| are 
ring, 
stem 
vices 
ment 
such 
ring 
k for 
land 
; of 
JS. 
iding 
1g to 
evant 
ıman 
oring 
oring 
vhose 
stem. 
] an 
vital 
rmed 
lavior 
of its 
ques, 
ring, 
status 
ce of 
nodel 
| are 
S are 
ed to 
resent 
| land 
ata to 
are 
's and 
natted 
entify 
which 
mechanism is used, the monitoring systems can be 
categorized into three types: hardware monitors, 
software monitors and hybrid monitors. 
Hardware monitors are separate objects that are used to 
detect events associated with an object or a group of 
objects. The detection is performed by observation of 
status of rural land use or by using remote sensing and 
‘manual survey techniques connected to a data 
acquisition system. Hardware monitors have been 
successfully used for monitoring various physical 
parameters of rural land use, where a great deal of 
information is collected and processed rapidly. 
On the other hand, software monitors usually makes 
use of simulation models to identify the occurrence of 
events. Software monitors present information in an 
application-oriented manner that is easy to understand 
and use, compared to the raw information generated by 
hardware monitors. Software monitors can easily be 
replicated and are more flexible, portable, and easier to 
design and construct than hardware monitors. The 
disadvantage of software monitors is that their 
simulation models require a great deal of input data 
and therefor interfere in both their application area and 
precision of monitored data. For this reason, pure 
software monitors are not adequate for on-line, real- 
time monitoring of rural land use. 
Hybrid monitors are designed to employ the advantages 
of both hardware and software monitors, while 
overcoming their disadvantages. Typical hybrid 
systems consist of an independent remote sensing 
hardware device that receives monitoring information 
from typically rural land components, and an 
independent software model that computes monitoring 
information for other similar rural land components. 
5. ALARM PROCESSING 
Alarm correlation technique is widely used in the 
monitoring system. The conceptual approach to alarm 
correlation was discussed in (Aloni, et al, 1991). 
Interpretation and correlation of events has been 
analyzed in other areas, such as electric power systems, 
nuclear-power-plant alarm management (Rellano, et 
al., 1991), and patient-care monitoring. 
5.1 Alarm Correlation 
The alarms are mediated by alarm messages about 
faults. A fault is an illegal state in rural land use or a 
disorder occurring in the hardware or software of the 
managed rural land system. Faults happen within the 
managed system components or rural land use, while 
alarms are external manifestations of faults. Alarms 
defined by designers and generated by managed rural 
Intemational Archives of Photogrammetry and Remote Sensing. Vol. XXXII, Part 7, Budapest, 1998 
lands and the monitoring system equipment are 
observable by managers of rural land use. 
There may be multiple alarms occurred at the same 
time. Alarm correlation techniques can be used in the 
situation. Alarm correlation is a conceptual 
interpretation of multiple alarms such that new 
meanings are assigned to these alarms. It is a generic 
process that underlies different management tasks of 
rural land use. Some typical operations relevant to 
alarm correlation are as the followings. 
(1)Compression: the reduction of multiple occurrences 
of an alarm into a single alarm. 
(2)Count: the substitution of a specified number of 
occurrences of alarms with a new alarm. 
(3)Suppression: inhibiting a low-priority alarm in the 
presence of a higher-priority alarm. 
(4)Boolean: substitution of a set of alarms satisfying a 
Boolean pattern with a new alarm. 
(5)Generalization: reference to an alarm by its 
superclass. Alarm correlation may be used for fault 
isolation and diagnosis, selecting corrective actions, 
proactive maintenance, and trend analysis in the 
monitoring system. 
5.2 Conceptual Framework 
One of the major applications of alarm correlation is 
the fault diagnosis in the monitoring system. Not all 
faults exhibit alarms. These faults can be recognized 
indirectly by correlating available alarms. Correlation 
between alarms due to a common fault is an 
equivalence relation. 
Alarm generalization is potentially very useful for 
management of rural land use. It allows one to deviate 
from a microscopic perspecjive of management events 
of rural land use and view situations from a higher 
level. There are two ways for alarm generalization. The 
first is subsumption of lower-level alarm classes by a 
higher-level class. This generalization process may 
utilize alarm class/subclass hierarchies. The second is 
interpretation of simultaneous events or events 
happening within a defined time interval as a 
qualitatively new complex situation. 
The conceptual framework of alarm correlation 
contains the structural and behavioral components. The 
relational approach to monitoring complex system 
(Snodgrass, 1988) can be used for design and 
implementation of the conceptual framework. The 
structural component is the description of the managed 
rural land system. It contains two major parts, the rural 
land system configuration model and the element class 
hierarchy of rural land system. The configuration 
model describes the managed objects, and the 
connectivity and containment relations between them. 
The element class hierarchy describes the managed 
193