Search

This paper proposes a synthetic inertia control (SIC) scheme for a permanent-magnet synchronous generator-based wind turbine generator (WTG). The proposed SIC scheme uses only the frequency deviation loop operating in conjunction with the maximum power point tracking loop. To improve the frequency nadir while preventing over-deceleration of a WTG, the proposed scheme employs a variable control gain for the frequency deviation loop, which varies linearly with the rotor speed. In the high-rotor-speed region, the gain is set to be high so that the frequency nadir can be improved. In the low-rotor-speed region, the gain decreases with the rotor speed so that over-deceleration can be prevented while improving the frequency nadir. To investigate the efficacy of the proposed gain, simulation case studies are conducted in the modified IEEE 14-bus system under varying wind speeds and event sizes. Simulation results indicate that the proposed scheme successfully improves the frequency nadir while preventing over-deceleration of the rotor speed by varying the control gain in a linear manner with the rotor speed.

Recently the idea of using deep learning algorithms in massive multiple-input multiple-output channel state information feedback has been studied in great detail. To use the deep-learning-based feedback schemes over the air, they must be able to operate under extremely large antennas. These schemes also need to be verified under a realistic channel model and propagation environment. Thus, we propose ChannelAttention, a deep-learning-based channel state information feedback scheme that utilizes attention layers and residual blocks. Simulations for a 64-by-64 QuaDRiGa channel model based on the 3GPP 38.901 urban microcell scenario show that ChannelAttention surpasses the normalized mean square error and cosine similarity performance of the conventional CsiNet+ scheme across all compression ratio ranges.

Physical-layer network coding (PNC) is an effective technique for enhancing wireless network throughput. Recently, it has been demonstrated that convolutional autoencoders effectively works in point-to-point communication systems, but their application to wireless relay networks is scarcely explored. In this letter, we propose a convolutional autoencoder for PNC in a two-way relay channel. The constellation mapping and demapping of symbols at each node are determined adaptively through a deep learning technique, such that the bit error rate performance is improved for high-order modulation. Simulation results verify the advantages of the proposed scheme over the conventional PNC scheme for various modulation types.

In a massive machine-type communication (MTC) environment, a traffic control algorithm is essential to solving traffic congestion problems. Additionally, as MTC applications have diverse quality-of-service requirements, traffic control should be applied in a way that guarantees diverse requirements. Therefore, we propose a priority-based access class barring (PACB) scheme, in which MTC devices are divided into multiple classes according to their delay requirements, and a different access class barring (ACB) factor is used for each class. This article suggests an approach to choose multiple ACB factors that can maximize the number of successful access events and can also guarantee the low delay requirement of the high-priority class. Then, an analytical model is proposed for the ideal case, and throughput, drop probability, and access delay are derived to evaluate the performance. For a realistic environment, we also suggest an estimation algorithm for the PACB scheme, in which a base station estimates the number of active devices for each class. Numerical and simulation results showed that the analytical model is consistent with the simulation results. Additionally, the performance of the estimation algorithm was similar to that of the ideal case. Furthermore, the proposed PACB scheme can decrease the delay of a high-priority class and improve the throughput and drop probability compared to existing algorithms.

In this paper, we consider a dynamic time division duplex system based on a cloud radio access network (RAN) to support the ultimate goal of efficient resource utilization. To optimize the performance of both remote radio heads (RRHs) and user equipments (UEs) which have multiple antennas, we propose an iterative scheme consisting of four types of beamforming, namely downlink (DL) transmit beamforming, DL receive beamforming, uplink (UL) transmit beamforming, and UL receive beamforming, in an attempt to minimize mean squared error (MSE). The proposed beamforming scheme makes it possible to allow DL and UL transmission be supported simultaneously by managing interference between RRHs and UEs. Our simulation results show that the MSE and the bit error rate (BER) are greatly improved under the proposed beamforming algorithm compared to conventional schemes. From intensive system level simulations, the proposed algorithm points out more than 3 dB performance improvement in terms of the required signal-to-noise-ratio (SNR) to achieve the BER less than $10^{-3}$ compared to conventional schemes.

In this paper, a means of transmit power control for underlaid device-to-device (D2D) communication is proposed based on deep learning technology. In the proposed scheme, the transmit power of D2D user equipment (DUE) is autonomously learned via a deep neural network such that the weighted sum rate (WSR) of DUEs can be maximized by considering the interference from cellular user equipment. Unlike conventional transmit power control schemes in which complex optimization problems have to be solved in an iterative manner, which possibly requires long computation time, in our proposed scheme the transmit power can be determined with a relatively low computation time. Through simulations, we confirm that the proposed scheme achieves a sufficiently high WSR with a sufficiently low computation time.

A transmit power control strategy using a deep neural network (DNN) is proposed for underlay device-to-device (D2D) communication where D2D user equipment (DUE) shares radio resources with cellular user equipment (CUE). In this scheme, a transmit power control strategy for DUE is found with the aid of a newly proposed DNN structure. Both the spectral efficiency (SE) of the DUE and the amount of interference at the CUE are taken into account, such that the SE of the DUE can be improved while alleviating any deterioration in the cellular transmission. Using simulations, we show that the proposed scheme can achieve a high SE of the DUE while properly regulating the interference caused to the CUE, with a low computation time.

In this letter, we propose a new synchronous duplexing scheme for a long distance wireless backhaul system. In conventional time-division duplexing, a backhaul system can transmit its data only after the transmission of the opposite backhaul system is finished. Thus, any propagation delay caused by long distances leads to a long guard time. However, in our proposed scheme, both local and remote backhaul systems transmit their down and up-link data to each other simultaneously. Thus, even for a long propagation delay, the time for which the channel is not used, which is called the feedback delay, can be effectively reduced. We conduct a mathematical analysis to derive the theoretical efficiency of the proposed scheme and show simulation results indicating the superiority of the proposed scheme in terms of synchronization error and propagation delay.

In this paper, we propose a novel method of managing a semidynamic cluster through the use of a reinforcement learning. We derive some concepts from reinforcement learning that could be suitable for cooperative networks. We also verify the performance of proposed algorithm by means of a simulation, in which we examined spectral efficiency and convergence properties. The proposed algorithm represents a considerable improvement for edge users in particular. In addition, we analyze the complexity of the clustering schemes. Our proposed algorithm is effective in the environment where there is a limited computational resource.

In energy-limited networks, battery-powered nodes suffer from energy famine, which can reduce network lifetime and affect the robustness of networks. To alleviate an energy problem, it is possible to harvest energy from ambient radio frequency signals. In this paper, we consider a proportional fair energy efficiency, which jointly considers energy efficiency and fairness in energy-harvesting-based wireless networks. We formulate a nonconvex optimization problem for solving subchannel and power allocation in order to maximize proportional fair energy efficiency. Using nonlinear fractional programming, we transform the optimization problem into a tractable convex problem. We also derive the solution of the transformed problem and propose a resource allocation algorithm using an iterative method. In addition, we prove the convergence of the proposed algorithm in view of a suboptimal point. Through intensive simulations, we compare the performance of our proposed algorithm with those of conventional algorithms. It is shown that the proposed algorithm improves fairness considerably while maintaining energy efficiency, compared with conventional algorithms.

In this paper, we try to solve the imbalance problem of communication data service and energy supply among receivers in energy harvesting based wireless networks, with receivers equipped with modules that could either decode information or harvest energy from radio frequency signals. We formulate a biconvex optimization problem and find the solutions for subchannel and power allocation. Based on the obtained solutions, we propose a proportional fair resource allocation algorithm using an iterative method, and also show the convergence of the proposed algorithm to the partial optimum solution. Simulation results show that the proposed algorithm improves, to a remarkable extent, fairness among the receivers in terms of both the data rate and the harvested energy. At the same time, the proposed algorithm is found to lead to a negligible loss of system throughput compared with other conventional algorithms.

We propose a new cooperative spectrum sensing for cognitive radio (CR) systems that is designed to detect low-power primary users (PUs). In previous works, it has been assumed that all cognitive terminals (CTs) are within the no-talk zone of PU, which is the region that a CT should not use the band of PU when it lies within the region. However, when the PU transmits at low power, only some CTs will lie within the no-talk zone. It is therefore clear that low-power PUs cannot be protected adequately when conventional sensing schemes are used. To solve this problem, we propose a new scheme in which CTs perform sensing in a number of different stages. Through performance analysis and simulations, we show that our proposed scheme can detect low-power PUs with high accuracy, with the result that the PUs can be protected properly. We also show that the throughput of CR systems can be improved.

When the cells in a location area are sequentially paged based on specific information, such as the last registered area or a mobile speed, the paging load may be non-uniformly distributed among the cells. This non-uniform paging traffic causes an additional paging delay due to the increased waiting time in cells that have a high paging load. In this paper, we introduce a new paging strategy in which the paging sequence in a location area is optimized according to both the location probability of a mobile terminal and the paging load distribution among the cells. In addition, we propose a simple polynomial-time heuristic algorithm to determine sub-optimal paging sequence. Numerical results show that our proposed strategy has almost an equivalent optimal performance and outperforms the conventional paging scheme with respect to a paging delay.When the cells in a location area are sequentially paged based on specific information, such as the last registered area or a mobile speed, the paging load may be non-uniformly distributed among the cells. This non-uniform paging traffic causes an additional paging delay due to the increased waiting time in cells that have a high paging load. In this paper, we introduce a new paging strategy in which the paging sequence in a location area is optimized according to both the location probability of a mobile terminal and the paging load distribution among the cells. In addition, we propose a simple polynomial-time heuristic algorithm to determine sub-optimal paging sequence. Numerical results show that our proposed strategy has almost an equivalent optimal performance and outperforms the conventional paging scheme with respect to a paging delay.

With developments in voice over IP (VoIP), IP-based wireless data networks and their application services have received increased attention. While multimedia applications of mobile nodes are served by Session Initiation Protocol (SIP) as a signaling protocol, the mobility of mobile nodes may be supported via Mobile IP protocol. For a mobile node that uses both Mobile IP and SIP, there is a severe redundant registration overhead because the mobile node has to make location registration separately to a home agent for Mobile IP and to a home registrar for SIP, respectively. Therefore, we propose two new schemes that integrate mobility management functionality in Mobile IP and SIP. We show performance comparisons among the previous method, which makes separate registration for Mobile IP and SIP without integration, and our two integrated methods. Numerical results show that the proposed methods efficiently reduce the amount of signaling messages and delay time related to the idle handoff and the active handoff.

This paper analyzes the other-cell interference of forward link channel in multicode CDMA system supporting a voice and data service. In this system, voice communication can be served by a basic code and data service can be supported by multiple codes. Using this analysis, we can get the approximated capacity of forward link. We describe the effect of data user activity and received power level ratio between a data service and a voice service. Also, this paper describes the forward link capacity and the power ratio of data power to voice power according to the change of FER threshold for data service.

A hierarchical group location tracking (HGLT) based on grouplocation tracking and hierarchical location update is proposed toreduce the location management cost for terrestrialtransportation systems (TSs) in personal communication networks. A global group location update (GGLU) and a local group locationupdate (LGLU) are performed to provide an accurate position for auser. The GGLU informs a virtual visitor location register of thecurrently camped-on location area and the LGLU informs the servingmobile switching center of the currently camped-on cell. Althoughthe LGLU is done each time a TS moves into a new cell, it causessmall increment in the signaling cost due to localized updates andgroup updates compared with the previous group location trackingscheme. The proposed scheme significantly reduces the paging costbecause the scheme pages only a camped-on cell using the LGLUinformation. Compared with the previous schemes, the HGLT reducesthe location management cost, and it is more efficient as thecost for paging a cell increases.

In this paper, the speed of Mobile Station (MS) is estimated based on the derivative and correlation properties of received wave envelope in additive noise and Rayleigh fading channel environments. We derive the effects of noise and nonzero sampling interval on the accuracy of the speed estimation. In addition, we propose a method to correct the estimation errors due to the noise and sampling interval. We compare the proposed method with the speed estimation using level crossing rate. The proposed estimation method shows better performance than the estimation method using level crossing rate when there is background noise.

Many schemes to reduce the inter-cluster handoff delay in wireless ATMnetworks have been proposed, but the previous schemes waste relativelylarge network resources to decrease the path rerouting delay. In thispaper, we propose the 2-layered cluster concept, where the seamlessinter-cluster handoff can be supported regardless of path reroutingtime. As a result, the waste of wired resources and the rate of theinter-cluster handoff can be reduced. From the performance analysisand simulation, the inter-cluster handoff rate for non-real-timetraffic is only about 1/3 of the conventional result. Suchadvantageous features of the proposed scheme neither incur anyincrease of the total handoff rate nor require additional ATMswitches.

In a wireless ATM network for mobile multimedia services, conventional signaling protocols generate heavy traffic because the signaling load must be handled in a HLR (Home Location Register). This centralized structure of the wireless ATM network causes critical connection setup delays. Thus, distributed processing based on a reduction of the connection setup delays is needed in wireless ATM networks. A cache strategy for call delivery with cache updates of registration based on ATM multicasting is introduced with a comparison of the cost of cache scheme with the cost of a conventional scheme. Results show that the cache scheme has better performance than conventional methods when portable mobility is low with large traffic density.

A hierarchical group location tracking (HGLT) is proposed to reduce the location management cost for transportation systems in personal communication services networks. The HGLT performs global group location updates and local group location updates (LGLUs) to obtain an accurate user location. Small increment in the signalling cost due to the LGLUs results in significant reduction in paging cost. The HGLT is an appropriate solution in high paging cost systems.

Energy efficiency is an important issue in ad hoc wireless networks since the battery power of mobiles is limited. The authors have previously proposed an energy efficient medium access control (MAC) algorithm based on reservation and scheduling for ad hoc wireless networks. Here, the fair threshold energy level is derived by mathematical analysis for pseudo base station reselection in the energy efficient MAC protocol.

A new distributed heuristic clustering scheme that minimises the required transmission power in two-tiered mobile ad hoc networks is proposed. This scheme, which can be implemented and executed in real-time, delivers performance similar to optimum results.

A new multiclass rate control method for real-time voice traffic transmission in packet networks is proposed. The class for rate control is defined according to the requested voice quality. In the proposed control mechanism, a source coding rate is adjusted based on network feedback information and class type. It is shown that, using the proposed method, multiple voice qualities can be supported without voice quality degradation in traffic with low priority.