). Thus, given a power budget at the IRS, the IRS-aided link can be enhanced by increasing the number of active REs as well as amplifying the incident signal. Specifically, we consider the use of an active IRS-aided single input multiple output (SIMO) system, in which the received signal-to-noise ratio (SNR) is maximized, by optimizing not only the reflecting coefficient matrix at the IRS but also the receive beamforming at the receiver. To solve this non-convex problem, we propose an alternating optimization algorithm, that iteratively optimizes the two design variables. In particular, the receive beamforming is founded to be in the form of a linear minimum mean square error (MMSE) detector, and the reflecting coefficient matrix is obtained via the Charnes-Cooper transformation and the semi-definite programming (SDP). Simulation results show that under a practical power consumption model, the proposed active IRS-aided system achieves better performance over the conventional passive IRS-aided system with the same power budget.

In this paper, we study the cross-layer performance of a symbiotic system comprising passive internet-of-things (IoT) devices and the ambient system using carrier sense multiple access (CSMA) with collision notification (CSMA/CN) MAC protocol. Different from the CSMA with collision avoidance (CSMA/CA), the CSMA/CN protocol can enhance the throughput of the ambient system notably by using physical layer techniques to detect collisions. We first evaluate the physical-layer outage probabilities for the ambient and the backscatter device (BD) system respectively, and then introduce the outage probability of the ambient system into the MAC-layer performance analysis. Thereafter, the crosslayer outage capacities of the ambient system and the BD system are derived by considering the available transmission time for the BD system. We find that the overall system performance can be improved by appropriately setting the reflection coefficient of the BD and the parameters of the MAC protocol. The simulation results are provided to verify our theoretical analysis and demonstrate the system performance.

Unmanned aerial vehicle (UAV) is considered to be a key component in the next-generation cellular networks. Considering the non-convex characteristic of the trajectory design and power allocation problem, it is difficult to obtain the optimal joint strategy in UAV-assisted cellular networks. In this paper, a reinforcement learning-based approach is proposed to obtain the maximum long-term network utility while meeting with user equipments' quality of service requirement. The Markov decision process (MDP) is formulated with the design of state, action space, and reward function. In order to achieve the joint optimal policy of trajectory design and power allocation, deep reinforcement learning approach is investigated. Due to the continuous action space of the MDP model, deep deterministic policy gradient approach is presented. Simulation results show that the proposed algorithm outperforms other approaches on overall network utility performance with higher system capacity and faster processing speed.

Intelligent reflecting surface (IRS), with a large number of reflective elements, is a promising technology to achieve both spectrum and energy efficient wireless communication. The IRS can reflect the incident electromagnetic wave passively and steer it to the desirable way before reaching the intended receiver by adjusting the phase shift on the reflective elements. In order to better improve communication quality, the beamforming vector at the base station (BS) and the phase shift induced by the IRS should be jointly designed carefully. However, thus far, previous works on IRS have assumed that the channel state information (CSI) is perfectly known at the BS, which is not available in the practical systems. In this paper, we study an IRS-enhanced multi-user multiple-input single-output (MISO) downlink communication system assuming imperfect CSI. An optimization problem is formulated to jointly optimize the beamforming vector at the BS and the phase shift at the IRS such that the total transmit power is minimized under the individual outage probability constraints. An algorithm based on alternating optimization (AO) and semi-definite relaxation (SDR) is proposed to solve this challenging non-convex problem. Finally, numerical results have validated the effectiveness of the proposed algorithm.

Traditionally, dynamic spectrum access (DSA) based on cooperative spectrum sensing relies on a centralized fusion centre to fuse and store the sensing results, which is vulnerable to single point of failure. In this paper, we propose a sensing-based DSA framework which is enabled by blockchain. The proposed DSA framework includes a protocol that specifies a time-slotted-based five-phase operations. In the proposed framework, each secondary user (SU) acts as both a sensing node for cooperatively sensing the spectrum and a node, i.e., a miner and a verifier, in the blockchain network for mining and updating the sensing and access results in a distributed and secure manner without the need for a fusion centre. In order to incentivize SUs for participating in such energy-consuming operations of the blockchain network, we reward them with tokens for sensing and mining, which can be used to bid for the access to the spectrum opportunities. The sensing and mining policies which they use to determine when to sense and mine affect the number of tokens they can obtain and subsequently how they bid for the spectrum. Hence, the performance of the system depends on their sensing-access-mining policy. Therefore, we consider a heuristic sensing-access-mining policy that determines whether to participate in sensing and mining in a probabilistic manner and that determines how much to bid based on its buffer occupancy and the number of available tokens. Simulation results show that although increasing sensing and mining probabilities can increase average transmission rate, it also leads to higher energy consumption. Moreover, there exists an optimal set of sensing and mining probabilities that maximize the system energy efficiency.

Heterogeneous cellular networks can offload the mobile traffic and reduce the deployment costs, which have been considered to be a promising technique in the next-generation wireless network. Due to the non-convex and combinatorial characteristics, it is challenging to obtain an optimal strategy for the joint user association and resource allocation issue. In this paper, a reinforcement learning (RL) approach is proposed to achieve the maximum long-term overall network utility while guaranteeing the quality of service requirements of user equipments (UEs) in the downlink of heterogeneous cellular networks. A distributed optimization method based on multi-agent RL is developed. Moreover, to solve the computationally expensive problem with the large action space, multi-agent deep RL method is proposed. Specifically, the state, action and reward function are defined for UEs, and dueling double deep Q-network (D3QN) strategy is introduced to obtain the nearly optimal policy. Through message passing, the distributed UEs can obtain the global state space with a small communication overhead. With the double-Q strategy and dueling architecture, D3QN can rapidly converge to a subgame perfect Nash equilibrium. Simulation results demonstrate that D3QN achieves the better performance than other RL approaches in solving large-scale learning problems.

Modulation classification using deep neural networks has recently received increasing attention due to its capability in learning rich features of data. In this paper, we propose a low- complexity blind data-driven modulation classifier. Our classifier operates robustly over Rayleigh fading channels under uncertain noise conditions modeled using a mixture of three types of noise, namely, white Gaussian noise, white non- Gaussian noise and correlated non-Gaussian noise. The proposed classifier consists of several layers of recurrent neural networks (RNN) which is well-suited for learning representations from time-correlated data. The classifier is trained using the labeled raw signal samples generated under different noise conditions. Simulation results show that the performance of our proposed classifier approaches that of maximum likelihood classifiers with perfect channel knowledge and outperforms existing expectation maximum (EM) and expectation conditional maximum (ECM) classifiers which iteratively estimate channel and noise parameters.

Heterogeneous networks (HetNets) can offload the traffic and reduce the deployment cost, which is regarded as a promising technique in next-generation cellular networks. Because of the non-convex and combinatorial features of the joint issue of user association and resource allocation, it is challenging to achieve an optimal solution. In this paper, a novel method is proposed to maximize the long-term overall network utility while ensuring the user equipments' quality of service requirements in the downlink of HetNets. Multi-agent reinforcement learning approach is developed to obtain the distributed optimal strategy. To solve the computationally expensive issue with the large action space, the multi-user deep reinforcement learning is presented. Double deep Q-network (DDQN) approach is introduced to achieve an optimal policy. Simulation results clearly indicate the better performance of DDQN than that of other reinforcement learning methods.

Blind modulation classification is a fundamental step before signal detection for cognitive radio networks where the users may not have the complete knowledge of the modulation scheme due to the flexibility of operating dynamically in multiple frequency bands. In this paper, a modulation-constrained (MC) clustering classifier is proposed for recognizing the modulation scheme with unknown channel matrix and noise variance for MIMO systems. By recognizing the fact that the received signals within an observation interval form into clusters and exploiting the intrinsic relationships of different digital modulation schemes, the modulation classification problem is transformed into a clustering problem without direct channel estimation for each modulation scheme and the maximum likelihood criterion is applied for the final classification decision. A central component of the proposed classifier is a method called centroid reconstruction, which exploits the structural relationships in constellation diagrams to reconstruct cluster centroids with fewer number of parameters. Furthermore, a method to initialize the cluster centroids is also proposed. The proposed MC classifier together with centroid reconstruction and initialization methods not only reduce the number of parameters to be estimated, but also help to initialize the centroids for enhanced convergence of expectation- maximization (EM) algorithm. Simulation results show that our algorithm can perform excellently even at low SNR and with very short observation interval length.

Ambient backscatter communication (AmBC) enables radio-frequency (RF) powered backscatter devices (BDs) (e.g., sensors and tags) to modulate their information bits over ambient RF carriers in an over-the-air manner. This technology, also called “modulation in the air,” has emerged as a promising solution to achieve green communication for future Internet of Things. This paper studies an AmBC system by leveraging the ambient orthogonal frequency division multiplexing (OFDM) modulated signals in the air. We first model such AmBC system from a spread-spectrum communication perspective, upon which a novel joint design for BD waveform and receiver detector is proposed. The BD symbol period is designed as an integer multiplication of the OFDM symbol period, and the waveform for BD bit “0” maintains the same state within the BD symbol period, while the waveform for BD bit “1” has a state transition in the middle of each OFDM symbol period within the BD symbol period. In the receiver detector design, we construct the test statistic that cancels out the direct-link interference by exploiting the repeating structure of the ambient OFDM signals due to the use of cyclic prefix. For the system with a single-antenna receiver, the maximum-likelihood detector is proposed to recover the BD bits, for which the optimal threshold is obtained in closed-form expression. For the system with a multi-antenna receiver, we propose a new test statistic which is a linear combination of the per-antenna test statistics and derive the corresponding optimal detector. The proposed optimal detectors require only knowing the strength of the backscatter channel, thus simplifying their implementation. Moreover, practical timing synchronization algorithms are proposed for the designed AmBC system, and we also analyze the effect of various system parameters on the transmission rate and detection performance. Finally, extensive numerical results are provided to verify that the proposed transceiver design can improve the system bit-error-rate performance and the operating range significantly and achieve much higher data rate, as compared with the conventional design.

Ambient backscatter communication (AmBC) enables a tag to modulate its information bits over ambient RF carriers by intentionally changing its reflection coefficient, thus has emerged as a promising technique to achieve green communications for future Internet-of-Things. In this paper, we model a cooperative AmBC system from a spectrum- sharing perspective, where a cooperative receiver (C-RX) decodes the information from both a multi-antenna primary transmitter (PT) and a single-antenna secondary transmitter (i.e., tag). We consider two scenarios: first, the tag-symbol period equals the PT-symbol period; second, the tag-symbol period is an integer multiple of the PT-symbol period. For each scenario, we analyze the data rate via successive- interference-cancellation (SIC) based decoding, and formulate a problem to maximize the sum rate by optimizing the beamforming vector at the PT. The problems are transformed into semi-definite programming (SDP), and solved by using the technique of semi-definite relaxation (SDR). Furthermore, a novel transmit beamforming structure is proposed to reduce the computational complexity of beamforming optimization. Numerical results show that the cooperative AmBC system can achieve a higher sum rate than a conventional point-to-point system without a backscatter tag.

Today multiple radio access technologies (RATs) coexist in wireless networks. Multi-mode mobile terminals (MMTs) which can switch between networks with different RATs can enjoy enhanced quality of service (QoS) by exploring the diversity among different radio access networks (RANs). Usually, it has been assumed in literature that an MMT can access only one network at a time. Multi-Radio Access (MRA) is a technology that allows an MMT to transmit and receive data via multiple RANs simultaneously. With MRA, users can combine the data streams from multiple networks to meet their throughput requirements and enjoy customized QoS. In this paper, we consider an MMT with MRA technology in a heterogeneous network environment. It is assumed that the connection session of the MMT will spread over multiple handover windows and that the MMT is allowed to switch from the present set of connected RANs to another set at each handover window. We are interested in designing an intelligent network switching strategy which maximizes the average cumulative utility function of the MMT. Taking into account the dynamics of the heterogeneous networks, we model the network selection problem as a Markov decision process (MDP). By using the value iteration algorithm, we obtain the optimal switching strategy. Simulation results show that the MDP method provides higher average cumulative utility function and higher average rate of minimum throughput satisfaction than greedy method.

Cooperative communication is a promising technique to mitigate channel impairment and improve spectrum efficiency. Due to the selfish nature of relay nodes, how to provide proper long-term incentives for relay nodes in dynamic communication environments is an essential issue. In this paper, a two-period dynamic contract is proposed under the dynamic asymmetric information scenario. Considering the relay nodes' types are independent in both periods with identical probability distribution, the contract-theoretic model for ability discrimination relay selection is formulated. And the necessary and sufficient conditions for the optimal contract are systematically characterized. To maximize the source's expected utility, a sequential optimization algorithm is proposed to obtain the optimal relay- reward strategy. Simulation results show that the optimal dynamic contract design scheme is effective in improving system performance for cooperative communication.

Probe vehicle data, also known as floating car data or connected vehicle data, is the data collected from GPS-enabled sensors on vehicles. With the advancement in wireless communications and localization technologies, more and more vehicles are expected to be equipped with such sensors. Existing studies only focus on using small-scale probe vehicle data. In this paper, we are interested in developing a real-time parallel stream processing framework to extract traffic flow KPIs from large-scale probe vehicle data. The developed framework is implemented using Apache Storm on Amazon AWS, and can process one million probe vehicle messages per second. Various design considerations, such as data partition and delay processing are discussed. To evaluate the performance of stream processing framework, simulated probe vehicle data based on the actual traffic flows in Jurong Lake District (JLD) of Singapore, is generated using the microscopic simulation software VISSIM. The JLD data is replicated multiple times to represent the one million population of vehicles in Singapore. GPS errors and communication delays are added to represent the real situations before the data is fed to stream processing module. The estimated KPIs from our stream processing model are validated against the ground truth values under different penetration levels.

In this paper, a TV white space (TVWS) network in which there is a centralized spectrum manager (SM) is considered. The SM is connected to the geo-location database (GLDB) to periodically obtain the available spectrum fragments, and is responsible for assigning the available spectrum to the white space devices (WSDs) in its coverage region. The spectrum available is usually fragmented with different bandwidths, and the bandwidth required by the WSDs can also be diverse. Unlike existing works which assume complete knowledge of all WSDs' bandwidth requirements before assignment, we consider the more practical settings that WSDs request bandwidth in a sequential manner. Upon the arrival of a bandwidth request, the SM has to determine which fragment the request should be assigned to, without a prior knowledge of the bandwidths of the future requests. We are interested in designing the spectrum assignment policy at the SM upon each arrival of bandwidth request. The above problem is formulated as a stochastic sequential decision-making problem. The optimal spectrum assignment policy to maximize the overall spectrum utilization of the TVWS network is computed through the value iteration method. We demonstrate the performance advantage of our proposed optimal spectrum assignment policy over two heuristic policies through numerical results. To the best knowledge of the authors, this is the first paper that addresses the modeling and design considering the sequential behaviors of WSDs for GLDB-based TVWS network.

In this paper, a spectrum sharing protocol is proposed for device-to-device (D2D) communication overlaying cellular networks. Specifically, the protocol allows the D2D users to communicate bi-directionally with each other while assisting the two-way communication between the cellular base station (BS) and the cellular user (CU) over the same time and frequency resources. The achievable rate region of the sum-rate of the D2D transmissions versus that of the cellular transmissions is evaluated. The Pareto boundary of the region is found by optimizing the transmit power at BS and CU as well as the power splitting factor at the relay D2D node. We find through numerical results that the proposed two-way protocol with power control at the BS and CU is effective to improve the sum rate for both the D2D and cellular communication.

In this paper we consider multi-input multi-output (MIMO) orthogonal frequency-division multiplexing (OFDM)-based systems under unknown narrow-band interference (NBI). We propose an iterative receiver to jointly estimate the channel information, which consists of channel coefficients and noise-plus-interference variances of each sub-carrier, and detect the transmitted signals. The simulation results show that our proposed receiver provides a close bit-error-rate (BER) to that of the case where perfect channel information is available at the receiver. Besides, Cramér-Rao lower bound (CRLB) of interested parameters are also derived. The mean-square-error (MSE) of the estimated parameters given by our proposed algorithm reaches the CRLB.

Device-to-device (D2D) communications has been proposed in the literature as an underlay approach to cellular networks to allow direct transmission between two cellular devices with local communication needs. In this paper, we consider a scenario of D2D communications overlaying a cellular network and propose a new spectrum sharing protocol, which allows the D2D users to communicate bi-directionally with each other while assisting the two-way communications between the cellular base station (BS) and the cellular user (CU). We derive the achievable rate region of the sum rate of the D2D transmissions versus that of the cellular transmissions. The Pareto boundary of the region is found by optimizing the transmit power at BS and CU as well as the power splitting factor at the relay D2D node. Since either of the two D2D users can be the relay and there can exist multiple pairs of D2D users, we also consider the relay selection from the potential D2D users. We find through numerical results that the proposed two-way protocol with power control at the BS and CU is effective to improve the sum rate for both the D2D and cellular users. In addition, relay selection can achieve further improvement in the sum rate of the cellular links.

Multiple antennas at the transmitter and the receiver can increase the channel capacity significantly. However, this is at the expense of linearly increasing circuit power consumption due to the use of multiple radio frequency (RF) chains to support the antennas, which is quite significant for large-scale MIMO systems but is largely ignored in the literature. Hence, in this paper, we assess the performance of a point-to-point large-scale MIMO channel considering the overall power consumption on both the transmitter and the receiver, and study the optimal RF chain configurations to maximize the transmission rate under such a total power constraint. Both configurations with and without the channel state information (CSI) of the channel matrix are considered. While the former requires the design of optimal selection of RF chains, the latter only needs to determine the optimal number of RF chains. We find through numerical results that the gain of configuration with CSI over that without CSI diminishes as the dimension of the channel gets large. We also provide guidelines for selecting the optimal number RF chains for configuration without CSI. In particular, for MISO case, we find that it is near-optimal to choose half of the maximum number of transmit RF chains, the circuits of which are affordable to be powered on by the total power budget.

When secondary users (SUs) in a cognitive radio network (CRN) are battery-powered wireless devices, energy resources become very precious. Therefore, it is important that their energies are used efficiently. In this paper, we define the energy efficiency as the ratio of the average throughput of the CRN over the average energy used by the CRN. In cooperative spectrum sensing, the fusion rule threshold, detector's thresholds at the SUs, length of the sensing time, and the number of cooperating SUs will affect both the average throughput and the average energy consumed by the CRN. Therefore, in this paper, we optimize and evaluate these parameters with the aim of maximizing the energy efficiency of the CRN.

Energy-efficient design has become increasingly important to battery-powered wireless devices. In this paper, we focus on the energy efficiency of a cognitive radio network, in which a secondary user senses the channels licensed to some primary users sequentially before it decides to transmit. Energy is consumed in both the channel sensing and transmission processes. The energy-efficient design calls for a careful design in the sensing-access strategies and the sensing order, with the sensing strategy specifying when to stop sensing and start transmission, the access strategy specifying the power level to be used upon transmission, and the sensing order specifying the sequence of channel sensing. Hence, the objective of this paper is to identify the sensing-access strategies and the sensing order that achieve the maximum energy efficiency. We first investigate the design when the channel sensing order is given and formulate the above design problem as a stochastic sequential decision-making problem. To solve it, we study another parametric formulation of the original problem, which rewards transmission throughput and penalizes energy consumption. Dynamic programming can be applied to identify the optimal strategy for the parametric problem. Then, by exploring the relationship between the two formulations and making use of the monotonicity property of the parametric formulation, we develop an algorithm to find the optimal sensing-access strategies for the original problem. Furthermore, we study the joint design of the channel sensing order and the sensing-access strategies. Lastly, the performance of the proposed designs is evaluated through numerical results.

In this paper, we investigate the optimal transmitter design, under the restriction of Gaussian signalling without preprocessing of information, for a secure multiple-input single-output (MISO) cognitive radio network (CRN), which consists of four terminals: a pair of secondary user transmitter (SU-Tx) and receiver, an eavesdropper and a primary user (PU). It is assumed that all the channel state information is not perfectly known at the SU-Tx due to the channel estimation errors and the loose cooperation between the SU and the PU. The design involves a nonconvex semi-infinite optimization problem, which maximizes the rate of the secondary link while avoiding harmful interference to the PU and preventing the eavesdropper from decoding the messages sent regardless of the uncertainties in the CSI. For this challenging optimization problem, we relate it with a sequence of semi-infinite capacity-achieving transmitter design problems in an auxiliary CRN without any eavesdropper, which can then be solved through transformations and using convex semidefinite programs. Finally, numerical examples are presented to evaluate the performance of the proposed algorithm.

In this paper, we address the physical-layer security issue of a secondary user (SU) in a spectrum-sharing cognitive radio network (CRN) from an information-theoretic perspective. Specially, we consider a secure multiple-input single-output (MISO) cognitive radio channel, where a multi-antenna SU transmitter (SU-Tx) sends confidential information to a legitimate SU receiver (SU-Rx) in the presence of an eavesdropper and on the licensed band of a primary user (PU). The secrecy capacity of the channel is characterized, which is a quasiconvex optimization problem of finding the capacity-achieving transmit covariance matrix under the joint transmit power and interference power constraints. Two numerical approaches are proposed to derive the optimal transmit covariance matrix. The first approach recasts the original quasiconvex problem into a single convex semidefinite program (SDP) by exploring its inherent convexity; while the second one explores the relationship between the secure CRN and the conventional CRN and transforms the original problem into a sequence of optimization problems associated with the conventional CRN, which helps to prove that beamforming is the optimal strategy for the secure MISO CR channel. In addition, to reduce the computational complexity, three suboptimal schemes are presented, namely, scaled secret beamforming (SSB), projected secret beamforming (PSB) and projected cognitive beamforming (PCB). Lastly, computer simulation results show that the three suboptimal schemes can approach the secrecy capacity well under certain conditions.

This paper studies the robust beamforming design problem for a multiple-input single-output (MISO) secrecy channel with a single-antenna eavesdropper. Due to the illegal nature, the eavesdropper may try to hide itself from being caught; thus, it could be difficult for the secrecy transmitter (S-Tx) to obtain accurate channel state information (CSI) of the eavesdropping link between S-Tx and the eavesdropper. Assuming that the CSI of the eavesdropping link belongs to a known uncertain set, this paper designs the optimal transmit strategy for the secrecy user to maximize the transmit rate under the condition that the eavesdropper cannot decode the secrecy message for all possible channel realizations of the eavesdropping link. This robust design problem is non-convex and cannot be solved by existing algorithms in the literature. By exploiting the relationship between the secrecy MISO channel and the cognitive radio (CR) MISO channel, this problem is transformed into a robust CR beamforming design problem, which can be solved efficiently by the interior point method. Numerical examples are provided to illustrate the effectiveness of the proposed algorithm.

In this paper, we study the sensing-throughput tradeoff problem for a multiple-channel cognitive radio (CR) network. In particular, using the sensing-throughput tradeoff metric, we investigate the design of the optimal spectrum sensing time and power allocation schemes so as to maximize the aggregate ergodic throughput of the cognitive radio network to guarantee the quality of service (QoS) of the primary users (PUs) without exceeding the power limit of the secondary transmitter. The optimal sensing time and power allocation strategies are developed under the average power constraint. Finally, numerical results show that, for a CR network with 3 channels, whose signal-to-noise ratio of PUs are -12dB, -15dB and -20dB, respectively, there is an optimal sensing time, and the optimal sensing time is almost insensitive to the total transmit power.

To improve the spectrum utilization efficiency, cognitive radio (CR) has been proposed by allowing a cognitive radio network (CRN) to coexist with a licensed primary network. The security issues, although critical, have been less explored in the literature of CRN. In this paper, we consider a secure CRN in which a multi-antenna secondary user (SU) transmitter sends confidential information to a SU receiver on the same frequency band with a primary user (PU) in the presence of an eavesdropper. All receive terminals are equipped with a single antenna. The capacity-achieving transmitter design is formulated as a quasiconvex optimization problem to maximize the rate of the secondary link while avoiding harmful interference to the PU and preventing the eavesdropper from decoding the messages sent. By exploring the inherent convexity, the original problem is solved efficiently by a single semidefinite program (SDP). Besides, two suboptimal algorithms are proposed to reduce the computational complexity, namely, scaled secret beamforming (SSB) and projected secret beamforming (PSB). It is shown through computer simulations that the two suboptimal algorithms can achieve close-to-optimal secrecy capacity under certain conditions.

In this paper, we consider a wideband cognitive radio network (CRN) which can simultaneously sense multiple narrowband channels and thus aggregate the perceived available channels for transmission. We study the problem of designing the optimal spectrum sensing time and power allocation schemes so as to maximize the average achievable throughput of the CRN subject to the constraints of probability of detection and the total transmit power. The optimal sensing time and power allocation strategies are developed under two different total power constraints, namely, instantaneous power constraint and average power constraint. Finally, numerical results show that, under both cases, for a CRN with three 6 MHz channels, if the frame duration is 100 ms and the target probability of detection is 90% for the worst case signal-to-noise ratio of primary users being -12 dB, -15 dB and -20 dB, respectively, the optimal sensing time is around 6 ms and it is almost insensitive to the total transmit power.

We consider a cognitive radio (CR) network that makes opportunistic access to a spectrum licensed to the primary users. During its operation, the CR network carries out spectrum sensing on a frame-by-frame basis to detect active primary users, thereby avoiding interfering with them. We formulate a collision-throughput tradeoff problem which, based on the sensing time requirement and the traffic pattern of primary users, finds optimal value for the frame duration of CR operation so that the throughput of the CR network is maximized, yet the collision probability of the primary users is not greater than a threshold. We derive the theoretical formula for achievable throughput of CR network and find the optimal solution for frame duration. Computer simulations are presented to evaluate the performance of our approach.

Interleaved frequency division multiple access (IFDMA), which has been adopted as the uplink air-interface for the third generation long term evolution (3G-LTE), outperforms its competing rival OFDM/OFDMA for its low peak-to-average power ratio (PAPR). To achieve future high-speed transmission rate, spatial multiplexed multiple-input multiple-output (MIMO) system is incorporated into IFDMA. Although no multiple access interference (MAI) exists, the system suffers heavily from both inter-symbol interference (ISI) and multi-stream interference (MSI). Block-iterative decision feedback equalizer (BI-GDFE) can be applied to mitigate the interference effectively. In this paper, a subcarrier-based processing technique is proposed to achieve same performance with significant complexity reduction over conventional BI-GDFE equalization approach in MIMO-IFDMA system. To achieve even further gain in computational efficiency, a suboptimal BI-GDFE equalizer is also designed with only slight performance degradation. Computer simulations are presented to illustrate the BER performances of MIMO-IFDMA system with the proposed equalizers.