NS3 example of channel allocation on 2.4GHz WIFI network
Abstract
The NS-3 Network Simulator is a powerful research tool that allows users to test packet forwarding protocols and algorithms without a need for underlying devices. It is typically a collection of source code that can be downloaded as program code and formed on top of an existing operating system. As computing device networks grow in size and complexity, the need for extremely accurate and customizable network simulation innovations becomes essential.
The 2.4 GHz WLAN structure is divided into 13 reciprocated streams, each with 22 MHz of frequency band. Only three of the thirteen networks are considered to be non with one another. Due to the rise of large-scale network testbeds, testing remains critical in terms of adaptability (both in experimental speed and size), rapid prototyping, education, and reproducibility. The framework can be studied in detail at varying scales, with varying data applications, and under differing field conditions, resulting in measurable and easily interpretable findings. The Wi-Fi standard has incorporated advanced and powerful PHY-layer methodologies in its evolvement to just provide ever higher data transfer, which has enhanced the sophistication of network-wide obstruction relationship issues. Proper modelling of the eventually results across device connections is critical in order to accurately estimate performance of the network Wi-Fi, particularly in today’s traffic and network densifying perspective. Simulators for event-driven, such as the ns-3 open-source, can in theory encapsulate these interconnections; nevertheless, it is critical to verify their underlying models in contrast to test findings to ensure that they accurately reflect network behaviour in practice. First, experiments were carried out in a indoor large-scale test platform to justify the IEEE 802.11ac Wi-Fi framework in network simulator 3 for various channel allocation configurations and width. The outcomes prove that ns-3 accurately models co-channel Wi-Fi interrelations but does not account for adjoined channel interference, which some previous tests display is precarious in complex systems. As a result, the author proposes and implements an ACI archetypal for ns-3. Importantly, the author’s model accurately predicts the contextual behavior of the CA/CSMA framework for data transmission occurs on end-to-end channels. Furthermore, when tried to compare to the fundamental model without ACI in ns-3 Wi-Fi, The ACI integration greatly enhances connectivity and per-device data transfer accuracy estimates for the IEEE 802.11ac network that is thought to be dense. Without ACI modeling, for example, ns-3 vastly overstates increase network bandwidth by approximately to 220 percent, while the estimate for ACI deployment is only 64 percent greater than the experiment conducted observations.
CHAPTER 1: INTRODUCTION
Wireless network development is always in sequence, and new standards are introduced. IEEE 802.11b/g/n standards are widely used in many devices. The most recent standard, 802.11ac, is used in new devices. The IEE is currently working on a new standard called 802.11ax. Simulations are one method for studying network behaviour. A network simulator can also be used for this objective. It enables the simulation of a computer network, including devices, deployment models, channel conditions, and implementations. The simulator can be set up to assess performance at different circumstances. The sector employed simulators to develop and test new features and expansions for the next generation of wireless guidelines. The reason for using a simulator rather than mathematical analysis is that when you try to analyse a larger circumstance with various nodes, the mathematical formula becomes rather sophisticated. If one wants to create scenarios with more than a handful of equipment, using a testbed is also more costly than to use a simulator. This is critical for the creation of future standards. It also offers benchmarks, numerical simulations, and assessments for various deployment instances.
1.2 Research Rationale
Because of its inherent advantages over diagnostics with real hardware, simulation is an important tool in networking research (Rahman et al., (2009)). Budget and time costs make simulation appealing for research, whilst also adaptability, expandability, and pretty nearly no cost advancement are also significant advantages above all other testing protocols (Obaidat & Boudriga, 2010). Network simulators enable us to study and implement various network elements in a virtual environment, allowing us to test new technologies, protocols, topologies and conceptual models with a high degree of flexibility. Centralized wireless local area networks (WiFi) based on the IEEE 802.11 standard, such as those found on university premises or airport terminals, typically necessitate meticulous planning and transmission assignment in order to achieve better performance and reduce co-channel jamming. Channel allocation is a critical issue in multiple channel wireless connections. It is a tactical cohesion and routing of the links and channels. The goal of channel allocation is to reduce the amount of noise and interference. Nevertheless, there are the following practical limitations in channel allocation: (i) The amount of radios per nodule is limited in relation to the sum of channels that are orthogonal 802.11 allows to mitigate the need for numerous diametrically opposed networks in a specified spectral range channel conflict while also minimizing Obstruction induced by other wireless traffic on the same circuit. (ii) There are a limited variety of possible channels. (iii) There is a continuous necessity keep the network connected. 801.11b allows for 3 channels in a 2.4GHz spectrum whereas 802.11a allows for 12 fully orthogonal channels in a 5GHz spectrum. Ns-3 was designed to be the successor to ns-2, with its development teams attempting to address or alleviate several of the well known ns-2 flaws while also incorporating new ideas, such as verification and software development strategies, to generate an extra trustworthy simulation platform to assist industrial and scholarly investigation. The Ns simulation software group has its origins in Unix-like settings, as it was created using standard UNIX tools and languages. Linux is now one of the most popular development environments for both ns-2 and ns-3. Though they’ve been mapped to all other operating systems (OS), like BSD, Cygwin, windows, and OSX for Mac, operating under Linux/GNU environments makes it far easier to configure, keep updating, and preserve instances between both simulation models. Ns-3 provides an all-in-one installation package that only contains compiled ns-3 source code. During the execute stage, the ns-3 code scripts will notify you of any missing software interconnections, which can be easily obtained from various sources. Ns-3 is a distinct simulation-based open computer network simulation environment. Each event in a simulator is linked to a specific runtime, and the modelling moves through successfully implementing actions in the simulation time direction. Once an incident is handled, it may result in the generation of one, more events or zero. Events are ingested as a model runs, but more incidents may (or may not) be produced. When a special “Stop” event is found or when there are no more events in the event queue, the simulation will automatically stop (Riley and Henderson (2010)). The Ns-3 is the culmination of a series observation that progressed across previous two simulation models (i.e. ns-2 and ns-1): notwithstanding, the developer crew resolved to finally relaunch the ns-3 design process on various principles, without any interoperability, coming up with totally new masterpiece while preserving and utilizing the past expertise variants. Ns-3 simulations can be coded in Python or C++ with their elementary execution and setup governed by a procedure recommended by the project teams. Ns-3 was created with scalability in mind. ns-3 is intended use for research, with the requirements of the scientific world in mind, and encourages a peer cooperation framework to aid in the creation of new functionalities as well as verification or peer assessment operations across various clusters. ns-3 was released as part of an open-source framework and the free software framework. The modelling interface was created to be used as well as a real-time emulator, allowing physical portions and simulation to coexist across that very same channel. A real-time scheduler, in addition to simulation support, provides this level of integration to allow for in the loop simulation. The ns3 simulator is designed for use through its API within algorithmic characterizations of configurations and users’ scenarios that can be created using standard processes and workflows. The API is employed to generate results in the form of marks that are consistent with those obtained by watching traffic in network systems using toolkits such as (TCPdump (https://www.tcpdump.org/) or Wireshark (https://www.wireshark.org/) or , so that they can be analysed using the same toolkits that have been in use for diagnostics and experiment operations. The researchers of ns-3 expressly declared that they intend to allow reusing of real world security policy implementations for ns-3, in order to minimize the effort related to the production and design of very detailed simulation modules as much as feasible. The aim should be to provide very accurate simulation help for a wide range of communication, protocol and network infrastructures, technologies, and layers, enabling network-enabled programs to run with minimal effort on highest part of the simulation.
1.3 Objectives and goal
The overall goal of this thesis work is to create an NS3 example of channel allocation on a 2.4GHz wifi network and undertake performance reviews under varying circumstances and with parameter settings. As such, the objectives of the study will be:
- To investigate the impact of NS3 channel allocation on a 2.4GHz wifi network
- To investigate performance reviews of NS3 channel allocation on 2.4GHz wifi network.
1.4 Thesis organization
This thesis is arranged as follows: Chapter 2 provides some background material and related work that is required to understand the subsequent discussions. The algorithm is introduced in Chapter 3 and the NS3 example of channel allocation on a 2.4GHz wifi network is developed. The fourth chapter contains extensive results from ns-3 simulations and field test results of the algorithm on a 2.4GHz wifi network. Chapter 5 presents the thesis’s conclusions and discusses potential directions for future research.
CHAPTER 2: LITERATURE REVIEW
2.1 Introduction
Simulation is a popular tool in the field of computer networks. The review of the literature includes both free and commercial simulation tools intended to assist in the design and analysis of communication systems at different levels and for diverse reasons. ns-3 is a discrete-event simulation tool intended primarily for education and research. The platform is an open – source that is sustained as determined by the public and is upgraded every three months with fresh content and numerical simulations. The software facilitates the establishment of modeling techniques that mimics many established network technologies using a real-time network emulator. Because it can penetrate walls and barriers better than 5.0 GHz, the current generation of WiFi uses the 2.4 GHz non-licensed frequency range. The 2.4 GHz frequency band is used by many other short-range wireless networks, including microwave ovens, ZigBee, Bluetooth, and cordless telephones. Local area wireless networks can be classified based on how they are managed, they are classified as either centralized or distributed (Surachai et al., 2010). A centralized system has a central controller who makes all network decisions and is assumed to know all network data. Each device in a distributed system makes decisions based solely on the information available to it locally. Because of increased processing knowledge and power at the central decision-making enterprise, centrally controlled systems outperform distributed networks, but they are not as scalable. Because of the constraint 15-AP and the CORAL units’ cognitive capacity, centralized network made it possible for the advancement of extra intellectual algorithms to greatly increase band usage.
2.2 Empirical Review
Wi-Fi makes use of the IEEE 802.11 standard (Wireless Fidelity). 802.11a, 802.11b, and 802.11g are some other Wi-Fi plugins. Wi-Fi is a radio communication technique that operates at a frequency of 2.4 GHz. A study by (Nicolaescu, 2008) shows Wi-Fi is a wireless communication system developed by the Wi-Fi Alliance that ensures data communication over a local network (WLAN) using IEEE (Institute of Electrical and Electronics Engineers) 802.11 communication standards. This system is now used by a variety of devices such as personal computers, video cameras, cell phones, televisions, and others. According to (Shi and LI, 2017), The preliminary 802.11 standard calls for a 2.4 GHz DSSS (Direct Sequence Spread Spectrum) system. A number of provisions have significantly increased WLAN functionality by trying to define supplemental advanced modulation techniques and bandwidth utilization. IEEE 802.11b was the first generally recognized WiFi protocol, functioning on the 2.4 GHz band, as contrasted to IEEE 802.11a, which used the unlicensed 5.0 GHz band. Several studies on current issues have been conducted recently. In Dolinska et al (2017) The researchers presented a number of findings concerning co-channel interruption in the 2.4 GHz bandwidth. The study investigated a variety of algorithms used to avoid non-overlapping channel allocation, and set to start with the Minimum Spanning Tree Inspired (MISTI) methodology, they created two channel selection methods that were assessed on four Python instances and compared in terms of Signal to Interference Power and Interference Proportion. In Sarbu et al., (2018) The authors test the Wi-Fi implementation in a residential facility with two other channel interferences: a Baby video monitoring system and a Bluetooth device. A study by (Weingartner et al., 2009) illustrates that the ns-3 Network Simulator handles networks which are huge magnificently in terms of drop probability, memory usage and simulation time. The study by (Rodrigues et al., 2000) indicates the taken steps during survey of a site are defined, and an optimization integer algorithm is provided for using the map on radio coverage to select the most effective channels premised on the interruptions detected because of the APs. This method relies heavily on the effectiveness of the site visit and appears to work well in networks with a gradually changing intervention surroundings. Cellular networks, function on a licensed band, for example, which means that no other intrusion contributions will be revealed unless installed by the business that possesses the bandwidths. (Kotz and Essien 2005; Chen et al., 2006) finalized a ground study displaying statistics traffic for a university and an urban network that is wireless. Their findings revealed that several considerations, such as location and time of day, influence the data distribution sent by customers’ networks. To achieve the finest outcomes, these diverse users with varying transmitting behaviours should be dealt with very inversely. Mishra et al., (2006) sought to develop a DCS method with there is no additional added expense in which each AP was designated a distinct channel series through which It stepped through time. The classifications were designed to maintain the APs causing the maximum intrusion from using the same platform. In Cao and Zheng, (2005), The researchers suggested a local negotiating method, which is a one-to-one negotiating technique that effectively transfers channels between adjacent subscribers, with the sum of the throughputs of the users as the optimal power flow objective. This algorithm was shown by Etkin et al., (2007) to choose suboptimal solutions. To determine the best approach, the writers suggested a repetitive model game in which a vibrant game circumstance is generated by repeatedly attempting to play static games that are equivalent. This enables operators to decide on choices based on previous flows in order to reach at a much more effective outcome. While it may reach an optimum position, the quantity of surplus necessary to finish numerous games will be extremely high, limiting adaptability. Ns-3 places all simulator core and model source code files in the src folder, separating them from the remaining auxiliary tools, building scripts, documentation, and examples. To accommodate the above categorization of abstract entities, the src folder is also divided up. As a result, all network device models can be found in src/devices. This tiered structure makes it simple to assist programmers in navigating the simulation software script. Succeeding the preceding structure, programmers of ns-3 are urged to identify a few elements in their conceptual framework in order to be able to fit their script into the multilateral simulation software framework, making sure a shared archetypal skeletal structure as a result, the overall code is more adaptable and flexible, as well as simpler to modify and recycle in the coming years. Because the src folder was designed to only comprise normal ns-3 modules that are managed, backed, and catalogued by the ns-3 project, new third-party components should be generated and founded into the ns-3 home/scratch folder. The waf tool searches the scratch folder for executables for each defined simulation. The ns-3 strategy overwhelms the aforementioned disadvantage by supplying a libns3.so library that should be the same those ns-3 scenarios connected to the identical rollout. If programmers place new simulations and models in the scratch folder, they will acquire basic codes that require the libns3.so shared library in order to use ns-3 standard features, and they should include the personalized models as part of their own binary code. These executables are only limited by the architecture of the targeted processor (amd64, ppc, i386, etc.). Doxygen, an open-source multi-language documentation generator that parses and extracts ns-3 documentation directly from source code, is used by Ns-3 to incorporate documentary evidence in source file codes. It permits you to specify certain elements of the documentation’s format, such as return values, parameters, and associates to other portions of citations. Furthermore, documentation can be produced in a variety of formats, including HTML, LATEX, XML, and PDF. As a result, generating ns-3 documentation is as straightforward as gathering the code’s source, and it eliminates the requirement to keep two separate documentation sources. Thus, code writing and commenting written documents are combined into a user and a single task can obtain this details in the layout that meets all the requirements (Ns-3, 2009). Ns-3 has a shorter lifespan compared to NS-2, but it has profited from ns-2 experience and ideas, so it cannot be considered to have been created from the ground up. ns-2, on the contrary, has passed across various phases throughout its extended presence and is It is now starting to reach the end of its lifespan, with code fixes and Its highest concerns are maintenance updates. Earlier as specified, this thesis shapes on the work of a prior similar artefact (Font et al., 2010), getting into detail into open source code analysis by calculating and interpreting software algorithm.
2.3 Summary and Research Gap
This section gives a review of the literature on the NS3 example of channel allocation on a 2.4GHz wifi network. A theoretical framework is included in the literature review. Furthermore, previous studies on this topic were evaluated in the literature review. According to several researchers, NS3 channel allocation on 2.4GHz wifi networks has a significant value. Nonetheless, some researchers claim that NS3 has a negative impact on 2.4GHz wifi networks. Nevertheless, some researchers discovered no effect. Furthermore, a review of the impact of the NS3 example of channel allocation on a 2.4GHz wifi network reveals that some researchers discovered a mixed impact. Some studies, on the other hand, indicate that the NS3 example of channel allocation on a 2.4GHz wifi network has a positive effect on wifi network performance. According to the theoretical and empirical reviews, there have been discrepancies in the arguments presented by theorists and previous researchers. This indicates that there is a research gap in the allotment of NS3 channels on 2.4GHz wifi networks.
CHAPTER 3: METHODOLOGY: NS3 EXAMPLE OF CHANNEL ALLOCATION ON 2.4GHZ WIFI NETWORK
3.1 Introduction
This chapter presents NS3 example of channel allocation on a 2.4GHz wifi network and adapts the APs’ channels to enhance utilization and spectral response effectiveness. Section 3.2 describes the ns-3 Simulator and shows how it responds to both external interference and intra-network. It also explains the overall customised considerations, which enable the methodology to adapt to different types of network situations. Section 3.2.1 discusses the ns-3 Wi-Fi Module model of the Channel Allocation on 2.4 GHz Wifi Network Dynamic Channel Selection algorithm (DCS 2.0), which uses an enhanced frame weight to make a distinction among both active and idle clients. Lastly, Sections 3.2, 3.3, and 3.4 explain Channel Allocation on a 2.4 GHz WiFi Network and data interaction, classifying the channel selection issue as NP-hard and proposing a binary integer optimal solution approach for resolving it. (Michael and David, 1979).
3.2 ns-3 Simulator
The bandwidth of 2.4 GHz is split into thirteen (13) streams with a width of 22 MHz and a spacing of 5 MHz. As shown in Fig.1, There are three non-overlapping networks in the IEEE 802.11 standard One(1), Six(6), and Eleven(11)). Channel 1 has a center frequency of 2.412 GHz, while channels six(6) and eleven(11) have a center frequency of 2.47 GHz and 2.46 GHz, correspondingly.
Figure 1: depicts a spectrum analyzer view of the frequency space occupied by these fourteen channels
Wireless communication have grown in popularity over wired networks due to features such as ease of maintenance, flexibility, mobility, and installation, and lower cost. Because (WLANs) Wireless Local Area Networks enable data transfer at high speeds without the use of wires, this know-how has grown in reputation in recent times and is already an inevitable component of our personal and business lives (LaRocca and LaRocca, 2002).
Simulators using the ns-3.26 simulator will be used to evaluate the techniques used in this research project. The ns-3 simulator is a discrete event network simulator designed for research, education, and experiment simulations. Because the simulation software is free to use, researchers can segment their offerings. It’s a brand-new simulation tool that is not a backwards-compatible extended version to the ns-2 network simulator. As a result, ns-3 applications are not supported (ns-3 Network Simulator, 2020). Even though ns-3 is a very sophisticated simulation tool, there are several considerations to use it, including conducting studies or experiments that would be impossible to function in real-world systems, attempting to regulate difficult systems and analyzing the actions of these systems controlled, and understanding precisely how the channels function. When tried to compare to certain other simulator networks, ns-3 has many distinguishing features. To begin, it is composed of numerous library packages that can be supplemented collectively or with independent software library collections whereas creating a only one graphical user interface setting in which all operations are supported. Second, independent data analysts, advanced analytics, and animations can be recycled in conjunction through ns-3. Third, consumers can use Python programming or/and C++ on the command – line interface, and modeling packages can be authored in C++ or Python. Fourth, while ns-3 is primarily utilized throughout Linux operating system, it can also be used in Cygwin (for Windows) and VSD Free. Ultimately, it is open source software with widespread assistance from the community. Furthermore, ns-3 is exceptionally appealing compared to ns-2 for a variety of reasons. To begin with, ns-2 has not seen major development in the main code, whereas ns-3 is actively maintained through a user mailing list. Second, ns-3 includes some functionalities that ns-2 does not, such as the ability to run real implementation code in the simulation model. Third, ns-3 has a large number of comprehensive simulations in various areas of research (such as Wi-Fi models and LTE). Subsequently, ns-3 has a minor level extrapolation layer compared to ns-2. Within ns-3, there are some foundational artifacts that are regarded the foundation for any ns-3 framework. The Node class represents the basic virtualization device abstract concept in ns-3, which would be the Router. The Node can be thought of as a computer that will be packed with different functionalities such as protocol suite, tools, and ancillary tokens. While a Channel is a physical connection among both two or more nodes. Channel variants include CsmaChannel, WifiChannel, and PointToPointChannel to name a very few. a network card is the NetDevice that can be integrated into a node’s output/input interface and contains both simulated hardware and a software driver. As a result, NetDevice is implemented in a Node to allow communication amongst Nodes in the model via Channels. There are several types of NetDevices, including CsmaNetDevice, WifiNetDevice, and PointToPointNetDevice. The Implementation, on the other hand, is a packet generator or consumer that can run on a node and communicate with a set of network slabs. As a direct consequence, several links among NetDevices, Channels and Nodes must be established in ns-3. There are numerous Assistants in ns-3 that assist in the creation of NetDevices, the assignment of IP addresses, the addition of MAC addresses, the installation of NetDevices on Nodes, the configuration of the protocol stack, and the connection of NetDevices to Networks.
3.2.1 Wi-Fi Module in ns-3
WifiNetDevice is a NetDevice in the ns-3 that generates features of 802.11-founded ad hoc systems and infrastructure. ns-3 includes simulations for various 802.11 flavors, including 802.11b, 802.11g, 802.11ac, 802.11a, 802.11s and 802.11n (both 5 GHz and 2.4GHz bands) (ns-3 Network Simulator, 2020). It includes various propagation loss frameworks, regeneration delay modeling techniques, and optimization techniques. Furthermore, in ns-3, WifiNetDevice can exist simultaneously with supplementary NetDevices, which is not possible in ns-2. The WifiNetDevice deployment includes three model sub-layers: the physical layer (PHY), which is in charge for modeling tracking energy consumption and packet reception, the MAC low layer, which designs features such as data transmission (CTS/RTS, ACK and DCF), and the high layer of the MAC, which integrates non-time important processes such as state machines and transmitter generation. However, there are a little restriction in the NS-3 Wi-Fi module, such as the lack of modeling of interference from other technologies, the lack of support for MIMO systems, and the use of only one channel model (YansWifiChannel:: ns3). As a result, only Wi-Fi nodes can be connected to the ns3::YansWifiChannel. As a result, supplementary knowhow, for instance as LTE, are not permitted, necessitating some improvements to the basic or default available prototypes in ns-3.
3.2 .2 Channel Allocation on 2.4 GHz Wifi Network
Because the maximum number of channels in the 2.4GHz ISM band that are not overlapping is three (channels 1, 6, and 11), each cell can have up to six different channel assignments. Because of the unique channel restriction placed on Aps (Access Points) in the same cell, the algorithm must determine channel assignments for 3-sector cells, which differs from previous single AP scenarios. The overwhelming majority of traffic recognition techniques presume that all obstruction sources are within the same network, enabling for the use of more comprehensive network data, for instance queue size transmit (Ihmig and Steenkiste, 2007). This assertion was inappropriate because the method’s primary objective is to recognize respectively external interference and intra-network. The researchers of (Rozner et al., 2007) suggested an estimation technique that provided traffic understanding by utilizing the burden of sending or receiving Access points. The effect of channel access is proportional to the number of streams. For instance, it has little effect at 5 GHz, where there are many channels (nine), making conflict-free assignment relatively easy, especially in comparison to 2.4 GHz (three channels), where it has a massive effect. When channels are assigned to cluster centres, they have an effect on the enabled load of the cluster based on the sharing of its channel (and thus capacity) with other clusters in its neighbourhood. While both channel assignment and clustering and have an impact on the amount of net traffic served, their contributions to is skewed, with clustering playing a significant role. Given the cross docking, it allocates channels to clusters in attempt to settle as many inter-cluster confusion as possible while retaining as much of the (traffic satisfaction) from the grouping process as possible. While this decoupled strategy is less complicated, it addresses the task that the clustering remedy that provides the maximum in the first step may not contribute to a final solution with the best traffic fulfilment (after channel assignment). In other words, a clustering remedy with a smaller (after clustering) may incur less loss during channel assignment, resulting in a higher. To resolve this concern, AmorFi employs a stepwise optimization technique. The controller receives the accumulated load requirements of each RRH at the start of each epoch (based on estimates from previous epochs). It then performs optimization iterative, solving a dual of the challenge in each iterative process to find the best network-wide structure.
The NS3 example of channel allocation on a 2.4GHz wifi network is grounded on a network detection segment that enables nodes in a multi-hop network to dynamically allocate channels. Nodes four hops apart can use the same channel without harmful disruption (While using MAC layer references, this is the case). In 2.4 GHz channel allocation, the coverage zone is relatively small, and wireless sensor nodes are only two travels apart from the sensor nodes. This was established by field dimensions that could be taken. NS3 was created for larger networks. As a result, the author streamlined the NS3 channel allocation procedure as follows. The sink node is outfitted with three radio crossing point that will operate on three orthogonal channels. This makes it possible for the sink to boost its response potential and obtain data on three different functionalities at the same time. With a data transmitted speed of 200 Kbps, three crossing point are satisfactory to collect all network data. Channels used by the sink node will be used by other network nodes as well. As a result, the entire beam network will use three frequencies. This enables the deployment of multiple sensor networks in the same physical space while using different channels for each network. It should be noted that IEEE 802.15.4’s physical layer supports the utilization of sixteen perpendicular channels in the 2.4 GHz band rate.
Because of the numerous considerations and complex scenarios that must be evaluated, channel assignment mechanisms are typically evaluated using simulations(Nezhad and Cerdà-Alabern, 2010). Nevertheless, most current network simulation models dont uphold multi-radio access points or active radio re-configuration interfaces, necessitating adjustments at the fundamental level to allow the assessment of CA methodologies(Aguero-Calvo and Perez-Campo, 2010).
3.3 Components for simulation of multi-radio mesh networks
The author introduces the necessary extensions to the ns-3 simulator (ns3 development team, 2011) for simulating hybrid Channel Allocation techniques in this section. These plugins include the fundamental work upon which the Channel Allocation framework can be built, as well as instructions on how to simulate interruption for responsive CAs.
3.3.1 Wireless nodes with multiple radios
The simulation scenarios require multi-radio wireless mesh nodes as fundamental building blocks. Fig 2 depicts a node which is wireless mesh with multiple radio integrations.
Figure 2: wireless mesh node with multiple radio functionalities
The diagram depicts an overall cross-layer channel assignment procedure that interacts with the simulation’s functionalities, which include a routing protocol, a channel sensing mechanism, and traffic generators. Program 1 demonstrates how to build a (MRNode) in ns-3 multi-radio node, where I symbolizes the radio number of access technologies that should stand implemented on a connection point. It should be noted that in ns-3, each radio interface is automatically assigned a unique ID beginning with 0.
Program One
Program 1 shows how to create a In ns-3, a multiradio node, where I stands for the number of radio access technologies that must be configured on an access point. It should be noted that in ns-3, each radio interface is assigned automatically a unique ID beginning with 0.
Table 1 Configuration of the physical and MAC layers
| Parameter name | Description | Sample value |
| Standard | MAC and physical layer standard | ns3::WIFI_PHY_STANDARD_80211a |
| PropagationDelay | The time it takes for data to travel amongst the stipulated destination and source. | ns3::ConstantSpeedPropagationDelayModel |
| PropagationLoss | Modalize the propagation delay throughout a medium of transmission. | ns3::RangePropagationLossModel |
| MaxRange | Radio range | 250 m |
| RemoteStationManager | Transmission rate of data and control packets | ns3::ConstantRateWifiManager |
Figure 3. The node’s class diagram
3.4 Network Devices
3.4.1 Radio Interface configuration
ns-3 assigns a unique ID to each channel, which begins with 0. The channel allotted to all radio systems in an access point is set to 0 by default. To reset a radio interface’s channel, a node must alter the ID of the de facto stream in the radio host controller higher layers. In ns-3, all channels are invented.
The second program demonstrations the progression of allocating c channel to the MRNode’s 1st radio link (net-Device).
Program Two
It should be noted that switching to another frequency is not possible if the functionality is overloaded as a result of receiving and sending. Before setting the new channel, Program 2 checks the device’s status (line 5).
When using dynamic or hybrid channel assignment, the available routing protocols in ns-3 are inapplicable for multi-radio wireless networks. Nonetheless, because we are only interested in fixed WMNs with limited or fixed motion nodes, a simple way to avoid this is to use static routing the complexity associated with dynamic routing protocols. By satisfying the routing tables at the start of the model, ns-3’s global routing simulation only does provide wired networks with static routing. There is no static routing in wireless networks because the A wireless network’s configuration is decided at run time as determined by the modelling technique and other variables (ns-3 Development Team, 2011)
3.4.2 An example of NS3 channel allocation implementation
The spectrum module in ns-3 handles the modeling of signal propagation through the wireless channel, which includes the abstract classes SpectrumPhy and SpectrumChannel. Devices interacting over the same wireless channel each have their own SpectrumPhy incidences in charge of calculating the Power Spectral Density (PSD) of the transceiver. The various SpectrumPhy instances are linked to the same SpectrumChannel object, which routes transmissions between the devices. SpectrumPhy invokes the method Spectrum- Channel:: at each transmission. StartTx sends a notification to each received signal and calculates the PSDs of the signals received. SpectrumChannel uses two standard interfaces to account for power attenuation and fading caused by signal transmission through the environment, namely PropagationLossModel and SpectrumPropagationLossModel. The former represents slow fading, in which the loss is constant across the signal’s frequency band, whereas the latter is used for fast fading models, which initiate frequency selective losses(fsl).
The author provides a momentary enlightenment of the channel allocation mechanism in this section (Nezhad and Cerda-Alabern, 2011; Nezhad et al, 2013). The author goes over each component and how it is implemented in ns-3 in great detail. From a theoretical standpoint, the author introduces a fresh model to authenticate the Channel Assignment outcomes. The author also creates a new automated test module to validate the simulator’s operation of Channel Allocation.
Figure 4: Channel Allocation class diagram
Channel Assignment is a protocol proposed and simulated in ns-3 for wireless mesh networks (Nezhad et al, 2013). The source code of NS3 example of channel allocation on 2.4GHz wifi network is available and can be accessed at https://www.nsnam.org (Nezhad, 2012).
3.4.3 Algorithm
Model validation warrants the correctness of the conceptual model’s computer programming and implementation (Sargent, 2000). One of the simulation dynamic verification techniquess is assertion checking. Unlike static models that analyze the program to see if it is correct (syntax analysis), verification dynamic checks the accuracy of the standards attained after running the code (Sargent, 2000; Balci,1998). To validate the ns-3 simulation, the author created a (SSE) state space explorer that is triggered by key proceedings in the models and accomplishes checking state to ensure that the simulation is correct.
The author presume that the simulation, as a framework, is made up of a collection of entities that work collectively to achieve the packet transmission target, and thus practically symbolizes the functioning of the actual structure.
We use the X coordinates to make a list of AP, the start AP has a number 0, then according to X coordinates ( from the lowest to the highest value) we have consequent numbers 1,2…8.).
EXAMPLE
INPUT FOR NINE AP
A(0,0),B(1,5),C(2,2),D(4,4),E(5,0),F(6,4),G(10,4),H(11,2),I(12,0)
The channel allocation is according to table 1
| AP number | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| AP channel | 1 | 6 | 11 | 1 | 6 | 11 | 1 | 6 | 11 |
Table 1 According to Table 1
Verification of Algorithm Efficiency:
Analysis of distance between any AP and AP with the same channel number after the allocation, what is the minimal distance without allocation and after allocation.
| AP | A | B | C | D | E | F | G | H | I |
| x | 0 | 1 | 2 | 4 | 5 | 6 | 10 | 11 | 12 |
| AP no | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Channel no | 1 | 6 | 11 | 1 | 6 | 11 | 1 | 6 | 11 |
Table 2: AP Numbering/Channel Allocation
| AP | A | B | C | D | E | F | G | H | I |
| A | 0 | 5.1 | 2.8 | 5.7 | 5 | 7.8 | 10.8 | 11.2 | 12 |
| B | 0 | 3.2 | 3.2 | 6.4 | 5.1 | 8.1 | 9.5 | 11.2 | |
| C | 0 | 2.8 | 3.6 | 4.5 | 7.3 | 8 | 9.2 | ||
| D | 0 | 4.1 | 2 | 5 | 6.3 | 7.9 | |||
| E | 0 | 4.1 | 5.7 | 5.8 | 6 | ||||
| F | 0 | 3 | 4.5 | 6.4 | |||||
| G | 0 | 2.2 | 4.5 | ||||||
| H | 0 | 2.2 | |||||||
| I | 0 |
Table 3: Distance between AP (Meters)
| Distance/AP | A channel no 1 | D channel no 1 | G channel no 1 |
| A channel no 1 | 0 | 5.7 | 10.8 |
| D channel no 1 | 0 | 5 | |
| G channel no 1 | 0 |
Table 4: Distance between AP with Channel 1 after Channel Allocation
| Distance/AP | B channel no 6 | E channel no 6 | H channel no 6 |
| B channel no 6 | 0 | 6.4 | 9.5 |
| E channel no 6 | 0 | 5.8 | |
| H channel no 6 | 0 |
Table 5: Distance between AP with channel 6 after Channel Allocation
| Distance/AP | C channel no 11 | F channel no 11 | I channel no 11 |
| C channel no 11 | 0 | 4.5 | 9.2 |
| F channel no 11 | 0 | 6.4 | |
| I channel no 11 | 0 |
Table 6: Distance between AP with channel 1 after channel allocation
Code/Algorithm
NetDeviceContainer apDevice, staDevices;
std::string channelStr (“{0, ” + std::to_string (channelWidth) + “, “);
if (frequency == 2.4)
{
wifi.SetStandard (WIFI_STANDARD_80211ax);
channelStr += ” BAND_2_4GHZ, 1}”;
}
else if (frequency == 2.4)
{
wifi.SetStandard (WIFI_STANDARD_80211ax);
channelStr += ” BAND_2_4GHZ, 6}”;
}
else if (frequency == 2.4)
{
wifi.SetStandard (WIFI_STANDARD_80211ax);
channelStr += ” BAND_2_4GHZ, 11}”;
}
else if (frequency == 2.4)
{
wifi.SetStandard (WIFI_STANDARD_80211ax);
channelStr += “BAND_2_4GHZ, 0}”;
Config::SetDefault (“ns3::LogDistancePropagationLossModel::ReferenceLoss”, DoubleValue (40));
}
else
{
std::cout << “Wrong frequency value!” << std::endl;
return 0;
CHAPTER 4: RESULTS
- Results Evaluation
4.1 Simulation parameters and application scenario
The author evaluated the results on NS3 simulator. Depending on the mode of the nodes, data samples are aggregated into data packets. It should be noted that each sample is a 1-byte data element. Data samples are concatenated into packets in order to avoid generating a packet for each data sample, which would overburden the network and node packet queues (packet queues size is fixed to 8 slots). In addition, in order to meet the application’s delay constraint of 500 ms, we decided not to keep data samples longer than 200 ms in nodes. Table 1 shows the packet generation rate and data sample rate for each mode of node. The following results are an average of 200 iterations. The author generated a new network topology with the same number of nodes with each iteration. For each iteration, the random behavior of CA/CSMA varies as well. Furthermore, we used a probabilistic propagation model (Log Distance model), which includes a random component that varies with each iterative process.
4.2 Simulation settings
To test the performance of channel allocation on 2.4GHz wifi network, A NS3 simulation design is produced. The simulated network follows the same configuration as seen in Figure 5.
Figure 5 depicts the simulation model’s topology
As previously stated, the nodes A, AP, C, and D are standard devices that use wifi protocols. On these nodes, only standard protocol stacks and applications are installed for simulation purposes. The gateway node B is a multi-purpose node that can communicate with both the AP and node C. As previously stated, because the NS3 system has no 802.15.4 protocol stack development, the two nodes C and D are alternated by two ad-hoc nodes with comparable circumstances.
4.3 NS3 model and result analyze
An NS3 model of the solution is created to test the performance of the application layer mechanism. This NS3 model has 5 nodes, and the topology is depicted in Figure 6.
Figure 6 depicts the simulation model’s topology.
Node A in this model is a standard wifi device. The gateway node, Node B, communicates with both networks. Both nodes A and B are linked to the AP node, and these three nodes form a small WLAN network. However, because there is no implementation of the NS3 system’s 802.15.4 standard stack, two alternative wifi nodes are assumed to be the two nodes. These two nodes form a basic WPAN network. The sender is node C, and the receiver is node D. The WLAN network’s bandwidth is set to 10 Mbps, while the WPAN network’s is set to the standard value of 250 kbps. The AP node continues to send UDP packets to node A. This deployment is used to model a real-world streaming data transmission scenario. Because the two networks use the same frequency channel, there is a chance of a collision. A control signal with a length of 1000 bits will also be sent as a request. The AP point has two applications installed. The first is the UDP application that runs between AP and node A. The second application is intended to halt an ongoing transfer of information of the AP node for a specified period of time. When the AP receives a request from the gateway node B, this application will be launched. This is the security application that has been installed on the AP node. As previously stated, the sender will send the 1000 bit packet after the wifi communication has been terminated. Some duration parameters are assumed in this simulation. The UDP application has a time range of 2s to 6s. The transmission request will be generated in 3s and the packet will be sent in 0.1s. This interval is set aside for the stopping process. To clearly see its performance, the cut-off duration is set to 1s. This time can be set to any length of time for a transmission. Figure 7 depicts the end result.
Figure 7 depicts the simulation’s node throughputs.
Figure 7 depicts the throughput of each node throughout the simulation process. It is clear that the AP node continues to send data to node A from 2s to 6s, with the exception of the 1s timeframe in the middle. The request and application schedules determine the cut-off interval. Following that, as illustrated in Figure 7, the 1000-bit packet is properly transferred, as is the data collected.
4.4 Conclusion based on simulation results
The system’s performance under 802.11a and b has been tested in this simulation. In particular, on both the AP and node A, a UDP data transmission application is installed. To simulate the streaming data transmission scene, this application will run from 2s to 5s. There is also a UDP implementation with its own policies and procedures installed between the AP and node B. To meet the reservation requirement, the wifi packet size is set to 1040 bytes. Transmission requests, on the other hand, are assumed to begin in the 3s to 4s. During this time, packets of 25 bytes in size are sent to test the system’s performance. In this NS3 simulation, specific sending times are transmitted from node C to gateway node B, presenting the process flow in the actual moment.
CHAPTER 5: CONCLUSIONS AND FUTURE WORK
5.1 Conclusions and future research
Chapter 5 sums up the research contribution of this thesis as the final chapter. The conclusions drawn from the research and simulations will be presented first, followed by recommendations for future work.
5.1.1 Conclusions
The objective of the present research is to show an NS3 example of channel allocation on a 2.4GHz wifi network, including potential problems and solutions based on simulation testing. First, the current development and prerequisites of the 2.4GHz wifi network industry have been presented, demonstrating the market demand for a stable NS3 channel allocation with good wireless techniques. This topic has also become a source of research interest. A critical issue concerning channel allocation on a 2.4GHz wifi network system is designed. This topic was thoroughly discussed in Chapter 1. As a result, the primary contribution of this thesis is the development of a solution to demonstrate channel allocation on a 2.4GHz wifi network using NS3. Second, background information about the problem has been provided, including a summary of wifi protocol stacks and a conversation of the existing ideal frequency-hop remedy.
Chapter 2 presents a literature review in detail. Research on this area has been reviewed.
In chapter 3, presents the ns-3 Simulator, Wi-Fi Module in ns-3, Channel Allocation on 2.4 GHz Wifi Network, Simulation components for multi-radio mesh networks, Network Devices. Because it necessitates devices with powerful embedded processors, the double method utilize has been shown to be infeasible for large and dense sensing networks. This also limits the method’s development. Second, the application layer cut-off method has been validated using an NS3 model. The simulation results of the channel allocation method were presented in Chapter 4. To put this method to the test, an NS3 model was created, and the results show that it works well. The gateway node could do reservation by virtualizing packets and sending them to the AP devices using this technique. Moreover, with the help of a two reference simulation, this method has been proven to be stable with varying bandwidths and data rates. However, if the packet size exceeds a predefined threshold, the reservation performance will suffer significantly and the wifi interference will be too strong for transmission. The author has presented the extensions that must be made in the ns-3 simulator to simulate channel allocation mechanisms for multi-radio wireless networks in this thesis. The simulation details of the Channel Assignment process proposed in (Nezhad and Cerda-Alabern, 2011; Nezhad et al, 2013). Furthermore, the Channel Assignment source code has been published and is available in (Nezhad, 2012). In the future, the author proposes expanding on the implementation by improving the data delivery mechanism to include differentiated priority scheduling, allowing higher priority traffic to be transmitted ahead of lower value traffic. It is also desirable to consider traffic rate adaptation for channel allocation, as well as developing a method to avert data queue concentration for high-rate traffic. The author also suggests that investigations be conducted to determine the necessary modifications to the routing protocol in order for it to work with blended channel assignment procedures.
5.1.2. Recommendations for future works
The frequency division multiplexing (ofdm) method is thought to be the most effective method for resolving the coexistence issue. To achieve automatic hopping, a more intelligent frequency hopping mechanism should be designed. When other protocols interfere, the devices should be able to change their transmission channel to a free one. Furthermore, the mechanism should be efficient because the frequency division multiplexing (ofdm) method may result in complex calculations and execution delays. In this thesis, the author introduced a channel model for ns-3, developed following the specifications in Study on NS3 example of channel allocation on 2.4GHz wifi network. This thesis research is expected to enhance ns-3 studies by allowing for more precise prediction of the complexities of 2.4 GHz wifi network wireless channels, thereby improving support for wireless system simulation. Furthermore, because of the modular design, this work can be easily extended to support other ns3 channel models, resulting in a common framework for wireless system simulation.
