Search

In vehicular edge computing (VEC), some tasks can be processed either locally or on the mobile edge computing (MEC) server at a base station (BS) or a nearby vehicle. In fact, tasks are offloaded or not, based on the status of vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication. In this paper, device-to-device (D2D)-based V2V communication and multiple-input multiple-output and nonorthogonal multiple access (MIMO-NOMA)-based V2I communication are considered. In actual communication scenarios, the channel conditions for MIMO-NOMA-based V2I communication are uncertain, and the task arrival is random, leading to a highly complex environment for VEC systems. To solve this problem, we propose a power allocation scheme based on decentralized deep reinforcement learning (DRL). Since the action space is continuous, we employ the deep deterministic policy gradient (DDPG) algorithm to obtain the optimal policy. Extensive experiments demonstrate that our proposed approach with DRL and DDPG outperforms existing greedy strategies in terms of power consumption and reward. [ABSTRACT FROM AUTHOR]

In high-speed railway communication systems, the distributed antenna systems are usually employed to mitigate frequent handover and enhance the signal-to-noise ratio to receivers. In this case, jointly optimizing downlink power allocation with antenna selection (PAWAS) can enhance system energy efficiency, while the channel state and traffic density are taken into account. Besides, two typical kinds of terrains with sparse and rich scatterings and three traffic patterns including delay-sensitive, -insensitive, and hybrid traffics are investigated in this paper. We show that severe small-scale fading decreases the ergodic capacity, which is quantitatively analyzed, and proved to be proportional to the number of selected transmit antennas. In addition, in case of delay-sensitive traffic, we show that the PAWAS can be viewed as generalized channel-inversion associated with transmit antenna selection. In case of delay-insensitive traffic, we show that when multiple antennas are selected, the power allocation can be viewed as channel-inversion, whereas when single antenna is selected, it is traditional waterfilling. In case of hybrid traffic, we prove that the optimal PAWAS method can be given by separately solving the PAWAS of its delay-sensitive and -insensitive parts. Simulation results validate our theoretical results and demonstrate that proposed PAWAS can minimize the average transmit power in cases of arbitrary traffic density and channel states. [ABSTRACT FROM AUTHOR]

Interference has been a key challenge to wireless networks. As a fundamental transmission unit, the Gaussian interference channel (GIC) provides much insight on understanding the optimal transmission policy and the transmission limit over wireless networks. This paper investigates the multiple-layer power allocation of GIC that maximizes the system sum-rate. First, we derive the optimal signal-layer power allocation and the corresponding sum-rate in closed form for all cases of weak interference GIC based on the rate splitting scheme. Theoretical result indicates that: 1) In low power and asymmetric power regimes, the rate splitting scheme degrades to the pure public or private message transmission at transmitters and simple successive decoding process is efficient enough at receivers and 2) the signal-layer sum-rate is not concave for weak interference GIC and a frequency division scheme brings strict positive sum-rate gain for some power constraints. Second, we specify the relationship between the optimal bandwidth-power allocation of the frequency division scheme and the concave envelope of signal-layer sum-rate in the subband-layer. Finally, considering general GIC with time-varying flat fading, we present an optimal channel-state-layer power allocation associated with rate splitting and frequency division. Numerical results demonstrate that the comprehensive power allocation over the signal-layer, the frequency subbands, and the channel states can largely increase the sum-rate of weak interference GIC in most scenarios. [ABSTRACT FROM PUBLISHER]

Decode and forward (DF) and compress and forward (CF) are two efficient cooperation strategies for relay networks. The combination of the DF and CF strategies is promising to improve the achievable rate of cooperative communication systems. To thoroughly aggregate the advantage of the DF and CF strategies, we put forward a joint power allocation and strategy selection (PASS) scheme for the half-duplex relay channel. The PASS scheme is shown to achieve a higher rate compared with a regular hybrid DF/CF scheme. The rate gain obtained in the static relay channel results from actively adjusting and optimizing the relay power, followed by selecting the better strategy between DF and CF. In particular, when the relay receives and transmits in sequential time slots, we characterize the positive rate improvement area of the PASS scheme analytically and give a near-optimal setting for approaching the best rate performance. The PASS scheme is extended to fading relay channels. We show that the rate increment obtained from the PASS scheme is further magnified by employing an advanced relay power allocation technique over channel states. The corresponding optimal power allocation is established based on the concavity of the rate achieved by the PASS scheme in static relay channel. Numerical results are presented to validate the analysis. [ABSTRACT FROM PUBLISHER]

Widespread deployment of high-speed railways in recent years has created huge demand for high-mobility broadband wireless communications. To provide broadband wireless access for passengers on the train, a well-acknowledged approach is to apply a two-hop architecture, under which passengers communicate with base stations (BSs) via an access point (AP) installed in the train. We consider the uplink transmission from the AP to the BSs on the ground along the railway. The key problem we discuss is how to match user-data arrival process and time-varying channel service process with a delay constraint at the AP. We first assume the constant-rate data arrival and the wireless channel affected only by large-scale fading in high-speed railway scenarios. We present the optimal power-allocation policy given the delay constraint. Based on the policy, we show that there exists a power–delay tradeoff in our system model, with the delay within a limited range. Besides, we find that the average transmit power and the train velocity have a tradeoff, which is similar to the power–delay tradeoff. Besides, given average transmit power constraint, the delay constraint is inversely proportional to the train velocity. Finally, when small-scale fading is considered, we modify the optimal power-allocation policy and provide a convenient approach to counteract Nakagami- $m$ fading in high signal-to-noise ratio (SNR) regimes. The modified power-allocation policy can be well adapted to different fading statistics in different terrestrial environments experienced by the high-speed train. [ABSTRACT FROM PUBLISHER]

In this paper, we consider one quality-of-service (QoS) criterion, average packet delay (queueing delay plus service time), in a multiaccess system and investigate the basic problem whether a multiaccess system can meet the different average packet delay requirements of all users by combining information theory with queueing theory. Two different cases of the central scheduler with and without queue state information (QSI) are discussed. If the QSI is not available to the central scheduler, we show that static rate allocation policies (SRAPs) can achieve better average packet delay performance than probabilistic rate allocation policies. Based on this conclusion, the delay feasibility checking process reduces to checking whether the required service rate vector lies in the multiaccess capacity region. We find that for users with equal transmit power, only $N$ inequalities are necessary for the checking process, whereas for users with unequal transmit powers, we provide a polynomial-time algorithm for such a decision. Furthermore, if the system cannot satisfy the average packet delay requirements of all users, we prove that as long as the sum power is larger than a threshold, there is always an approach to adjust the transmit powers of different users to satisfy the average packet delay requirements. On the other hand, if the QSI is available to the central scheduler, we propose two dynamic scheduling algorithms to achieve proportional average packet delay and compare their performances with optimal SRAP by simulations. [ABSTRACT FROM AUTHOR]