Deadline-Aware Adaptive Packet Scheduling and Transmission in Cooperative Wireless Networks Lu Zhang * , Yao Yu *‡ , Fei Huang * , Qingyang Song * , Lei Guo * , and Shiqiang Wang † * School of Information Science and Engineering, Northeastern University, Shenyang 110819, P. R. China † Department of Electrical and Electronic Engineering, Imperial College London, SW7 2AZ, United Kingdom Email: luzhang199007@gmail.com, yuyao@ise.neu.edu.cn, huangfeinf@gmail.com, songqingyang@ise.neu.edu.cn, guolei@ise.neu.edu.cn, shiqiang.wang11@imperial.ac.uk Abstract—We study scheduling and transmission of packets with deadline constraints in cooperative wireless networks. The packets which miss their deadlines become useless and have to be dropped. To minimize packet dropping probability, we consider multiple transmission methods and integrate packet scheduling with adaptive transmission method selection. We first introduce an exhaustive search method to obtain the optimal scheduling sequences and the corresponding transmission methods, under different channel conditions. Through observing the optimal results, we propose a heuristic method based on a dynamic graph. Simulation results show that the proposed heuristic method can obtain results which are similar to those achieved with the exhaustive search method, but with low computational complexity. Index Terms—Cooperative wireless networks, packet dropping probability, scheduling, transmission method adaptation. I. I NTRODUCTION With the growth of computational capability and various breakthroughs in wireless technologies, a variety of services with strict delay requirements emerged in commercial wireless networks, such as real-time video communications in mobile cellular systems. In these applications, packets have strict deadline constraints and must arrive at their destinations before their deadlines. Otherwise, they become invalid and will be dropped. The problem becomes more challenging when supporting real-time services in cooperative wireless networks, because the diversity of transmission rates and meth- ods increases the complexity of packet scheduling. Therefore, it is necessary to develop deadline-aware adaptive packet scheduling and transmission schemes in cooperative wireless networks to reduce packet loss. In the literature, there are many studies focusing on tackling the packet scheduling and transmission problem with deadline constraints in cooperative communications [1]–[7]. One of the promising solutions is the adaptive utilization of conventional network coding (CNC). It is because, on one hand, CNC can reduce transmission time by permitting a relay to encode at least two packets, which are received separately from different source nodes, into one packet and broadcast it to destinations for the decoding of the intended packets; on the other hand, the decoding delay in CNC may become serious if the destination nodes cannot receive sufficient packets for decoding [1]. The authors in [2] proposed an adaptive network coding scheme ‡ The corresponding author of this paper is Yao Yu. to optimize the block size of coded packets while scheduling multimedia traffic flows in a time-slotted downlink system in order to maximize the throughput and reduce packets’ waiting time. Instead of emphasizing on network throughput, the authors in [3] focused on dealing with the scheduling problem in the single-hop broadcast system to minimize the number of dropped packets, and an encoding algorithm was investigated based on CNC to find a proper transmission order and the number of encoded packets. However, they assumed a fixed transmission rate. Noticing that more network coding opportunities appear through adjusting the transmission rate, an efficient algorithm was proposed to determine the coding strategy and transmission rate of each transmission in [4] based on [3]. An immediately decodable network coding (IDNC) scheme was developed in [5] to reduce packet loss in video streaming. However, another efficient transmission method, analog network coding (ANC), was ignored in these works. Compared with CNC, ANC can further reduce transmission time by allowing two signals to be transmitted simultaneously from their source nodes and superpose at the relay [8]. Howev- er, ANC has more stringent restrictions on channel conditions and network topologies. Therefore, ANC should be applied adaptively and jointly with other transmission methods, such as CNC, plain routing (PR), and direct transmission (non- relaying, NR). In [9], an adaptive relaying method selection scheme was developed for multi-rate wireless networks, but without considering the deadline constraints of packets. In order to minimize the packet dropping probability in multi-rate cooperative wireless networks, in this paper, we integrate transmission method selection with packet scheduling and propose two deadline-aware methods (exhaustive search and heuristic methods) to decide the optimal transmission methods and sequences of packets. Packet transmissions with- in two-hop topologies are considered in this paper, and the proposed methods can be easily extended to general multi- hop networks. The rest of this paper is organized as follows. In section II, we formulate the optimization problem. Section III shows two schemes to resolve the deadline-aware packet scheduling and transmission problem. Simulation results are shown in Section IV and Section V draws conclusions. © 2014 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.