Logiciel d'étude WiFi : Comment cartographier et optimiser votre réseau sans fil

Welcome to the Purple Intelligence Briefing. I'm your host, and today we're tackling a topic that sits right at the intersection of network engineering and business performance: WiFi survey software — what it is, how to use it properly, and how the data it generates can transform the way you design and manage wireless networks across large, complex venues. Whether you're responsible for a hotel with three hundred rooms, a retail estate with fifty branches, a university campus, or a conference centre that turns over ten thousand visitors a day, the quality of your wireless network is no longer a back-office IT concern. It is a direct driver of guest satisfaction, operational efficiency, and increasingly, revenue. And yet the majority of organisations we speak to are still running networks that were designed once, deployed, and never properly validated. That is a significant risk — and it is entirely avoidable. So let's get into it. Let's start with the fundamentals. WiFi site survey software is a category of tools that allows network engineers to measure, map, and model the radio frequency environment within a physical space. The output is typically a heatmap — a visual overlay on your floor plan that colour-codes signal strength, signal-to-noise ratio, channel utilisation, and other key RF metrics across every square metre of your venue. There are three distinct types of survey you need to understand. The first is a passive survey. Your laptop or survey device listens to the RF environment without connecting to any network. It captures beacon frames, measures RSSI — that's Received Signal Strength Indicator — across all visible access points, and logs the data against GPS or floor plan coordinates. This gives you a picture of what is actually being broadcast in your space, including interference from neighbouring networks. This is your baseline. The second is an active survey. Here, your survey device connects to the network and performs real throughput tests — UDP and TCP — measuring actual data rates, packet loss, and latency at each survey point. This is where you move from "can devices see the network" to "can devices use the network effectively." For venues running real-time applications — video conferencing, point-of-sale systems, IoT sensor networks — active survey data is non-negotiable. The third is a predictive survey, sometimes called a virtual survey. You import your floor plan into the software, define the construction materials — concrete, glass, plasterboard — assign attenuation values, and the software models how RF signals will propagate before you install a single access point. This is invaluable for greenfield deployments and major refurbishments. It reduces the risk of over-provisioning or under-provisioning your infrastructure before you've committed capital expenditure. Now, what are the key metrics you're actually measuring? Let me give you the five that matter most in a commercial deployment. RSSI, as I mentioned, is your signal strength indicator, measured in dBm. For general connectivity you want a minimum of minus 70 dBm at the client device. For voice and video applications, you want minus 67 dBm or better. Anything below minus 80 dBm and you will see degraded performance and frequent roaming events. Signal-to-Noise Ratio, or SNR, is arguably more important than raw signal strength. SNR measures the difference between your signal level and the background noise floor. You need a minimum of 25 dB SNR for reliable operation; 30 dB or above for high-density environments. A strong signal in a noisy environment is still a bad network. Channel utilisation tells you how busy each radio channel is. In a dense urban environment or a conference centre with hundreds of devices, you may have excellent signal strength but terrible throughput because every device on the channel is competing for airtime. Your survey software should be capturing this. Roaming behaviour is critical in large venues. IEEE 802.11r — fast BSS transition — and 802.11k and 802.11v together form the trifecta of enterprise roaming standards. Your survey needs to validate that client devices are handing off cleanly between access points without dropping connections. Poor roaming is the number one complaint in hotel and hospitality WiFi deployments. Finally, co-channel and adjacent-channel interference. In a multi-AP environment, overlapping coverage cells on the same channel create contention. Your survey software will identify these conflicts and allow you to adjust channel assignments and transmit power to resolve them. Now, let's talk about the software itself. The market broadly divides into two categories. Professional-grade tools — Ekahau Site Survey and NetSpot Pro are the most widely deployed — offer full floor plan import, active and passive survey modes, predictive modelling, and detailed reporting. These are the tools your network architects will use for formal deployments. Then there are lightweight mobile tools — apps like WiFi Analyser on Android — which are useful for quick spot checks but lack the rigour for enterprise design work. When evaluating WiFi site survey software, look for four capabilities: accurate floor plan scaling and calibration, multi-floor support for multi-storey buildings, the ability to export data in formats your network management platform can consume, and integration with your access point vendor's planning tools. Cisco's DNA Spaces, Aruba's AirWave, and Juniper Mist all have native integrations with the leading survey platforms. One area that is increasingly important — and often overlooked — is the integration between your survey data and your guest WiFi analytics platform. When you layer analytics on top of a well-surveyed network, you move from knowing where your signal is strong to understanding where your users actually are, how long they dwell, and how that correlates with business outcomes. That is a fundamentally different conversation. Let me give you the practical guidance that separates a successful deployment from one that generates a support ticket every Monday morning. First: always conduct a pre-deployment predictive survey before you order hardware. I have seen organisations install access points based on a vendor's generic coverage calculator, only to discover that the concrete pillars in their atrium create RF shadows that the calculator never accounted for. A predictive survey costs a few hours of an engineer's time. Ripping out and reinstalling access points costs significantly more. Second: survey at representative load. An empty venue at nine in the morning on a Tuesday is not representative of a stadium at full capacity or a hotel during a conference. Your active survey should be conducted with a realistic number of client devices on the network. Some survey tools support simulated client load; use that capability. Third: document everything. Your survey report is a living document. Every time you add an access point, change a channel plan, or modify transmit power, you should re-survey the affected area and update your baseline. Networks that are not documented are networks that cannot be troubleshot efficiently. Fourth: do not ignore the 6 GHz band. WiFi 6E and WiFi 7 deployments are introducing the 6 GHz spectrum, which offers significantly less interference but also shorter range due to higher frequency attenuation. Your survey methodology needs to account for tri-band environments. The most common pitfall I see is organisations treating the site survey as a one-time event rather than an ongoing operational practice. Your RF environment changes. Tenants move in next door. New construction materials are introduced. Seasonal changes in occupancy alter the interference profile. A quarterly survey cadence for high-density venues, and an annual survey for standard office environments, should be your baseline operational standard. Let me address the questions I get most often. "How many access points do I need?" — The honest answer is: it depends on your density requirements, not your square footage. A 500 square metre open-plan office with 50 users needs a very different AP count than a 500 square metre conference room with 300 delegates all on video calls. Survey first, then size. "Can I use free WiFi survey software?" — For a home office or a small retail unit, yes. For anything with more than two access points and a compliance requirement, no. The reporting and validation capabilities of professional tools are worth the licence cost. "How does this relate to GDPR and PCI DSS?" — Your survey data itself is not personally identifiable, so GDPR is not directly in scope. However, the network design decisions you make based on survey data — segmentation, guest network isolation, encryption standards — absolutely are. WPA3 and IEEE 802.1X are your baseline for any network handling payment card data or personal information. To bring this together: WiFi survey software is not an optional extra for enterprise network design. It is the foundation of a network that performs reliably, scales predictably, and can be troubleshot efficiently when issues arise. The three things I want you to take away from this briefing are: one, conduct a predictive survey before deployment, not after. Two, treat your survey as an ongoing operational practice, not a one-time project. And three, connect your RF performance data to your business analytics — because a well-mapped network is also a network that can tell you something meaningful about how your venue is being used. If you want to go deeper on any of this — particularly on how guest WiFi analytics and footfall data layer on top of a well-designed network — head to purple dot ai. The guides and case studies there will give you the implementation detail you need. Thanks for listening. Until next time.