Opportunistic delay‐margin‐based resource allocation for next‐generation wireless networks

Abstract

This paper studies and develops efficient traffic management techniques for downlink transmission at the base station (BS) of multi-service IP-based networks by combining quality-of-service (QoS) provision and opportunistic wireless resource allocation. A delay-margin-based scheduling (DMS) for downlink traffic flows based on the delays that each packet has experienced up to the BS is proposed. The instantaneous delay margin, represented by the difference between the required and instantaneous delays, quantifies how urgent the packet is, and thus it can determine the queuing priority that should be given to the packet. The proposed DMS is further integrated with the opportunistic scheduling (OPS) to develop various queueing architectures to increase the wireless channel bandwidth efficiency. Different proposed integration approaches are investigated and compared in terms of delay outage probability and wireless channel bandwidth efficiency by simulation.are presented in Sections 4 and 5, respectively. Illustrative results on performance of these proposed schemes are presented and discussed in Section 6. Finally, Section 7 provides concluding remarks.

Maximization of user satisfaction in OFDMA systems using utility‐based resource allocation

“…In order to increase the percentage of satisfied users in a scenario with RT services, RRA techniques should take into account efficiency in the resource usage and QoS guarantees (delay bounds). We have classified some works that dealt with both factors in three main approaches: heuristics [4,9], cross-layer packet scheduling (PSC) [10][11][12][13][14], and utility theory [15][16][17].…”

“…The opportunistic PSC algorithms suitable for RT services found in the literature have priority functions that use an efficiency indicator, such as the instantaneous transmission rate (rate maximization policy) [10][11][12] or the ratio between the instantaneous transmission rate and throughput (proportional fairness policy) [12][13][14], and a QoS indicator based on delay. The idea behind these algorithms is not only using the resources in the most efficient way but also giving priority to users with poorer QoS (higher delays).…”

“…The proposed framework utilizes a suitable combination between the efficiency and QoS indicators, which is governed by the chosen utility function. In contrast to other RRA techniques based on heuristics [3,4,9], or common sense [5,[10][11][12][13][14][15], our framework uses an RRA policy derived from a mathematical formulation.…”

“…The corresponding DRA algorithm, which is given by (8), must use the particular expression of w j presented in (10). Note that expression (10) is easy to calculate, which has a positive impact on the computational complexity of the utility-based RRA framework.…”