Real time date compression method and apparatus for a data recorder

• US 6,950,041 B2
• Filed: 07/21/2003
• Issued: 09/27/2005
• Est. Priority Date: 12/26/2002
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Chu, Yao-Tung
• Primary Examiner(s): WAMSLEY, PATRICK G
• Application Number:
  US10/623,671
• Publication Number:
  US 20040128330A1
• Time in Patent Office: 799 Days
• Field of Search: 341/76, 341/95, 341/60, 341/51, 360/15, 386/65, 386/98, 704/260, 707/4, 707/101, 375/240.12, 375/250, 708/203, 714/758
• US Class Current: 341/51
• CPC Class Codes: H03M 7/30 Compression speech analysis-synthesis for redundancy reduction G10L19/00; for image communication H04N; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A real time data compression method examines whether a present read-in data point is in a predicted tolerable error range. If yes, the previous data point is deemed as redundant and is ignored and after which a new data point is read in for comparing again. When a predetermined amount of data points is continuously ignored, it means the data is steady and only the final data point of these ignored ones is recorded. Otherwise, when data is varied greatly during a period, only the total amount of these varying data points and their value are recorded so the memory capacity for storing the compressed data is saved. Furthermore, the compressed data is stored in a form of a data structure in which the compressed data are expressed by multiple blocks. The block form allows increase of the efficiency of the searching process of the compressed data.

Claims

1. A real time data compression method for a data recorder that sequentially reads a plurality of sampled data points, the method comprising the steps of:

setting the first sampled data point as a starting point X₀;

sequentially examining the plurality of data points, wherein a present data point X_i being examined is determined whether it is in a predetermined tolerable error range;

wherein if the present data point X_i is in the predetermined tolerable error range, a previous data point X_i-1 is ignored;

wherein the previous data point X_i-1 and the total amount of data points accumulated from the starting point X₀ to the previous data point X_i-1 are recorded, and the previous data point X_i-1 is then set as a new starting point, when (a) the present data point X_i exceeds the predetermined tolerable error range or (b) the amount of the ignored data points reaches to a first predetermined amount;

wherein during a period that the sampled data points are varied quickly, and when the amount of the continuously examined data points is less than a second predetermined amount, the total amount of the data points that exceed their respective predetermined tolerable error ranges and the value of each of these data point are recorded,

wherein a data format is provided to store all recorded data points, the data format comprising;

a main region composed of multiple segments, wherein each segment has a predetermined length and is provided to store the value of a presented recorded data point and an amount value of the data points calculating from a previous recorded data point to the present recorded data point; and

a secondary region for recording a start time of the examining process, a value of sampling time interval and multiple blocks, wherein each block corresponds to several segments in the main region and includes a time value and a position value respectively representing a start time of the block and the start position of the block.

2. The method as claimed in claim 1, wherein the time parameter of each block is calculated relative to the start time of the examining process and is expressed by multiples of sampling time interval.

3. The method as claimed in claim 2, wherein data of all recorded data points is firstly stored in the main region, and then in the secondary region.

4. The method as claimed in claim 1, wherein in a first segment of the main region, the amount value is set to zero.

5. The method as claimed in claim 4, wherein data of all recorded data points is firstly stored in the main region, and then in the secondary region.

6. The method as claimed in claim 1, wherein the second predetermined amount is less than the first predetermined amount.

7. The method as claimed in claim 1, wherein the total amount of these data points that are temporarily stored when the sampled data points are varied quickly is expressed by a negative number.

8. The method as claimed in claim 1, wherein data of all recorded data points is firstly stored in the main region, and then in the secondary region.