曝光臺 注意防騙 網曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費者

track pulse code estimate for the position is the same as
the code detected in that position in the second reply
provided that was not also a "low" confidence ONE. If
the second was also a "low" confidence ONE, the initial
track pulse code estimate is the same as the code
detected in that position in the first reply.
3.1.2.9 MODE A/C SWRVEILWCE EXTENSION
3.1.2.9.1 General. The equipment should continue to pass
position reports for a target to the collision avoidance
algorithms only if:
a) the track has not been identified as an image (see
3.1.2.9.6); and
b) the reply altitudes occur within an altitude window of
k60 m (200 ft) centred on the altitude predicted from
previous reply history; and
C) all replies used for threat assessment after the initiation
procedure occur within a range window centred on the
range predicted from previous reply history.
Annex 1 U - Aeronautical Telecommunications Volume ZV
3.1.2.9.2 Range correkiion. The folluwir~gi s an example
of an acceptable set of rules for determining the size of the
range window:
a) The tracks are processed individually in increasing range
order with input rangc precision of at least 15 m (50 ft)
and retained computational accuracy of at least 1.8 m
(6 ft), Range is estimated and predicted by a recursive
(alpha-beta) tracker with alpha of 0.67 and beta of 0.25.
b) After each surveillance update a new range
measurement is available for each target. Since the
measurement includes errors, it must be smoothed based
on previous measurements to obtain improved estimates
of the current target position and velocity. The range and
range rate estimation equations are as follows: .
r(t) estimate = r(t) prediction +
[alpha x (r(t) measurement - r(t) prediction)]
;(t ) estimate = ;(t - T,) estimate + [(beta/Tp) x
(r(t) measurement - r(t) prediction)],
where T, is the time difference between the current and
previous measurements.
c) The gains, alpha and beta determine the relative degree
of reliance on current and previous measurements; gains
of unity would place complete reliance on the current
measure~nent and result in no smoothing.
d) The estimates obtained from the above equations are
subsequently used to predict the range at the time of the
next measurement as follows:
r(t + T,) prediction = r(t) estimate + [; (t) estimate x T,]
where T, is the time difference between the next
measurement and the current measurement.
e) The range correlation window is centred at the predicted
range and has a half-window width as follows:
760 ft if coasted
last interval
If track Is not establlshed:
0
if track Is establlshed:
570 ft a updated + I 2 000 It, if 0.00 NM I r < 0.17 NM
last interval 1 OM] R, if 0.17 NM < r <0.33 NM
600 ft, if 0.33 NM 5 r s 1.00 NM
3.1.2.9.3 Altitude correlarion. For the purposes of altitude
correlation, altitude is estimated and predicted by an alphabeta
tracker with alpha of 0.28 and beta of 0.06. The tracker
has retained computational accuracy of (30 m) (100 ft) divided
by 16. The altitude prediction is rounded to the nearest 30 rn
(100 t't) increment and converted to grey code. The grey codes
of the predicted altitude k30 m (100 ft) are also computed. The
longer-term altitude predictions performed by the threat
detection logic require a more accurate aMtude tracking
procedure (see 3.5.3). The reply(ies) that lies in the range
correlation window is tested for altitude correlation in
increasing range order. The track is updated with the first reply
that has exact agreement (in all bits) with any of the three grey
codes computed above. If no reply matches, two additional
grey codes are computed and the process tried again. The two
codes are the predicted altitude *60 m (200 ft).
3.1.2.9.4 Track updating - establishment. The updating
reply (if any) is eliminated from further consideration in
updating other tracks, or in the track initiation process. If there
is no updating reply, the range and altitude estimates are set
equal to the corresponding predicted values. If this is the sixth
consecutive interval having no updating reply, the track is
dropped. If there is an updating reply, and if the track is not
identified as an image (see 3.1.2.9.6), the track is flagged as
established, that is, it is now available for use by the threat
detectiori logic. Once established, a track remains established
until it is dropped, even if it subsequently satisfies the
conditions for an image track.
3.1.2.9.5 Test for track splits. When a\l tracks have been
　
中國航空網 www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：附件10--航空電信an10_v4_3ed(98)