International Journal of Interactive Mobile Technologies (iJIM) – eISSN: 1865-7923 – Vol 16 No 22 (2022) Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application Response Time Using a Distributed Model https://doi.org/10.3991/ijim.v16i22.35675 Ibrahim A. Lawal1, Adamu A. Ibrahim2(), Nuha A. Zammarah3 1 Department of Information Technology, Bayero University Kano, Kano, Nigeria 2 Department of Computer Science, International Islamic University Malaysia, Kuala Lumpur, Malaysia 3 Department of Computer Science and Information Systems, Taibah University, Medina, Saudi Arabia adamu@iium.edu.my Abstract—In the most recent few years, there has been a meteoric rise in the number of different wireless networks, as has the demand for wireless data ser- vices and multimedia applications. An improvement in quality of service (QoS) is required in order to meet the ever-increasing demand while also providing a service that is of a higher standard. One of the most cutting-edge technologies, known as World Wide Interoperability for Microwave Access (WIMAX), was developed with the quality of service in mind during its development. Despite utilizing the most recent advancements in technology, the WiMAX network con- tinues to struggle with QoS performance issues. A brand new distributed Client- Server model was designed for the fixed WiMAX Network in order to reduce application response time. This was done in order to improve the quality of ser- vice performance. In order to make a comparison with the existing centralized model, the performance of the proposed model was analyzed with OPNET Mod- eler 16.0. According to the findings, the newly proposed distributed Client-Server system is capable of satisfying the requirements set forth by its users owing to an improvement in QoS performance in terms of application response time when compared to the conventional model. The performance of the network was in- creased by the deployment of more Base Stations (BSs), Subscriber Stations, and the utilization of client-server Base Stations that were chosen by the Nearest Neighborhood Algorithm that was devised. Keywords—WiMAX, QoS, application response time, OFDM, FDM base sta- tion, subscriber station and OPNET modeler 16.0 1 Introduction WiMAX has become the clear frontrunner among cutting-edge technologies for es- tablishing high-speed data links over long distances. WiMAX's minimal infrastructure makes it a viable option for resolving issues with last-mile wireless connections caused iJIM ‒ Vol. 16, No. 22, 2022 75 https://doi.org/10.3991/ijim.v16i22.35675 mailto:adamu@iium.edu.my Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… by things like multipath fading, environmental conditions (such heavy rains), interfer- ence, and differing service-level agreements (SLAs). It works particularly well in re- mote locations where it can be challenging to construct wired infrastructure. One Base Station (BS) and one or more Subscriber Stations (SS) make up the bare bones of a WiMAX network [1]. Downlink (DL) traffic refers to the direction of communication from base station (BS) to serving station (SS), while uplink (UL) traffic refers to the opposite direction of communication (from SS to BS). WiMAX's primary architectures are Point to Multipoint (PMP) and Mesh Architecture. In PMP, a single BS supplies all of the neighboring SSs [2]. While SSs do talk to one another, they often talk to the BS first. Each time the SS makes a connection request, the BS functions as a gateway to the network and is responsible for establishing and managing those connections. How- ever, the mesh architecture is developed when communications between SSs are re- quired. Since connections can be made over multiple hops in a mesh topology, a tree network topology is theoretically possible. The PMP architecture is incompatible with the mesh architecture since it only permits one hop transmission and has less signaling overhead than mesh mode does [3]. Due to the inherent unpredictability and significant variability of wireless infrastruc- tures in comparison to conventional networks [4], the provisioning of QoS in WiMAX networks is done at the Medium Access Control (MAC) layer. WiMAX employs an association-oriented MAC architecture, where all downlink and uplink connections are controlled by the serving Base Station, and each connection is recognized with a con- nection identifier (CID), for data transmissions within the specific link [5]. The Support for Quality of Service (QoS) is an integral part of this design. For packets with a specific set of QoS parameters, such as traffic priority, maximum sustained traffic rate, maxi- mum burst rate, minimum tolerable rate, scheduling type, ARQ type, the maximum delay, tolerated jitter, service data unit type and size, and bandwidth request mechanism to be used [6], the MAC layer allocates traffic to a service flow identifier (SFID). In recent years, Quality of Service has emerged as a crucial component of data routing in wireless sensor networks (WSNs). When the outcome of recognizing a task depends not only on proper recognizing in the environment but also on timely delivery, QoS necessitates legitimate time data transfer. In order to send voice, images, or data in real time (for example, to notify someone of a time-sensitive, high-priority incident), a spe- cific delay and bandwidth are required [7]. The APM metric known as "Application Response Time" measures how long it takes for an application process to process a user's request for a service. It took into account the user's perspective when measuring. Actually, the user is the most important factor, and there are three main states of performance [8]: satisfaction, tolerance, and frustra- tion. Users frequently engage with the app throughout the day, therefore their overall impression of the app is likely the result of a compilation of many different interactions. Frustration on the network is to be expected if response times are poor, but persistent delays, especially if the user's task is serial in nature, are guaranteed to annoy. For this reason, it is essential that network applications can respond quickly enough to keep up with the new requirements [9]. 76 http://www.i-jim.org Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… The proposed new distributed Client-Server model aims to decrease the Application Response Time for the Fixed WiMAX Network, which will improve the QoS perfor- mance. In this setup, a Client-Server BSs is chosen to distribute the network using a Nearest Neighborhood Algorithm. OPNET Modeler 16.0 will be used to analyze the results of the comparison between the suggested model and the current Centralized model. While the current centralized approach employs the FDM method of transmis- sion, the proposed Client-Server architecture will use the OFDM method. To find out how much of an upgrade to the network can be expected from using the new distributed Client-Server model, we'll compare its performance to that of the current models. All of the components, including the Base Stations (BSs), the Subscriber Stations (SSs), and the client-server BSs, will combine to establish the quality of service (QoS) levels. 2 Related work In recent years, both the availability of and demand for wireless data services and multimedia applications have increased dramatically, spurring the development of a plethora of new wireless networks. There has been a lot of study into Quality of Service (QoS) to figure out how to improve service to keep up with rising demand. Many of the articles have dealt with QoS concerns as the IEEE 802.16 standard has been developing and expanding. There is a brief overview of recent research in this area here. Basic mechanisms for achieving QoS in packet networks are reviewed by Guerin and Peris [10]. The mechanisms for providing differentiated services are discussed, as are the control path techniques required to facilitate agreement between users and the network on service definition. The IEEE 802.16 standard's QoS features are based on these ideas. The integrated QoS Control for IEEE 802.16 is described by Chen et al. [11]. For Point-to-Multi-Point (PMP) mode, a quick signaling system is planned to supply a cross-layer integrated QoS. IEEE 802.16 networks are supported by a cross-layer QoS framework proposed by Mai et al. [12]. This paradigm proposes two novel ways for enhancing performance. Recent developments in network modelling, QoS mapping, and QoS adaptation in terms of delivering end-to-end QoS for video distribution over the wireless internet are described by Zhang et al. [13], who also give a broad frame- work of a cross-layer network-centric approach. Similarly, Chu et al. [14] propose a same structure for the 802.16 MAC protocol. It consists of a traffic classifier, an up- stream scheduler for the SS, and upstream and downstream schedulers for the base sta- tion. To provide quality of service for 802.16 networks, a combination of priority sched- uling and dynamic bandwidth allocation forms the backbone of the network's architec- ture. As an added bonus, it suggests productive methods for schedulers to use. Alavi et al. [15] also offer an open architecture to support QoS mechanisms in IEEE 802.16 standards, which is similar to what is proposed in this study. They argue that the difficulty is in creating an efficient design to achieve the QoS requirements, despite the fact that the IEEE 802.16 standard offers various techniques to do so. That's why it's so hard to guarantee quality of service. They provide a design strategy to implement the suggested architecture for all types of traffic classes as specified by the standard, which would be a significant step toward resolving the problem. Cicconetti et al. [16] iJIM ‒ Vol. 16, No. 22, 2022 77 Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… discuss the topic of quality-of-service support in IEEE 802.16 networks. Using a pro- totype simulation of the IEEE 802.16 protocol, they assess the networks' efficiency. In [17], Nair et al. detail the media access control techniques deployed in WiMAX networks. Discuss how this MAC protocol’s capability can be used to facilitate Wi- MAX deployments, including the sorts of provisioning and Quality of Service (QoS) that can be attained. In this article, we discuss the difficulties of implementing the Wi- MAX MAC to meet quality-of-service requirements. In [18], Sayenko et al. describe a scheduling method for the WiMAX backbone. WiMAX specifications don't specify the scheduling policy, or the algorithm to allocate slots. The door is open for creative en- actment. From what they've seen in simulations, it's clear that the suggested scheduling algorithm can meet the needs of all WiMAX service types in terms of quality of service. 3 Research methodology Simulations were the preferred method utilized for this study's analysis as the key tool employed for the measurement of the response time. The research conceptualized “response time” as the primary unit of measurement. Hence the study's overall goal was to better understand the relationship between the application response time and the ap- plication response time model in the context of the network scenario that was used for this study. The context for this research was the network that was utilized for the Fixed WiMAX network. In order to validate the model, the experimental simulations analysis was carried out. 3.1 Conceptual framework The "Application Response Time" and the "Application Response Time model" in this research established a tradeoff, and the model that was proposed in this study was considered to be more effective as a result of this tradeoff. This is the foundation for the conceptual framework that has been proposed. From the point of view of the user, the Application Response Time (ART) is the most significant quality of service char- acteristic. It is the period of time that elapses between the sending of a request and the time at which the user is presented with the response to the request. A response time is the amount of time that has passed since an enquiry was made before receiving a re- sponse. The formula for calculating the Application Response Time is shown in Equa- tion (1) [7]. n ART r Tthink = − (1) where n is the number of concurrent users r is the number requests per second the server receives Tthink is the average think time (in seconds). This study proposed to model the application response time because there is a need to improve the situation. In order to do so, we used the average value of the response time that an application uses to respond to a number of requests per second. This value was determined by using the data from the study, and the calculated by Equation (2). 78 http://www.i-jim.org Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… ( )_ Pr 1 i n SS App server oc i avg T T ART r − = + = ∑ (2) where n= number of SS requesting the network information. Tssi-App_server = Time to send a request from SSi to Application server. Tproc = Time to process a single request from the application server r = number of requests per second application server is receiving. Another way to model Application Response Time is to use network response time. ART NRT TRT= + (3) ( ) payload NRT APP Turns RTT bandwith = + × (4) where Network Response Time (NRT) = the time between a user’s action and the Network’s response to the action. Transaction Response Time (TRT) = the time taken for the application to complete the transaction. Payload = the information content in bytes. Bandwidth = the minimal link speed between client and server. APP Turns = the number of interactions needed between the client and server to pro- vide a response to the user. Round Trip response Time (RTT) = the propagation time for data between the client and server. TRT SRT CRT= + (5) where TRT= Transaction Response Time STR= Server Response Time CRT = Client Response Time where Server response time (SRT) = the processing time required by the server. Client response time (CRT) = the processing time required by the client. The proposed Distributed Client-Server model will be having less application re- sponse time as compared to the Centralized Model because; the number of SS request- ing from the central server will be less. Thus, the Client-Server BSs selections Algo- rithm, Client-Server BSs communication Algorithm as well as the Base Station and Subscribers Stations Communications Algorithm are presented. The Nearest Neighbor- hood Algorithm was used to describe the process of selecting clients and servers, and the flow chart was used to illustrate the process as shown in Figure 1. In this regard, some of the closest BSs will be chosen to serve as Server BSs in order to deliver net- work information to the closest BSs that do not already have the information. iJIM ‒ Vol. 16, No. 22, 2022 79 Paper—Improving the Quality of Service of Fixed WiMAX Networks by Decreasing Application… StartStart i ≤ K BS={ BS1, BS2, BS3, ..., BSn } Register with the central server P= {P1, P2, P3,..., Pm } BS be the set of m Base Stations (Candidates for Master) and m