Testing Wi‑Fi and Bluetooth on Office PCs: a Measurement Plan

Why test Wi‑Fi and Bluetooth on office PCs at all

Wi‑Fi and Bluetooth in offices are often treated as “secondary”: if websites load, everything seems fine. But issues appear at the worst moments: a call drops, throughput falls in waves, latency spikes, or a headset starts crackling and disconnecting. The result is simple: people lose time and start blaming the ISP, the router, a “bad laptop” or a particular PC.

Testing wireless next to the router or in a server room is almost pointless. There the signal is usually strong, there’s little interference, and every adapter looks good. A real office has walls and glass, elevator shafts, neighboring access points, dozens of smartphones and headsets, and densely seated employees. So tests must be done in the actual rooms where people work.

For an office, what matters is not peak numbers “in ideal conditions” but stability and repeatability. If everything is fine today but the same room has stuttering calls tomorrow, that connection can’t be considered good even if it had a high peak speed.

Pay special attention to places where failures are immediately noticeable: meeting rooms, call centers, reception, and executive offices.

A simple example: Wi‑Fi in an 8‑person meeting room “holds up”, but when a Teams call starts the video freezes and a Bluetooth headset crackles if someone puts a phone near the PC. Such things are revealed only by field measurements, not by ticking a box next to an access point.

What to consider quality: simple metrics without extra theory

Wi‑Fi and Bluetooth serve different purposes. Wi‑Fi provides network access and service operation. Bluetooth is often used for headsets, mice, keyboards, scanners and sometimes calls. Therefore quality criteria differ.

For office Wi‑Fi, a small set of metrics that are easy to measure in rooms and then turned into clear requirements for the specification is usually enough. Don’t chase records — focus on stability.

Minimal Wi‑Fi metrics to record: signal level (RSSI) and its stability, real throughput (separately download and upload), latency (ping) and its variation (jitter), packet loss, and also disconnects, reconnects and throughput drops.

Bluetooth is better evaluated via real work scenarios rather than standalone numbers.

Minimal Bluetooth metrics for a workstation: range in the actual room, voice quality on headsets (no stuttering or robotic voice), connection stability when moving, and whether delay is noticeable.

To keep results honest, always record test conditions: PC model, OS, wireless driver version, access point (model and band), distance to it, obstacles (wall, cabinet), and network load (how many people are present at the time).

Set acceptable values according to your scenarios. Accounting may need stable access to ERP and printing for accounting, while meeting rooms require good call quality and low loss. It’s better to agree once on thresholds for typical zones (by the window, in the center, behind a wall) than to compare everything to “average numbers from the internet.”

Preparing for measurements in real rooms

To avoid a “works for me” debate, define in advance where and under which conditions you will check connectivity.

List rooms and mark zones within them: desks by windows and interior walls, meeting rooms, corridors, reception, storage or archive. Issues often appear in transition areas and in rooms with glass and metal.

Next plan measurement points. A practical approach is 3–5 points per room: by the entrance, in the center, in the far corner and next to specific workstations with office PCs. If you accept particular seats, tie points to desk numbers so checks can be repeated later.

Agree separately which networks you will test and who will provide access. Don’t mix up the corporate network, guest network and networks for telephony or corporate devices. Clarify whether accounts, certificates or VPN are needed, and if you can temporarily disable auto‑connect to “favorite” networks during the test.

Prepare a measurement kit in advance: a smartphone (quick checks and some Bluetooth tests), a second PC or laptop (for file transfers and calls), a Bluetooth headset and one or two simple peripherals (mouse or keyboard), a tape measure or pedometer, and a template table for records.

Test time affects results. Schedule two windows: peak hours (when most staff are present) and a quiet period. That shows where the issue is coverage‑related and where it’s load‑related.

Step‑by‑step Wi‑Fi measurement plan

Measure Wi‑Fi on the actual office PCs and in the rooms where they are used. For each room mark 2–3 key points in advance: by desks, in the meeting room, near the printer or MFP.

Main pass through points

Follow the same route each time. At every point perform a series of measurements rather than a single snapshot — this better shows stability.

• Connect to the target network and note which band the PC uses: 2.4 GHz or 5 GHz. Note if it switches bands automatically.
• Measure signal level and its stability: 3–5 consecutive readings at 10–15 second intervals. Watch for sudden drops.
• Run a speed test: one inside the LAN (for example copying a test file to an internal resource), another to the internet. Record not only peak numbers but repeatability.
• Check latency and loss to an internal server or gateway (often the router or nearest office server). Loss or sudden spikes are a reason to investigate.
• Test roaming: walk the corridor between rooms while on an active call in a messenger or with continuous data transfer. Handover between access points should not cause noticeable dropouts.

After the pass, mark where results were worse: a specific workstation, a corner, a section of corridor. These zones often become “control points” in acceptance documents.

Repeat the same plan during peak hours and compare to quiet‑time measurements. This approach makes the check reproducible: another specialist should be able to repeat measurements and obtain comparable figures.

Bluetooth measurement plan for a workstation

Bluetooth issues on office PCs often don’t show up in “ideal” conditions but in a normal room: headsets stutter on calls, mice jump, or devices need re‑pairing after sleep. So Bluetooth checks should follow real scenarios.

Devices to include

Take 2–3 typical devices that employees actually use. Usually a call headset and a mouse or keyboard are enough. If there’s a wireless barcode scanner or a small printer in the department — add it, but don’t try to cover every model.

Five quick checks

Run tests on one PC in each typical room: open space, private office, meeting room.

1. Pairing. Pair devices from scratch and note how many steps are needed and whether errors occur. Then remove the device and pair again to check repeatability.

2. Voice 5–10 minutes. Make a test call or record voice. Note stutters, dropouts, noticeable latency and desync.

3. Range. Test at the desk and toward the room boundary: walk 5–10 m, turn your back on the PC, close the door (if it’s a private office).

4. Interference. Create typical office load: active Wi‑Fi nearby, multiple Bluetooth devices in the room (colleagues’ mice and headsets). See if dropouts increase.

5. Reconnection. Put the PC to sleep and wake it, then reboot. The device should reconnect quickly without manual actions.

Example: a headset works fine at the table in a meeting room, but at 7 m audio drops when the projector is on and 6–8 laptops are nearby. That means the spec should require testing under interference, not just “in an empty room.”

How to record results so they are usable

If results are recorded “however it turns out,” disputes are almost inevitable: everyone has their own version of where the problem is and who’s at fault. A proper protocol makes wireless acceptance repeatable.

A convenient table template (minimal but sufficient)

The best format is a simple table where each row is one measurement at one point. In each row record:

• room and point (e.g. “by window”, “by door”, “meeting room table 2”);
• date and time, and a short note: “doors closed/open”, “many people/quiet”;
• PC model and wireless adapter model, driver version;
• Wi‑Fi: signal level (dBm), throughput (download/upload), latency (ping), packet loss (if measured);
• Bluetooth: scenario (headset/mouse/transfer), stability (dropouts, latency).

Add a “comment” column for details that help later: a microwave is running nearby, wireless presentation systems are active, many laptops are around, the PC is under the desk near a metal cabinet.

How to mark issues so they can be discussed

A table works best together with an office plan. A simple printout with bad points marked and row numbers referenced links issues to specific locations rather than “somewhere it’s bad.”

To separate a random glitch from a systematic problem, repeat measurements: if a metric “dipped”, repeat 2–3 times with 1–2 minute intervals. A systematic problem usually repeats in the same place and similar conditions.

Save screenshots and logs selectively — for contentious events (a disconnect, band switch, or driver oddity). For most table rows, numbers and a short comment are sufficient.

Example real measurement scenario in an office

Imagine a single‑floor office: 20 workstations, 2 meeting rooms, a reception area and a long corridor. The goal is to ensure Wi‑Fi and Bluetooth on office PCs behave predictably where people actually work and walk, not only near access points.

Run a “live” scenario. One employee with a work PC (or laptop) puts on a Bluetooth headset and for 20–30 minutes performs typical actions: a meeting call, walking the corridor, working by the window, quick network access at reception. Repeat the route twice to check result stability.

Use the same points and actions:

• meeting room 1: 5 minutes of voice call in the company’s standard app;
• meeting room 2: same but with the door closed;
• reception: connect to Wi‑Fi from scratch and open 2–3 web services;
• desk by the window: 10 minutes of email and corporate documents;
• corridor: walk 30–40 meters with an active headset call.

“Bad” usually looks like: headset audio drops for 1–2 seconds periodically, the PC hops between 2.4 and 5 GHz, or network connection drops for 10–20 seconds then recovers. Sometimes the symptom is “it kind of works but is annoying” — that’s exactly what you need to catch.

“Good” means repeated passes at the same points look the same: quick connection, uninterrupted calls, and small signal fluctuations don’t turn into dropouts. This scenario is useful for acceptance, especially when office PCs are supplied centrally and in batches.

Wording requirements that are easy to put into a specification

To avoid disputes, list exactly where the connection must work, which values are acceptable, and how acceptance is performed. Below are typical formulations that usually suffice.

First define the boundaries: list rooms and key zones (workstations, meeting rooms, reception, corridor by elevator, printer areas). Mark “critical” points separately — places used for video conferencing or ERP.

Then add a block of requirements (adjust values to your office and tasks):

1) Покрытие Wi-Fi
Wi‑Fi соединение должно быть доступно и стабильно во всех помещениях, перечисленных в Приложении А (карта точек замеров).

2) Качество Wi‑Fi по метрикам
Для каждой точки замера:
- уровень сигнала (RSSI): не хуже -67 dBm в переговорных и зонах ВКС; не хуже -70 dBm в рабочих кабинетах.
- потери пакетов при тесте: не более 1%.
- задержка до корпоративного шлюза/сервера: средняя не более 50 мс.

3) Стабильность при перемещении
При обычном перемещении сотрудника по коридору между кабинетами связь не должна обрываться; допустимый разрыв при переключении между точками доступа - не более 2 секунд.

4) Bluetooth для гарнитур
При использовании типовой Bluetooth-гарнитуры голос должен передаваться без заметных искажений/заиканий в пределах рабочего места (радиус 3 м). Допускается кратковременное ухудшение при перекрытии корпуса, но без разрыва соединения.

5) Совместимость и удобство
ПК должен работать без ручных настроек с типовыми устройствами Заказчика (перечень в Приложении Б: гарнитуры, мыши, клавиатуры). Первое сопряжение - до 60 секунд, автоподключение после перезагрузки - без действий пользователя.

To make requirements truly measurable, add a test procedure: who measures, which apps to use, how many repeats and what counts as a failure. For example: at least 10 measurement points per floor, 3 repeats at different times; “accepted” if at least 95% of points meet thresholds, and 100% in meeting rooms and executive offices.

Require the report in a table format: point, time, RSSI, loss, latency, comment (e.g. “near microwave”). This saves time when comparing different batches of PCs, including desktops and all‑in‑ones often bought for government or corporate projects.

Short checklist for quick pre‑acceptance testing

A checklist helps ensure tests were run in a repeatable scenario in real rooms. Pick 2–3 typical points: by an access point, at a regular workstation and in a place where connectivity is often weaker (partition, corridor, remote office).

Test on the same PC with the same scenario, at least 3 repeats per point. If results vary widely, that’s a reason to reject.

• Connecting to the corporate Wi‑Fi must be quick and stable: on first connection and on reconnect the network appears without long waits or manual steps.
• Metrics should not behave randomly: speed and latency in the same point should not jump wildly between runs.
• A real call test lasts 3–5 minutes with no noticeable dropouts, robotic voice or pauses.
• Moving between office zones (office — corridor — meeting room) should not cause long network losses: at most a brief pause during roaming.
• After sleep or reboot wireless modules should return to working state automatically: Wi‑Fi connects to the correct network and Bluetooth sees paired devices without manual re‑pairing.

If any item regularly fails, record place, time, PC model and conditions (how many people nearby, which network, which Bluetooth device). Such a protocol is easy to turn into a remark and a requirement for fixes.

Common mistakes and traps during measurements

The most common mistake is testing only next to an access point and concluding that “everywhere is the same.” In practice issues appear near distant walls, in meeting rooms, behind double doors or in rooms with dense furniture.

Another trap is confusing the problem source. Slowness can be due to a congested network or AP settings, or due to a PC Wi‑Fi driver, power‑saving mode or poor case/antenna placement. If you don’t separate “network or PC”, results get disputed and requirements blurred.

What usually ruins results

Avoid typical biases so measurements can be repeated and compared:

• testing only in one point (often by the router) without walking through rooms and workstations;
• measurements without recording time, network load, doors open/closed and PC placement;
• comparing different PCs with different drivers, firmware and power settings (especially USB Wi‑Fi);
• testing Bluetooth without a real headset or without a 5–10 minute call;
• ignoring physics: a system unit under the desk, next to a metal cabinet or radiator often worsens both Wi‑Fi and Bluetooth.

A mini‑scenario that often misleads

Two identical PCs in one room show different results: one is on the desk, the other under the desk by the wall. It seems like “the Wi‑Fi module is at fault.” After moving the system unit and turning off aggressive power saving the speeds and stability align. So note case position, antenna orientation and any setting changes in the protocol.

What to do after measurements: fixes and next steps

After measurements, turn numbers into actions quickly. A practical approach is to split findings into two groups: fixes that are configuration changes (Wi‑Fi settings, drivers, power saving) and those that require hardware or design changes (an AP can’t handle the load, weak PC module, poor AP placement, strong interference).

Typical steps:

• confirm the issue repeats on several PCs in the same area;
• fix quick items: update drivers, align Wi‑Fi settings, disable aggressive power saving on adapters;
• check the network: channels, channel width, transmit power, roaming, AP placement, and peak‑hour load;
• if unresolved, record exactly what to change: replace an AP, redesign placement or specify a stronger wireless module for PCs.

Agree final acceptance criteria and include them in the procurement specification. Tie wording to your measurements: which rooms, what time, what distance, which applications (video calls, file copies, headset work). After any changes repeat measurements with the same method and keep the report as a baseline for future purchases.

If the project covers both workstations and network, it can be easier to involve a system integrator. For example, GSE.kz (gse.kz) as a manufacturer of office PCs and a system integrator can help align workstation and Wi‑Fi infrastructure requirements and make acceptance less contentious, especially for large deliveries.