Collection
zero Useful+1
zero

real-time data

The carrier of information obtained at the same time during the occurrence and development of something
Real time data is the carrier of information obtained at the same time during the occurrence and development of something, and it is the raw material used to represent the objective things.
Chinese name
real-time data
Foreign name
real-time data

object

Announce
edit
1. Image object
In RTDB, real world objects (REWO) are created by sensor For monitoring, its value is sampled periodically or irregularly and written into the database. The data object being written is called the image object (IMO). So an IMO is a RWO image
It can be seen that an IMO has a corresponding RWO sampling time associated with it, which is called "sampling/event timestamp". The time zone from this beginning to the next sampling is the "external validity period" or "effective time" of the IMO.
Once an IMO is recorded in the database, it is generally not changed, but the sampling value of the corresponding Rw (_) at a later time is written data base A new IMO for this object. Whether and how to retain the "old" IMO of the same RW depends on the application semantics. Generally, the "old" IMO is stored in archives, so RTDB needs to maintain the instantaneous images of data at different time points. [1]
2. Export Object
Exporting an object is a transaction between a group of IMO and/or other data object The data object of m is obtained from the calculation of. Associated with this is the transaction time from which it was exported. Unlike IMO, the value of DEO may be updated in the database, and its archive storage can be maintained or not.
3. Invariant object
An IVO is a time invariant value. It can be regarded as real-time data or not. If it is, it does not change with time. Its sampling time is always "current" and its validity period is arbitrary. It is a special case of real-time data. [1]

Resettlement principle

Announce
edit
(1) Real time data. The current valid and short-term valid data can only be saved in the "memory version" of the database, otherwise it may have expired after I/O.
(2) Data activity. Active data with high access frequency shall reside in memory.
(3) Data permanence. File storage data is placed in external storage and retrieved into memory when it needs to be accessed.
(4) Criticality. The key data critical to hard real-time transaction processing is best placed in memory. If it is permanent at the same time, of course, there must be an external copy.
(5) Transaction type. For example, those "data receiving" transactions that receive data from sensors are often hard real-time "write only" transactions, so their data should be placed in memory, and then whether to transfer to external storage should be determined as needed.
(6) Transaction priority. High priority transactions are generally urgent, and their data must reside in memory and cannot be exchanged.
(7) Log. Transaction processing is always accompanied by frequent log operations. Log placement greatly affects the satisfaction of transaction deadlines. Therefore, "memory" logs must be designed. [1]

Real time data processing

Announce
edit
If real-time data processing is required, decoder It must be able to adapt to the relevant data rate. This can be used delayed As a cost, for example, when decoding one sequence, the next sequence is buffered. In some cases, the decoding delay becomes apparent, especially if it is variable.
Forward error correction The decoding delay is required. In most cases, the delay depends on the specific error sent. However, there is usually a certain maximum delay, which cannot be exceeded. The decoded information is buffered until the maximum delay expires. This method can generate a smooth flow of information to the terminal. The two main factors that determine the delay are Data rate And yard length. Information theory tells us that long codes are ideal, but for many applications, long delays are not ideal. Therefore, the maximum acceptable delay may limit the code length that can be used.
If the maximum decoding delay cannot be determined, then decode Post information will experience variable delay, which will cause serious damage to real-time information. One of the main error control strategies showing variable delay is ARQ, because we cannot guarantee that any retransmission will succeed. These problems can be minimized by using the appropriate ARQ/FEc hybrid scheme. [2]