NIA Conference Call

June 14, 2000

Good afternoon Gentlemen:

 We are participating in a very important event for the telecommunications industry that without doubt will be a cornerstone to future developments in wireless telecommunications.

 Not only the wireless licensing process was changed to involve all parties but for the first time the industry decided to tackle the peer-to-peer interference issue.

 The FCC and the engineers that specified what we call the Methodology or Appendix D brilliantly addressed this challenge.

 We at CelPlan very quickly realized the importance of this process and accepted the challenge of developing software tools that could materialize the Methodology. Today, we consider this methodology an important part of our broadband technology and this concept has been extended to the whole broadband universe.

 It was not an easy task due to the difficulties introduced by the peer-to-peer interference. Much of the ground was already addressed in the Methodology and we just had to build on top of it. CelPlan had the benefit of its existing tools that provided the framework to add the new functionality.

 We started our development in September 1999 and had our first release in January 4, 2000. The availability of the tool was publicly announced February 17 at the NIA conference in Long Beach and at the CTIA trade show February 28. On April 10 the tool was made commercially available.

 We licensed more than 10 entities, totaling more than 150 licenses. We trained more than 100 engineers and those trained more engineers. Up to now the software has been used for more than 100,000 hours. Our software is very user friendly and engineers can start producing applications after two or three days of training even without the existence of user manuals. Proficiency is achieved over a period of a month.

 All this activity resulted in many software revisions to address all the findings and functionality requests. These resulted in a very powerful tool, extremely friendly and easy to use. We provided very powerful analysis features that slashed the design time considerably. Several optimizations were done to increase the processing speed.

 We have actively contributed to improvements in the Methodology revisions, which were frozen on end of April. On June 2 we felt comfortable enough to freeze the formats and calculations until the filing window (release V3.A07). Any interference analysis since then does not need to be re-worked with the exception of the generation of output files, which impact very little in time. Some previous studies that fall into specific cases may have to be re-analyzed.

 To the best of our knowledge the software is fully compliant with the methodology, it is very stable and mature. It was independently tested by several parties and experimented on different OS and many machine brands. To help in the accuracy and consistency analysis by independent parties we introduced in the tools additional text files containing intermediate and final results.

 We and other entities have designed hundreds of markets from simple super cell designs to complex multicell ones.

 We will continue to improve the tool, to increase its efficiency and to automate some procedures. This is the case with the limited exception rule, also known as grand fathered interference. Today we generate a text file with all the information and all the user has to do is to analyze the data in a spreadsheet. The importation of the Methodology text files is done manually today.  We are in the process of making those two steps fully automatic. The calculation of the protection of response station hubs is not required until the filing window and we gave lower priority to it.

 The software has been optimized in terms of speed and is extremely fast if we consider the calculations that are involved. An RSA – PSA pair has generally more than 10 million paths each one with more than 5000 mathematical calculations. This gives about 50 billion calculations per pair. Each pair is processed today in 1 to 9 minutes, depending on the size of the RSA. A complete study of a small market can be processed in 4 hours and for a large one in 30 hours. The major part of this time is computer-processing time and the human intervention is limited to about 10 % of it. Of course we need to consider that there is preparation time and many iterations may be required to clean the interference in a market.

 The Methodology is the most compete issued to date, defining many parameters to assure that everyone gets the same results in terms of interference analysis. This does not mean that every step in the process will be exactly the same when different parties do the analysis. There are several factors that can differentiate the results, like truncating, circle approximations, unit conversions, sampling rates, secondary parameters, and bounding, etc. Those parameters can cause variations in the intermediate results but should not affect the final analysis.

 The interoperability issue has been addressed by the FCC with the text file output and is natural that small adjustment may be required when doing the importation. Those adjustments will have to be addressed as they arise and this is being done today.

 We sincerely believe that our software is sufficiently developed to design any ITFS/MMDS market according to the rules of the Methodology.

 Leonhard Korowajczuk

This site powered by Contentor.