Networking Systems and Social Computing

Disruptive Technology

🎯 (EC-11-01, EC-11-06, EC-11-07)

📌 Investigate the effects of disruptive technology on everyday computing

• effects on the individual
• innovations in infrastructure, hardware and software
• trends and technology buzzwords

Disruptive technology in context

A disruptive technology changes how people or organisations do something so much that older products, services or business models lose their place. The idea comes from the business writer Clayton Christensen, who described new products that start out simpler or cheaper, win over customers the market has overlooked, then improve until they displace the established leaders. In everyday speech the term is used more loosely for any technology that shifts habits quickly. To investigate a disruptive technology, ask four questions: what did it replace, who gained, who lost, and what new problems did it create?

Effects on the individual

Effects are rarely all good or all bad. On the positive side, disruptive technology can bring convenience (banking, shopping and telehealth from home), access (a student in a remote community can join an online class), new work (app developers, data analysts, content creators) and lower costs for some services. On the negative side, it can remove jobs or change the skills needed, create a digital divide between people with and without fast internet or devices, erode privacy as more personal data is collected, and affect wellbeing through constant connection. Individuals must keep learning new skills, a pattern often called lifelong learning.

Innovations in infrastructure, hardware and software

Disruption usually needs several layers to advance together.

• Infrastructure: fibre and the national broadband network (nbn), 4G and 5G mobile networks, low Earth orbit (LEO) satellites for regional areas, and large cloud data centres. Faster, cheaper connections make new services practical.
• Hardware: powerful, low-cost smartphones, wearables, tiny sensors and microcontrollers, and specialised chips for graphics and artificial intelligence.
• Software: app stores, web and mobile apps, application programming interfaces (APIs) that let systems connect, and machine learning that turns large datasets into predictions.

Ride-sharing shows all three layers working together: smartphones with GPS (hardware), mobile data networks (infrastructure) and a matching app with online payments (software) combined to challenge the taxi industry.

Trends and technology buzzwords

Technology marketing is full of buzzwords: cloud, AI, big data, blockchain, the Internet of Things (IoT), edge computing, 5G, digital transformation, metaverse. Some name genuine advances; others are used to make ordinary products sound new. The Gartner Hype Cycle is a well-known model that describes how a new technology often passes through a peak of inflated expectations and a trough of disappointment before reaching steady, practical use. A useful test for any buzzword is: what does it actually do, who benefits, and what evidence is there?

Graph Theory in Social Networks

🎯 (EC-11-04, EC-11-06)

📌 Investigate the application of graph and network theory in the design of social networks

Nodes, edges and NESA's notation

A social network can be modelled as a graph: a set of nodes (also called vertices) joined by edges (links). In NESA's notation, a circle is an individual or a category, and a straight line is a link that connects nodes, groups of nodes or clusters. Circle size may show the size of a dataset or how important the node is, and labels on the links describe the relationship. NESA's worked example shows VB at the centre with friends and interest categories around them.

Adjacency

Two nodes are adjacent if an edge joins them. Adjacency can be recorded as an adjacency list (each node lists its neighbours) or an adjacency matrix (a grid with 1 where two nodes are adjacent and 0 where they are not). In an undirected friendship graph the matrix is symmetrical. On a platform with "follow" relationships, edges are directed (an arrow from follower to followed), so the matrix is no longer symmetrical. Designers choose the model: mutual friendship suits an undirected graph; following a celebrity suits a directed one.

The same graph as an adjacency list: A: B, C, D   B: A, C   C: A, B   D: A, E   E: D, F   F: E. Lists use less memory for large, sparse networks such as a social platform with millions of users, where most pairs of people are not friends.

Centrality: who matters most?

Centrality measures how important or influential a node is. There are several measures, and they can disagree.

• Degree centrality is the number of edges a node has. It is often normalised by dividing by the maximum possible degree, n − 1. For the six-person graph, A has degree 3, so its normalised degree centrality is 3 ÷ 5 = 0.6; F has 1 ÷ 5 = 0.2. Degree centrality finds the best-connected people, such as an influencer with many followers.
• Betweenness centrality counts how often a node lies on the shortest paths between other pairs of nodes. People with high betweenness are bridges between groups: information must pass through them. In the six-person graph, A and D each lie on six pairs' shortest paths, even though D has a lower degree than A.
• Closeness centrality measures how near a node is to every other node (the inverse of its total shortest-path distance). A's distances are 1, 1, 1, 2 and 3, a total of 8, so closeness is 5 ÷ 8 = 0.625, whereas F's total is 14 and its closeness is about 0.36. Nodes with high closeness can spread news the fastest.

Connectedness, paths and clusters

A path is a sequence of edges joining two nodes. A graph is connected if there is a path between every pair of nodes; otherwise it is disconnected and splits into separate components. Dense groups inside a network are clusters or communities, such as a year group or a sports team. Platforms use these ideas in their design: friend suggestions often propose a friend of a friend (in the six-person graph, B and D are not friends, but they share the friend A, so D is a likely suggestion for B); community detection groups similar users; and measuring how far a post spreads along paths helps show how information, and misinformation, moves.

Web Apps and Mobile Apps

🎯 (EC-11-01, EC-11-06)

📌 Describe how web applications and dedicated apps for mobile devices influence the success of social networking

Web applications and dedicated mobile apps

A web application runs in a browser and is delivered from a server, so nothing needs to be installed and every device gets the same, current version. A dedicated (native) mobile app is built for one operating system, installed from an app store, and can use device features such as the camera, microphone, GPS, contacts and push notifications. Many social networks offer both, and the two work together.

How they influence the success of social networking

• Network effects: a social network is only valuable if your friends are on it, so anything that makes joining easy grows the network, and a bigger network attracts more users. A web link that opens instantly in any browser lowers the first barrier to joining; a well-built app keeps people coming back.
• Always with you: phones bring the camera, location and contacts into every post, story and message. Sharing a photo takes seconds, which increases content creation and activity.
• Notifications: push notifications pull people back into the service. They drive engagement but also contribute to distraction and pressure.
• Cross-platform consistency: users expect to move from phone to laptop mid-conversation, so successful networks keep accounts, messages and content in sync across web and mobile.
• User experience and performance: slow loading, confusing layouts or intrusive permissions make people leave. Good UX, including accessibility, widens the audience.
• Distribution and rules: app stores are the main channel for mobile apps, and their policies, review process and fees shape what an app can do. Web apps avoid store rules but have less visibility.

IoT and the Internet of Me

🎯 (EC-11-03, EC-11-06, EC-11-07)

📌 Investigate how the development of hardware and software has influenced the adoption of the Internet of Things (IoT) and the Internet of Me (IoMe)

• impact on the individual
• impact on enterprise, including edge computing
• hardware infrastructure
• software protocols
• data backup
• regulation, compliance and cybersecurity

IoT and IoMe defined

NESA's glossary describes the Internet of Things (IoT) as smart devices and smart objects that send and receive information using the internet and communication infrastructure. The Internet of Me (IoMe) loosely refers to technology that connects our minds and bodies with the online world, turning biological and cognitive life into streams of data that can be monitored, shared and shaped. In practice IoT covers thermostats, cameras, farm sensors and factory machines, while IoMe covers the devices on or in a person: smartwatches, fitness trackers, sleep monitors, continuous glucose monitors and hearing aids that connect to phones.

Impact on the individual

Benefits include convenience (lights, heating and door locks controlled from a phone), safety (fall detection, smoke alarms that message you) and health insights (heart rate and activity trends that support better choices or alert a doctor). The costs are privacy (a home device may record audio, video and daily routines), security (a weakly protected camera can be taken over), dependence (a smart lock that fails when the internet drops) and data ownership (who can use your health data, and for what?). Health information is sensitive information under the Privacy Act 1988, which means the Australian Privacy Principles require stronger protection.

Impact on enterprise, including edge computing

Enterprises use IoT to track assets (fleet vehicles and shipping containers), monitor equipment for faults, manage energy in buildings and improve farming with soil and weather sensors. Thousands of sensors produce a huge flow of data, and sending all of it to a distant cloud is slow and costly. Edge computing, in NESA's words, is a framework that distributes processing capabilities and hardware so applications can be closer to their data source. An edge gateway on the farm or factory floor filters and analyses readings locally, acts immediately (for example, switching a pump on), keeps working if the internet drops, and sends only summaries to the cloud.

Hardware infrastructure

An IoT system needs sensors (measure the world) and actuators (act on it, such as valves and relays), microcontrollers or small computers (such as Arduino, ESP32 or Raspberry Pi boards) to run them, a network (Wi-Fi, Bluetooth Low Energy, cellular, LoRaWAN) to carry the data, gateways to link local devices to the internet, and servers or cloud services to store and analyse it. Power is a design limit: many sensors run for months or years from a battery, so they use low-power hardware and radios.

Software protocols

IoT devices talk using lightweight protocols suited to small devices and unreliable links: MQTT, CoAP, HTTP and the radio-level standards Zigbee, Thread and Bluetooth Low Energy. These are compared in detail in the section on IoT communication protocols. The important point here is that protocols are agreed rules (NESA's glossary calls them "generally accepted standards") that let devices from different makers exchange data.

Data backup

IoT data and settings need backing up too: the readings themselves (usually copied to cloud storage), device configurations and firmware images so a failed device can be replaced quickly, and the gateway's data if it buffers readings locally. The common 3-2-1 rule keeps three copies of important data, on two different types of storage, with one copy off site.

Regulation, compliance and cybersecurity

Personal data collected by IoT and IoMe devices falls under the Privacy Act 1988 and the Australian Privacy Principles, and an eligible data breach must be reported under the Notifiable Data Breaches scheme run by the OAIC. IoT in critical infrastructure such as energy and water is covered by the Security of Critical Infrastructure Act 2018. The Cyber Security Act 2024 provides for mandatory security standards for smart devices, based on baseline practices such as no universal default passwords, a published way to report vulnerabilities, and a stated period of security updates. Weak default passwords on IoT devices have been a common way attackers build large botnets, so cybersecurity has to be designed in from the start. See the Cyber Security Act 2024 and the OAIC's Notifiable Data Breaches page.

What Makes a Start-Up Succeed

🎯 (EC-11-01, EC-11-06)

📌 Outline the business and individual cultural characteristics that contribute to the success of start-ups

What a start-up is

A start-up is a new business, usually built around a new product, service or technology, that aims to grow quickly. Start-ups often begin with a small team and limited money, so their success depends on business decisions and on the people running them. Australian technology companies such as Atlassian and Canva both began as small start-ups.

Business characteristics that contribute to success

• A real problem and a clear customer: the product solves a problem people have and will pay to fix (often called product-market fit).
• A scalable business model: the business can serve many more customers without costs rising at the same rate, which is why software and online services suit start-ups.
• Testing early: building a minimum viable product (MVP), collecting feedback and improving in short cycles before spending heavily.
• Willingness to pivot: changing direction when evidence shows the first idea is not working.
• Funding and cash management: a plan for money coming in, from founders' savings, grants, angel investors, venture capital, crowdfunding or early sales, and awareness of how long the cash will last.
• Protecting ideas: using patents, trade marks, copyright and confidentiality where appropriate.
• Timing and networks: launching when technology and customers are ready, with mentors, advisers and partners who open doors.

Individual cultural characteristics

Start-up culture refers to shared attitudes and habits in the team. It is usually informal and flat (few management layers), fast-moving and experimental. Individuals who help a start-up succeed often show:

• Resilience: recovering from setbacks and rejections.
• Adaptability and a growth mindset: learning quickly and treating mistakes as feedback.
• Creativity and curiosity: looking for new solutions to problems.
• Initiative and ownership: acting without waiting to be told, and taking responsibility for results.
• Collaboration and communication: small teams cannot afford silos, so people share knowledge and take on different roles.
• Comfort with risk and uncertainty: accepting that plans change and many start-ups do not survive.

Connectivity and Work Practices

🎯 (EC-11-01, EC-11-06)

📌 Investigate how the developments in network connectivity and speed have influenced work practices within an enterprise

• remote working opportunities
• file sharing
• task delegation and monitoring

Faster, more reliable connections

Two measures describe how good a connection is. Bandwidth is how much data a connection can carry per second (in megabits per second, Mbps), and latency is how long a signal takes to make the trip (in milliseconds). Video calls, cloud files and remote desktops need both. In Australia, the national broadband network (nbn) brought fibre, hybrid fibre-coaxial, fixed wireless and satellite connections to homes and businesses, and 4G and 5G gave mobile connections that can carry video. Together these made it practical for many jobs to be done away from the office. Regional and remote areas still depend more on fixed wireless and satellite services, so the benefits are not shared equally.

Remote working opportunities

Remote working means doing your job from home or another location, using a network to reach files, systems and colleagues. It relies on a virtual private network (VPN) or secure cloud sign-in, video conferencing (Teams, Zoom, Google Meet), messaging and shared documents.

File sharing

Teams share files through shared network drives or a network-attached storage (NAS) device on the local network, and through cloud file services such as Microsoft OneDrive and SharePoint or Google Drive. Good file sharing needs permissions (who can view, edit or share), version history (to undo changes or recover a deleted file), co-authoring (several people edit the same document at once) and sync so every device has the latest copy. Risks include sharing links that are too open, accidental deletion, and confidential files being copied to personal devices.

Task delegation and monitoring

Networked project and task tools (such as Microsoft Planner, Trello, Asana and Jira) let managers delegate tasks by assigning an owner, due date and priority, and monitor progress through boards and dashboards, comments and automatic notifications. Team members see what depends on their work, and managers see bottlenecks early. Some employers also use software that tracks logins, web use or location. In NSW, the Workplace Surveillance Act 2005 requires employers to give employees notice before starting computer, camera or tracking surveillance. Monitoring outcomes (tasks completed) is generally more accepted and more useful than monitoring every keystroke, which can damage trust.

Digital Workflows

🎯 (EC-11-01, EC-11-10)

📌 Examine the benefits and limitations of digital workflows operating in an enterprise

What a digital workflow is

A workflow is a series of tasks that move work from a start to a finish, with rules about who does each step and when. A digital workflow uses software and a network to carry the work, its data and its approvals automatically from person to person instead of by paper forms, emails and phone calls. Examples: staff leave requests, expense claims, purchase orders, student enrolments, insurance claims and helpdesk tickets. Tools include forms and approvals in Microsoft Power Automate, ServiceNow-style ticketing systems, and simple automations such as "when a form is submitted, email the manager".

Benefits

• Speed: requests are routed instantly, and reminders stop them stalling.
• Accuracy: forms can validate data at the start, so there are fewer transcription errors.
• Consistency and compliance: every request follows the same rules, and an audit trail records who did what and when.
• Visibility: managers and staff can see the status of any request at any time.
• Lower costs and less paper: less printing, storage and re-keying of data.
• Works with remote teams: approvals can happen from anywhere.

Limitations

• Set-up cost and training: designing workflows, integrating with existing systems and training staff takes time and money.
• Rigidity: unusual cases that do not fit the rules need a manual workaround.
• Dependence on the network and system: if the server, cloud service or internet is down, work stops unless there is a backup process.
• Security and privacy: workflows carry personal and financial data, so they need access control and protection.
• Automating a poor process: a slow or confusing process just becomes a faster confusing process, so review the process before automating it.
• Resistance and exclusion: some staff may find new tools difficult, and roles can change or disappear.

Enterprise Data Storage Requirements

🎯 (EC-11-02, EC-11-03)

📌 Investigate data storage requirements for an enterprise

• capacity
• accessibility
• location
• security

Before choosing storage, an enterprise works out what it must store, who needs to reach it, where it will be kept and how it will be protected. These four requirements shape every decision.

Capacity

Capacity is how much data the enterprise must store now and in the future. Estimate current data, growth per year, backups and headroom. Storage is measured in gigabytes (GB) and terabytes (TB).

Accessibility

Accessibility is how easily and reliably authorised people can reach the data. Consider:

• Availability: the share of time the data can be reached (uptime), and what happens if a disk, server or link fails.
• Method of access: local network, VPN or internet; and which devices (laptops, phones, kiosks).
• Performance: solid-state drives are faster than hard disk drives but cost more per gigabyte; distance and bandwidth to the server affect speed.
• Permissions: role-based access so staff see only what they need.
• Usability: including access for people with a disability, and offline access when the network is down.

Location

Data may be stored on premises (a server or NAS in the building), in a data centre or in the cloud, in a specific country or region. Location matters for speed (closer is faster), disaster risk (a flood that affects the office should not also destroy the backup) and law. The data sovereignty of information depends on the country where it is stored, because that country's laws apply to it. NESA's glossary also describes data sovereignty as the right of Indigenous Peoples to determine the creation, collection, ownership and application of data that is for and about Indigenous Peoples. In Australia, the Maiam nayri Wingara Indigenous Data Sovereignty Collective sets out principles for how data about Aboriginal and Torres Strait Islander Peoples should be governed. Under the Privacy Act 1988, Australian Privacy Principle 8 places responsibilities on an organisation that discloses personal information to an overseas recipient, so many Australian organisations choose cloud regions located in Australia. Government agencies often require services assessed by ASD's Information Security Registered Assessors Program (IRAP).

Security

Storage must protect the confidentiality, integrity and availability of data. Controls include:

• Encryption at rest (on the disk) and in transit (across the network).
• Access control with strong authentication and least privilege.
• Backups and tested recovery plans.
• Physical security such as locked server rooms, or a cloud provider's data-centre controls.
• Logging and monitoring to detect unusual access.
• Retention and secure disposal so data is not kept longer than needed.

Health, education and financial data attract particular obligations, and a breach likely to cause serious harm must be reported under the Notifiable Data Breaches scheme.

Cloud Computing Services

🎯 (EC-11-01, EC-11-06)

📌 Explore cloud computing services

• Infrastructure as a Service (IaaS)
• Software as a Service (SaaS)
• Platform as a Service (PaaS)

What cloud computing offers

NESA describes cloud computing as distributing computing over a network so that storage of files, processing of data and access to software occurs on interconnected server computers to which the user's device is connected. Instead of buying hardware, the enterprise rents what it needs, usually paying for what it uses. Benefits include lower upfront cost, scalability (add or remove capacity on demand) and access from anywhere. Risks include dependence on the provider and internet, ongoing subscription costs, and the need to understand where the data is held.

Infrastructure as a Service (IaaS)

The provider supplies the basic computing infrastructure: virtual servers, storage and networks. The customer chooses the operating system and installs and maintains their own software. It is the most flexible model and needs the most technical skill. It suits an enterprise that wants to move existing servers to the cloud or needs a custom environment.

Platform as a Service (PaaS)

The provider supplies a ready-to-use platform including the operating system, runtime, databases and development tools, so developers deploy their own code without managing servers. It suits software teams who want to build and scale web apps and APIs quickly.

Software as a Service (SaaS)

The provider runs a complete application that users access through a browser or app, usually on a subscription. The provider handles updates, servers and security of the platform. It suits common needs such as email, accounting, customer records and video meetings.

Types of Cloud Storage

🎯 (EC-11-02, EC-11-03, EC-11-10)

📌 Compare different types of cloud-based data storage in terms of access and services

• public cloud
• private cloud
• hybrid cloud
• non-cloud on premises

The four options are compared on access (who can reach the storage and how) and services (what the provider offers, such as self-service, scaling and management).

Public cloud

Storage and services are owned by a provider and shared by many customers over the internet (Google Drive, Microsoft OneDrive, Dropbox, Amazon S3). Anyone with an account and internet access can use it, usually with a subscription or pay-as-you-go pricing. Capacity scales quickly, and the provider maintains the hardware. The trade-offs are less control over where and how data is handled, and shared infrastructure that some regulations do not allow for the most sensitive data.

Private cloud

Cloud-style infrastructure used by one organisation only, either in its own data centre or hosted for it by a provider. Access is through the organisation's own network or a VPN. It offers self-service and scaling within the capacity the organisation buys, with greater control over security, location and compliance, but at higher cost and needing skilled staff.

Hybrid cloud

A mix of private (or on-premises) and public cloud, connected so data and applications can move between them. A hospital might keep patient records on a private system and use public cloud for research computation on de-identified data, or a business might use public cloud for backups and extra capacity during busy periods. Hybrid gives flexibility but adds complexity, since two environments must be secured and managed together.

Non-cloud on premises

Servers and storage (for example, a NAS) owned and operated inside the organisation's building, accessed over its local network. The organisation has full control and can work offline, but pays for hardware, power, cooling and staff, must plan capacity in advance, and must protect against fire, theft and hardware failure itself. It usually does not offer on-demand self-service scaling.

Enterprise IT Infrastructure

🎯 (EC-11-01)

📌 Describe key components of an organisation’s information technology infrastructure

• networking, including servers, storage and end-point devices
• on-premises computing
• cloud computing

An organisation's information technology (IT) infrastructure is the combined hardware, software, networks and services that let it store, process and share information. NESA's list groups it into networking (including servers, storage and end-point devices), on-premises computing and cloud computing.

Networking, including servers, storage and end-point devices

A network joins devices so they can share data and resources. A local area network (LAN) covers a building or campus, a wide area network (WAN) joins sites over long distances (the internet is the largest WAN), and a wireless LAN (WLAN) uses Wi-Fi. Most LANs use a star topology: every device connects to a central switch or access point. Network diagrams show this structure; NESA asks for each device to be drawn as a symbol (node), joined by lines, with every device labelled.

A server provides a service and a client requests it: this is the client-server model. End-point devices are the most numerous and the most exposed part of the network, so organisations manage them with security software and policies, including for bring your own device (BYOD) arrangements.

On-premises computing

With on-premises computing, the organisation owns and runs its servers and network equipment in its own building or data room. It controls everything and can keep working without the internet, but must buy hardware up front, provide power, cooling, an uninterruptible power supply (UPS) and physical security, and employ staff to maintain and upgrade it.

Cloud computing

With cloud computing, some or all of the infrastructure is rented from a provider (see the cloud services section). The organisation still needs its own LAN, Wi-Fi, end-point devices and a good internet connection. Many organisations use a mix, for example an on-premises file server plus cloud email.

Transmission Media

🎯 (EC-11-01, EC-11-08)

📌 Describe how transmission media is used in networks

Transmission media are the physical paths, or the airwaves, that carry signals between devices. They are grouped into guided (wired) media and wireless (unguided) media.

Wired media

• Twisted pair: pairs of copper wires twisted together to reduce interference. Unshielded (UTP) cable is the standard for Ethernet in buildings, and telephone lines use it too. It is cheap and easy to install, but a single Ethernet run is limited to about 100 metres.
• Coaxial cable: a copper core with an insulating layer, a metal shield and an outer jacket. The shield protects the signal better than twisted pair. It is used for pay-TV, antenna connections and the hybrid fibre-coaxial (HFC) nbn connections.
• Fibre optic: data is sent as pulses of light along glass strands. It has the highest bandwidth, loses very little signal over distance and is unaffected by electrical interference. It is used in nbn fibre connections, city and regional backbones, and the undersea cables that carry most international internet traffic. It costs more to install and needs care because the glass is fragile.

Wireless media

• Wi-Fi (IEEE 802.11): radio waves in the 2.4, 5 and (with newer standards) 6 GHz bands for wireless LANs. It offers mobility and cheap installation, but has a range of tens of metres indoors and is shared and easily disrupted.
• Microwave: a point-to-point radio link between two dishes with a clear line of sight. It is used to link remote towers and sites where laying cable is too costly, and to connect mobile towers to the core network.
• Satellite: a dish links to a satellite in space. Geostationary satellites orbit about 36 000 km above the equator and cover huge areas, but the long trip creates high latency (noticeable delay); nbn's Sky Muster service uses satellites like this for remote Australia. Low Earth orbit (LEO) satellites orbit from a few hundred kilometres up, so latency is much lower, but many satellites are needed to give continuous coverage, and users need a dish that tracks them.
• Cellular (3G, 4G, 5G): mobile networks divide areas into cells served by towers. 3G introduced practical mobile data (and has now been shut down in Australia). 4G provided mobile broadband suitable for video. 5G adds higher capacity, lower latency and support for many more connected devices, which suits IoT and real-time uses. Coverage and speed depend on distance from the tower and the radio band used.
• Bluetooth: short-range radio (about 10 metres for many devices) for personal accessories such as headphones, watches and sensors. Bluetooth Low Energy uses very little power.

Choosing a medium

Selection depends on distance, bandwidth needed, cost, mobility, interference and security. The decision tree below shows a simplified way to choose. Real designs often combine media: fibre from the ISP to the building, Ethernet to desktops and Wi-Fi for mobile devices.

Interference with Data Transmission

🎯 (EC-11-01, EC-11-10)

📌 Explain factors that interfere with the transmission of data across a computer and social network

• distance
• topography
• physical barriers
• environmental barriers

Signals rarely arrive exactly as they were sent. Interference is anything that weakens, delays or distorts a signal. NESA lists four factors, and each affects some media more than others.

Distance

As a signal travels, it loses strength. This is called attenuation. Copper cable loses signal quickly, which is why one Ethernet run is limited to about 100 metres and longer links use a switch, repeater or fibre. Radio signals spread out as they travel, so a Wi-Fi signal weakens quickly with distance. Distance also adds delay (latency): signals to a geostationary satellite travel roughly 36 000 km each way, so the round trip takes a noticeable fraction of a second. On a social network, the same effect shows up when a platform's servers are on the other side of the world.

Topography

Topography is the shape of the land: hills, valleys and mountains. Microwave links, mobile towers and satellites need a clear line of sight or a clear path, so a hill between a tower and a farm can block a signal. The curve of the Earth also limits how far apart two microwave dishes can be. This is why towers stand on high ground, why regional and rural Australia has coverage gaps, and why remote communities often depend on satellites.

Physical barriers

Walls, floors, metal, glass and water absorb or reflect radio waves. Brick, concrete and metal weaken Wi-Fi far more than plasterboard. Other equipment also interferes: microwave ovens, cordless phones and Bluetooth devices share the 2.4 GHz band with Wi-Fi, and neighbouring networks on the same channel compete for airtime. In cables, electromagnetic interference (EMI) from motors, fluorescent lights and power cables can create noise, and crosstalk occurs when signals leak between neighbouring wires (twisting the pairs and shielding help reduce both).

Environmental barriers

Weather and natural conditions affect signals too. Heavy rain can weaken satellite and microwave links (rain fade), trees and foliage absorb some radio signals, lightning and power surges damage equipment, and flooding, fire, heat and wind can damage cables, towers and data centres. Undersea cables can be cut by anchors, landslides or volcanic activity. In January 2022 an eruption near Tonga damaged the submarine cable that connected the country to the internet, cutting most of its international connection for several weeks. Networks reduce the risk by using diverse routes and backup links.

Improving Data Flow

🎯 (EC-11-08, EC-11-10)

📌 Investigate ways to improve data flow within a system, considering proximity and modes of connectivity

• size, location and power of antennae when using satellite and microwave
• traffic filtering
• flow scheduling and traffic monitoring on telecommunications, including wireless broadband for mobile devices (3G→), Bluetooth and wi-fi

Good design considers proximity (how far the data has to travel) and the mode of connectivity (wired, Wi-Fi, cellular, satellite). Moving devices closer, or from wireless to wired, often gives the biggest improvement. The techniques below improve the flow of data when the medium is already chosen.

Size, location and power of antennae (satellite and microwave)

• Size: a larger dish or antenna gathers more of the weak signal and focuses it into a narrower beam. This increases gain and improves reliability and speed, but a narrow beam must be aimed accurately.
• Location: mount the antenna high, with a clear line of sight and away from trees, buildings and hills, and keep the cable between the antenna and equipment short. A satellite dish must also point at the correct part of the sky.
• Power: a higher transmit power extends range and overcomes weak signals, but uses more energy, can interfere with other services, and is regulated. In Australia the Australian Communications and Media Authority (ACMA) sets limits on radio equipment and licences.

Traffic filtering

Traffic filtering means allowing only wanted data through. A firewall or router applies rules based on addresses, ports and protocols, and can block unwanted or harmful traffic, such as unused services or known malicious sites. Content and DNS filters block categories of sites. Filtering improves flow by keeping unnecessary traffic off the network and improves security.

Flow scheduling and traffic monitoring

Flow scheduling decides when and in what order data is sent. Quality of Service (QoS) rules give priority to time-sensitive traffic such as voice and video calls over bulk downloads, and traffic shaping limits how much bandwidth a low-priority service can use. Large jobs, such as cloud backups and software updates, can be scheduled for overnight. On mobile networks, the tower's scheduler shares radio time between many phones; Wi-Fi devices take turns on a channel.

Traffic monitoring measures what is happening so problems can be found and fixed. Useful measures are utilisation (how full the link is), throughput, latency, jitter (variation in delay) and packet loss. Tools include the router's usage dashboard, speed tests, ping and traceroute (which show delay and where a path breaks), and packet analysers such as Wireshark. A baseline of normal performance makes unusual patterns easy to spot.

Wireless broadband, Bluetooth and Wi-Fi

• Wireless broadband for mobile devices (3G to 5G): each generation increased capacity. Better flow comes from newer standards, using more or wider radio channels, more antennas at the tower and phone, smaller cells with fewer users each, and moving closer to the tower or to an outdoor antenna.
• Bluetooth: keep devices close, avoid obstacles, and reduce clashes with Wi-Fi. Bluetooth hops between channels to avoid busy ones.
• Wi-Fi: choose a clear channel, use the 5 GHz band for speed at short range and 2.4 GHz for range, place access points centrally, add access points or a mesh system in large buildings, and connect fixed devices such as TVs and desktops by cable to free up airtime.

Graph Theory in Network Design

🎯 (EC-11-05, EC-11-08)

📌 Investigate the application of graph theory and network theory in the design of optimised computer networks

• adjacency
• centrality
• connectedness
• weighted graphs

Graph theory helps network designers because a computer network is a graph: devices are nodes and cables or wireless links are edges. The same measures used for social networks (adjacency, centrality and connectedness) apply, and weights on the edges let designers compare routes. NESA's notation for graphs uses circles for nodes and straight lines for links, with labels on the links.

Adjacency

Two nodes are adjacent when a link joins them directly. The adjacency matrix below records the network in the figure: the number in each cell is the cable length in metres, and 0 means there is no direct cable. The matrix is symmetrical because each cable works both ways. Adjacency tells the designer which devices need a port on which switch, and how many ports each device needs.

Centrality

Centrality shows which devices matter most. Degree centrality counts the links: in this graph A has 2, B has 3, C has 4, D has 3 and E has 2, so C is the most central. In network design, a highly central node is a natural place for a core switch or server, but also a single point of failure: if C fails, every route through it is lost. Betweenness centrality shows which node carries the most shortest-path traffic, so it needs the highest-capacity links. Closeness centrality shows which node has the shortest total distance to all the others, which suits a server that everyone must reach quickly.

Connectedness

A network must be connected, so every device can reach every other. A disconnected graph has separate components that cannot communicate. Designers add redundant links, so that if one cable fails there is still a path. A graph that contains cycles (like A-B-C above) has redundancy; a tree, which has no cycles, has none, and cutting any link disconnects part of the network.

Weighted graphs and the shortest path

A weighted graph attaches a number to each edge. Routers use link costs to choose routes; the OSPF routing protocol, for example, applies Dijkstra's shortest path first algorithm. The worked example below finds the shortest cabling route from A to E by hand. Dijkstra's algorithm keeps a table of the best-known distance to each node and repeatedly finalises the closest unfinished node.

1. Set A = 0 and all other distances to ∞ (unknown).
2. Finalise the nearest unfinished node, then update its neighbours: if (its distance + link weight) is less than a neighbour's current distance, replace it.
3. Repeat until the destination is finalised, then trace back through the nodes that gave each best distance.

Tracing back from E: E was reached from D, D from C and C from A, so the shortest route is A → C → D → E = 25 + 20 + 10 = 55 m. The direct-looking route A → C → E costs 70 m, and A → B → D → E costs 80 m.

Try it: the campus network lab

Use the tool below to test the ideas on a bigger graph. Pick two buildings, cut cables and watch the shortest path, the centrality measures and the connectedness change. The step table follows the same steps as the hand-worked example.

IoT Interoperability and SCADA

🎯 (EC-11-01, EC-11-06)

📌 Explore device interoperability within an Internet of Things (IoT) based enterprise network, including supervisory control and data acquisition (SCADA)

What interoperability means

Interoperability is the ability of devices and systems from different manufacturers to work together and exchange data without special effort. In a home, it means a light bulb from one brand, a speaker from another and a hub from a third can be controlled from one app. In an enterprise, it means sensors, controllers, software and cloud services that were bought at different times can share data. IoT devices often fail to interoperate because they use different radio technologies, different messaging protocols, different data formats and different cloud accounts.

The Matter standard from the Connectivity Standards Alliance was created to reduce these problems in smart homes: certified devices can work with different brands' ecosystems and can be controlled locally on the home network over Wi-Fi, Thread or Ethernet. In industry, OPC UA and Modbus are widely used standards for connecting machines and controllers.

Supervisory control and data acquisition (SCADA)

SCADA systems monitor and control industrial and infrastructure processes spread over large areas, such as water supply, electricity, gas pipelines, railways and factories. Sensors and actuators at each site are controlled by a programmable logic controller (PLC) or remote terminal unit (RTU). A communications network carries data to a SCADA server at the control centre, where operators use a human-machine interface (HMI) to see live readings and alarms and to send commands. A historian database stores the readings.

SCADA and IoT overlap. Traditional SCADA runs on private networks with strict reliability and equipment that can last decades. IoT adds many more, cheaper, internet-connected devices and cloud analytics. Connecting the two (often called the industrial IoT) brings efficiency but exposes systems that were not designed for the internet to cyber attack. Critical infrastructure operators in Australia have obligations under the Security of Critical Infrastructure Act 2018.

IoT Communication Protocols

🎯 (EC-11-02, EC-11-06)

📌 Investigate communication protocols between devices in an Internet of Things (IoT) network

A protocol is a set of agreed rules for exchanging data. IoT networks use several protocols working together at different levels: radio and network protocols that carry data between nearby devices, messaging protocols that structure the data, and interoperability standards that let different brands cooperate.

Messaging and application protocols

• MQTT (Message Queuing Telemetry Transport): a lightweight publish/subscribe protocol. Devices publish messages to named topics on a central broker, and other devices or applications subscribe to the topics they need. It runs over TCP (port 1883, or 8883 with TLS) and has three levels of delivery guarantee (quality of service 0, 1 and 2). It suits sensors on unreliable links.
• CoAP (Constrained Application Protocol): a lightweight, web-like protocol for very small devices. It uses the same request types as HTTP (GET, POST, PUT, DELETE) but runs over UDP (port 5683) with less overhead, and can be secured using DTLS.
• HTTP/HTTPS: the protocol of the web, based on requests and responses. It is well understood and works with existing tools and APIs, but its headers are larger and it is heavier for tiny battery-powered devices. Devices often use HTTPS to talk to cloud services.

MQTT example: a soil sensor publishes to a topic; the irrigation controller subscribes.

          topic:    farm/paddock3/soil/moisture
          payload:  {"value": 23.4, "unit": "%"}
          QoS:      1  (delivered at least once)

          Subscriber:  farm/paddock3/soil/#     (# matches every sub-topic)

Radio and network protocols

• Bluetooth Low Energy (BLE): short-range, very low power, widely built into phones and wearables; suits fitness trackers, beacons and device set-up.
• Zigbee: a low-power mesh protocol built on the IEEE 802.15.4 radio standard, used in smart lighting and sensors. Each mains-powered device can relay messages for others, extending range.
• Z-Wave: a low-power mesh protocol used in home automation. It works in a sub-gigahertz radio band, which avoids crowded Wi-Fi channels, and certification helps different brands work together.
• Thread: a low-power IPv6 mesh network built on 802.15.4. Devices get IP addresses, and border routers connect them to the home network, so the network is designed to avoid a single point of failure.
• LoRaWAN: a low-power wide-area network (LPWAN) protocol using unlicensed radio bands (in Australia, the 915-928 MHz range). It carries small messages many kilometres, which suits farm, mining and city sensors that send only a few readings each hour. Its data rate is low.
• Wi-Fi and cellular (including NB-IoT and LTE-M): higher data rates; cellular reaches wherever there is coverage but each device needs a subscription.

Interoperability standards

Matter is an application-layer standard for smart-home devices that runs over Wi-Fi, Thread and Ethernet (Bluetooth Low Energy is used for set-up). It defines a common way for a light, lock or sensor to describe itself and be controlled, regardless of brand.

Machine Learning and IoT

🎯 (EC-11-04, EC-11-05, EC-11-06)

📌 Explore the benefits of interfacing machine learning (ML) with IoT

• data-driven optimisation to enable predictive maintenance and logistics in manufacturing
• data acquired from medical imaging to improve accuracy of diagnosis
• improving traffic control through ML and sensors
• automating soil moisture data collection to manage farm irrigation

Machine learning (ML) finds patterns in data and uses them to make predictions. IoT supplies the raw material: a steady stream of sensor readings from the physical world. Together they let systems move from collecting data to acting on it. Models are usually trained in the cloud on historical data, then run either in the cloud or on an edge device close to the sensor.

Predictive maintenance and logistics in manufacturing

Sensors on machines measure vibration, temperature, current and pressure. ML learns what normal readings look like and what patterns come before a breakdown, so the factory can repair a machine before it fails, instead of on a fixed schedule or after a fault. This cuts unplanned downtime and waste. The same data-driven optimisation helps logistics: predicting demand, planning routes and scheduling deliveries and stock to reduce delays and costs.

Medical imaging

Imaging devices such as X-ray, CT, MRI and retinal cameras generate digital images. ML models trained on many labelled images can highlight suspicious areas, helping clinicians improve the accuracy and speed of diagnosis. Connected devices send images securely to the systems where they are analysed. Human clinicians remain responsible for the diagnosis. Risks include bias (a model trained on limited patient groups may perform less well for others), privacy, and errors. In Australia, the Therapeutic Goods Administration (TGA) regulates software used for medical purposes.

Improving traffic control

Traffic sensors, cameras and vehicle detectors report how many vehicles are waiting and moving. Adaptive signal systems, such as SCATS, which was developed in Sydney, adjust signal timing to real conditions. ML can add prediction, forecasting congestion and suggesting timing or routes before queues build, and can help prioritise emergency vehicles and buses. Limitations include the cost of sensors, the need for reliable communication, and privacy where cameras are used.

Automating soil moisture collection for irrigation

Soil-moisture sensors in the ground send readings over a low-power network such as LoRaWAN to a gateway. ML combines the readings with weather forecasts and crop type to decide when and how much to water, and can control valves and pumps automatically. It saves labour and water, an important benefit in Australia's dry climate, and can improve yields. Limitations are sensor calibration, connectivity in remote areas, and the cost of equipment.

Controlling Access to Networks

🎯 (EC-11-03, EC-11-07)

📌 Investigate security measures used to control access to networks

• password systems
• biometric measures
• CAPTCHA
• trusted Platform Module (TPM)
• automatic software updates

Access control asks two questions: who are you? (authentication) and what are you allowed to do? (authorisation). The measures below work best in combination, an approach called defence in depth.

Password systems

A password proves you know a secret. Systems should store only a hash of each password (a one-way scrambled value, ideally with a random salt), never the password itself. Good practice:

• Use long passwords or passphrases. The Australian Cyber Security Centre recommends a passphrase of four or more random words, because length gives more protection than clever substitutions.
• Use a unique password for each account, with a password manager to remember them.
• Turn on multi-factor authentication (MFA), which adds something you have (a phone app or key) or something you are (a fingerprint).
• Systems should limit login attempts, lock accounts after repeated failures and check for passwords known to have been leaked.

Biometric measures

NESA's glossary describes biometric as relating to a source of data being human or biological. Biometric authentication uses fingerprints, faces, irises or voices. Strengths: nothing to remember or lose, and it is quick. Limitations: a biometric cannot be changed if it is stolen; systems make errors (a false accept lets the wrong person in, a false reject locks out the right person); they can be fooled by spoofs (photos, copies); accuracy can vary between groups of people; and biometric information is sensitive information under the Privacy Act 1988, so it needs extra care. Phones typically store biometric templates on the device, not on a central server.

CAPTCHA

CAPTCHA stands for Completely Automatic Public Turing Test to Tell Computers and Humans Apart. It asks the user to do something that is easy for a human and harder for a program, such as choosing images, typing distorted text or ticking a box while the site analyses behaviour. It reduces automated abuse such as fake sign-ups, spam and credential stuffing (bots trying leaked passwords). Limitations: it can frustrate users, be difficult for people with vision or motor impairments unless an alternative is offered, and modern AI can solve many puzzles, so it is one layer, not a complete defence.

Trusted Platform Module (TPM)

A TPM is a dedicated security chip (or firmware equivalent) on a device's motherboard. It generates and stores cryptographic keys in hardware so they never leave the chip, can check that the device started with trusted software, and supports full-disk encryption (for example Windows BitLocker) and sign-in features such as Windows Hello. If a laptop is stolen, its drive is unreadable without the TPM and the user's credentials. Windows 11 requires TPM 2.0.

Automatic software updates

Attackers exploit known weaknesses (vulnerabilities) in software. Vendors release patches to fix them, so applying updates quickly closes the gap. Automatic updates remove reliance on people remembering. Patching operating systems and applications is part of the Australian Signals Directorate's Essential Eight mitigation strategies. The drawbacks are that an update occasionally causes problems or a restart at an inconvenient time, so organisations test updates and roll them out in stages, while home users can schedule them.

Securing a Smart Home Network

🎯 (EC-11-03, EC-11-07)

📌 Examine data security for an intelligent home network

• access levels, including universal plug and play (UPnP)
• passwords and firewalls
• port management on routers and switches
• wi-fi device positioning
• virtual private networks (VPNs) and hidden networks
• backup and disaster recovery

An intelligent (smart) home network links computers, phones, TVs, cameras, doorbells, speakers, lights and appliances through a home router. Many of these devices are low-cost, rarely updated and always on, so the network needs deliberate protection.

Access levels, including universal plug and play (UPnP)

Give each person and device only the access it needs (least privilege). Use a separate administrator account for the router with a strong, unique password; ordinary users should not have it. Create a guest network for visitors and a separate network for IoT devices so a compromised camera cannot reach the laptops. UPnP is a feature that lets devices ask the router to open ports automatically so that consoles, cameras and apps work without manual set-up. The convenience is a risk: malware or a poorly designed device can open ports to the internet without the user knowing. Turn UPnP off unless it is needed, and open ports manually if required.

Passwords and firewalls

Change every default password (the router's admin login, the Wi-Fi password, and each device's). Weak default passwords on IoT devices were used to build large botnets such as Mirai. Use WPA3 (or WPA2 with AES) for Wi-Fi with a long passphrase. A firewall filters traffic and blocks unsolicited connections from the internet; routers include one, and computers have their own software firewalls. Keep both switched on.

Port management on routers and switches

A port is a numbered doorway for a network service (for example, 443 for HTTPS). Open ports are places an attacker can knock. On a router, keep port forwarding to a minimum, close unused ports and disable remote administration. On a managed switch (common in businesses), disable unused physical ports, limit the number of devices per port and assign devices to separate virtual LANs (VLANs). A basic home switch has no such settings, so the router and access point do most of the work at home.

Wi-Fi device positioning

Placing the router centrally and high gives every room a good signal (see the interference section). It also limits how far the signal leaks outside the home, where a stranger could try to connect. Avoid placing it next to windows and metal objects. Position Wi-Fi-connected devices, such as cameras, within good range; a weak signal causes dropouts that can leave a camera or alarm offline.

Virtual private networks (VPNs) and hidden networks

A VPN creates an encrypted tunnel between your device and a VPN server, protecting data on untrusted networks such as public Wi-Fi, and letting you reach your home network remotely without opening ports for each service. A VPN does not make you anonymous or fix an insecure device, and you must trust the VPN provider. A hidden network is a Wi-Fi network with its name (SSID) not broadcast. It may deter casual visitors, but the name is still sent when devices connect, and free tools can find it, so it is not real security. Strong encryption (WPA3) and a long passphrase matter far more.

Backup and disaster recovery

Data can be lost through hardware failure, theft, fire, mistakes and ransomware. Follow the 3-2-1 rule (three copies, two types of storage, one off site), automate backups, and test that a restore works. Keep a copy of router and device configurations so replacement is quick. A disaster recovery plan lists what to restore first, and how long the household or business can afford to be without it (the recovery time objective) and how much recent data it can afford to lose (the recovery point objective).

Designing and Modelling a Network

🎯 (EC-11-08, EC-11-11)

📌 Design and model a network of interconnected devices for a specific purpose

To design and model a network you decide what it must do, choose the devices and connections, plan how they will be addressed and secured, and represent the design in a diagram that can be checked before anything is bought. The worked example below is one you can reproduce for a school club, a small office or your own project.

Step 1: Define the purpose and requirements

Step 2: Choose the topology, media and devices

A star topology puts a central switch in the middle, with each wired device on its own cable, so a failed cable affects only one device. Wired connections use twisted pair Ethernet (each run under 100 m); wireless devices connect through access points.

Step 3: Plan the addressing

Devices on a private network use private IP address ranges (10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16). A /24 network has the mask 255.255.255.0 and up to 254 usable addresses. Servers and printers get fixed addresses; other devices are given addresses automatically by the router's DHCP service.

Step 4: Model the design

A model lets you test the design on paper or in software before building. Draw a network diagram using NESA's notation: a symbol for each device (node), lines showing how they connect, and every device labelled. Free simulators such as Cisco Packet Tracer or GNS3 let you build the same network virtually and test it.

Step 5: Test the model and evaluate

Walk through each requirement with a test:

1. Can Reception computer 1 reach the NAS and printer? (It should.)
2. Can the visitor phone reach the NAS? (It should not.)
3. Can staff laptops and the tablet reach the internet and the NAS over Wi-Fi?
4. If a reception cable is unplugged, do the other devices keep working? (Yes, in a star topology.)
5. If the router fails, what is lost? (Internet and guest access; this single point of failure is accepted, with a spare router on the shelf.)

Record the results, compare them with the requirements, and revise the design where a test fails. This cycle of design, model, test and improve is what makes a design reliable.

Project Management Tools

🎯 (EC-11-08, EC-11-09)

📌 Apply appropriate project management tools to develop a project

Building a network is a project: it has a goal, tasks, a deadline, limited resources and risks. NESA's Course Specifications describe two tools for managing a project: Gantt charts and process diaries (log books). Other tools, such as task lists and Kanban boards, support them.

Apply a Gantt chart

A Gantt chart shows each task on an estimated timeline, with self-explanatory task names, the estimated time for each task, its dependencies, a clear dated time scale and milestones. To apply it to the network installation for the vet clinic:

1. List the tasks using self-explanatory names (survey, design, order, cable, configure, test).
2. Estimate durations from experience or supplier advice, in working days.
3. Identify dependencies: which tasks must finish before another can start? Equipment cannot be ordered until the design is approved.
4. Place tasks on the timeline with a dated scale, starting each as soon as its dependencies allow. Tasks that do not depend on each other can run in parallel.
5. Mark milestones, such as "Network live", with a diamond.
6. Update the chart as work is done, comparing actual with estimated times.

NESA's style also allows allocating resources (team members' names) to tasks and showing percentage complete by shading part of each bar. Update the chart regularly so it reflects actual versus estimated times.

Apply a process diary (log book)

A process diary records what actually happened. NESA lists what each entry should contain: the date, the person making the entry, progress since the last entry, tasks achieved, stumbling blocks or issues and how they were managed, possible approaches for upcoming tasks, reflective comments and resources used.

Supporting tools

A task list or Kanban board (Trello, Planner) shows work as cards moving from To do to Doing to Done, which suits day-to-day tracking. A risk list notes what could go wrong (a late delivery, a cable fault) and what you will do about it. Choose the tool for the purpose: a Gantt chart for planning time and dependencies, a process diary for recording and reflecting, and a board for daily work.

Configuring Network Devices

🎯 (EC-11-03, EC-11-08)

📌 Configure devices within a network

• naming the device
• updating the device
• configuring security protocols
• connecting to the internet

To configure a device is to set it up so it works correctly and safely on the network. The steps below configure the clinic's router and access point, and the same order works for most small networks. Screens differ between brands, but the settings are the same.

1. Naming the device

Give each device a clear, consistent name so it can be found in a list of connected devices and in logs. Use a pattern such as SITE-ROLE-NUMBER: WCVC-RTR-01 for the router, WCVC-NAS-01 for the file server, WCVC-REC-01 for a reception computer. Set the hostname in the device's settings. Choose a network name (SSID) that does not reveal personal information or the device model, for example WattleCreekVet for staff and WattleCreekVet-Guest for visitors.

2. Updating the device

1. Note the current firmware version from the router's status page.
2. Check the manufacturer's site for a newer version, and read the release notes for security fixes.
3. Back up the current configuration.
4. Download the update from the manufacturer only, apply it, and let the device restart without interruption.
5. Check that settings survived, and turn on automatic updates if available.

Do the same for operating systems on computers and for other network devices such as access points and printers.

3. Configuring security protocols

1. Replace the default administrator username and password with a unique, long passphrase.
2. Set Wi-Fi security to WPA3-Personal (or WPA2/WPA3 mixed mode if an older device needs it) with a strong passphrase. Do not use WEP or an open network.
3. Turn off WPS (push-button set-up), which can be attacked, and turn off UPnP and remote administration unless needed.
4. Enable the router's firewall.
5. Create the guest network and enable client isolation, so guests cannot see each other or the staff network.
6. Use HTTPS to reach the router's admin page.

4. Connecting to the internet

1. Connect the nbn connection box (network termination device) to the router's WAN port with an Ethernet cable, and connect the switch to a LAN port.
2. In the router's internet settings, choose the connection type your internet service provider (ISP) specifies. Some connections are configured automatically with DHCP; others need a username and password (PPPoE).
3. Set the LAN address and DHCP range from the addressing plan.
4. Restart and check the WAN status shows an address from the ISP.

Configuration record (Wattle Creek Vet Clinic)
          Router hostname:     WCVC-RTR-01
          LAN address:         192.168.10.1 / 255.255.255.0
          DHCP range:          192.168.10.100 - 192.168.10.199
          Reserved addresses:  NAS 192.168.10.10, printer 192.168.10.20
          Staff Wi-Fi:         SSID WattleCreekVet, WPA3-Personal
          Guest Wi-Fi:         SSID WattleCreekVet-Guest, isolated, 192.168.20.0/24
          Router admin:        default password changed; remote admin OFF; UPnP OFF; WPS OFF
          Firmware:            updated, automatic updates ON
          Backup:              configuration file saved to NAS and off-site storage

5. Test the configuration

From a computer on the staff network, run these commands (on Windows use ipconfig /all; on macOS or Linux use ip addr or ifconfig):

ping 192.168.10.1      (can I reach the router?)
          ping 1.1.1.1           (can I reach the internet by address?)
          ping example.com       (does name lookup by DNS work?)

If the second works but the third does not, the fault is in DNS. If the first fails, check the cable or Wi-Fi connection. Testing one step at a time locates faults quickly.

Security Procedures and Protocols

🎯 (EC-11-03, EC-11-09)

📌 Implement procedures and security protocols considering cybersecurity

To implement security means to put controls into practice on the network. Two kinds work together: protocols are technical rules that protect data as it moves or as devices connect, and procedures are the steps people follow. Both are needed, because the best technical protection fails if people share passwords or ignore updates.

Security protocols to implement

Security procedures to implement

1. Accounts: create an individual account for each person, give only the access needed (least privilege), turn on multi-factor authentication, and remove accounts promptly when someone leaves.
2. Patching: update operating systems, applications and firmware on a set schedule, and automatically where possible.
3. Backups: follow the 3-2-1 rule, and test restoring from a backup on a regular schedule.
4. Segmentation: keep guest and IoT devices apart from staff systems.
5. Monitoring and logging: review router and server logs for repeated failed logins or unknown devices.
6. Acceptable use and training: staff learn to spot phishing and use strong passwords; a written policy states what is allowed.
7. Physical security: lock the comms cabinet and server area.
8. Incident response: agree who does what if something goes wrong. Under the Notifiable Data Breaches scheme, an organisation covered by the Privacy Act 1988 must notify affected individuals and the OAIC when a breach is likely to cause serious harm.

These procedures line up with the Australian Signals Directorate's Essential Eight, which includes patching applications and operating systems, multi-factor authentication, restricting administrator privileges and regular backups. The Cybersecurity page of these notes covers threats, risk and legal obligations in more depth: see Principles of Cybersecurity.

Check that it works

Test each control: try to join the network with a wrong password, connect a phone to the guest network and confirm it cannot reach the NAS, restore a test file from backup, and confirm updates are switched on. Only test networks you own or have permission to test: unauthorised access to computer systems is an offence under the Cybercrime Act 2001 (Cth).

Optimising Network Performance

🎯 (EC-11-08, EC-11-10)

📌 Explore opportunities for optimising network performance

• improving bandwidth
• updating drivers and firmware

A network's performance is judged by bandwidth (how much data per second the link can carry), throughput (how much actually gets through), latency (delay) and reliability. To optimise a network, measure first, find the bottleneck (the slowest part), change one thing, and measure again.

Improving bandwidth

• Upgrade the internet plan or connection type: a higher nbn speed tier, or a fibre connection if available.
• Use wired connections for fixed devices such as desktops, servers and TVs to leave Wi-Fi capacity for mobile devices.
• Fix bottlenecks: a switch with slow ports, an old hub, or a cable that limits speed will limit the whole path. Replace hubs with switches, and use cables that support the speed needed.
• Improve Wi-Fi: use a newer Wi-Fi standard, the 5 GHz band at short range, wider channels where there is little interference, a clear channel and a better access point position.
• Manage traffic: prioritise voice and video with QoS, and schedule large transfers for off-peak times.
• Reduce the data sent: compress files and use caching, so common content is stored closer to users.

Updating drivers and firmware

A driver is software that lets the operating system control a hardware device such as a network adapter. Firmware is software stored inside a device such as a router or access point. Out-of-date drivers and firmware can cause dropped connections, poor speeds, compatibility problems and security holes, and updates often add support for newer, faster standards.

1. Identify the device and its current driver or firmware version (Device Manager on Windows; the status page on a router).
2. Download the update from the manufacturer's official site only.
3. Create a restore point or save the configuration so you can roll back.
4. Install the update, restart, and confirm the device works.
5. Test performance again and record the result.

Hardware and Software for Data Transmission

🎯 (EC-11-03, EC-11-10)

📌 Evaluate the role of hardware and software related to the transmission of data

• unsecured data
• encrypted data
• infrastructure

Data crosses a network as packets, passing through hardware and software that can protect it or leave it open. To evaluate, judge how well each option meets criteria such as security, performance, cost and reliability, and reach a supported conclusion.

Unsecured data

Unsecured (unencrypted) data is sent as readable plain text. Anyone able to capture packets on the path (on an open Wi-Fi network, or a compromised router) can read or change it using a packet-capture tool. Plain HTTP, FTP, Telnet, open Wi-Fi and unencrypted email are examples. Risks include stolen passwords and personal information (eavesdropping), altered messages (tampering) and fake websites or servers (impersonation). Unsecured transmission is only acceptable for public information, such as a public weather page.

Encrypted data

NESA's glossary defines encryption as encoding data with mathematical formulas into a form that only the intended recipient can decode. The main methods are:

• Symmetric encryption (for example AES): one shared secret key encrypts and decrypts. It is fast, so it protects bulk data.
• Asymmetric (public-key) encryption: a public key locks data and a matching private key unlocks it. It solves the problem of sharing a key safely and supports digital signatures and certificates.
• TLS (HTTPS) uses both: public-key methods set up the connection and check the site's certificate, then a fast shared key protects the data. WPA3 protects Wi-Fi, VPNs protect whole connections, and SFTP and SSH protect file transfer and remote access.

Encryption has costs: it needs some processing power (usually small, since most modern devices have hardware support), keys must be managed and protected, security tools cannot inspect encrypted traffic unless configured to, and a lost key can make data unrecoverable. Encryption protects confidentiality; it does not on its own stop malware, weak passwords or a compromised device at either end.

Infrastructure

Hardware and software work together to carry and protect data:

Evaluating a choice

Judgement: Option B is clearly better. The performance and effort costs are small compared with the risk of exposing personal information, and the data is sensitive. Unencrypted transmission cannot be justified for client records.