Juan J. Alcaraz* and Fernando Cerdán

Transcription

1 Int. J. Wireless and Mobile Computing, Vol. X, No. Y, XXXX Using buffer management in 3G radio bearers to enhance end-to-end TCP performance Juan J. Alcaraz* and Fernando Cerdán Department of Information Technologies and Communications, Technical University of Cartagena, Plaza del Hospital,,, Spain *Corresponding author Abstract: TCP performance over 3G links can be enhanced by means of Active Queue Management (AQM) techniques at the downlink buffer of the mobile network link layer (RLC). In this paper we describe Slope Based Discard (), a novel deterministic buffer management technique with a discarding policy based on the observation of the filling rate of the buffer. Compared with the classic Random Early Detection () scheme, has a simpler operation and is best suited to multiplex short- and long-lived flows simultaneously. By means of extensive simulation experiments we compare the performance of TCP over three different RLC buffer management schemes: drop-tail (no AQM), and. The simulation scenarios are configured to evaluate how the end-to-end goodput and delay are influenced by the number of multiplexed flows, the TCP flavour, the buffer size and the wireless channel error rate. Keywords: TCP over 3G links; active queue management; radio link control; cross-layer design. Reference to this paper should be made as follows: Alcaraz, J.J. and Cerdán, F. (XXXX) Using buffer management in 3G radio bearers to enhance end-to-end TCP performance, Int. J. Wireless and Mobile Computing, Vol. X, No. Y, pp.xx xx. Biographical notes: Juan J. Alcaraz obtained his engineering degree in telecommunication from the Polytechnic University of Valencia in 999. His professional experience includes working at Siemens S.A. and at the Spanish 3G operator Xfera Móviles S.A. in the area of network design. He obtained his PhD in 7 from the Polytechnic University of Cartagena, Spain. His research interests include cellular access networks design and performance evaluation. Fernando Cerdán obtained his engineering degree in telecommunication in 994 from the Polytechnic University of Catalonia and received his PhD from the same university in. He is a Full Associate Professor at the Polytechnic University of Cartagena. Most of his published papers are in the field of service integration and QoS in wired IP networks, although his current interest includes security and QoS in 3G and heterogeneous networks. Introduction Third generation cellular networks (3G) provide internet access at data rates of up to 384 Kbit/s for wide area coverage. In contrast with wired networks, wireless links may suffer frequents frame losses due to signal corruption. The link layer protocol of the 3G radio access network, the Radio Link Control (RLC), can recover from frame losses by means of an Automatic Repeat request (ARQ) algorithm, when configured in Acknowledged Mode (TS 4.3). Many internet applications like , web surfing and file downloading rely on TCP for end-to-end data transport. TCP congestion control, originally developed for wired networks, considers a packet loss as an indication of congestion at the network routers that should be handled with a rate reduction in the TCP source. To avoid bandwidth underuse in wireless connections, where propagation errors cause frequent frame losses (Tsaoussidis and Matta, ), the RLC protocol of 3G links should operate in Acknowledged Mode, providing a reliable radio bearer to TCP connections. Many previous works showed that a reliable link layer benefits TCP in this environment (e.g. Xylomenos and Polyzos, 3). However, several characteristics of 3G radio bearers like high and variable latency and buffer overflow of the downlink buffers (Yavuz and Khafizov, ; Inamura et al., 3; Chakravorty et al., 5), have undesired effects on TCP performance. There are several approaches to enhance the TCP RLC interaction. Some of them, surveyed by Barakat et al. () propose the modification of the TCP protocol itself or the introduction of an intermediate proxy (e.g. Meyer et al., Copyright X Inderscience Enterprises Ltd.

2 J.J. Alcaraz and F. Cerdán 3; Chakravorty et al., 5). Many of these solutions are not practical in the short term because they may need changes on the protocol stack at the end users, and some proposals break the end-to-end semantics of TCP. A less developed approach is the application of Active Queue Management (AQM) techniques at the downlink RLC buffers. Simulation experiments have shown that this solution improves TCP performance over 3G links (Agfors et al., 3; Alcaraz et al., 6) with a small change at the RLC operation in the Radio Network Controller (RNC) nodes. One of the most extended AQM mechanisms in the internet routers, Random Early Discard () (Floyd and Jacobson, 993), can be adapted to the particularities of 3G links (Alcaraz et al., 6). However, Agfors et al. (3) argues that the complex parameter configuration of in 3G links is a drawback that deterministic approaches may overcome. It should be mentioned that although is extensively used for internet routers, its implementation in wireless link layer buffers is relatively novel. Thus, the effect of its parameter configuration on TCP performance is still not fully known. In this paper we also contribute by providing further insight on parameter setting for this environment. In this paper we describe and evaluate a novel deterministic AQM algorithm called Slope Based Discard () especially suitable to the characteristics of 3G radio bearers. By means of simulation, is compared to and to the simple drop-tail scheme in several scenarios. The protocols involved are implemented in detail in the simulator which was developed and configured according to the recommendations of previous works. Analysing the results, we disclose the influence of TCP flavours, the number of concurrent TCP flows and the wireless error rate. deterministic operation and lower parameter sensitivity compared to makes it easier to implement in a highly variable environment like a 3G network. Moreover, is best suited to multiplex long-lived and short-lived TCP flows together, thanks to its faster reactions against changes in the buffer occupancy. The rest of the paper is organised as follows. Section describes the characteristics of a 3G radio bearer that may degrade TCP performance. Section 3 surveys previous work about buffer management in radio bearers. Section 4 explains the algorithm in detail. Section 5 provides a brief description of the simulation environment, compares the performance figures of, and drop-tail schemes in several scenarios and provides configuration guidelines for each scheme. Section 6 deals with the performance of shortand long-lived flows multiplexed together at the RLC. The findings of this paper are summarised in Section 7 where future improvements are suggested. Characteristics of 3G links Previous experimental (Chakravorty et al., 5) and simulation (Bestak et al., ) results provide a clear view of the characteristics of 3G wireless links. The behaviour of the link buffer occupancy has shown a great impact on TCP performance.. Buffer overflow 3G links employ per-user buffering. Flows going to a single mobile terminal share a single buffer. Therefore, radio bearers are expected to multiplex a number of simultaneous connections ranging form to 4 TCP flows (Gurtov and Floyd, 4). At a reliable RLC layer, the upper layer packets will be stored in the downlink buffer until they are fully acknowledged by the receiver side. The consequence is that, as described by Chakravorty et al. (5) and Rossi et al. (3), frame losses in the downlink channel result in higher buffer occupancy at the RLC network side. Considering that current RLC implementations use drop-tail buffers, the buffer may overflow causing consecutive packet losses. This situation is especially harmful in the first stages of a TCP connection (slow start) and has a higher impact in TCP Reno, which can only recover from consecutive losses with a Retransmission TimeOut (RTO). An RTO reduces TCP transmission window to one, causing the highest reduction of the source rate.. Overbuffering The buffer should be large enough to avoid frequent overflow. However, excessive queueing cause some additional problems (Chakravorty et al., 5) like Round Trip Time (RTT) inflation, unfairness between competing flows and viscous web surfing. Figure illustrates the effect of the buffer size on TCP performance over an RLC connection without AQM and a Frame Error Ratio (FER) in the wireless channel of %. The buffer size is given in RLC Service Data Units (SDU) of bytes. The figure shows the end-to-end goodput, defined as the number of successfully received packets at the receiver per time unit, and the delay of a 384 Kbit/s radio bearer in four different scenarios. Each scenario is defined by the number of simultaneous TCP flows at the user terminal, ranging from to 4. We present the goodput and delay metrics for the most extended TCP flavours in the internet (Gurtov and Floyd, 4): Reno, New Reno and SACK. Table shows the parameter configuration for the RLC and TCP protocols. More details about the simulation environment are given in Section 5.

3 Using buffer management in 3G radio bearers Figure TCP performance over RLC. Goodput: () Reno, () New Reno (3) SACK. Delay: (4) TCP, (5) TCP Reno (6) SACK 3 7 TCP flow TCP flows TCP flows 4 TCP flows Buffer Size (SDU) 7 Buffer Size (SDU) () () (3) (4) (5) (6) Table Simulation parameters TCP parameters Setting Maximum TCP/IP packet size bytes Maximum allowed window 64 Kbytes Initial window Fixed Round Trip Delay ms 3G link parameters Setting PDU payload size 3 bits PDUs per TTI TTI (Transm. Time Interval) ms RB nominal bit rate 384 Kbit/s Transmission window 47 PDUs maxdat 5 In-order-delivery True Status Prohibit Timer ms Missing PDU detection true Poll Timer ms Wireless Round Trip Delay ms Normalised doppler frequency, Poll window % Last PDU in buffer Poll yes Last retransmitted PDU Poll yes From Figure we conclude that larger buffer sizes benefit the goodput performance, especially when the radio bearer only carries one TCP flow. On the other hand, an oversized buffer causes higher latency values. These results also show the improvement achieved by the more advanced TCP versions, New Reno and SACK, compared to TCP Reno. This better performance is in agreement with the recommendations of Inamura et al. (3). Figure shows the trace of a TCP connection over a 384 Kbit/s radio bearer multiplexing a single TCP flow. The curve at the top shows the RLC buffer occupancy. The buffer size is limited to SDUs. The TCP congestion window (cwnd) is depicted below, and the curve at the bottom shows the sequence number of the packets when they are sent, received and dropped. At the first stages of the connection, multiple packets are dropped due to buffer overflow, triggering an RTO. The buffer is drained because of the rate reduction, with the consequent underutilisation of the resources. After that, the rate is recovered slowly and the overbuffering appears again, causing high delay and additional consecutive packet losses.

4 J.J. Alcaraz and F. Cerdán Figure Trace of a TCP Reno connection over RLC BO cwnd Sequence Number Time (s) data sent data rx drop 3 Related work The objective of AQM techniques is to detect congestion before the buffer is completely filled, so that certain measures can be taken to avoid overflow. In general, these measures consist of sending a congestion signal (e.g. a packet discard) to the source in an early stage of congestion, allowing the TCP congestion control algorithm to reduce the data sending rate. AQM techniques were developed for internet routers, but now they are considered a feasible strategy to enhance TCP performance over 3G links because they are able to maintain the buffer occupancy around certain level, thus avoiding consecutive packet losses and reducing the delay. One of the most extended AQM schemes for internet routers is Random Early Detection () (Floyd and Jacobson, 993). Alcaraz et al. (6) shows how can be adapted to 3G links. Previously, Agfors et al. (3) had pointed out the possibility of using deterministic discard arguing the difficult parameter tuning of at an RLC buffer. Following this trend, we presented, in a shorter version of this paper (Alcaraz and Cerdán, 6) an evaluation of a deterministic algorithm,. According to the useful definition given by Srivastava and Motanim (5) the use of AQM at RLC to enhance TCP over 3G links is a clear case of cross-layer design, because it implies the design of an algorithm at a layer based on the details of how other layer operates. Section 6 addresses other cross-layer issues related to parameter configuration. 4 Slope-based discard algorithm Slope-Based Discard () is a novel deterministic AQM algorithm that can be implemented at the downlink RLC buffer. is based on the following ideas: A packet discard is a congestion signal directed to the TCP sender side that takes a certain amount of time, T S, to arrive to the TCP source (see Figure 3). The rate reduction caused by this signal is perceived at the buffer after the propagation time, T f, of the fixed (wired) network. The RLC buffer occupancy process reflects the quality of the channel. When the frame loss rate at the channel increases, the rate at which the buffer is drained decreases. 3 The discarding policy is driven by the buffer filling rate, r. When r exceeds a critical value, r c, a packet will be drop. The buffer occupancy level determines the value of r c. 4 After a packet drop, additional packet discarding should be avoided until the rate reduction at the TCP source can be noticed at the RLC buffer. In Figure 3, this reaction time equals T S + T f. Only after this period it is possible to know if the rate reduction is enough to prevent buffer overflow. 5 The value of r c represents the filling rate that, if sustained, would fill the buffer entirely in a time equal to the reaction time.

5 Using buffer management in 3G radio bearers 6 The packet chosen for discard will be as close as possible to the front of the queue, thus the signalling time T S does not include the queueing time, T Q. Additionally, the algorithm should not discard a packet whose transmission over the RLC link has already started. Otherwise, upon a packet discard, the RLC would start the signalling procedure required to synchronise RLC sender and receiver sides (TS 5.3, 6). Consequently, our protocol discards the first packet whose transmission has not yet started, thus reducing complexity and avoiding changes in the 3GPP specification itself. Figure 3 Fixed network Source Schematic diagram of the end-to-end connection with the delay times of each section T f T Q RLC entity T S Packets already sent in RLC frames that are still not acknowledged. 3G link User The following parameters control operation: minth is the occupancy level above which packets can be dropped. maxth is the maximum occupancy allowed in the buffer. T r is the estimated reaction time. α is the occupancy interval used for the estimation of the slope of BO process curve, r. 4. Description of the algorithm Figure 4 shows a graphical example of the algorithm. For the sake of clarity, the curve that represents the buffer occupancy is always increasing until a certain point, where it starts to decrease. In reality, there will be a fast oscillation around the average occupancy caused by the instantaneous arrival and departure of packets. Figure 4 Buffer occupancy curve of a buffer implementing T S maxth When the buffer occupancy (BO) is higher or equal to minth α, the algorithm sets the value of the measuring interval threshold (th n = BO + α), and calculates with equation () the value of r c corresponding to this th n. r maxth th n c = () TR The time C n that it will take the buffer to store α additional bits at r c is C n = α/r c. A timer for C n is started. If the timer expires and the buffer occupancy is below the threshold th n, then the actual filling rate is lower than r c, and no packet is dropped. On the other hand, if the threshold th n is reached before the expiration of the timer, the current filling rate surpasses r c and therefore a packet will be discarded. In Figure 4, the dotted segments starting at each measuring interval represent the buffer filling at the critical rate (critical curve). In the intervals C, C and C, no packet is discarded because the filling curve is below the critical curve. In contrast, in the C 3 period, the filling rate is above r c. Hence, the threshold th 3 is reached before the timer C 3 expires. When th 3 is reached, a packet is dropped. The monitoring and discarding algorithm is deactivated for a period T r, avoiding consecutive packet discards. A description of the algorithm in pseudocode can be seen in Figure 5. Figure 5 Pseudocode of algorithm for each packet arrival update BO if timer C n on if (BO th n ) drop packet deactivate timer C n start timer T r if timer C n off if (BO minth α) calculate C n th n = BO + α start timer C n upon timer (C n or T r ) expiration if (BO minth α) calculate C n th n = BO + α start timer C n Buffer Occupancy th 3 th th th α C DROP C C C 3 T f minth time 4. Implementation issues One of the main advantages of compared to other -like mechanisms is that it does not need to generate random numbers to compute the discarding probability because of its deterministic operation. This reduces the computational cost of the algorithm, and makes it more feasible for its implementation at the RLC level where the buffering is done in a per-user basis. The maintenance of a new timer, C n, does not add too much complexity to the RLC operation, which already handles several timers, e.g. Poll Timer and Status Prohibit

6 J.J. Alcaraz and F. Cerdán Timer (TS 5.3, 6). The RLC can synchronise C n to the Transmission Time Interval (TTI), which is equivalent to a clock signal with a granularity of ms (similar to that of other RLC timers). The protocol could be further simplified in terms of implementation with the use of pre-calculated C n values. Therefore, the buffer could be considered as divided in gaps, each one having a corresponding C n value. In our simulations, the number of C n values equals the number of SDUs fitting in the buffer space delimited by minth and maxth. 5 Performance evaluation 5. Simulation environment The simulation environment employed in this research was developed in OMNeT++ ( and comprises a complete implementation of TCP and RLC protocols. Similar simulators were described by Bestak et al. () and Rossi et al. (3). The simulation scenario consists of one or several TCP sources connected to their respective receivers in a mobile equipment (see Figure 6). The end-to-end connection consists of two sections, the fixed network and the radio bearer. The fixed network considers the internet and the 3G core network. The radio bearer has a round trip time of ms (Holma and Toskala, 4) and a bidirectional nominal rate of 384 Kbit/s, representing the bottleneck link, which is the situation expected in most cases (Gurtov and Floyd, 4). The fixed network is modelled with a fixed rate of Mb/s and a round trip delay of ms. Figure 6 Network Side Schematic representation of the simulation environment TCP sources # # #3 #4 RTT Internet: ms RLC Wireless Channel # Frame Error Rate: % - 5% - % Normalizad Doppler Freq.:. TCP receivers # #3 #4 RLC User Side The wireless channel generates error bursts according to the model described by Chockalingam and Zorzi (999) where the Doppler frequency, f d, of the wireless link determines the average burst length. Lower f d causes longer bursts of errors. It is usual to employ the normalised Doppler frequency, equal to the product of f d and the radio frame duration ( ms). In order to obtain more realistic results, the error probability in our model is the same in the uplink and in the downlink direction. The FER is fixed to %, a typical UMTS design value (Meyer et al., 3), except for scenarios evaluating the effect of higher FER values (see Section 6). The simulation results are obtained averaging runs per sample. Each run is a -s download session, except for the performance evaluation of short-lived TCP flows (see Section 7). The radiuses of the confidence intervals are obtained with a confidence degree of 9% according to a t-student distribution. RLC and TCP configuration is shown in Table. The description of RLC parameters can be found in (TS 5.3, 6). RLC parameter values have a considerable impact on TCP performance. Our RLC configuration follows the guidelines provided by previous works (Rossi et al., 3; Alcaraz et al., 6) to achieve the best performance at the RLC layer and to avoid protocol stalling. 5. Simulation scenarios Our performance simulation analysis compares two metrics, goodput and delay, for the simple drop-tail scheme () and two AQM mechanisms: and. Each AQM scheme was evaluated for several combinations of parameter values, in order to find the optimum configuration of the algorithms in terms of end-to-end performance in each scenario. For the simulations implementing the algorithm, the following values for parameters were evaluated: minth: from 5 to 55 SDUs in steps of 5 SDUs. maxp: from.5 to. It should be mentioned that, as stated by Agfors et al. (3), an implementation at the RLC layer should not perform buffer occupancy averaging. Using the instantaneous queue size instead makes the reaction of the algorithm to sudden increases faster, simplifies its configuration and reduces computing load. For the simulations with buffer, the parameter values evaluated are: minth: from 5 to 55 SDUs in steps of 5 SDUs. Tr: from ms to 9 ms. α: from to SDUs. The different scenarios are generated with the combination of the following values of the simulator parameters: TCP flows multiplexed:,, 3 and 4 TCP flows. TCP flavours: Reno and New Reno. FER in the wireless channel: %, 5% and %. The RLC buffer size is fixed to SDUs (9 Kb). For most AQM configurations, this size is enough to prevent buffer overflow. 5.3 parameter setting We start by selecting the value of α. Figure 7, shows the goodput and delay performance for different values of α, minth = and T r = ms. Lower values of α imply a fast detection of congestion but excessive sensitivity to occupancy oscillations, causing more packet drops than

7 Using buffer management in 3G radio bearers needed. If α is too large, the reaction against buffer increments is too slow and therefore is less effective against overbuffering. For the rest of the experiments in this paper, α is set to 5, which is a good compromise value. Figure 7 3 Influence of α on performance goodp. flows= goodp. flows= goodp. flows=3 goodp. flows=4 α delay flows= delay flows= delay flows=3 delay flows=4 We define the aggressiveness of the algorithm as the tendency to reduce the source s rate given certain BO level. It is clear that, if the protocol discards a packet at low BO values (low minth) or if T r is large (r c value is small), the aggressiveness is high. Figure 8 shows the performance figures for one and several TCP Reno connections to illustrate the effect of T r in (minth = ). Higher T r values yield lower packet latency, because the buffer occupancy is kept at lower levels. If T r is too small, the algorithm operation approaches a drop-tail scheme. Considering that the goodput for a single TCP Reno flow over a drop-tail RLC buffer is 76.8 ± 4. Kbit/s and the delay is.7 ±. s, there is a range of configuration values for where the goodput and the delay performance are simultaneously improved in the single flow scenario. The goodput improvement is more noticeable for T r values near the Round Trip Propagation Delay of the connection. If T r is too high, the frequency of packet drops is excessive, causing goodput degradation. If the number of concurrent TCP connections in the RLC buffer increases, the effect of T r is slightly different. If the number of multiplexed flows augments, the goodput reduction is less severe for aggressive configurations. For example, within the range of T r values considered in Figure 8, the goodput is almost constant for the case of four TCP connections. This fact is easily explained considering TCP cwnd dynamics. Using AQM, an early packet discard halves the window of the TCP connection. Obviously, in a single flow scenario this measure halves the overall user rate. This avoids buffer overflow but limits the goodput improvement if packet drops are too frequent. In a multiple flow scenario the overall user rate reduction is less severe because it only affects one connection upon each packet discard. In the multiple flow scenario, the goodput improvement compared to that of droptail is less noticeable, but the delay reduction can be of up to % End-to-end delay (s) Figure 8 3 Influence of T r on performance goodp. flows= goodp. flows= goodp. flows=3 goodp. flows=4 Tr (ms) delay flows= delay flows= delay flows=3 delay flows=4 In, the impact of minth is lower than in. Figure 9 shows the effect of minth (T r = ms). In this figure, the goodput is almost constant. The reason is that discards packets if the buffer s filling rate surpasses the critical rate, r c, on each BO level. Assuming that there is a maximum rate r max, at which packets directed to a single user can arrive to the RNC, the maximum buffer filling rate is equal to r max. Therefore, although BO is above minth, no packet is discarded if r c is higher than r max, i.e. if r c can not be reached. T r, determines the BO value where r c equals r max. We may consider this BO value as the effective minimum buffer threshold, minth eff, above which a packet discard may occur (see Figure ). For T r = ms, and r max = Kbit/s (83 SDU/s), according to (), minth eff = maxth T r r c = 39 SDUs. For more aggressive T r values, minth acts as as security limit, assuring that the goodput does not fall bellow certain threshold. Figure 9 3 Influence of minth on performance goodp. flows= goodp. flows= goodp. flows=3 goodp. flows=4 minth (SDUs) delay flows= delay flows= delay flows=3 delay flows= End-to-end delay (s) End-to-end delay (s)

8 J.J. Alcaraz and F. Cerdán Figure Critical filling rate r c respect the Buffer Occupancy r c No packet is dropped in this area with the bottleneck link rate. Our results show that the optimum minth value in terms of goodput for a single TCP flow is around 4 BDP. Figure Influence of minth on performance 3 r max minth minth eff 5.4 parameter setting maxth BO In, the aggressiveness is determined by the combination of maxp and minth. Figure shows the performance of respect maxp, with minth =. As in the case, there are several configuration values that improve goodput while reducing the delay respect to those of a drop-tail scheme. For the single flow scenario, these values are around maxp =.5. When more connections are multiplexed, maxp can be set to higher values. Like in, when more flows are multiplexed, a more aggressive setting provides a better overall performance. Figure Influence of maxp on performance goodp. flows= goodp. flows= goodp. flows=3 goodp. flows=4.5 maxp.. delay flows= delay flows= delay flows=3 delay flows=4.8 Figure shows the influence of minth in performance for maxp =.5. It can be seen that performance is more dependant on minth than, especially for the single-flow case. In this case, minth determines the minimum cwnd to allow a packet discard. To avoid link underuse, minth should be above the Bandwidth Delay Product (BDP) of the connection, defined as the product of the Round Trip Propagation Delay (RTPD = RTTw + RTTf) End-to-end delay (s) goodp. flows= goodp. flows= goodp. flows=3 goodp. flows=4 minth (SDUs) 5.5 Performance comparison delay flows= delay flows= delay flows=3 delay flows=4 In order to compare the performance of both AQM schemes and evaluate its improvement respect to the drop tail buffer, we have selected different parameter settings regarding the number of multiplexed TCP flows and the FER. These configuration values, shown in Table, do not provide, in general, the highest goodput in each scenario, because our objective is to balance goodput and delay improvement. For this reason, we choose configurations providing a goodput near the optimum (98% or more) with the lowest delay. Table FER AQM parameter setting for each scenario Flows %,,,, SDB * 5%,,,, % 5, 5, 5, 5, %,.5,.5 5,.5 5,.5 ** 5%,.5 5,.8 5,.8,.8 Notes: % 5,. 5,.5 5,.5 5,.8 * parameter values: minth (SDUs), Ts (ms). ** parameter values: minth (SDUs), maxp. Figure 3, shows that the improvement in goodput and delay is noticeable for the case of TCP Reno, while for TCP New Reno the improvement is focused on the delay reduction. TCP New Reno is less affected by buffer overflow than TCP Reno. Therefore, the avoidance of consecutive packets losses helps especially Reno sources. These results are extensible to TCP SACK, which operation is similar to that of TCP New Reno. End-to-end delay (s)

9 Using buffer management in 3G radio bearers Figure 4 shows the performance of TCP Reno for different FERs. Obviously, higher FER values cause worse performance in every scenario. AQM buffers helps to increase the goodput even for higher FER values, although the main improvement is the delay reduction. AQM algorithms counteract the buffer occupancy increment caused by a higher frame loss ratio in the channel. However, it is advisable to relax slightly the aggressiveness of AQM schemes, as shown in Table. Obviously, the average BO tends to increase for higher FER. In consequence, an AQM scheme should reduce its tendency to discard a packet at each BO level in order to avoid excessive packet drops. Figure 3 Performance figures of different buffer schemes for different scenarios (FER = %) (a) Goodput TCP Reno (c) Delay TCP Reno flow flows 3 flows 4 flows flow flows 3 flows 4 flows (b) Goodput TCP New Reno flow flows 3 flows 4 flows (d) Delay TCP New Reno flow flows 3 flows 4 flows Figure 4 Performance figures of different buffer schemes for different FERs in the radio bearer (a) FER % flow flows 3 flows 4 flows (d) FER % flow flows 3 flows 4 flows 3 8 (b) FER 5% flow flows 3 flows 4 flows (e) FER 5% flow flows 3 flows 4 flows 8 8 (c) FER % flow flows 3 flows 4 flows (f) FER % flow flows 3 flows 4 flows

10 J.J. Alcaraz and F. Cerdán It should be remarked that the configuration proposal of this section relies on a cross-layer approach. According to Srivastava and Motanim (5), the joint tuning of parameters across layers is a type of cross-layer design. The configuration of AQM based on the FER requires an upward information flow from the physical layer. In the 3GPP architecture, this is already possible thanks to the Radio Resource Control (RRC) protocol that handles the resource management algorithms, setting up and modifying layer and layer protocol entities (TS 5.33, 6). On the other hand, the number of multiplexed flows can be provided in a downward information flow from the Packet Data Convergence Protocol (PDCP) which is above RLC (TS 5.3, 6). It should be noticed that these information flows do not require additional signalling because they operate only on the RNC side, and therefore could be implemented without introducing changes in 3GPP specifications. 6 Performance with short-lived flows In this section we show how each scheme performs when long- and short-lived flows are multiplexed simultaneously in the RLC buffer. The scenario for the simulations consists of a TCP connection that is continuously transmitting (longlived flow). After s, the second TCP connection starts the transference of a 64 kb file (short-lived flow). Considering Reno sources, the file s transfer time using a drop-tail buffer is 35 ±.99 s. Figure 5 shows the transfer time for different configurations and Figure 6 considers. The average transfer time of drop-tail () configuration is shown as a reference value. It is clear that outperforms in this task. If T r is equal or above ms (as recommended in Section 6) the transfer time is below that of most configurations. The reason is that is more responsive against sudden increases in BO. Figure 5 Transfer time for a 64 Kb file with Transfer Time (s) minth=5 Ts (ms) minth= minth=5 minth= In order to illustrate this, Figure 7 shows a trace of one run of the experiment using and Figure 8 shows a trace for with the same channel conditions. Both and are configured for the one flow scenario. As can be seen, is not able to avoid buffer overflow when the short transfer starts, which suffers multiple packet drops, with the consequent RTO. In, which is conceived to react against BO increments, the number of packet drops during the first stages of the second connection is higher. Some of these packet discards affect to the new flow which reduces its rate preventing buffer overflow. Figure 6 Transfer time for a 64 Kb file with Transfer Time (s) minth=5..5 maxp minth= minth=5.. minth=.8 Figure 7 Transference of a 64 Kb file over a buffer with a pre-existing connection SN Flow SN Flow Time (s) data sent data rx drop BO cwnd cwnd

11 Using buffer management in 3G radio bearers Figure 8 Transference of a 64 Kb file over a buffer with a pre-existing connection BO cwnd SN Flow SN Flow Time (s) data sent data rx drop cwnd 7 Conclusion and future work By means of simulation experiments we show that AQM techniques implemented at RLC downlink buffers improve goodput and delay performance of TCP over 3G radio bearers. The relative improvement of AQM schemes respect to droptail depends on several features. For example, AQM avoidance of buffer overflow benefits especially the goodput performance of TCP Reno because New Reno and SACK are more robust to consecutive packet losses. On the other hand, the delay is reduced in every scenario because AQM maintains the buffer occupancy around a certain level. AQM is more effective for large buffers and when the radio bearer multiplexes more than one TCP flow. performance is similar to that of, although in certain scenarios show better figures. The main advantages of are its easier implementation, its lower parameter sensitivity and its better performance for short-lived flows when multiplexed with long-lived connections. is deterministic; hence it does not need the generation of random numbers required for the computation of a discarding probability. This fact makes more feasible for buffer management at RLC level, where the buffering is done in a per-user basis. This simplicity can be applied in the design of automatic configuration algorithms for. The wireless links are highly variable; the delay and the bandwidth can experience sudden changes due to several mobile network phenomena, e.g. handovers, outages and scheduling. TCP-based applications may have different QoS objectives (e.g. higher goodput or less delay). Following a cross-layer approach, a dynamic algorithm could rearrange parameters adapting it to the changing link characteristics while meeting the application requirements. Acknowledgements This work was supported by project grant TEC /TCM (CON-PARTE-) and it was also developed in the framework of Programa de Ayudas a Grupos de Excelencia de la Región de Murcia, Fundación Seneca. References 3GPP TS 5.3 (6) Radio Link Control (RLC) protocol specification, v GPP TS 5.33 (6) Radio Resource Control (RRC) protocol specification, v GPP TS 5.3 (6) Packet Data Convergence Protocol (PDCP) specification, v Agfors, M., Ludwig, R., Meyer, M. and Peisa, J. (3) Queue management for TCP traffic over 3G links, Proceedings of IEEE WCNC 3, Vol. 3, pp Alcaraz, J.J. and Cerdán, F. (6) Using buffer management in 3G buffers to enhance end-to-end TCP performance, Proceedings of IEEE AINA 6, Vol., pp Alcaraz, J.J., Cerdán, F. and García-Haro, J. (6) Optimizing TCP and RLC interaction in the UMTS radio access network, IEEE Network, Vol., No., pp Barakat, C., Altman, E. and Dabbous, W. () On TCP performance in a heterogeneous network: a survey, IEEE Communications Magazine, Vol. 38, No., pp. 46. Bestak, R., Godlewski, P. and Martins, P. () RLC buffer occupancy when using a TCP connection over UMTS, Proceedings of IEEE PIMRC, Vol. 3, pp Chakravorty, R., Clark, A. and Pratt, I. (5) Optimizing web delivery over wireless links: design, implementation and experiences, IEEE Journal on Selected Areas in Communication, Vol. 3, No., pp. 46. Chockalingam, A. and Zorzi, M. (999) Wireless TCP performance with link layer FEC/ARQ, Proceedings of IEEE ICC, Vol., pp. 6. Floyd, S. and Jacobson, V. (993) Random early detection gateways for congestion avoidance IEEE/ACM Transactions on Networking, Vol., pp Gurtov, A. and Floyd, S. (4) Modeling wireless links for transport protocols, ACM SIGCOMM Computer Communication Review, Vol. 34, No., pp Holma, H. and Toskala, A. (4) WCDMA for UMTS: radio access for third generation mobile communications, 3rd ed., Wiley.

12 J.J. Alcaraz and F. Cerdán Inamura, H., Montenegro, G., Ludwig, R., Gurtov, A. and Khafizov, F. (3) TCP over second (.5G) and third (3G) generation wireless networks IETF RFC 348. Meyer, M., Sachs, J. and Holzke, M. (3) Performance evaluation of a TCP proxy in WCDMA networks, IEEE Wireless Communication, Vol., No. 5, pp Rossi, M., Scaranari, L. and Zorzi, M. (3) On the UMTS RLC parameters setting and their impact on higher layers performance, Proceedings of IEEE 57th VTC, Vol. 3, pp Srivastava, V. and Motanim, M. (5) Cross-layer design: a survey and the road ahead, IEEE Communications Magazine, Vol. 43, No., pp. 9. Xylomenos, G. and Polyzos, G.C. (3) Multi-service link layer enhancements for the wireless internet, Proceedings of IEEE ISCC, pp Yavuz, M. and Khafizov, F. () TCP over wireless links with variable bandwidth, Proceedings of IEEE 56th VTC, Vol. 3, pp.3 37.