IEEE 802.11e-2005: A Study on Qos and Network Performance

IEEE 802.11e-2005 was designed to improve the quality of service (QoS) in wireless networks. This standard introduced the concept of Enhanced Distributed Channel Access (EDCA) to prioritize traffic.

The EDCA mechanism uses four access categories (ACs) to classify traffic into different priority levels. These ACs are Voice, Video, Best Effort, and Background.

The Voice AC has the highest priority, ensuring that voice traffic is transmitted reliably and with minimal delay. The Video AC has the next highest priority, followed by the Best Effort and Background ACs.

The IEEE 802.11e-2005 standard also introduced a new parameter called AIFS (Arbitration Inter-Frame Space) to control the access to the medium. AIFS is used to determine when a station can transmit a frame.

A unique perspective: IEEE 802.11d-2001

EDCA is a supported QoS mechanism in 802.11e that provides service differentiation by traffic prioritization. It ensures high-priority traffic has a higher chance of being sent than low-priority traffic.

EDCA uses a variation of CSMA/CA called TCMA, which has a shorter arbitration inter-frame space (AIFS) for higher priority packets. The exact values of AIFS depend on the physical layer used to transmit the data.

The TCMA protocol uses a contention window (CW) that can be set according to the traffic expected in each access category (AC). The CWmin and CWmax values are calculated from aCWmin and aCWmax values, respectively, that are defined for each physical layer supported by 802.11e.

The levels of priority in EDCA are called access categories (ACs), and there are four ACs: Background (AC_BK), Best Effort (AC_BE), Video (AC_VI), and Voice (AC_VO). ACs map directly from Ethernet-level class of service (CoS) priority levels.

Here are the default EDCA parameters for each AC:

These parameters are used to determine the contention window (CW) and arbitration inter-frame space (AIFS) for each AC. The CWmin and CWmax values are used to calculate the contention window, while the AIFSN value is used to determine the arbitration inter-frame space.

EDCA also provides contention-free access to the channel for a period called a Transmit Opportunity (TXOP). A TXOP is a bounded time interval during which a station can send as many frames as possible.

Broaden your view: Short Interframe Space

IEEE 802.11e HCCA is a coordination function that allows for contention-free transmission, initiated by the AP at any time during the contention period.

In HCCA, the AP controls the access to the medium during the Controlled Access Phase (CAP), while stations function in EDCA during the Contention Period (CP).

The AP can provide a per-session service, coordinating traffic streams in any fashion it chooses, and giving priority to one station over another based on queue length information.

Stations are given a TXOP, allowing them to send multiple packets in a row for a given time period selected by the AP.

The AP allows stations to send data by sending CF-Poll frames during the CAP.

HCCA is generally considered the most advanced coordination function, allowing for QoS configuration with great precision.

QoS-enabled stations can request specific transmission parameters, such as data rate and jitter, which is beneficial for advanced applications like VoIP and video streaming.

Check this out: Wireless Access Point

HCCA support is not mandatory for 802.11e APs, and few APs currently available are enabled for HCCA.

Implementing HCCA on end stations uses the existing DCF mechanism for channel access, with no change to DCF or EDCA operation needed.

On the AP side, a scheduler and queuing mechanism are required to support HCCA.

Here are some key differences between HCCA and PCF:

A simulation model was designed and implemented in MATLAB Simevent environment to study the IEEE 802.11e-2005 protocol. The model was divided into three blocks: sources, access point, and sink, which were replicated into four places to represent different access categories (ACs).

The simulation was run for 3600 seconds to achieve normalization, with source rate varied in steps of 100 kbps from 0 to 1000 kbps. The model's performance was evaluated using throughput and delay, with probes placed strategically to collect data for calculations.

The first simulation scenario used default EDCA access parameters for all four ACs, and the result was validated using Bahi Hour et al.'s simulation parameters. The validation result showed similar throughput and delay results as the simulation model, with a slight difference of 0.01 Mbps throughput.

The MATLAB SimEvents EDCA simulation model is a valuable tool for evaluating the performance of the IEEE 802.11e wireless LAN EDCA protocol.

This model was designed to handle all QoS parameters, as there is no single analytical model that can do so.

It was implemented in the MATLAB Simevent environment and divided into three blocks: sources, access point, and sink.

The sources are made up of packet generator's arrival rate, packet length, set attribute, and first in first out (FIFO) queue blocks.

The access point consists of input switch, get attribute, FIFO queue, server, and output switch blocks.

It also contains a signaling loop that informs the sources about the busy and idle state of the input switch.

These blocks are replicated into four places, each representing a different AC.

The packet generation and arrival pattern depict bursty ON-OFF, showing the random nature of packet arrival patterns which depict Poisson process.

The simulation was run for 3600 s to achieve normalization.

During the simulation, source rate was varied in steps of 100 kbps from 0 to 1000 kbps.

The simulation model was used to verify access parameters as it affects some QoS parameters in the IEEE 802.11 wireless LAN EDCA protocol.

The model performance was evaluated using throughput and delay.

Probes were strategically placed to collect data for calculations.

In the first simulation scenario, four ACs were modeled using default EDCA access parameters to test the service differentiation ability of the developed model.

The results were validated using Bahi Hour et al.'s simulation parameters, ensuring the validity of the research data, ethics, and reliability of the design.

VO and VI traffic recorded 38% and 29% of the mean throughput, while BE and BK traffics recorded 19% and 14% at a 600 kbps source rate.

The EDCA parameters provided optimal performance at 600 kbps source rate, and the validation result showed almost the same throughput as the simulation model.

VO and VI traffic experienced 4.2% and 16.3% mean delay, while BE and BK traffic recorded 33.6% and 45.9% of mean delay.

The validation result showed that VO and VI traffic experienced 8% and 20% of mean delay, while BE and BK traffic recorded 30% and 42% of mean delay.

AIFS number and CW size were analyzed to determine their impact on QoS parameters, and the results showed that high AIFSN for BE traffic improved the throughput of VO traffic and vice versa.

The AIFS number has a significant impact on throughput, as seen in the simulation results. AIFS stands for Arbitration Inter-Frame Space, and it determines how long a device waits before attempting to access the channel.

In the study, the AIFS number was varied to examine its effect on throughput. The results showed that VO traffic achieved 61% of the mean throughput, while BE traffic recorded 39%. This indicates that the AIFS number plays a crucial role in service differentiation.

The AIFS number also affects the throughput of BE traffic, with an increase in AIFS number resulting in a decrease in throughput. For example, when the AIFS number of BE traffic was increased from 3 to 4, the throughput decreased by 4%. Conversely, an increase in AIFS number for VO traffic resulted in an increase in throughput.

The impact of AIFS number on throughput is more significant than the CW size. Increasing the AIFS number of AC3 by one had a more substantial effect on optimal throughput performance than increasing the CW size.

The results of the simulation provide valuable insights into the impact of AIFS number on throughput. By understanding these effects, network administrators can optimize their network settings to achieve better performance.

VO traffic at a high AIFS number achieved higher throughput, while BE traffic at a low AIFS number achieved higher throughput. This suggests that the AIFS number should be adjusted based on the type of traffic being transmitted.

The AIFS number should be carefully tuned to achieve optimal throughput performance. AIFS numbers that are too high can result in reduced throughput, while numbers that are too low can lead to increased collisions and reduced performance.

VO traffic achieved higher throughput at a fixed CW size when the AIFS number was increased. This indicates that the AIFS number plays a crucial role in determining the throughput of VO traffic.

See what others are reading: Size Speakers

The impact of AIFS number on throughput is a critical consideration in network design. By understanding the effects of AIFS number on throughput, network administrators can design networks that meet the demands of their users.

In the study, the AIFS number was varied to examine its effect on throughput. The results showed that VO traffic achieved 61% of the mean throughput, while BE traffic recorded 39%. This indicates that the AIFS number plays a crucial role in service differentiation.

The AIFS number also affects the throughput of BE traffic, with an increase in AIFS number resulting in a decrease in throughput. For example, when the AIFS number of BE traffic was increased from 3 to 4, the throughput decreased by 4%. Conversely, an increase in AIFS number for VO traffic resulted in an increase in throughput.

The impact of AIFS number on throughput is more significant than the CW size. Increasing the AIFS number of AC3 by one had a more substantial effect on optimal throughput performance than increasing the CW size.

VO traffic at a high AIFS number achieved higher throughput, while BE traffic at a low AIFS number achieved higher throughput. This suggests that the AIFS number should be adjusted based on the type of traffic being transmitted.

The AIFS number should be carefully tuned to achieve optimal throughput performance. AIFS numbers that are too high can result in reduced throughput, while numbers that are too low can lead to increased collisions and reduced performance.

The simulation results provide valuable insights into the impact of AIFS number on throughput. By understanding these effects, network administrators can optimize their network settings to achieve better performance.

In conclusion, the AIFS number has a significant impact on throughput, and its effects should be carefully considered in network design. By adjusting the AIFS number based on the type of traffic being transmitted, network administrators can achieve optimal throughput performance.

Increasing the Contention Window (CW) size can have a significant impact on the performance of both Voice (VO) and Best Effort (BE) traffic in a network.

A larger CW size can lead to reduced backoff and collision in the network, resulting in higher throughput for BE traffic and lower throughput for VO traffic.

In fact, a 2% reduction in throughput was observed for VO traffic when the CW size was increased, while a 2% increase in throughput was observed for BE traffic under the same conditions.

The optimal CW size for VO traffic was found to be between 15 and 31, with a maximum throughput of 63.7, while BE traffic experienced a significant increase in throughput when the CW size was increased to 1023.

In terms of delay, a larger CW size can also have a positive impact on lower priority traffic, such as BE traffic, by reducing collision in the network.

Conversely, VO traffic experienced the least delay at low CW sizes due to reduced collision.

The system's optimum performance remains stable at 700 kbps source rate despite changes in AIFSN and CW size.

Increasing the AIFSN for BE traffic improves the throughput of VO traffic and vice versa. This shows that a high AIFSN for BE traffic has a positive effect on VO traffic throughput.

High CW size, especially under saturation conditions, improves the performance of lower priority traffics by decreasing collisions in the network.

A low CW size results in VO traffic experiencing the least delay due to the collision effect.

The mean delay of BE and VO traffic is sufficiently small to satisfy their specified QoS when the network is working under unsaturated conditions.

The delay experienced by VO traffic is much smaller compared to the BE traffic due to prioritization and the priority value of the individual traffic.

Increasing the AIFSN of AC1 from 3 to 4 and AC3 fixed at 2 results in a 9% decrease in VO traffic delay and a 9% increase in BE traffic delay.

The 3% impact introduced by high reduction of CWmax of BE traffic from 1023 in the first case to 63 in the third case is advantageous to traffic with higher priority and against the traffic with lower priority.

VO achieved 61% while BE recorded 39% of the mean throughput at a fixed CW size.

Increasing the CW size can have a significant impact on throughput. VO traffic at CWmin1,3 = 15,3 and CWmax1,3 = 63,7 recorded the highest throughput.

As CW size increases, VO throughput decreases, while BE throughput increases. This is due to reduced backoff and collision in the network.

An increase in CW size from 15,3 to 31,7 resulted in a 2% reduction in VO throughput and a 2% increase in BE throughput.

The effect of CW size on throughput is more pronounced when comparing the second case to the third case, where CWmin1,3 = 31,7 and CWmax1,3 = 1023,15. In this scenario, BE throughput increased significantly, while VO throughput decreased.

Increasing AIFSN of AC3 by one has a more significant effect on optimal throughput performance than CW size. This is evident in the comparison of Figures 12 and 13, where the effect of AIFS of service differentiation is more severe.

The estimated TXOP duration has a significant impact on the overall system performance. For overestimation, the system is underutilized, while underestimation can lead to longer delays, packet drops, and delay variation.

Here's an interesting read: Wireless Distribution System

Correctly estimating TXOP duration is quite challenging. The admission control typically accepts each flow with a mean data rate, but this may not be sufficient to support fluctuated traffic, such as Variable Bit Rate (VBR).

To accommodate VBR traffic, the admission control should accept each flow with a mean data rate plus a small extra value. This extra value is less than the standard deviation (SD) in case of known arrival rate traffic, such as playback video.

For your interest: Control and Provisioning of Wireless Access Points Protocol

The TXOP limit is a crucial aspect of wireless networks, allowing stations to transmit packets for a specified duration after winning access to the medium.

This limit covers the entire frame exchange sequence, including SIFS periods, RTS, ACKs, and CTS.

The allowable duration for transmission is directly tied to the TXOP limit, which means that stations can transmit multiple frames within a single TXOP as long as the transmission period is lower or equal to the limit.

Non-zero TXOP limits enable EDCA functions to transmit multiple frames in a single TXOP, provided the frames belong to the same AC.

Different ACs have their default TXOP limits, which are presented in Table 2.

A non-zero TXOP limit allows for efficient transmission of multiple frames, reducing the need for repeated access to the medium.

This can result in improved network performance and reduced latency.

The proposed mechanism is designed to dynamically adjust the TXOP duration for each Station (SI) based on the feedback queue size. It implements a finite state machine to allocate TXOP duration.

The mechanism can support various video types with different characteristics in IEEE 802.11e HCCA mode. It's specifically tailored to cope with burst traffic by allocating various TXOP durations according to the state.

The proposed mechanism is different from the ARROW mechanism, which precisely adjusts the TXOP duration for the next SI based on the feedback queue size. Our mechanism, on the other hand, adjusts the TXOP duration according to an event and the current state of the particular flow.

The mechanism uses a coefficient factor δk to bound the range for the particular event. The δ values are obtained from a fine-tuning process of trial-and-error adjustment.

Four states have been defined in the mechanism, with each state specifying the amount of TXOP duration granted for each flow with an extra duration. The state transition is defined as shown in Figure 1.

The mechanism aims to provide an extra duration for clearing the occurred burst, especially when a new I-frame arrives. The system should provide only the minimum amount of TXOP duration when the burst has been served.

The mechanism can support both uplink and downlink traffic flows, with the uplink traffic flow requiring feedback information from QSTA. The mechanism independently keeps the state of each traffic flow, with each state changing according to the event defined by the queue size information and certain threshold values.

The IEEE 802.11e-2005 standard provides a mechanism for service differentiation, which is strongly dependent on channel access parameters such as AIFS and CW sizes.

Tuning AIFS requires caution to avoid starving best-effort traffic. The CW size should be tuned dynamically in response to varying load.

A smaller AIFSN is recommended for real-time service or BE traffic to reduce collision, while a larger CW size can help reduce the collision at high network loads.

For networks with high volumes of BE traffic, the EDCA protocol is not considered efficient. Admission control or an appropriate scheduling scheme is needed to guarantee channel access to real-time traffic.

The proposed ATMV mechanism for video transmission in IEEE 802.11e HCCA at QAP can support up to 8 concurrent flows with slight movement video.

However, the video quality degrades with other video categories, such as gentle walking movement and rapid movement.