1 Distributed End-to-end Proportional Delay Differentiation in Wireless LAN Yuan Xue, Kai Chen and Klara Nahrstedt Abstract— Providing service differentiation in wireless net- works has attracted much attention in recent research. Existing studies so far have focused on the design of MAC layer protocols to achieve distributed priority scheduling. While these works have been shown to achieve certain differentiation (higher throughput or lower delay for higher priority packets) at MAC layer, they did not consider the problem of service differentiation from end- to-end perspective. Moreover, these approaches did not present a formalized service differentiation model that quantifies the degree of differentiation, which is critical for service class selection and provision. This paper addresses above problems by introducing a clear service differentiation model, proportional service differentiation, to the domain of wireless network which targets at providing proportional delay differentiation in wireless LAN. Due to the unique characteristic of distributed medium sharing, the scheduling algorithm employed in wireline networks can not be applied directly to the context of the wireless network. We argue that delay differentiation in wireless LAN can be only achieved through a joint packet scheduling at network layer and distributed coordination at MAC layer. Hereby, we present a distributed waiting time priority scheduling (DWTP) algorithm which is deployed on a cross-layer architecture. DWTP consists of two tiers: an intra-node predictive WTP scheduler at network layer and an inter-node priority-based distributed coordination at MAC layer. These two tiers coordinate via cross-layer priority mapping to achieve proportional delay differentiation. Simulation results show that the cross-layer DWTP algorithm can effectively achieve proportional delay differentiation in wireless LAN. I. I NTRODUCTION With advances in wireless communications and emergence of diversified multimedia applications, there is a growing need to support quality of service (QoS) in wireless networks. As an important approach to support network QoS, service dif- ferentiation divides network traffic into classes with different priorities, from which applications can choose to meet their QoS requirements. Existing studies on service differentiation in wireless net- works have focused on the design of MAC layer protocols to achieve distributed priority scheduling (e.g. [1], [2], [3], [4], [5], [6], [7]). While these works (e.g. [1], [2], [5], [6], [7]) have been shown to achieve certain differentiation (higher throughput or lower delay for higher priority packets) at MAC layer, they did not consider the problem of service differentia- tion from end-to-end 1 perspective. This is an important issue, Yuan Xue, Kai Chen and Klara Nahrstedt are affiliated with the Department of Computer Science, University of Illinois at Urbana-Champaign. Their email addresses are {xue,kaichen,klara}@cs.uiuc.edu. The Board of Trustees of the University of Illinois is the recipient of a Motorola Partnerships in Research Grant. 1 Here “end-to-end” means the performance measurement between higher layers at the end hosts, not just between their MAC layers. since many crucial performance metrics of an application are defined as end-to-end. Moreover, these approaches (e.g. [3]) did not present a formalized service differentiation model that quantifies the degree of differentiation, which is critical for service class selection and provision. This paper approaches these problems in the context of wireless LAN in two steps. First, it introduces a service differentiation model, proportional service differentiation [8], to the domain of wireless LAN. In the proportional service differentiation model, the performance of a service class is proportional to another according to the ratio of their differentiation parameters. Such a predictive and consistent differentiation behavior can greatly facilitate class selection for applications and provide great flexibility for class provisioning and management. Based on this model, this paper studies the problem of providing proportional delay differentiation in wireless LAN. Here delay is an end-to-end QoS metric which includes both queueing delay and MAC delay. In this paper, we study a fully distributed wireless LAN without the support of a base station. All nodes in the wireless LAN are within the transmission range of each other. Traffic is sent among dif- ferent pairs of nodes and are divided into different service classes. At any time, only one node can access the channel to transmit a packet. Such packet transmission is coordinated in a distributed manner at MAC layer (e.g. via IEEE 802.11 DCF). As a result of such distributed medium sharing, service differentiation is impossible unless nodes cooperate with each other. This is in contrast to wireline networks where a node’s scheduling decision only considers its own packets. In fact, delay differentiation in wireless LAN can only be achieved through a joint packet scheduling at network layer and dis- tributed coordination at MAC layer. Hereby, based on a cross- layer architecture, this paper presents a distributed waiting time priority scheduling (DWTP) algorithm. DWTP consists of two tiers – an intra-node predictive WTP scheduling algorithm at network layer and an inter-node priority-based distributed coordination at MAC layer. These two tiers coordinate via cross-layer priority mapping to achieve the proportional delay differentiation. The contributions of this paper as follows: (1) It is the first paper that considers the proportional differentiation model in wireless LAN from an end-to-end perspective, instead of the performance measurement only at the MAC layer. (2) It presents a solution to this problem by designing a distributed waiting time priority scheduling algorithm, which is deployed on a cross-layer architecture via priority mapping. The rest of the paper is organized as follows. Section II gives background of proportional service differentiation