Summary of Mathematics doctoral thesis: Improving the performance of throughput and fairness in IEEE 802.11 EDCA wireless Adhoc networks

Chia sẻ: _ _ | Ngày: | Loại File: PDF | Số trang:27

Thêm vào BST

Báo xấu

33
lượt xem 3
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

The propose a new method aiming to adjust the TXOP parameter according to a dynamic mechanism that suits the priority of each data type from the existing limitations with TXOP parameter in IEEE 802.11 EDCA; the propose a novel fuzzy logic approach for enhancing the fairness of low priority data flows in IEEE 802.11 EDCA.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Summary of Mathematics doctoral thesis: Improving the performance of throughput and fairness in IEEE 802.11 EDCA wireless Adhoc networks

MINISTRY OF EDUCATION VIETNAM ACADEMY OF AND TRAINING SCIENCE AND TECHNOLOGY GRADUATE UNIVERSITY OF SCIENCE AND TECHNOLOGY ----------------------------- LUONG DUY HIEU IMPROVING THE PERFORMANCE OF THROUGHPUT AND FAIRNESS IN IEEE 802.11 EDCA WIRELESS AD HOC NETWORKS Major: Mathematical Foundations for Computer Science Code: 9 46 01 10 SUMMARY OF MATHEMATICS DOCTORAL THESIS Ha Noi – 2020
This thesis was done at: Graduate University of Science and Technology, Vietnam Academy of Science and Technology. Supervisor 1: PGS.TS. Thai Quang Vinh Supervisor 2: PGS.TS. Pham Thanh Giang Reviewer 1: … Reviewer 2: … Reviewer 3: …. The dissertation will be defended at Graduate University of Science and Technology, 18 Hoang Quoc Viet street, Hanoi. Time: .............,.............., 2020 This thesis could be found at National Library of Vietnam, Library of Graduate University of Science and Technology, Library of Vietnam Academy of Science and Technology.
1 INTRODUCTION 1. The urgency of this thesis In recent years, the field of the wireless network in general and ad hoc networks in particular has been increasingly paid attention to by the broad application domain, which directly impacts important fields such as military, security, health care, aviation, etc. The applications of the Adhoc network span many areas of life from military to civilian with many different variations. An example in the mili- tary field with FANET [44] allows intelligence to be collected, constructing a battle map. In the civil field with VANET [35, 15] network allows the deployment of services for intelligent traffic. The BAN [9] network is used in the medical field to enable the implementation of community healthcare monitoring services. In recent years, the field of research about Adhoc networks at home [5, 1, 2] and abroad [51, 7, 24, 36] is interested by researchers because it brings practical value and scientific value, opens up many opportunities for application and enhancement of technology potential for each country. Many important problems are still being solved by scientists such as ac- cess control, route finding problem, reliable communication, service quality assurance [4, 3]. In particular, a subclass that is currently interested in the research community is to deeply solve key problems at the MAC layer on IEEE 802.11 standard to improve performance and improve service quality for applications in Adhoc wireless networks. This classification problem covers many problems to solve with complex technical constraints. It is a matter of resource contention and transmission collision between network nodes and between data flow in the network. The wireless links with high-bit errors are caused the packets to be retransmitted many times, lowering the effective bandwidth. The external factors such as radio channel noise, interference, fading have affected the quality of the transmitted signal. In the Adhoc network, a station has to transmit both the direct data flow generated by the station itself and the forward flows generated by neighboring flows. The effect of this dispute affects network performance. In the face of the great increase in wireless-based applications, the need
2 for research towards improving performance for Adhoc networks becomes an urgent issue. Foreseeing that trend, Ph.D. students choose the topic: "Improving the performance of throughput and fairness in IEEE 802.11 EDCA wireless Adhoc networks". To ensure the feasibility of the research with limited time conditions, the existing equipment, and infrastructure. The Ph.D. student focuses on research direction to improve the throughput and fairness index at the MAC layer on the IEEE 802.11 standard. 2. Goals of the thesis (1) The propose a new method aiming to adjust the TXOP parameter according to a dynamic mechanism that suits the priority of each data type from the existing limitations with TXOP parameter in IEEE 802.11 EDCA. (2) The propose a novel fuzzy logic approach for enhancing the fairness of low priority data flows in IEEE 802.11 EDCA. 3. Objectives of the thesis The thesis focuses on the following objects of the adhoc network: (1) Focus on Medium Access Control layer. (2) Focus on Enhanced Distributed Channel Access in IEEE 802.11 EDCA. 4. Scope of thesis Reseach the parameters in the Enhanced Distributed Channel Access of IEEE 802.11 EDCA based on throughput and fairness. 5. Research methods (1) Addressing existing problems of research subjects. (2) Analyzing and surveying factors are affected by research subjects. (3) Study the related topic of the subject and give a new author’s approach. (4) The propose the algorithms and models to solve identified problems. (5) Checking the results by using reliable techniques. 6. Meaning of the research topic (1) Practical value: Many applications on wireless network platforms,
3 especially in adhoc networks such as smart traffic, smart medical, smart agriculture, smart city are expected to become popular and have a great impact on social life in the coming decades. Solving the problem of im- proving wireless performance adhoc is a current issue and makes a lot of sense in deploying applications that require high throughput, reliability, and availability. (2) Scientific significance: The results of the research contribute to sup- porting the training of higher education, promoting the application of basic research results, and making practical contributions to the socio-economic development of the country; contribute to enhancing basic research poten- tial in the field of high technology, promote technological innovation in the fields of the technological revolution 4.0. 7. New contributions of the thesis (1) The thesis propose a new method aiming to adjust the TXOP pa- rameter according to dynamic mechanism that suits the priority of each data type from the existing limitations with TXOP parameter in IEEE 802.11 EDCA. (2) The thesis propose a novel fuzzy logic approach for enhancing the fairness of low priority data flows in IEEE 802.11 EDCA. The proposed method also provides the rate among flows more balanced than the IEEE 802.11 EDCA standard. (3) In addition to two theoretical proposals, the thesis has done the survey and analysis, evaluating the impact of parameters in IEEE 802.11 EDCA on wireless Adhoc networks. The thesis also introduces an experi- mental application in multimedia data transmission applied to the VANET network.
CHAPTER 1. OVERVIEW WIRELESS ADHOC NETWORK In this chapter, the thesis presents factors related to the network per- formance of wireless ad hoc networks, main approaches to the research, evaluate the pros and cons of domestic and international researches in or- der to determine existing issues that need to be studied more in the future. 1.1. Introduction adhoc networks There are many definitions for wireless adhoc networks, each with dif- ferent expressions, but all of the basic features of adhoc networks are in- dependence and self-connectivity. 1.1.1. The features of adhoc networks Besides the advantages, wireless adhoc networks have properties that make them face many challenges when applying to practice. [51, 45, 52]. (1) Dynamic topology (2) Limited energy (3) Liminted bandwidth and short boardcasting radius (4) Many security challenges (5) Unreliable transmission 1.1.2. Application of adhoc network (1) Civil field: A variant of the adhoc network is the VANET network [35, 15] which is developed to be applied in Intelligent Transport System (ITS). Automation Guided Vehicle is a typical example of VANETs in which the system will automatically connect, contact, and independently perform actions, such as detecting road lanes, identifying obstacles. In re- search [CT3] author conducted an experiment on automation guide vehicles with the aim of managing multimedia database processing that serves in driving automation guided vehicles. This is also a contribution in the di- rection of the research to improve the performance of the adhoc network applications implemented and published by the graduate student at [CT3] [CT4] [CT8]. (2) Military field:In most military operations it is usually fast. There isn’t an available infrastructure network. Because of that, wireless ad hoc
5 networks meet the demand for flexibility, quick deployment, mobile con- nection. A variant of wireless ad hoc networks is Flying Adhoc Networks (FANET)[44] which are developed for connection with drones (UAV). Each vehicle can exchange information with each other and collect controlling data from network nodes to the ground to perform missions such as col- lecting information on forest resources, water resources, climate change, intelligence, and building up area maps. 1.2. The method evaluation of performance adhoc network There are three common methods used to evaluate network perfor- mance: experimental evaluation, evaluation methods by analytical models, and evaluation methods by simulation models. 1.3. Approach to solving problems for improving performance adhoc network There are currently many approaches to solve the problem of perfor- mance improvement in adhoc networks, in which there are three main approaches: Approach of routing protocols [30, 80, 70, 32, 53]; Approach of the process queue [12, 61, 63] ,and Approach of the media access [32, 13, 18, 50, 34, 57]. 1.3.1. Approach of routing Finding the optimal route through the establishment and maintenance of routing information at network nodes. 1.3.2. Approach of the process queue Focus on solutions to control inbound and outbound queues so that traffic flows between the flows and nodes appropriately 1.3.3. Approach of the media access Focus on solving the concurrency problem at the MAC layer. In which, improving distributed access protocols (DCF, EDCA) and improving pa- rameters in 802.11 EDCA. 1.4. Approach and research orientation On the basis of analyzing the main approaches, the thesis chooses the method of transmission medium access control for improving the perfor- mance adhoc network in IEEE 802.11 EDCA. This approach has been
6 selected based on existing issues in establishing IEEE 802.11 EDCA. (1) Problems of fixed parameters in IEEE 802.11 EDCA Many researches [19, 20, 21] are shown that that the IEEE 802.11e standard has partly met QoS assurance for the multimedia data type, but in terms of fairness, this standard is still limited very much because it only gives fixed values for the control parameter set in EDCA. (2) Problems of Fairness index in EEE 802.11 EDCA The competition for transmission medium by different priorities presents challenges with increasing fairness in IEEE 802.11 EDCA adhoc networks. Hình 1.1: Enhanced Distributed Channel Access of 802.11 EDCA [43] The thesis studies the following main contents: Content 1: Analysis and evaluation of the parameters in IEEE 802.11 EDCA have an effect on the throughput of data flows adhoc network. Content 2: The thesis propose a method for improving the QoS of the data flows according to different priority based on dynamic TXOP param- eter tuning mechanism in IEEE 802.11 EDCA. Content 3: The thesis propose a novel fuzzy logic approach for enhanc- ing the fairness of low priority data flows in IEEE 802.11 EDCA. 1.5. Consclusion Chapter 1, The Thesis systematizes the basic theories of adhoc net- works. The thesis is engaged in research according to the main approaches that domestic and international researchers have achieved. The thesis presents a research orientation to solve the chosen problems.
7 CHAPTER 2. ANALYSIS AND EVALUATION SETS OF PARAMETER IN IEEE 802.11 EDCA In this chapter, the parameters of IEEE 802.11 EDCA are analyzed through the investigation of transmission environment access control pa- rameters to evaluate the impact of the quality of Voice, Video, and Best- effort flows. 2.1. Enhanced Distributed Channel Access 2.1.1. Overview IEEE 802.11 EDCA The EDCA mechanism installed on IEEE 802.11e standard entities uses the multiple queues to receive and process the frames that need to be transmitted, classified according to each type of AC (Access Category). The IEEE 802.11 EDCA applies independent sets of parameters for each queue. The IEEE 902.11 EDCA is considered an upgraded version of IEEE 802.11 DCF using CSMA / CA and back-off function, but based on AC- specific parameters [79, 65]. 2.1.2. The format of IEEE 802.11 EDCA The format of the information field structure for the set of EDCA parameters is shown in Figure 2.1. In which, the field for the types AC (AC_BE, AC_BK, AC_VI, AC_VO) uses 4 bytes, each AC includes a set of parameters is shown in Figure 2.1. Hình 2.1: The format of IEEE 802.11 EDCA [43] The EDCAF function is responsible for assigning access to the medium based on priority values set in each AC. 2.1.3. Mechanism of access channel IEEE 802.11 EDCA The mechanism for channel access is performed by a set of parameters set for each AC. Meanwhile, the contention window (CW), transmit op- portunity (TXOP), arbitration interframe space (AIFS) are parameters
8 for changing the priority level of each kind of flow. These parameters are set permanently in EDCA AIF S[AC], CWmin [AC], CWmax [AC] và T XOP [AC]. (1) CWmin , CWmax parameters: are the maximum and minimum limits of the contention window (CW), used in the back-off algorithm. In which, CWmin [AC], CWmax [AC] are allowed to determine the factor Backoff[AC] by equation (2.1). Backof f [AC] = Random[0, min(2k (CWmin [AC] + 1) − 1, CWmax [AC]) (2.1) Which k is the number of the collision occur. (2) Tham số AIFS[AC]: ) is the waiting time before transmitting the next packet or initiating back-off algorithm. AIF S[AC] = AIF S[AC] × Te + SIF S (2.2) Where, Te is the variable time for 1 slot time. (3) TXOP(Tranmission Opportunity): is the maximum transmission time when the flow gains the right to participate in data transmission. T XOP [AC] = TDAT A + 2 × SIF S + TACK (2.3) Where, the time of transmission interval covers the entire frame exchange, such as waiting time SIFS, transmission time ACK, time of transmission data and time to send and receive RTS/CTS if use the mechanism RT- S/CTS. 2.2. Simulation and analysis of the results 2.2.1. Topology and simulation environment The thesis is evaluated the performance of the proposed method using Network Simulator (NS-2) [6]. The simulation parameters are listed in bel- low R=11Mbps, Antenna type=Omni directio, transmission range=250m, Carrier sensing range=500m. MAC protocol=MAC 802.11 EDCA, Packet siz=512bytes. Simulation time=150s, Connection type=UDP. This topol- ogy includes two nodes, node S and node D. The source (node S) sends three types of data (best effort, video, and voice) to node D.
9 2.2.2. Scenario of TXOP parameter Bảng 2.1: Scenarios of TXOP parameter TXOP AC TXOP Case 1 TXOP Case 2 TXOP Case 3 BE 3.264(ms) 6.016(ms) 10(ms) VI 6.016(ms) 6.016(ms) 6.016(ms) VO 3.264(ms) 3.264(ms) 3.264(ms) 2.2.3. Scenario of CW paramete Bảng 2.2: Scenario of CW parameter CW Case 1 CW Case 2 CW Case 3 Min Max Max Min Max Min BE 15 31 7 15 2 7 VI 15 31 15 31 15 31 VO 7 15 7 15 7 15 2.2.4. Simulation Analysis Simulation results of BE throughput flow according TXOP parameter are shown in Table 2.1 and Figure 2.2a. Simulation results of BE through- put flow according CW parameter are shown in Table 2.2 and Figure 2.2b. (a) (b) Hình 2.2: (a) Throughput of BE flow for each scenario in Table 2.1; (b) Throughput of BE flow for each scenario in Table 2.2 The fairness index evaluation for TXOP parameter are shown in Table 2.1 and Table 2.3. The fairness index evaluation for CW parameter are shown in Table 2.2 and Table 2.4.
10 Bảng 2.3: The fairness index in Table 2.1 802.11 EDCA TXOP Case 1 TXOP Case 2 TXOP Case 3 0.6 0.67 0.8 0.78 Bảng 2.4: The fairness index in Table 2.2 802.11 EDCA CW Case 1 CW Case 2 CW Case 3 0.6 0.67 0.79 0.77 Some comments from the simulations: (1) If the fairness index is raised, the throughput value will be narrowed. These two parameters are considered to be inversely proportional. (2) The value of TXOP (CW) parameter at least-priority flow is in- creased, (decreased) it will lead to increased throughput, but the band- width will break. (3) The TXOP, CW parameter in Enhanced Distributed Channel Ac- cess of IEEE 802.11 EDCA set to the default value given by IEEE 802.11 not suitable for adhoc networks [75, 10, 25, 49] because the network topol- ogy is always changing in terms of bandwidth sharing with neighboring nodes. 2.3. Conclusions In this chapter, the thesis focuses on analyzing the working mecha- nism of IEEE 802.11 EDCA. The simulation scenarios are constructed for two key parameters in IEEE 802.11 EDCA to evaluate the impact on net- work performance. The TXOP and CW parameters are evaluated based on the throughput and fairness index. The evaluation aims to determine the specific roles of TXOP and CW parameters for Voice, Video, and Best Effort streams according to two performance criteria: throughput and fair- ness. The thesis proposes a method for adjusting the parameters of 802.11 EDCA with the changing network topology to ensure the fairness of low priority flows in the case of large loads, thereby contributing to the im- provement of QoS for applications in ad hoc network.
11 CHAPTER 3. IMPROVING THE PERFORMANCE OF DATA FLOWS IN 802.11E EDCA WIRELESS AD HOC NETWORK BY ADJUSTING THE DYNAMIC TXOP PARAMETER 3.1. Introduction The IEEE 802.11 EDCA protocol has now become the de facto standard for media access control in the ad hoc wireless network. The parameters in EDCA is related to the probability of accessing the channel of each flow. The TXOP (Transmission Opportunity) parameter is the maximum transmission time when the flow gains the right to participate in data transmission. Although 802.11 EDCA has good support for multimedia data flows through the access parameter set according to the priority. However, many studies [75, 10, 25, 49] show that when the network load reaches saturation state, the high priority flows will tend to occupy the entire bandwidth of low priority flows, leading to unfairness in the network. The problem is how to choose the optimal TXOP parameter. The thesis proposes a solution that allows sharing bandwidth in a flexible manner among the different types of data in IEEE 802.11e by adjusting the TXOP value for each flow at the station, thereby improving the fairness index among data flows (Voice, Video, Best effort) in 802.11e EDCA. The simulation results show that the proposed method will help to improve the throughput, and the fairness index. 3.2. Proposal of a method according to dynamic TXOP param- eter 3.2.1. The idea of the proposed algorithm The goal of the method is to prevent the unfairness from happening when flows with high priority tend to occupy the entire bandwidth. In order to divide the bandwidth according to the desired ratio of 3: 2: 1 with priority, Voice, Video, Best-effort [14]. 3.2.2. Proposed method The thesis propose three modules that undertake the following functions
12 (1) TXOP-Flow module: Oper- ating at MAC layer, which func- tions to count the number of flows in the transmitter domain. A flow is determined based on the source IP address, destination IP address, source MAC address, destination MAC address and AC in the be- ginning of the frame. The symbol of the number of flows is n, ki is the weight for each type of data flow. The thesis set up kV O = 3, kV I = 2, kBE = 1. The module Hình 3.1: Mô hình IEEE 802.11 gives the total weight of the flows EDCA với các module đề xuất at the survey node according to the formula. n X W = (ki ) (3.1) i=1 Where, ki is the weight for each type of data flow, n is the number of flows. (2) TXOP-Flow-Active-Time module: There is a function to evaluate the true linkage performance of a flow within the Estimation Period EP (Estimation Period). The link performance is determined by analyzing the time used to deliver information packets in i flow with 80% of the current send and receive time and 20% of the previous send and receive time. TActime−time[i] U= (3.2) EP In which, U is the actual linkage performance of i. TActiveT ime[i] is the total time to send the current information packets of i flow i. TActive[i] is the total time of sending and receiving the previous packet of i flow i. (3) AdaptiveTXOP: Module contains algorithm to adjust TXOP pa- rameters based on actual bandwidth sharing ratio and fair bandwidth sharing ratio. TXOP parameter adjustment value i will be determined by formula (3.3). F SR[i] T XOP 0 [i] = × T XOP [i] (3.3) RSR[i]
13 3.2.3. Proposed algorithm Algorithm 3.1 The fair bandwidth sharing ratio 1: Init EP = 2s, W = 0 2: Setup kV O = 3; kV I = 2; kBE = 1. 3: for (each interval time EP) do 4: //Decode the packet 5: flow = Decode(packet) ; Flows.append(flow) 6: //Call the function for sums the weight of the data flows 7: W = F lows.T XOP − F low.W eight(f low) ki 8: F SR = . W 9: end for 10: Return F SR. Algorithm 3.2 The Fair bandwidth sharing ratio RSR 1: Init EP = 2s, W = 0 2: Setup: kV O = 3; kV I = 2; kBE = 1. 3: for (each interval time EP) do 4: //Call the function for Calculate the link utilization 5: TActiveT ime[i] = F lows.T XOP − F low.Active − T ime(f low) TActime−T ime[i] 6: RSR = . EP 7: end for 8: Return RSR. Algorithm 3.3 Thuật toán điều chỉnh tham số TXOP 1: Init EP = 2s, W = 0, RSR = 0, F SR = 0 2: Setup: kV O = 3; kV I = 2; kBE = 1. 3: for (each interval time EP) do 4: //Call the function for FSR 5: F SR[i] = F air_Share_Ratio : Get(f low) 6: //Call the function for RSR 7: RSR[i] = Real_Share_Ratio : Get(f low) 8: //Caculate the new TXOP parameter F SR[i] 9: T XOP 0 [i] = × T XOP [i]. RSR[i] 10: end for 11: Return T XOP 0 [i].
14 3.3. Simulation 3.3.1. Single hop topology The throughput results between Voice, Video, and Best Effort flows in Figure 3.2a and Figure 3.2b. (a) (b) Hình 3.2: (a) Throughput of 802.11 EDCA with standard parameters; (b) Throughput of flows with proposed method. The graph of comparing the total throughput in Figure 3.3b. The fair- ness index between the flows is shown in Figure 3.3a. (a) (b) Hình 3.3: (a) Comparing the fairness index of the two methods; (b) Com- paring the total throughput of the two methods 3.3.2. Multi-hop topology Simulation results of total throughput between data flows between the proposed method and the method of fixed TXOP parameter according to 802.11 EDCA are shown in Figure 3.4a and Figure 3.4b. Figure 3.5a is shown the total throughput of the two methods. Figure 3.5b is shown the
15 fairness index of the two methods. (a) (b) Hình 3.4: (a) The total throughput of data flows with standard EDCA parameters; (b) Total throughput of data flows with proposed method (a) (b) Hình 3.5: (a) Comparing the total throughput of the two methods; (b) Comparing the fairness index of the two methods 3.4. Analysis of the results 3.4.1. Evaluation of Throughput (1) Single hop topology: Comparing the throughput results in Figure 3.2a and Figure 3.2b, the thesis see that the throughput of Best-effort flow according to the proposed method is significantly improved while the throughput of Voice flow and Video flow is not degraded compared to the 802.11 EDCA standard. (2)Multi-hop topology: The comparison of the results in Figure 3.4a using the 802.11 EDCA standard and the result in Figure 3.4b by the proposed method, the total throughput of flows obtained from the proposed method makes the ratio among flows more balanced. 3.4.2. Evaluation of Fairness (1)Single hop topology: The fairness index between the flows is shown in Figure 3.3a. When the network load is small, the fairness of the two methods is equivalent, but when the network load is large, the fairness of
16 the proposed method is better than 802.11 EDCA standard. (2) Multi-hop topology: The fairness between the flows is shown in Fig- ure 3.5b, the results show that the proposed method gives a higher fairness than 802.11 EDCA standard. 3.4.3. Evaluation of delay (1) Single hop topology: The delay Độ trễ between the flows is shown in Figure 3.6a for 802.11 EDCA and Figure 3.7b for proposed method. The results show that the dynamic proposed TXOP parameter method gives a higher delay than the fixed TXOP parameter setting method. (a) (b) Hình 3.6: (a) Delay of data flows with standard EDCA parameters; (b) Delay of data flows with proposed method (2) Multi-hop topology: Total delay of data flows from S2 to D by pro- posed method and 802.11 EDCA is shown in Figure 3.7a. The results show that the proposed method is better than 802.11 EDCA standard. (a) (b) Hình 3.7: (a)Total delay of data flows between 802.11 EDCA and pro- posed method; (b) Total delay of data flows in the Multi-hop topology
17 3.5. Comparison of published results The thesis use two researchs [8]1 , [56]2 for comparing the two methods. (1)The throughput: the Table 3.1 is shown the result of research [8], [56], the authors are compared between IEEE 802.11 EDCA and proposed method. Bảng 3.1: Table the result of low priority data flows [8], [56] Throughput in [8] Throughput in [56] 802.11 EDCA In [8] 802.11 EDCA In [56] 0.06 (Mbps) 0.07 (Mbps) 0.3 (Mbps) 0.4 (Mbps) Table 3.2 is compared the throughput between the research [8], [56] and proposed method. The result is shown that the proposed method is better than 802.11 EDCA standard. Bảng 3.2: Comparison result of ratio throughput of low priority data flows Ratio throughput Proposed method In [8] In [56] 36.7% 16.6% 33.3% (2) The comparisons of fairness index: Table 3.3 is compared the fair- ness index between the two methods. The fairness index of the proposed method of Fuzzy logic is higher than the IEEE 802.11e EDCA. Bảng 3.3: The comparisons of fairness index Fairness index Proposed method In [8] In [56] 0.8 0.6 0.6 3.6. Conclusions The important contribution of the thesis is to propose a method to adjust TXOP parameters according to dynamic mechanism based on actual bandwidth sharing, to improve the fairness between multimedia data flows suitable for ad hoc networks. 1 Alam et al, "Enhancements of the Dynamic TXOP Limit in EDCA Through a High-Speed Wireless Campus Network," Wireless Pers Commun, vol. 90, p. 1647–1672, 2016. 2 Namazi, Mohammad, and Moghim, “Dynamic TXOP Assignment in IEEE802.11e Multi-hop Wireless Networks Based on an Admission Control Method”, Springer Science Business Media New York, vol .8, pp.6–17, 2017
18 CHAPTER 4. IMPROVING FAIRNESS IN IEEE 802.11 EDCA ADHOC NETWORKS BASED ON FUZZY LOGIC 4.1. Introduction Adhoc networks consists of a number of mobile devices that come to- gether to form a network as needed, without any support from any existing Internet infrastructure or any other kind of fixed stations. Most of wire- less Adhoc network architectures are currently based on the random access method of IEEE 802.11 EDCA (Enhanced Distributed Channel Access ) in CSMA/CA. In addition, the EDCA’s performance of voice and data traffic based on the type of access in IEEE 802.11e EDCA can be adjusted Content window (CW), Transmit Opportunity (TXOP), Arbitration inter- frame spacing (AIFS) are parameters to change the priority level of each kind of flow. However, this method is still fixed, merely changing the values of these parameters compared to the default setting in EDCA. 4.2. Relevant theory 4.3. Proposed method 4.3.1. The idea of the proposed method The thesis propose a method for adjusting the CW, TXOP parameter based on fuzzy logic with the changing network topology to ensure the fairness of low priority flows in the case of large loads, thereby contributing to the improvement of QoS for applications in ad hoc networks. 4.3.2. Fuzzy logic for control TXOP parameter Fuzzy processing includes three sequential steps: fuzzification, process- ing (inference system), and defuzzification. The center of the fuzzy con- troller is the fuzzy rule base. The fuzzy logic decision system is shown in Figure 4.1