Search

Intelligent reflecting surface (IRS) is deemed as a promising solution to improve the spectral and energy efficiency of wireless communications cost-effectively. In this paper, we consider a wireless network where multiple base stations (BSs) serve their respective users with the aid of distributed IRSs in the downlink communication. Specifically, each IRS assists in the transmission from its associated BS to user via passive beamforming, while in the meantime, it also randomly scatters the signals from other co-channel BSs, thus resulting in additional signal as well as interference paths in the network. As such, a new IRS-user/BS association problem arises pertaining to optimally balance the passive beamforming gains from all IRSs among different BS-user communication links. To address this new problem, we first derive a tractable lower bound of the average signal-to-interference-plus-noise ratio (SINR) at the receiver of each user, termed average-signal-to-average-interference-plus-noise ratio (ASAINR), based on which two ASAINR balancing problems are formulated to maximize the minimum ASAINR among all users by optimizing the IRS-user associations without and with BS transmit power control, respectively. We also characterize the scaling behavior of user ASAINRs with the increasing number of IRS reflecting elements to investigate the different effects of IRS-reflected signal versus interference power. Moreover, to solve the two ASAINR balancing problems that are both non-convex optimization problems, we propose an optimal solution to the problem without BS power control and low-complexity suboptimal solutions to both problems by applying the branch-and-bound method and exploiting new properties of the IRS-user associations, respectively. Numerical results verify our performance analysis and also demonstrate significant performance gains of the proposed solutions over benchmark schemes. [ABSTRACT FROM AUTHOR]

It has been a challenging task to derive the complementary cumulative distribution function (CCDF) of aggregate interference as closed-form in Poisson point process (PPP). In this paper, we consider wireless communication networks where transmitters are all distributed according to homogeneous PPP and study the CCDF of the interference. Mainly, we derive generalized bounds on the CCDF of aggregate interference from whole homogeneous PPP based on discretization of interference, which considerably enhance tightness compared to the previous bounds. These bounds can be used to the analysis requiring the distribution for aggregate interference, such as deployment and activation of BSs, massive MIMO, and millimeter wave communication networks. As examples of utilizing the derived bounds, we derive tractable coverage probabilities of uplink non-orthogonal multiple access (NOMA) users and downlink users in cellular networks under line-of-sight(LoS)/non-line-of-signt (NLOS) channels, respectively. In NOMA, we derive the closed-form bounds on coverage probability when successive interference cancellation is adopted for NOMA signal demodulation. Using the proposed bounds, we also obtain the bounds on coverage probability of a cellular system with LoS/NLOS channels. The tightness of bounds in this work is shown via numerical comparison. Moreover, we mathematically prove the convergence of bounds to exact CCDF. [ABSTRACT FROM AUTHOR]

This paper aims to investigate the benefit of using intelligent reflecting surface (IRS) in multi-user multiple-input single-output (MU-MISO) systems, in the presence of eavesdroppers. We maximize the weighted sum secrecy rate by jointly designing the secure beamforming (BF), the artificial noise (AN), as well as the phase shift of the IRS. An alternating optimization (AO) method is proposed to deal with the formulated non convex problem. In particular, the secure beamforming and AN jamming matrix are optimally designed via the successive convex approximation (SCA) approach for given phase shift, which can be derived by considering the alternating direction method of multiplier (ADMM) and element-wise block coordinate decent (EBCD) methods. Finally, simulation results are presented to show the benefit of the IRS in terms of improving the secrecy performance, when compared to other methods. [ABSTRACT FROM AUTHOR]

Automated and coordinated vehicle platoon driving is challenging as the complexity of vehicle control and the presence of unreliable wireless inter-vehicle communication. This subject has an interdisciplinary nature, involving control theories, vehicle dynamics, and communication. Due to the constant interaction between control logic and communication topology, a co-design approach for the controller and for reliable communication protocol becomes necessary. This paper describes the AddP-CACC - a consensus-based solution to vehicle platoons. The proposed solution turns the communication topology into a design parameter that dynamically reconfigures the controller. In addition, the controller automatically compensates for outdated information caused by network losses and delays. AddP-CACC was implemented in the PLEXE tool and was compared with other related works. Simulation results have shown the robustness and performance of the proposed solution in various scenarios that would include realistic propagation conditions with interference caused by other vehicles. The results have also evidenced the ability of the proposed solution to maintain a stable vehicular platoon by using different communication topologies, different vehicle flows, even in the presence of strong interferences, delays, and fading conditions. [ABSTRACT FROM AUTHOR]

To increase the feasibility of wireless powered communication networks (WPCNs), the synergy of efficient multiple access protocols and diverse types of resources, e.g., multiple antennas, memories, and energy storage devices is of paramount importance. To this end, in this paper, we introduce the use of buffered sources in WPCNs with uplink non-orthogonal multiple access, in which a multi-antenna hybrid access point can both transmit energy and receive information. The proposed scheme aims at maximizing the long-term utility function, which is determined by the achieved average data rate and fairness among sources. The formulated long-term time-average optimization problem is converted into instantaneous ones by exploiting the Lyapunov optimization framework. The derived analytical solutions reveal the impact of data buffers’ length on the decoding order of sources’ messages. Finally, it is shown that the proposed scheme achieves superior long-term utility compared to the considered baseline schemes with non-buffered sources and buffered sources with orthogonal multiple access. Additionally, it is demonstrated that our proposed scheme can ensure fairness without reducing the average data rate. [ABSTRACT FROM AUTHOR]

Complex building structures constrain indoor wireless communications. The significant dependence between the indoor wireless network performance and the building structures can be considered as an intrinsic property of a building, which means that the building has to be designed to achieve desirable indoor wireless network performance. The power gain (PG) and the interference gain (IG) have been defined as the figures of merit (FoMs) to evaluate the wireless performance of a building. Employing open space as the benchmark, the PG and the IG respectively quantify the effective change of the intended signal power and the undesired power, i.e., the interference and thermal noise power, received in a targeted user equipment (UE) location due to the presence of the building. In this paper, a tractable approach is proposed to obtain the PG and the IG with a partition-based path gain model. It enables the PG and the IG to capture the impact of both building materials and the building layout on indoor wireless network performance. Numerical results show that the analytical models in closed-form expressions facilitate the wireless performance assessment to a given building computational-efficiently. This work is a critical complement to the initial framework of the building wireless performance (BWP) evaluation. The proposed method can be employed in future wireless-friendly building design. [ABSTRACT FROM AUTHOR]

In the near future, Vehicle-to-Vehicle (V2V) transmission will enable wireless broadcast communication among nearby vehicles. Benefits for users include improved driver safety and potentially optimized traffic. However, this communication is vulnerable to cyber attacks involving message manipulation. Research aimed at tackling this problem has resulted in the proposal of multiple authentication protocols. The security, privacy, and other desirable features of authentication in vehicular networks have been widely studied. However, the efficiency of such authentication schemes has not been suitably addressed. There is no model to evaluate the efficiency of the proposals in a practical context, where the delay introduced by V2V authentication may impact on driver safety. In this paper, we provide such a model for evaluation. We explicitly present the key factors involved to evaluate the computational delay in the V2V authentication protocols. Our model has clearly defined metrics for computing the delay and evaluating the impact. Developing this model enables future research in the design of secure and efficient V2V authentication protocols suitable for practical application. Applying the model to assess proposed authentication protocols permits categorization based on safety service they can support. We demonstrate the applicability of our model through case studies. Our model can effectively analyze the delay introduced by an authentication protocol, and determine whether this would result in a crash, in the real world vehicular environments. [ABSTRACT FROM AUTHOR]

Consider a (multiple-access) wireless communication system where users are connected to a unique base station over a shared-spectrum radio links. Each user has a fixed number k of bits to send to the base station, and his signal gets attenuated by a random channel gain (quasi-static fading). In this paper we consider the many-user asymptotics of Chen-Chen-Guo’2017, where the number of users grows linearly with the blocklength. Differently, though, we adopt a per-user probability of error (PUPE) criterion (as opposed to classical joint-error probability criterion). Under PUPE the finite energy-per-bit communication is possible, and we are able to derive bounds on the tradeoff between energy and spectral efficiencies. We reconfirm the curious behaviour (previously observed for non-fading MAC) of the possibility of almost perfect multi-user interference (MUI) cancellation for user densities below a critical threshold. Further, we demonstrate the suboptimality of standard solutions such as orthogonalization (i.e. TDMA/FDMA) and treating interference as noise (i.e. pseudo-random CDMA without multi-user detection). Notably, the problem treated here can be seen as a variant of support recovery in compressed sensing for the unusual definition of sparsity with one non-zero entry per each contiguous section of 2k coordinates. This identifies our problem with that of the sparse regression codes (SPARCs) and hence our results can be equivalently understood in the context of SPARCs with sections of length 2100. Finally, we discuss the relation of the almost perfect MUI cancellation property and the replica-method predictions. [ABSTRACT FROM AUTHOR]

The new ways to consume and deliver media content in a wireless environment have led to a significant increase in the demand for spectrum resources. In addition, the arrival of next-generation wireless communication technologies has encouraged the creating of new verticals targeting novel applications, such as industry IoT, e-health, distance learning, smart city, smart-metering. All these new use cases are centred on an IP-based infrastructure. A promising alternative to satisfy the demand for resources is to incorporate a new delivery mechanism that can alleviate the congestion of classical cell-based point-to-point broadband wireless infrastructures. The best-positioned candidate is the next-generation Digital Terrestrial Television system, which offers a very spectrum efficient one-to-many PHY, and for the first time, native IP support. In ATSC 3.0, the whole system was designed with an IP-centric view, whereas DVB has created standards to deliver services to generic IP devices (DVB-I). Nevertheless, there is still a last missing piece to bring the broadcast infrastructure to the IP-connected world: a Broadcast Core Network (BCN). In this paper, the basics of a BCN are presented with new use cases that are attractive to Broadcast Network Operator (BNO). A possible roadmap towards implementation is described, and a service-based architecture for a BCN is presented and discussed. [ABSTRACT FROM AUTHOR]

This paper describes systems, devices, and methods to implement a bi-directional integrated inter-tower wireless communications network (IITWCN). The described technology can be implemented in combination with the Broadcast Core Network (BCN) in next generation broadcast eco-system and, therefore, support new business cases for broadcast operators such as the delivery of flexible datacasting services or support broadcast or point-to-point Internet services. The introduced bi-directional inter-tower communications network (ITCN) extends the previous unidirectional in-band distribution links (IDL) and adopts the on-channel repeater (OCR) as a simplified backhaul solution in single frequency networks (SFN). The concept of the coordinated ITCN is also presented, aiming at future broadcast Internet services. The ITCN provides a scalable and configurable network solution embedded in a broadcast system, which becomes independent from any non-broadcasting telecommunication infrastructure. The described technology partially relies on the infrastructure of the underlying broadcast/multicast network, using the allocated service channels without requiring additional frequency bands or a separate frequency band. The bi-directional inter-transmitter communication links are therefore referred to as integrated transmission links and the corresponding network as an integrated network. [ABSTRACT FROM AUTHOR]

Low-latency communication is expected to play a key role in the Industrial Internet of Things (IIoT). Although there has been considerable effort on reducing latency, few attention has been focused on the low-latency transmission of distortion-tolerant data, e.g. IIoT’s analog samples, over fading channels. In this paper, we present a cross-layer approach to jointly adapt the transmission power, rate, and compression ratio based on the instantaneous buffer and channel states. In particular, to minimize the average delay under both average power and distortion constraints, we formulate a Constrained Markov Decision Process (CMDP) with multi-dimensional state and action spaces. A novel solution is then presented by judiciously decomposing the multi-dimensional CMDP into a deterministic hierarchical optimization consisting of a linear programming and a convex optimization problem. Furthermore, we show the optimality of a threshold-based scheduling policy that highly reduces the complexity and present an optimal delay-power-distortion tradeoff that characterizes a fundamental performance tradeoff between the physical, network, and application layers. The joint lossy compression and power allocation scheme realizes a deep cross-layer optimization of source coding, queuing control, and wireless transmission, which outperforms the traditional cross-layer design consisting of only two layers, not to mention layered protocols. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the performance of a mixed radio-frequency-underwater wireless optical communication (RF-UWOC) system where an unmanned aerial vehicle (UAV), as a low-altitude mobile aerial base station, transmits information to an autonomous underwater vehicle (AUV) through a fixed-gain amplify-and-forward (AF) or decode-and-forward (DF) relay. The analysis accounts for the main factors that affect the system performance, such as the UAV height, air bubbles, temperature gradient, water salinity variations, detection techniques, and pointing errors. Employing the fixed-gain AF and DF relays, expressions for some key performance metrics are derived, e.g., outage probability (OP), average bit error rate (ABER), and average channel capacity. Moreover, in order to get further insights, asymptotic analyses for the OP and ABER are also carried out. Additionally, for the two relaying systems, we derive analytical expressions for the optimal UAV altitude that minimizes the OP. Simulation results show that the UAV altitude influences the system performance and there is an optimal altitude which ensures a minimum OP. Furthermore, it is demonstrated that the diversity order of the fixed-gain AF relaying depends on the RF link and pointing errors, while the diversity order of the DF relaying depends on the detection techniques and pointing errors. [ABSTRACT FROM AUTHOR]

This paper considers optimal joint time, frequency, and power allocation for energy-efficient device-to-device (D2D) communications in an overlay cellular network, where multiple D2D transmitters communicate with their corresponding D2D receivers while utilizing the energy harvested from a cellular base station (BS). Due to channel estimation error and feedback delays, the channel state information (CSI) available at the transmitters is usually imperfect, which can significantly deteriorate the energy efficiency (EE) of the considered system. To mitigate the impact of imperfect CSI, we formulate a non-convex robust optimization problem for the maximization of the EE (in bits/Joule) of the system while ensuring quality of service (QoS) requirements on the minimum data rata and the minimum amount of harvested energy for cellular and D2D users, respectively. We reveal that the optimal energy-efficient D2D transmission should exhaust all the harvested energy of each D2D transmitter. Exploiting this fact, we propose an efficient iterative algorithm for obtaining the optimal solution. To strike a balance between performance and computational complexity, we further propose a low-complexity suboptimal algorithm. Simulation results show that compared with several baseline schemes, both the proposed optimal and suboptimal algorithms can significantly improve the EE of wireless-powered D2D communications under imperfect CSI. Moreover, the proposed suboptimal algorithm is close-to-optimal when the number of cellular users are small or the channel estimation error is large. [ABSTRACT FROM AUTHOR]

In recent years, wireless communication has experienced a massive shift from a single service (i.e., voice) to an interconnected web of networks. Although many techniques have been developed improving the offered services to mobile users, still the demand for high-quality services cannot be reached. Therefore, this paper proposes a joint non-orthogonal multiple access (NOMA)-enabled optimization framework for small-cell network (SCNet) by utilizing the concepts of multi-objective problem. In particular, the transmit power of base station (BS) in each small-cell simultaneously optimizes to maximize the sum-capacity and total energy efficiency (EE) of SCNet. The multi-objective optimization problem is formulated as non-convex subject to several practical constraints, i.e., individual quality of service requirement, maximum power budget of small-cell BS, and efficient decoding of superimposed signal using successive interference cancellation. Based on the nature of the problem, the optimal solutions are provided using sequential quadratic programming, and Karush-Kuhn-Tucker approaches. The obtained results show significant performance gains over conventional orthogonal multiple access technique in terms of sum-capacity and total EE. [ABSTRACT FROM AUTHOR]

Adaptive optics (AO) has the potential to mitigate the effect of atmospheric turbulence and improve the performance of orbital angular momentum (OAM)-based optical wireless communication (OAM-OWC) links. Here, we propose a single-intensity-measurement phase retrieval algorithm (SPRA)-based AO technique of compensating for the distortion of the OAM beam. The only parameter required by the SPRA wave-front sensor is the intensity of the probe beam in the Fourier domain, which substantially simplifies the AO system. We first derive an analytical expression to characterize the expansion of probe beam in OAM-OWC links and then determine the diameter constraints as the apriori information of the SPRA required for guaranteeing a certain compensation performance. The simulation results illustrate that the SPRA-AO approach can indeed correct a distorted OAM beam both in a single-channel scenario and in multiplexed OAM-OWC systems. The bit error rate can be improved by orders of magnitude with the aid of SPRA-AO compensation. Furthermore, we establish noise models of AO-based OAM-OWC systems and analyze the robustness of the SPRA-AO technique. In a nutshell, this paper provides new insights for the applications of AO and forms the theoretical basis of employing probe beams in OAM-OWC systems. [ABSTRACT FROM AUTHOR]

In this paper, we consider a hybrid satellite-terrestrial network (HSTN) where a multiantenna satellite communicates with a ground user equipment (UE) with the help of multiple cache-enabled amplify-and-forward (AF) three-dimensional (3D) mobile unmanned aerial vehicle (UAV) relays. Herein, we employ the two fundamental most popular content (MPC) and uniform content (UC) caching schemes for two types of mobile UAV relays, namely fully 3D and fixed height. Taking into account the multiantenna satellite links and the random 3D distances between UAV relays and UE, we analyze the outage probability (OP) of considered system with MPC and UC caching schemes. We further carry out the corresponding asymptotic OP analysis to present the insights on achievable performance gains of two schemes for both types of 3D mobile UAV relaying. Specifically, we show the following: (a) MPC caching dominates the UC and no caching schemes; (b) fully 3D mobile UAV relaying outperforms its fixed height counterpart. We finally corroborate the theoretic analysis by simulations. [ABSTRACT FROM AUTHOR]

Orthogonal Time Frequency Space (OTFS) modulation has been recently proposed to be robust to channel induced Doppler shift in high mobility wireless communication systems. However, to the best of our knowledge, none of the prior works on OTFS have derived it from first principles. In this paper, using the ZAK representation of time-domain (TD) signals, we rigorously derive an orthonormal basis of approximately time and bandwidth limited signals which are also localized in the delay-Doppler (DD) domain. We then consider DD domain modulation based on this orthonormal basis, and derive OTFS modulation. We show that irrespective of the amount of Doppler shift, the received DD domain basis signals are localized in a small interval of size roughly equal to the inverse time duration along the Doppler domain and of size roughly equal to the inverse bandwidth along the delay domain (time duration refers to the length of the time-interval where the TD transmit signal has been limited). With sufficiently large time duration and bandwidth, there is little interference between information symbols modulated on different basis signals, which allows for joint DD domain equalization of all information symbols. This explains the inherent robustness of DD domain modulation to channel induced Doppler shift when compared with Orthogonal Frequency Division Multiplexing (OFDM). [ABSTRACT FROM AUTHOR]

This paper presents collaborative spectrum sensing for optical wireless scattering communications. The received signal is characterized by the conditional Poisson distribution, given the mean photon count following Málaga distributed correlated fading. We consider multi-scattered non-coplanar links and use the stochastic analytical method to model the channel. A system model is developed, utilizing the cooperative diversity drawn from the fact that the users are randomly distributed. A novel centralized cooperative spectrum sensing technique is developed that all the raw data available at each collaborative user are combined at the fusion center to make a decision. Another novel decentralized technique is also proposed where only one-bit information is required to be sent to the fusion center. For this decentralized technique, we propose a unique voting rule that leverages the location flexibility of randomly distributed users with each user experiencing correlated fading. Results demonstrate their effectiveness and provide insights into the proposed cooperative spectrum sensing system. Interestingly, it is also found that the use of the proposed voting rule requires fewer than the total number of collaborative users while satisfying a given error rate, and yields the Bernoulli distributed output of the fusion center. [ABSTRACT FROM AUTHOR]

The use of channel state information (CSI) at the transmitter significantly enhances the performance of wireless communication systems. However, the requirement of CSI feedback places an undue burden on the reverse link, especially in links that employ multiple-input multiple-output (MIMO) and orthogonal frequency division multiplexing (OFDM), where CSI takes the form of a precoding matrix (precoder) for each subcarrier. Typical deployments use quantization and feedback of CSI at certain subcarriers, with interpolation to fill in missing CSI at the transmitter. Past work has used the orthogonal structure of precoders with Flag manifolds for quantization and interpolation of CSI, although interpolation is complicated due to the absence of analytic expressions for geodesics on Flag manifolds. Other approaches have involved the parameterization of the precoder into scalar parameters that are amenable to quantization and interpolation. In this paper, we present efficient methods to quantize and interpolate on Flag manifolds, using both optimal algorithms as well as simplified suboptimal algorithms. Further, we unify these with the parameterization based approaches and show that these translate directly to low-complexity quantization and interpolation on Flag manifolds. Simulations reveal that the proposed precoder quantization and interpolation effectively enhance achievable rates with limited complexity. [ABSTRACT FROM AUTHOR]

We consider two users A and B wanting to exchange their messages MA and MB via a relay in a wireless communication network. To this aim, we have two methods: direct forward (DF) and computation and forward (CAF). In the first method, after an orthogonal multiple access (OMA) or non-orthogonal multiple access (NOMA) phase, the receiver decodes the two messages. In the second method, after a NOMA phase, the receiver decodes only the (modulo) sum. In the latter case the capacity region (computation region) is not known. This paper compares the two methods. To such aim, we propose a unified coding framework and identify a lower bound of achievable computation rates. We consider three constellations for the symmetric two-user Gaussian channel with Eisenstein constellation showing highest performance. We also show that a minimum signal-to-noise ratio is required for our CAF method to outperform DF, and we obtain the numerical values of such minimum ratios for the three constellations. Finally, we show achievability of our bounds, which is proved to require affine codes for computation, which has not been proved before. Overall, our results are useful for communication system designers towards practical implementation of multiple access methods in integrated wireless systems. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the problem of wireless service provisioning through a rotary-wing UAV which can serve as an aerial base station (BS) to communicate with multiple ground terminals (GTs) in a boost demand area. Our objective is to optimize the UAV control for maximizing the UAV.s energy efficiency, in which both aerodynamic energy and communication energy are considered while ensuring the communication requirements for each GT and backhaul link between the UAV and the terrestrial BS. The mobility of the UAV and GTs lead to time-varying channel conditions that make the environment dynamic. We formulate a nonconvex optimization for controlling the UAV considering the practical angle-dependent Rician fading channels between the UAV and GTs, and between the UAV and the terrestrial BS. Traditional optimization approaches are not able to handle the dynamic environment and high complexity of the problem in real-time. We propose to use a deep reinforcement learning-based approach namely Deep Deterministic Policy Gradient (DDPG) to solve the formulated nonconvex problem of UAV control with continuous action space that takes into account the real-time of the environment including time-varying UAV-ground channel conditions, available onboard energy of the UAV, and the communication requirement of the GTs. However, the DDPG method may not achieve good performance in an unstable environment and will face a large number of hyperparameters. We extend our approach to use the Trust Region Policy Optimization (TRPO) method that can improve the performance of the UAV compared to the DDPG method in such a dynamic environment. [ABSTRACT FROM AUTHOR]

This paper focuses on the way to protect privacy of clients requesting datasets stored in data servers while keeping communication efficiency. To this end, we introduce a novel communication-efficient and privacy protecting framework termed crowded information acquisition (CIA), well suited to a large number of clients scenario. We investigate the CIA under various conditions addressing possible communication scenarios. Contrary to the conventional belief, the results claim that a large number of clients demanding private services can enhance the privacy protection while providing low latency services and generating a small amount of traffic. [ABSTRACT FROM AUTHOR]

In this paper, a mixture Gamma shadowed (MGS) model is proposed as a unified composite distribution via representing the shadowing impact by an inverse Nakagami- $m$. The exact expression and the asymptotic behaviour at high average signal-to-noise ratio (SNR) regime of the fundamental statistics of a MGS distribution are derived first. These statistics are then applied to analyze the performance of the wireless communication systems over double shadowed $\kappa$ - $\mu$ fading channels. In particular, the outage probability (OP), average bit error probability (ABEP), average channel capacity (ACC), effective capacity (EC) and average area under the receiver operating characteristics curve (AUC) of energy detection (ED) are provided. The numerical and simulation results as well as a comparison with previous exact works are presented to verify the validation of our analysis. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) have been widely applied for classification tasks in wireless communications such as link scheduling, user association and base station (BS) muting. When training a DNN using datasets with highly-skewed class distribution where most data belong to a few majority classes, learning performance for minority classes will degrade since the imbalanced data forces the training process to be biased towards the majority classes. In many wireless problems, multiple decisions for classification are made jointly, say deciding which BSs in a network should be muted to avoid interference. In this paper, we employ DNN to learn an optimal predictive interference coordination policy, and strive to avoid biased learning in such a multi-label multi-class classification problem. The major contribution is proposing a training method to encourage fairness among classes by minimizing the maximal cost of decisions among classes, which is converted into a problem to optimize the weighting factors on the training cost of each class. We provide a gradient descent algorithm to optimize the weighting factors and the model parameters of the DNN by alternative updates. Simulation results show that the training method can improve the learning performance for the minority class and can achieve higher network utility than existing training methods. [ABSTRACT FROM AUTHOR]