any nation where there are 100 ISM machines all operating
cofrequency with a given television station.
The probability of television interference, as used here, is
the likelihood that one television receiver will receive
enough noise from one ISM machine to have a signal-tonoise
ratio less than 50 dB. To calculate this probability,
the S/N is calculated for the interaction of every television
receiver with every ISM machine. The number of times
this interaction results in S/N < 50 dB is divided by the
total number of interactions (170 000 or 400 000, depending
on the number of TV receivers in the series) to arrive
at the probability of interference.
For those cases where more than one ISM machine is assumed
operating simultaneously, the noise received by a
television receiver from each simultaneously operating
ISM machine is summed as the root sum square. This is
treated as a single primary interaction and the number of
times this primary interaction resulted in S/N < 50 dB is
divided by the total number of primary interactions
(170 000 or 400 000) to arrive at the probability of interference.
This model was written in BASIC for the IBM-PC. Parts
were compiled using the BASIC compiler by Microsoft.
The model is "table driven." That is, significant parameters
are put into a table, which is a file on a diskette, and can
be changed easily. During this study, the following parameters
were fixed:
1. A 64 km diameter (3217 km2 or 1257 mi2) coverage
area for a television transmitter located in the center
of a 64 km square area. (See Figure 1.)
2. 1 700 (4 000 in a second series) television receivers
randomly located in the 64 km diameter circle.
3. Ten ISM machines, randomly located in each of nine
"Industrial Zones" (500 m x 750 m rectangles) distributed
randomly in the 64 km square (4096 km2 or 1600
mi2) area concentric with the television service area.
No industrial zone is permitted within 2 km of another
industrial zone.
4. Ten additional individual ISM machines located randomly
throughout the 64 km square area.