This paper proposes a network design model, considering data consistency for a delay-sensitive distributed processing system. The data consistency is determined by collating the own state and the states of slave servers. If the state is mismatched with other servers, the rollback process is initiated to modify the state to guarantee data consistency. In the proposed model, the selected servers and the master-slave server pairs are determined to minimize the end-to-end delay and the delay for data consistency. We formulate the proposed model as an integer linear programming problem. We evaluate the delay performance and computation time. We evaluate the proposed model in two network models with two, three, and four slave servers. The proposed model reduces the delay for data consistency by up to 31 percent compared to that of a typical model that collates the status of all servers at one master server. The computation time is a few seconds, which is an acceptable time for network design before service launch. These results indicate that the proposed model is effective for delay-sensitive applications.

Machine-to-Machine (M2M) communication plays a pivotal role in the evolution of Internet of Things (IoT). Cellular networks are considered to be a key enabler for M2M communications, which are originally designed mainly for Human-to-Human (H2H) communications. The introduction of M2M users will cause a series of problems to traditional H2H users, i.e., interference between various traffic. Resource allocation is an effective solution to these problems. In this paper, we consider a shared resource block (RB) and power allocation in an H2H/M2M coexistence scenario, where M2M users are subdivided into delay-tolerant and delay-sensitive types. We first model the RB-power allocation problem as maximization of capacity under Quality-of-Service (QoS) constraints of different types of traffic. Then, a learning framework is introduced, wherein a complex agent is built from simpler subagents, which provides the basis for distributed deployment scheme. Further, we proposed distributed Q-learning based autonomous RB-power allocation algorithm (DQ-ARPA), which enables the machine type network gateways (MTCG) as agents to learn the wireless environment and choose the RB-power autonomously to maximize M2M pairs' capacity while ensuring the QoS requirements of critical services. Simulation results indicates that with an appropriate reward design, our proposed scheme succeeds in reducing the impact of delay-tolerant machine type users on critical services in terms of SINR thresholds and outage ratios.

This work addresses the problem of a fast packet retransmission scheme intended for transporting delay-sensitive flows in a Cooperative Diversity (CD) environment. This cooperative fast retransmission scheme exploits the advantages of the CD environment and hybrid Automatic-Repeat-reQuest (ARQ), while allowing retransmission just one time via a cooperating user (i.e., partner) or via both the sender and the partner simultaneously. Complementary link packets are used for the retransmission whose policy can be adjusted on the basis of the qualities of channels among the sender, the partner and the receiver, as well as the application layer protocol data unit size, using the application throughput as the objective. For this scheme, we first derive the application throughput formulas which are then verified via simulations. Next, the CD-based optimized fast retransmission scheme is shown able to achieve better effective throughput (goodput) than other CD-based or non-CD-based ARQ schemes in various Nakagami-m slow-fading environments. As a result, the proposed scheme should be an excellent fast retransmission mechanism for real-time multimedia transport in many CD environments.

Optical Burst Switching (OBS) is one of the most promising switching technologies for next generation optical networks. As delay-sensitive applications such as Voice-over-IP (VoIP) have recently become popular, OBS networks should guarantee stringent Quality of Service (QoS) requirements for such applications. Thus, this paper proposes an Adaptive Loss-aware Flow Control (ALFC) scheme, which adaptively decides on the burst offset time based on loss-rate information delivered from core nodes for assigning a high priority to delay-sensitive application traffic. The proposed ALFC scheme also controls the upper-bounds of the factors inducing delay and jitter for guaranteeing the delay and jitter requirements of delay-sensitive application traffic. Moreover, a piggybacking method used in the proposed scheme accelerates the guarantee of the loss, delay, and jitter requirements because the response time for flow control can be extremely reduced up to a quarter of the Round Trip Time (RTT) on average while minimizing the signaling overhead. Simulation results show that our mechanism can guarantee a 10-3 loss-rate under any traffic load while offering satisfactory levels of delay and jitter for delay-sensitive applications.

Delay jitter degrades the quality of delay-sensitive live media streaming. We investigate the use of multipath transmission with two paths to reduce delay jitter and, in this paper, propose a nearly equal delay path set configuration (NEED-PC) scheme that further improves the performance of the multipath delay jitter reduction method for delay-sensitive live media streaming. The NEED-PC scheme configures a pair of a maximally node-disjoint paths that have nearly equal path delays and satisfy a given delay constraint. The results of our simulation experiments show that path sets configured by the NEED-PC scheme exhibit better delay jitter reduction characteristics than a conventional scheme that chooses the shortest path as the primary path. We evaluate the performance of path sets configured by the NEED-PC scheme and find that the NEED-PC scheme reduces delay jitter when it is applied to a multipath delay jitter reduction method. We also investigate the trade-off between reduced delay jitter and the increased traffic load incurred by applying multipath transmission to more flows. The results show that the NEED-PC scheme is practically effective even if the amount of additional redundant traffic caused by using multipath transmission is taken into account.

This letter presents a new approach for obtaining the expected waiting time for packets under the drop-head (also called a drop-from-front) scheme for buffer management. The results show that the drop-head scheme is more effective in reducing queueing delays than the drop-tail scheme.

This paper presets a fast computation algorithm for connection admission control for heterogeneous delay sensitive traffic in ATM networks. Cell loss probability is adopted as the measure of quality of service. In our study, cells of each connection are allowed to have two different loss priorities and the aggregate traffic can have more than two. To cope with multiple quality of service requirements, the link capacity is divided into several bands. For simplicity, each traffic source is assumed to alternate between active and idle periods. However, the results can be extended to traffic sources having more than two states. Upper bounds of actual cell loss probabilities are derived based on the bufferless fluidflow model. Numerical results show that the upper bounds are close to the actual cell loss probabilities.