Two training strategies exist for radial basis networks. One is to .x the bias and create one neuron for each training vector. This way, the network is sure to classify all the training vectors correctly. An alternative approach is to limit the number of neurons, and adjust neuron positions and biases so that the total miss-classi.cation of the training set is minimized. The latter approach will generate a network with fewer coe.cients / neurons, and in some cases better generality.
Appendix B

IT Notations
B.1 Databases
Enterprise business solutions typically generates complex data storage needs, such as centralized storage for multiple users involving simultaneous multi users reading and writing. This creates problem that goes far beyond the scope of the .le locking functions provided by most .le systems. Standard Query Language (SQL) databases are solutions that are capable of providing such functions.
An SQL database is in all simplicity a patch of shared memory. SQL is the languages used for reading and writing to this memory. Commands in SQL format are sent from a user to the database engines, which return the result (if any) to the user. The mode of communication is speci.c to each database engine. Most database engines are background processes accessible through TCP/IP and / or platform speci.c data pipe systems. Database engines accessible through TCP/IP are typically accessible over network, and provides client connectivity implementations for several platforms. A client connectivity API is necessary even for TCP/IP based database engines, because there exist no common application layer protocol for SQL database (the layer between TCP and SQL).
The data in an SQL database is organized in tables. There exist SQL commands for inserting, retrieving, updating and deleting table data, as well as crating, modifying and deleting tables. Most database engines are capable of serving multiple clients at the same time, and provides rigorous mecha-nisms for ensuring data consistency in multi user environments. Enterprise scale database engines also provide sophisticated data mining functions and integrated backup solutions.
In scenarios involving multi user data access and data tra.cking over
139
networks, developers can save substantial time by using third party database servers rather than developing their own proprietary solutions. Further, us-ing open standards like SQL makes solutions more scalable and more easily accessible through other third party systems.

B.2 Object Oriented Programming
Object oriented programming has received increasing attention over the last decade, and has become the defacto standard for complex software projects. Object oriented programming means in essence nesting functions that belong together into objects. Some object oriented languages also deploys packages or namespaces, for nesting classes that belong together.
An illustrative problem example is .le i/o operations. Using classical C-style function calls, .le opening, reading, writing and closing are managed by a set of global functions. All of these functions use a .le handle as input or output, where the handle is a reference to a place in memory where the .le is stored. An object oriented approach to the same problem is creating an object containing the same functions, and with the .le handel being part of the internal memory of object.
　
中國航空網 www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：OPTIMIZATION OF FAULT DIAGNOSIS IN HELICOPTER HEALTH AND USA(74)