Box structure for net chamber specification and design (1)

The box structure (1) of the net chamber specification and design is not allowed. It is strictly forbidden to reproduce this column. This paper is reproduced from the Software Engineering Expert Network www.21cmm.com. Do not reprint without CBS license, thank you for the cooperation box structure Description of the external basic properties of the real system in the specification and design. Figure 3.1 depicts three boxes: black box, state box, bright box. These boxes not only show external behavior, but also increased internal visibility. The black box determines the external behavior of a system or system component. The status box further specifies the status data required to complete external behavior. The bright box further applies the status box, it determines the process design of completing the status box behavior. It uses existing black boxes or introduces new black boxes, which will be refined in the future (the bright box is constructed by the program control structure), and each step is verified according to the previous step. Such a box structure is separated from three aspects (behavior, data, and procedures for behavior, data, and procedures), but couples them into a fake and verified consolidation process. 1. Black box behavior black box defines an external behavior of a system or system component. When the system accepts incentives S (see input from its environment), it will generate a response R (as an output of its environment), the response is not only related to the current motif, but also related to the current receivable. For example, we consider an external behavior of a calculator. Suppose it is ongoing a calculation, the current excitation is to press the key 5; if the previous historical excitation is C718 (C represents clear). Then the response will be 7185. That is, the calculator shifts one bit left left and inserts 5 in unit position. If the previous historical excitation is C718 , the calculator will display a new string 5 in unit location. Therefore, the current incentive of the system and the previous historical incentive uniquely determine its response. We can look at the system and its components from external behavior. For example, the workstation accepts the keyboard and the mouse to change the current window content or display a new window to respond. Users may not know anything about internal operations of the workstation, only feeling its external behavior. Of course, internal operations may be completed by other machines that support this external behavior than it. Figure Box Structure Refrigeration and Verification BB = Mathematical Semantics of Black Box Black Box Behavior You can write as follows: History Incentive → Response Notes SH → R SH Represents all historical excitation of the current incentive. The black box definition does not include status data and process implementation. It defines the external visible behavior that can only be felt by users depending on the history. Therefore, the black box cares about the problem of system behavior from the user's perspective, and does not consider the design of status and procedures. The black box specification defines the behavior required for all possible use. That is, in the black box specification, all possible current excitation and historical incentives and their combinations define the correct response, and the following three principles of the black box specification in the clean room project are critical. 1. For system owners and users, the black box defines what they analyze and negotiate, which is the premise they prepare resources, start development and testing. 2. For system developers, the black box defines the required behaviors to be designed and implemented. 3. For system testers, the black box defines the required behaviors to be confirmed during the test process. The black box can be defined in the form of form, and the three columns in the table correspond to motivation, historical conditions, and responses, respectively. These tables can be described in any suitable form (from natural language to aggregation theory). Please see the non-form black box definition in Table 3.1. This low-level black box defines a predictive system based on 12 months for thousands of product forecasts for a product list control system. The system accepts the excitation of the monthly sales of products. And produce a corresponding response. The rule 1 of the black box specifies the correct response when a product is less than 12 months. Rule 2 specifies the correct behavior when there is at least 12 months. This table describes system behavior in an informal manner, where the contents of the angle brackets are important entities to further define and refine.

This table may be early in the early days of system specification, and the analysis and discussion of users have made a consistency in demand, but there is still no further refining work. For example, discussions on rule 1 may result in more calculation requirements, such as 3 or 6 months average sales. The discussion of rule 2 may lead to such a need: a monthly sales make incentives, before calculating the average, it is determined that the sales value is accidentally neglected in the collection of sales values. These changes will designate new behaviors. Such discussions are promoted in the black box specification. This avoids waste resources and prevents the development that does not meet the requirements of the user. 2. The status box behavior status cartridge specification is preliminarily refined to the system or its components, defines the status space. The status box puts the motivation history into state data, but still does not involve specific processes. It maps the old OS (OLD State) and excitation S to new status NS (New State) response R. The new state has become an old state in the next transform. The semantics of the status box behavior are one of the following transformation functions: (old status, incentive) → (new status, response) or short-write (OS, S) → (NS, R) status box according to black boxes according to black boxes. Status information is to incentive history that must be saved in accordance with black box specifications, so information comes from black boxes without definition. Because each historical motif can be represented, each black box has a state box description. Moreover, for a black box, many different state boxes can be designed because there may be a variety of different representations and access methods to a state. The status box can be defined, and the corresponding columns include old state, excitation, new status, response, and corresponding black box rules. Let's consider the status box of the sales prediction system, as shown in the following table, it corresponds to the previous table. In this example, the incentive historical analysis of the black box leads to the status item to save the last 11 months of each product. Each record in the file is identified by and contains an array that can save 11 months. Because the current excitation plus the previous 11 can ask for average, it is enough to save 11 values. It is not necessary to keep earlier, so the next incentive will result in deleting sales earlier than 11 months. amount. The transformation in the table below defines the behavior required when energizing the introduction of a new product. Transform 2 Defines the behavior required for the product known in the status box that has not reached the 11-month sales, transform 1 and 2 are only generated text messages, and the excitation is displayed to the user. Finally, the transform 3 defines the steady state of constantly calculating the average plant. Note that each transformation must be updated accordingly, preparing for the subsequent incentive. For example, the transform 3 is the accumulation process of such processing states, deleting the earliest and putting the current excitation as the latest Add to records of . The verification of the status box is to export the corresponding black box behavior and to see if it is equivalent to the original black box. This export is to convert the status operation into an incentive history. 3, the bright box design of the mandatory system or its components defines the process of state box behavior. The bright box is a computer program or assembly, based on the program-based internal state OS, which accepts the incentive S, generates a new internal state NS, and generates response R. These processes are defined by a control structure (order, selection, cycle) based on a structured programming (order, selection, and parallel structure if the concurrent mechanism is introduced). The bright box uses these control structures to complete the calculation of new status and response. A variety of different bright boxes can be defined for the given state cartridge.