RSOFT Design Group's WirelessSWAT
Page 1 of 1
Wireless providers are implementing citywide outdoor wireless networks with great intensity. This increasing tempo coupled with a soft economy places powerful demands on wireless Internet service providers (WISPs) to plan deployments carefully.
For those WISPs hoping to deploy additional markets and needing tools to model their deployments, they can trye RSOFT Design Group's WirelessSWAT strategic analysis software. RSOFT designed the program to provide a best-of-class statistical modeling tool for broadband wireless voice and data deployments.
- Very comprehensive. A professional level tool
- Simple interface for studies
- Lots of graphical reports
- Powerful RF efficiency and capacity modeling capability
- Personalized support
- Expensive for the casual user
- Cumbersome licensing procedure
- Modeling process could be more linear
WirelessSWAT is clearly a professional level tool for those serious about wireless deployment modeling. The software models both potential multiple spectrum wireless and DSL deployment results. Its basic interface is simple and straightforward. With a modest time investment users can generate sophisticated studies incorporating multiple markets.
The software looks at a market as the user defines it for a typical case.
The software makes some assumptions. For example, network cost factors are front-loaded at the start of the project rather than throughout a growth curve as actual deployment costs typically are factored. This serves a statistical model just fine, but could frustrate those wanting more concrete results for a single market.
Setting up the Software
Software setup was very simple. The software comes in a classy briefcase folio which includes excellent four-color documentation and tutorials. The CD self launched and the installation on my Windows 2000 Professional system went smoothly. The software currently runs on Windows 2000 Professional or NT 4 only. It requires a Pentium III processor with at least 128MB of RAM and 20MB of available hard drive capacity.
The only real knock on installing WirelessSWAT is a cumbersome registration process. RSOFT controls the registration by requiring users to submit a text file generated by the installation to receive an activation text file in return. This has to be copied to the WirelessSWAT folder. This is to deliberately validate only legitimate installations, but it is cumbersome.
What it Can Do
The interface is simple, deceptively so. There are numerous variables the user can tweak to fit particular business cases.
The basic paradigm revolves around the study. A study consists of three elements, the canonical area (basically market type), service options and the technology. Studies can incorporate up to two canonical areas and up to three customer segments. Canonical areas (CA) can range from dense urban to sparse rural. The database has in-built statistical tools to model these areas.
One suggestion to RSOFT would be to increase the number of canonical areas that could be incorporated into an overall study. However, it is simple to create multiple studies based on varying combinations of CAs.
The software models multiple types of spectrum deployments and users can select studies using MMDS, 2.4GHz or 5GHz service or others in combination for both rural and suburban markets simultaneously.
WirelessSWAT allows users to define a large set of variables including those for the core network, the business case elements, location aspects and link parameters. Within each of these essentials the user can further refine specific variables for things such as churn rates, bandwidth, RF propagation technology (FDD or TDD) and many others too numerous to name. It is quite comprehensive. Only a wireless professional can really take full advantage of this area.
Once variables are tweaked, then studies can be run and saved. The results of a modeling study are provided mostly in graphical format over the length of the study (up to 18 years). Summary results for cash flow, capital expense and risk/reward analysis are provided. A more traditional summary table is also included. Interpreting and understanding these reports requires an investment of time.
What it Doesn't Do
As mentioned before, WirelessSWAT looks at typical markets from a statistical viewpoint. It's not designed to globally model multiple additional and unrelated services that WISPs may choose to offer on a staggered basis such as Web hosting, security products and network services for example. While each element can be modeled, introducing them in staggered fashion throughout the study length isn't possible.
My recommendations to RSOFT would be to incorporate more customer segments in each study. It would also be helpful to add some linear elements to its model to more realistically reflect real world ROI results.
Clearly this kind of sophisticated analysis tool is new to the broadband wireless community. For those needing to build a business plan and financials for only a single specific market then spreadsheets may make more sense.
For companies headed down the expansion path, WirelessSWAT will be an invaluable and cost saving tool to conduct feasibility studies for multiple types of deployments in various markets. Likewise firms serving such clients could find it very helpful in providing modeling options to their clients.
Tim Sanders is founder of The Final Mile, a fixed wireless consulting group. His experience was gained running a multi-state fixed wireless ISP. He can be reached at www.thefinalmile.net.
Want to model your WISP setup? Tim Sanders can show you how! Join him at the 802.11 Planet Conference & Expo, Dec. 3-5 in Santa Clara, CA. He'll be speaking and moderating a number of panels for WISPs including Refining Your WISP Business Plan.