Major projects I have been involved in
Areas of expertise I have gained
What does Laser-Scan do?
Originally started around 30 years ago using laser scanning methods to analyze giant photos generated from hydrogen bubble chambers at Cambridge University, the company applied the laser scanning methods to the mapping world and its entry to the GIS world was marked by its success. Now, as well as supplying banknote design software, on which some 40% of the world's banknotes (including the UK) have been designed, the company also supplies software solutions to many parts of the GIS market, from mapping companies such as the AA to superstore positioning to telecommunications and utilities companies.
I joined Laser-Scan to work with one other software engineer on a telecommunications analysis application (Telecomms Analysis). This product uses geographical information (eg land height, roads, buildings and other natural features) and antenna/receiver information to predict signal strengths in any given area for which the geographical data exists.
The product was used by many telecomms companies, such as Philips Telecom, Hutchinson Telecom, and Ionica. My job was to add functionality and customizations to the application as required by various customers. One of the more important customers was Ionica, whose site in cambridge I visited regularly, and with whom I was almost a personal consultant, working with them to implement many of their requested features.
One of my major aims with the product was to add support for use with the GSM mobile cellular phone market, and for the alternative digital system used in the USA and much of Asia - the CDMA system. To this end I spent a month of consultancy with Korean Mobile Telecomms in Taejon, Korea, and a further month training and working with the Koreans in the UK.
After nine months at Laser-Scan my project manager and partner on the project left and I took over responsibility for the product at all stages. Coding was in C and an in-house compilable run time language called LULL. We used ORACLE as our GIS database. I also had to give training courses in specific and general applications of the software, in the UK and abroad (Korea and Hungary).
Major projects I have been involved in
As well as adding considerable amounts of functionality to the product and managing the project for half my time at Laser-Scan I was responsible for the following sizable projects:
User Prediction Methods
Telecomms Analysis had several prediction methods, ie given the antenna and geographical information there were several mathematical and statistical methods by which signal strength could be calculated. In readiness for the next major project, the product had to be adapted to be able to use a customer's own specific prediction calculation method since they would be better tailored to the customers own requirements. The customer would be supplied with a number of interface functions by which they could register a prediction method, and alter its characteristics. That method would be used in the signal strength calculations, and the results exported back to the user as an image or data in a requested format if necessary.
This project meant working very closely with an engineer from Ionica; firstly to fully understand their requirements, secondly through development to make sure that we understood the engineering side, and that they understood the software side, and thirdly to train prospective users in fully utilizing the system. I also worked with a Philips Telecomm field engineer (often at their site in Cambridge) who also wanted to add their own prediction methods to the product.
Cellular Analysis
Previously the product was designed to give signal strength predictions over a given area from a single transmitter based on the type of receiver. A considerable amount of change was necessary to adapt the product to cater for the cellular networks that were increasingly important with the popularity of the mobile phone. The product had to take account of many transmitters across the country and how they either interferred with each other or combined to give coverage of areas of population and road networks. The most likely server (ie transmission antenna with the strongest signal) had to be calculated, and service areas (that area for which a transmission antenna would provide the strongest coverage relative to adjacent antennas) was also important.
CDMA - Code Division Multiple Access
Over Europe and many other parts of the world, GSM communication has been taken as the accepted form of digital mobile telecomms. However in the USA and much of Asia, another new digital technology system called CDMA is used which calls for a completely different technique of calculating signal strengths based on the antenna and geographical information. I worked very closely with a team of four telecomms engineers from Korea mobile Telecom (KMT) from Taejon, Korea. Firstly I had to give a three day training course to the team to explain about the current methods for adding and using a customized prediction method. I then spent a four week period helping them to write their prediction method, for which they had very little English paperwork. After they had returned to Korea, I spent a further month adding some more functionality that would be required by the CDMS prediction method - mostly raster combination methods for extremely large data sets. Finally, I was sent out to KMT's headquarters in Korea for a month of consultancy, getting their system fully implemented and working. This was an intense period of work, working six or seven day weeks and often sixteen hour days. It was a very rewarding project with which KMT were particularly pleased, and I received a letter of commendation from Laser-Scan thanking me for my hard work in Korea. I was also proud to be the only person in the world who had implemented a software CDMA prediction method at that time, and my methods were used for a cellular system which now covers 85% of the population of Korea.
Training in Hungary
Although not such a long project, I was sent alone to Budapest, Hungary to install Telecomms Analysis on a SUN sparc10. I then had a couple of days to train an engineer in using the software, who was then due to demonstrate the software to the Hungarian military (I believe!) who were interested in the product. This was a big test of fending for myself to get over installation problems; in conveying as much information as possible to someone who spoke little English; and in flowing with the wishes of the important Hungarian people with whom I was working. The five days I spent in Budapest was a rewarding ordeal, and the demonstration to the military went well.
