Planning a wireless network is a fundamental step that determines the stability of an entire business or the comfort of a smart home. Mistakes at the design stage often lead to dead zones, channel interference, and the inability to connect new devices when the physical limits of the equipment are reached. Proper Calculating the number of Wi-Fi access points avoids costly rework and ensures uniform signal coverage.
Unlike wired networks, where only cable length matters, a wireless environment requires consideration of many variables, from wall materials to the type of client devices. Many people mistakenly believe that a single powerful access point can cover an entire floor, but in reality, user density and throughput are far more important than simple range. Below, we'll discuss a methodology that will help you design a professional-grade network.
Before diving into the math, it's important to clearly define the project's goals. Will this be a network for a warehouse, where coverage and signal penetration through racks are more important, or an open-space office with hundreds of employees actively using video conferencing? The key factor is not the floor area of the room, but the expected user density and the required bandwidth per device. It is these parameters that dictate the choice of equipment and the placement scheme.
Analysis of the room and wall materials
The first step is always to study the physical environment. Wi-Fi signals, especially at 5 GHz, are extremely sensitive to obstacles. Concrete walls with reinforcement, mirrored surfaces, and metal structures can completely block radio waves. Meanwhile, drywall and wood have minimal impact. For accurate calculations, a floor plan indicating the partition materials is necessary.
There's a concept called free space path loss, which increases exponentially with distance. However, in real-world conditions, material attenuation is added. For example, a signal can lose up to 10-15 dBm passing through a single brick wall, and up to 20-30 dBm through concrete and metal. Ignoring this fact will result in unstable network performance in adjacent rooms or even a complete loss.
⚠️ Attention: Don't rely solely on visual inspection. Metallized thermal insulation materials in modern office partitions or tinted glass may be invisible to the eye, but they will provide an impenetrable barrier to radio waves.
To account for building characteristics, use the following attenuation data:
- 🏢 Open space (office without partitions) - minimal attenuation, the signal propagates freely.
- 🧱 Plasterboard partitions - low attenuation, the signal passes reliably.
- 🪵 Wooden walls and doors - medium attenuation, speed reduction possible.
- 🚧 Concrete walls and brick – strong attenuation, direct line of sight or installation of repeaters/additional points is required.
Capacity Planning
In today's environment, it's bandwidth, not coverage, that dictates the amount of equipment needed. Wi-Fi 5 (802.11ac) and Wi-Fi 6 (802.11ax) standards allow for the transmission of large amounts of data, but airtime is limited. If 50 people gather in a conference room and everyone wants to stream 4K video, a single access point physically won't be able to handle the data flow, even with an excellent signal.
To calculate the required speed per user, you need to determine the required bandwidth. For typical office work (email, web surfing), 1-2 Mbps is sufficient. For working with large files, video editing, or VoIP telephony, the requirements increase to 5-10 Mbps or more. The total bandwidth is divided by the actual throughput of a single access point, taking into account the load factor (usually no more than 50-60% of the theoretical maximum).
The calculation formula is as follows: Number of APs = (Number of Users × Required Speed) / (Actual Speed of One AP × Load Factor)For example, if you have 100 users, each requiring 5 Mbps, and the access point actually delivers 300 Mbps with a safe load of 50%, then: (100 × 5) / (300 × 0.5) = 500 / 150 ≈ 3.3. Round up to 4 access points.
It is important to take into account that older devices (legacy clients) are slower and consume more airtime, reducing overall network efficiency. Therefore, if you have a large fleet of outdated devices, it's best to increase the number of access points.
Placement density and coverage area
The area-based calculation method is basic, but the least accurate without reference to traffic. It is suitable for initial assessments or for spaces with low user density (warehouses, production facilities). The standard recommendation for office spaces with average occupancy is one access point per 150-200 square meters. For high-density spaces (classrooms, cafeterias), this figure drops to 50-80 square meters per access point.
When planning placement, it's important to adhere to the cell pattern. Access points shouldn't be placed in a line or just along the perimeter. A staggered or uniform grid pattern is optimal, allowing for seamless roaming. The distance between access points should be such that their coverage areas overlap by approximately 15-20%, but no more than that to avoid interference.
Below is a table of approximate coverage areas depending on the type of room:
| Type of premises | User density | Recommended area per 1 AP | Frequency range |
|---|---|---|---|
| Open office | Average | 150-200 sq.m | 2.4 GHz + 5 GHz |
| Conference hall | High | 50-80 sq.m | Mainly 5 GHz |
| Warehouse | Low | 300-500 sq.m | 2.4 GHz (long range) |
| Classroom | High | 60-100 sq.m | 5 GHz (channel width 40/80 MHz) |
Keep in mind that the antenna's height also affects coverage. The higher the antenna is raised, the larger the coverage area, but the lower the signal strength near the floor (where clients are located). The optimal height for office ceiling-mounted antennas is 2.5-3.5 meters.
Counting the number of client devices
The limiting factor is often not the channel speed, but the number of simultaneous connections. Enterprise-class access point (Enterprise grade) can serve 50 to 250 clients depending on the model and load. Home routers used in businesses often fail after just 15-20 active devices due to insufficient RAM and weak processors.
When calculating, consider the concurrency factor. In an office, 80-90% of employees can work simultaneously, in a warehouse, 30-40%, and in a shopping mall, the load is peak and unpredictable. If you're designing a network for a smart factory, where every machine and sensor is connected via Wi-Fi, the number of access points is determined by the client limit per radio module port.
☑️ Check network requirements
There's a simple heuristic: for comfortable operation, divide the maximum number of clients per access point by two. This will give you the number of devices required to ensure the network operates quickly and without lag. If the estimated number of users exceeds this limit, new access points must be added, even if they technically overlap in range.
Interference and channel selection
Installing a large number of access points without proper configuration will lead to chaos in the airwaves. Adjacent access points should not operate on the same or overlapping channels. In the 2.4 GHz band, only channels 1, 6, and 11 are non-overlapping. In the 5 GHz band, there are significantly more channels, allowing for dense networks to be created without interference.
When planning a site, it's essential to conduct a site survey using specialized software. This will allow you to see existing neighboring networks and noise sources (microwaves, Bluetooth devices, CCTV cameras). Based on this data, a frequency plan is created, assigning each point its own channel and transmitting power.
⚠️ Attention: Never set the transmit power of all access points to "Maximum." This creates "long-range" cells that interfere with each other, causing clients to become stuck on the farthest access point, not switching to the closest one. The power should be reduced to a level sufficient for reliable reception (-65...-70 dBm) at the cell edge.
Modern Wi-Fi controllers (Wireless Controllers) can do this automatically, dynamically changing channels and power depending on the situation. However, the base plan still needs to be designed manually.
Final calculation and power reserve
The final stage is the synthesis of all the obtained data. The maximum value is calculated based on coverage, throughput, and client count. A capacity reserve (usually 20-30%) is added to the resulting figure. This reserve is necessary for staff expansion, the introduction of new services, and compensation for aging equipment.
It's also worth considering the project budget and the capabilities of the cabling infrastructure. Sometimes, installing new cable to an additional access point is more expensive than the access point itself. In such cases, mesh systems or access points with radio uplinks can be considered, although this will reduce the overall performance of the segment.
Hidden calculation of safety margin
If you estimate you need 10 access points, purchase 12-13. Keep one in stock as a hot spare (ZTP), and distribute the rest to the highest-load areas. This is cheaper than ordering equipment urgently during outages.
Don't skimp on cabling. Wi-Fi 6 and higher access points require Category 1 cable. Cat.5e or Cat.6 and switches that support the standard 802.3at/af (PoE+). Insufficient power over the twisted pair will disable the second radio module or USB ports, and the access point will not be able to operate at full capacity.
Frequently Asked Questions (FAQ)
Can home routers be used instead of professional access points?
Technically possible, but a bad idea for business. Home routers don't support seamless roaming (802.11r/k/v standard), lack centralized management, and quickly overheat under 24/7 load. In an office, 10 such devices will create 10 different networks with constant interruptions.
How often should the network be recalculated?
A project review is necessary when the office layout changes (new glass partitions or shelving), the number of employees increases by more than 20%, or new demanding applications are introduced (Wi-Fi video surveillance, VR).
Which is better: many low power points or few high power points?
Definitely the first option. The "multiple low-power points" (cell sizing) strategy provides higher overall network throughput since the airtime is divided among fewer clients in a single cell.