A Cross-layer Dual Queue Approach for Improving TCP Fairness in Infrastructure WLANs

Fairness is one of the most important performance measures in IEEE 802.11 Wireless Local Area Networks (WLANs), where channel is accessed through competition. In this paper, we focus on the fairness problem between TCP uplink and downlink flows in infrastructure WLANs from the cross-layer perspective. First, we show that there exists a notable discrepancy between throughput of uplink flow and that of downlink flow, and discuss its root cause from the standpoint of different responses to TCP data packet drop and TCP ACK packet drop at the access point (AP) buffer. In order to mitigate this unfairness, we propose a dual queue scheme, which works in a cross-layer manner. It employs two separate queues at the AP, one for the data packets of downlink TCP flows and another for the ACK packets of uplink TCP flows, and selects these queues with appropriate probabilities so that TCP per-flow fairness is improved. Moreover, we analyze the behavior of the dual queue scheme and derive throughputs of uplink and downlink flows. Based on this analysis, we obtain the optimal queue selection probabilities for fairness. Extensive simulation results confirm that the proposed scheme is effective and useful in resolving the TCP unfairness problem without deteriorating overall utilization.     

Fair and efficient Transmission Control Protocol access in the IEEE 802.11 infrastructure basic service set

When the stations in an IEEE 802.11 infrastructure basic service set employ Transmission Control Protocol (TCP), this exacerbates per‐flow unfair access problem. We propose a novel analytical model to approximately calculate the maximum per‐flow TCP congestion window limit that prevents packet losses at the access point buffer and therefore provides fair TCP access both in the downlink and uplink. The proposed analysis is unique in considering the effects of varying number of uplink and downlink TCP flows, differing round trip times among TCP connections and the use of delayed TCP acknowledgment (ACK) mechanism. Motivated by the findings of this theoretical analysis and simulations, we design a link layer access control block to be employed only at the access point in order to resolve the unfair access problem. The proposed link layer access control block uses congestion control and ACK filtering approach by prioritizing the access of TCP data packets of downlink flows over TCP ACK packets of uplink flows. Via simulations, we show that the proposed algorithm can provide both short‐term and long‐term fair accesses while improving channel utilization and access delay. Copyright © 2013 John Wiley & Sons, Ltd.     

iVoIP: an intelligent bandwidth management scheme for VoIP in WLANs

Voice over Internet Protocol (VoIP) has been widely used by many mobile consumer devices in IEEE 802.11 wireless local area networks (WLAN) due to its low cost and convenience. However, delays of all VoIP flows dramatically increase when network capacity is approached. Additionally, unfair traffic distribution between downlink and uplink flows in WLANs impacts the perceived VoIP quality. This paper proposes an intelligent bandwidth management scheme for VoIP services (iVoIP) that improves bandwidth utilization and provides fair downlink–uplink channel access. iVoIP is a cross-layer solution which includes two components: (1) iVoIP-Admission Control, which protects the quality of existing flows and increases the utilization of wireless network resources; (2) iVoIP-Fairness scheme, which balances the channel access opportunity between access point (AP) and wireless stations. iVoIP-Admission Control limits the number of VoIP flows based on an estimation of VoIP capacity. iVoIP-Fairness implements a contention window adaptation scheme at AP which uses stereotypes and considers several major quality of service parameters to balance the network access of downlink and uplink flows, respectively. Extensive simulations and real tests have been performed, demonstrating that iVoIP has both very good VoIP capacity estimation and admission control results. Additionally, iVoIP improves the downlink/uplink fairness level in terms of throughput, delay, loss, and VoIP quality.     

移动WiMAX网络下行链路TCP吞吐量公平性的跨层优化

该文提出了一种移动WiMAX网络中的TCP公平性跨层优化模型,设计了基于此模型的cross-layer TCP改进协议。利用端节点链路层的速率信息和在基站BS(Base Station)处预分配ACK分组所需带宽的策略改善了下行终端的QoS (Quality of Service)指标,保证了下行终端和上行终端的吞吐量公平性。仿真结果表明cross- layer-TCP改进协议能够在保持原有系统吞吐量的前提下改善下行终端的公平性。     

TCP-aware bidirectional bandwidth allocation in IEEE 802.16 networks

In this paper, we propose a bidirectional bandwidth-allocation mechanism to improve TCP performance in the IEEE 802.16 broadband wireless access networks. By coupling the bandwidth allocation for uplink and downlink connections, the proposed mechanism increases the throughput of the downlink TCP flow and it enhances the efficiency of uplink bandwidth allocation for the TCP acknowledgment (ACK). According to the IEEE 802.16 standard, when serving a downlink TCP flow, the transmission of the uplink ACK, which is performed over a separate unidirectional connection, incurs additional bandwidth-request/allocation delay. Thus, it increases the round trip time of the downlink TCP flow and results in the decrease of throughput accordingly. First, we derive an analytical model to investigate the effect of the uplink bandwidth-request/allocation delay on the downlink TCP throughput. Second, we propose a simple, yet effective, bidirectional bandwidth-allocation scheme that combines proactive bandwidth allocation with piggyback bandwidth request. The proposed scheme reduces unnecessary bandwidth-request delay and the relevant signaling overhead due to proactive allocation; meanwhile, it maintains high efficiency of uplink bandwidth usage by using piggyback request. Moreover, our proposed scheme is quite simple and practical; it can be simply implemented in the base station without requiring any modification in the subscriber stations or resorting to any cross-layer signaling mechanisms. The simulation results ascertain that the proposed approach significantly increases the downlink TCP throughput and the uplink bandwidth efficiency.     

Improving TCP/IP Performance over Third-Generation Wireless Networks

As third-generation (3G) wireless networks with high data rate get widely deployed, optimizing the transmission control protocol (TCP) performance over these networks would have a broad and significant impact on data application performance. In this paper, we make two main contributions. First, one of the biggest challenges in optimizing the TCP performance over the 3G wireless networks is adapting to the significant delay and rate variations over the wireless channel. We present window regulator algorithms that use the receiver window field in the acknowledgment (ACK) packets to convey the instantaneous wireless channel conditions to the TCP source and an ACK buffer to absorb the channel variations, thereby maximizing long-lived TCP performance. It improves the performance of TCP selective ACK (SACK) by up to 100 percent over a simple drop-tail policy, with small buffer sizes at the congested router. Second, we present a wireless channel and TCP-aware scheduling and buffer sharing algorithm that reduces the latency of short flows while still exploiting user diversity for a wide range of user and traffic mix.     

Rate selection based medium access control for full-duplex asymmetric transmission

The realization of full-duplex wireless communication is predictable. And asymmetric transmission is a practical and low-cost application scenario, where full-duplex access point (FD_AP) can communicate with two users simultaneously to receive and send packets. While, in an asymmetric transmission, the transmit power of uplink sender decides the uplink and downlink rates because of the inter-client interference, which accordingly restricts the throughput. Besides, the size of packets in uplink and downlink is generally unequal. Therefore, a WIFI network with a FD_AP and half-duplex users is studied in this paper, and a medium access control (MAC) protocol based on power control and rate selection (PCRS) is proposed. PCRS MAC employs a received signal strength based rate selection strategy to select different rates and power for uplink and downlink transmission. Then, FD_AP can establish efficient and reliable full-duplex asymmetric transmission. Simulation results show that PCRS can improve the throughput and the probability of successful asymmetric communication as compared to the distributed coordination function (DCF) and a simple full-duplex MAC protocol without PCRS. Besides, PCRS MAC also maintains a high level of fairness.

     

LTE系统上行链路调度算法研究

长期演进(Long Term Evolution,LTE)已经成为4G无线技术标准。目前,LTE分组调度的下行链路调度被大多数研究者研究,上行链路的研究相对较少。针对上行链路调度无法保证实时业务分组在延迟期限内传输,存在公平性较差、分组丢弃多的问题。因此,提出了一种新的上行链路调度算法。该算法根据实时业务的延迟约束条件建立目标整数线性规划模型,再根据目标整数线性规划模型进行调度。实验结果表明,该算法能保证实时业务分组在延迟期限内传输,适用于实时业务,能确保公平性,最小化分组丢弃,具有较好的适用性。     

Enhancing Fairness for Short-Lived TCP Flows in 802.11b WLANs

The problem of providing throughput fairness in a wired-cum-wireless network where the wireless portion is an 802.11 wireless local area network (WLAN) is addressed. Due to the distributed nature of the primary 802.11 media access control protocol and the unpredictability of the wireless channel, quality of service guarantees in general and fairness in particular are hard to achieve in WLANs. This fact seriously compromises the interaction between 802.11-based networks and well-established architectures such as DiffServ. The focus of this paper is on transmission control protocol (TCP) traffic, and two fundamental problems related to throughput fairness are identified. First, the basic requirement of providing fair access to all users conflicts with the nature of TCP, which is fair only under certain conditions and hardly met by 802.11b WLANs. Second, short-lived TCP flows that are sensitive to losses during the early stages of TCP window growth need to be protected. To address these issues, a logical-link-control-layer algorithm that can be implemented at both access points and wireless stations is proposed. The algorithm aims at guaranteeing fair access to the medium to every user, independent of their channel conditions. At the same time, the proposed scheme protects short-lived flows, while they strive to get past the critical "small window regime." A simulation study that shows the effectiveness of the new algorithm in comparison to the standard 802.11b implementation is presented     

Optimizing uplink TCP-ACK transmission in WiMAZ OFDMA systems

In WiMAX networks, a bandwidth request-grant process is required for uplink operation. However, the process is not optimized for TCP, since the uplink ACK stream is disrupted due to the following: (a) The process consists of several stages which in turn introduce big transmission delay; and (b) it requires additional uplink bandwidth which is significant compared to the ACK packet size. In this paper, the authors propose a new ACK transmission scheme, where ACK packets are combined with a bandwidth request (BR) header. Through simulation, it is demonstrated that the proposed scheme reduces the overheads of TCP-ACK transmissions effectively.     

Performance improvement of TCP in ad hoc networks by mitigating channel contention

In ad hoc networks, the spatial reuse property limits the number of packets which can be spatially transmitted over a path. In standard Transmission Control Protocol (TCP), however, a TCP sender keeps transmitting packets without taking into account this property. This causes heavy contention for the wireless channel, resulting in the performance degradation of TCP flows. Hence, two techniques have been proposed independently in order to reduce the contention. First, a TCP sender utilizes a congestion window limit (CWL), by considering the spatial reuse property. This prevents the TCP sender from transmitting more than CWL number of packets at one time. Second, a delayed ack (DA) strategy is exploited in order to mitigate the contention between the TCP ACK and DATA packets. Recently, although TCP‐DAA (Dynamic Adaptive Acknowledgment) attempts to utilize a CWL‐based DA strategy, TCP‐DAA overlooks a dynamic correlation between these two techniques. This paper, therefore, reveals the dynamic correlation and also proposes a protocol which not only reduces the frequency of the TCP ACK transmissions but also determines a CWL value dynamically, according to network conditions. Simulation studies show that our protocol performs the best in various scenarios, as compared to TCP‐DAA and standard TCP (such as TCP‐NewReno). Copyright © 2009 John Wiley & Sons, Ltd.     

A Novel Bandwidth Allocation Algorithm for IEEE 802.16 TDD Mode Wireless Access Networks

In this paper, we propose a novel bandwidth allocation algorithm for a two-tier hierarchy in IEEE 802.16 time division duplex mode wireless access networks under symmetric and/or asymmetric uplink and downlink traffic input. We demonstrate the performance of the new bandwidth allocation algorithm in terms of accumulated throughput (cumulative bandwidth) and fairness in both infinite and finite buffer cases compared with others by simulations. The simulation results show that the proposed algorithm not only can provide much better fairness and maintain satisfactory QoS support and high cumulative bandwidth but also in the case of finite buffer depth is less buffer-consuming than the others, meaning that the hardware cost can be reduced by employing the proposed algorithm.     

Design of an enhanced access point to optimize TCP performance in Wi-Fi hotspot networks

In the last years, the number of Wi-Fi hotspots at public venues has undergone a substantial growth, promoting the WLAN technologies as the ubiquitous solution to provide high-speed wireless connectivity in public areas. However, the adoption of a random access CSMA-based paradigm for the 802.11 MAC protocol makes difficult to ensure high throughput and a fair allocation of radio resources in 802.11-based WLANs. In this paper we evaluate extensively via simulations the interaction between the flow control mechanisms implemented at the TCP layer and the contention avoidance techniques used at the 802.11 MAC layer. We conducted our study considering initially M wireless stations performing downloads from the Internet. From our results, we observed that the TCP downlink throughput is not limited by the collision events, but by the inability of the MAC protocol to assign a higher chance of accessing the channel to the base station. We propose a simple and easy to implement modification of the base station’s behavior with the purpose of increasing the TCP throughput reducing useless MAC protocol overheads. With our scheme, the base station is allowed to transmit periodically bursts of data frames towards the mobile hosts. We design a resource allocation protocol aimed at maximizing the success probability of the uplink transmissions by dynamically adapting the burst length to the collision probability estimated by the base station. By its design, our scheme is also beneficial to achieve a fairer allocation of the channel bandwidth among the downlink and uplink flows, and among TCP and UDP flows. Simulation results confirm both the improvement in the TCP downlink throughput and the reduction of system unfairness.     

Queuing Scheme for Improved Downlink Throughput on WLANs

A transmission queuing scheme is described that increases downlink throughput on wireless local area networks (WLANs) while also increasing the total throughput. When the amount of uplink traffic increases on a WLAN, the carrier sense multiple access with collision avoidance (CSMA/CA) protocol, which is the prescribed scheme for IEEE 802.11 WLAN channel access, may substantially reduce the rate of downlink data frame transmission. This results in severe throughput degradation for mobile stations with downlink traffic. The proposed scheme comprises a transmission control function based on consecutive transmission, as described in the IEEE 802.11e standard, and a dynamic queue prioritization algorithm. Simulation results demonstrate that the proposed scheme increases the maximum total throughput for uplink and downlink traffic by 17% compared with the conventional distributed coordination function (DCF) scheme and that it reduces the difference between uplink and downlink throughput. In an environment where transmission errors occur, the difference in throughput is reduced by about 50% compared with the conventional schemes.     

Adaptive downlink and uplink channel split ratio determination for TCP-based best effort traffic in TDD-based WiMAX networks

In this paper, we study the determination of downlink (DL) and uplink (UL) channel split ratio for Time Division Duplex (TDD)-based IEEE 802.16 (WiMAX) wireless networks. In a TDD system, uplink and downlink transmissions share the same frequency at different time intervals. The TDD framing in WiMAX is adaptive in the sense that the downlink to uplink bandwidth ratio may vary with time. In this work, we focus on TCP based traffic and explore the impact of improper bandwidth allocation to DL and UL channels on the performance of TCP. We then propose an Adaptive Split Ratio (ASR) scheme which adjusts the bandwidth ratio of DL to UL adaptively according to the current traffic profile, wireless interference, and transport layer parameters, so as to maximize the aggregate throughput of TCP based traffic. Our scheme can also cooperate with the Base Station (BS) scheduler to throttle the TCP source when acknowledgements (ACKs) are transmitted infrequently. The performance of the proposed ASR scheme is validated via ns^-2 simulations. The results show that our scheme outperforms static allocation (such as the default value specified in the WiMAX standard and other possible settings in existing access networks) in terms of higher aggregate throughput and better adaptivity to network dynamics.     

Joint utility function-based scheduling for two-way communication services in wireless networks

A type of joint utility function-based scheduling is proposed for two-way communication services in wireless networks. The scheduling of uplink and downlink services is done jointly so that the base station selects a user efficiently and fairly while considering the channel state of both the uplink and the downlink. Because a user generally has two communication links, an uplink and a downlink, the overall satisfaction with a communication service can be formulated as the sum of the quality of the uplink and downlink services. However, most of the previous types of scheduling for the uplink and downlink were designed separately and independently. This paper proposes a joint scheduling algorithm for integrated uplink and downlink services: a base station selects a user while simultaneously considering both the uplink channel state and the downlink channel state. An analytical model is developed for the purpose of determining the scheduling metric, the system throughput, and the level of fairness. The numerical and computer simulation results show that in comparison with conventional proportional fair scheduling the proposed joint scheduling achieves a better throughput while satisfying the fairness among users.     

Differentiated TCP User Perception over Downlink Packet Data Cellular Systems

Current downlink scheduling algorithms in the (enhanced) third-generation (3G) cellular packet systems exploit instantaneous channel status of multiple users, but most of them are blind to traffic information. To improve TCP users' perception of quality-of-services (QoSs), characterized by response delay, goodput, and always-on connectivity, we propose a cross-layer hierarchical scheduler with traffic awareness and channel dependence to properly prioritize buffer and radio resource allocation among different TCP classes. The scheduler has two tiers: at the IP layer, an intrauser scheduler enhances a common practice, i.e., the DiffServ-based buffer management, by dequeuing same-user TCP packets according to per-class specified and measured responsiveness; at the MAC layer, an interuser scheduler transmits the dequeued packets by considering the opportunistic channel states, mean throughput, and class ID of all users. Both tiers consider the online measured throughput, a cross-layer metric, to achieve resource and performance fairness and TCP classification. Experiments show that, compared with (variations of) proportional fairness (PF) and other schemes, our scheduler can notably speed up time-critical interactive TCP services (HTTP and TELNET) or TCP slow-starts with minor cost to bulk file transfer (FTP) or long-lived flows. It offers scalable and low-cost TCP performance enhancement over the emerging cellular systems     

A Network Coding Scheme to Improve Throughput for IEEE 802.11 WLAN

In IEEE 802.11 infrastructure wireless local area network (WLAN), the communication between any two nodes is relayed by an access point (AP), which becomes the bottleneck of WLAN and severely restricts the overall throughput. It is well known that network coding technique is able to greatly improve the throughput of wireless networks. But, the available coding schemes do not make full advantage of channel capacity due to the fact that they pick at most one packet from each data flow for coding and the picked packets may have a great difference in packet size, wasting some channel capacity. To remedy the problem, in this paper, we propose the coding scheme that combines multiple buffered packets in one flow into a larger packet for coding so that the packets participating in coding have close sizes. We formulate an integer programming problem to find the optimal packet coding, which is solved by an optimal algorithm with relative high time complexity together with a heuristic algorithm with relative low time complexity. Simulation results show that the proposed coding scheme is able to greatly improve the throughput of WLAN and the throughput gain increases with the growth of the number of coding flows.     

Using TCP rate control for queue input‐output rate matching

In explicit TCP rate control, the receiver's advertised window size in acknowledgment (ACK) packets can be modified by intermediate network elements to reflect network congestion conditions. The TCP receiver's advertised window (i.e. the receive buffer of a TCP connection) limits the maximum window and consequently the throughput that can be achieved by the sender. Appropriate reduction of the advertised window can control the number of packets allowed to be sent from a TCP source. This paper evaluates the performance of a TCP rate control scheme in which the receiver's advertised window size in ACK packets are modified in a network node in order to match the generated load to the assigned bandwidth in the node. Using simulation and performance metrics such as the packet loss rates and the cumulative number of TCP timeouts, we examine the service improvement provided by the TCP rate control scheme to the users. The modified advertised windows computed in the network elements and the link utilization are also examined. Copyright © 2002 John Wiley & Sons, Ltd.     

An Adaptive Delayed Acknowledgment Strategy to Improve TCP Performance in Multi-hop Wireless Networks

In multi-hop wireless networks, transmission control protocol (TCP) suffers from performance deterioration due to poor wireless channel characteristics. Earlier studies have shown that the small TCP acknowledgments consume as much wireless resources as the long TCP data packets. Moreover, generating an acknowledgment (ACK) for each incoming data packet reduces the performance of TCP. The main factor affecting TCP performance in multi-hop wireless networks is the contention and collision between ACK and data packets that share the same path. Thus, lowering the number of ACKs using the delayed acknowledgment option defined in IETF RFC 1122 will improve TCP performance. However, large cumulative ACKs will induce packet loss due to retransmission time-out at the sender side of TCP. Motivated by this understanding, we propose a new TCP receiver with an adaptive delayed ACK strategy to improve TCP performance in multi-hop wireless networks. Extensive simulations have been done to prove and evaluate our strategy over different topologies. The simulation results demonstrate that our strategy can improve TCP performance significantly.