Specifications
47
5.7 Data format 
Datawell uses two types of message formats: the real-time format and the message format. The 
Datawell real-time format refers to the long-existing hexadecimal vectors with vertical, north 
and west displacements and words ultimately forming the spectrum and system files. It is 
mainly used for relatively low-speed, continuous, near real-time data transmission over the 
standard HF radio link. More recently, the Datawell message format has been introduced. This 
format has been prompted by an ever increasing number of datalinks and sensors. It is more 
suited to data bursts over relatively high-speed links or expensive links. The message format is 
independent of the link, easy to extend and easy to process in software. 
5.7.1 Datawell real-time format 
The real-time format is organized at four levels 
(1)  Vectors of 64 bits with real-time data together with cyclical data 
(2)  Blocks of 18 vectors assembling cyclical data with spectral and system data 
(3)  Spectral data of 16 blocks complete a spectrum file 
(4)  System data of 16 blocks complete a system file 
More precisely one or two samples of the real-time data are framed in one 64-bit vector, but the 
64-bit vectors also contain fractions of cyclical data. After collecting 18 vectors all fractions of 
cyclical data form a complete block of cyclical data. Similarly, one cyclical data block contains 
a fraction of a complete spectrum and system file, which require 16 blocks or 288 vectors to 
assemble. The full set of cyclical data in 16 blocks makes up a system file and a spectrum file. 
  Below all four levels are explained in detail. Timing and the derived compressed 
spectrum are also discussed in the following subsubsections. 
5.7.1.1 Real-time displacements 
Each 64-bit vector is subdivided into three parts 
•  cyclical data (sync word or system file word or spectral data words) 
•  real-time displacements (one vertical, north and west sample or two vertical samples) 
•  parity bits for transmission error detection and correction 
The parity bit algorithm uses a Galois code table to encode 63 bits of a vector. In the receiver 
these bits are processed to check data integrity and correct for transmission errors. All 
displacements are given as 12 bit signed integers in cm. The most significant bit (MSB) is the 
sign bit, MSB = 1 means negative. Table 5.7.1 shows the organization of the 64-bit vector. On 
DWR-G buoys the least significant bit (LSB) of the north is used as a GPS flag (0 = ok, 1 = 
error). 
Table 5.7.1. Organization of real-time displacements in 64-bit vector. 
Cyclic data  Real-time displacements (1 cm/bit)*  Parity bits
WR-SG  vertical (even sample) vertical (odd sample) dummy (−628 cm, 
101010101010) 
DWR-MkIII, 
DWR-G 
vertical north** west 
see text 
16 bits  12 bits  12 bits  12 bits  12 bits 
Bits 63-48  bits 47-36  bits 35-24  bits 23-12  bits 11-0 
*MSB is sign, 1 means negative 
**LSB is GPS flag (DWR-G only) 










