U.S. patent application number 14/529446 was filed with the patent office on 2015-02-12 for wireless communication system.
The applicant listed for this patent is Tyco Fire & Security GmbH. Invention is credited to David Harif, Mordechai Mushkin, Miri Ratner, Shimon Zigdon.
Application Number | 20150043569 14/529446 |
Document ID | / |
Family ID | 39944095 |
Filed Date | 2015-02-12 |
United States Patent
Application |
20150043569 |
Kind Code |
A1 |
Mushkin; Mordechai ; et
al. |
February 12, 2015 |
WIRELESS COMMUNICATION SYSTEM
Abstract
A method for communication includes transmitting a first uplink
message from a first remote node (200, 300, 400) to a central node
(100) in a wireless communication system according to a first
frequency hopping scheme, and transmitting a second uplink message
from a second remote node to the central node in the wireless
communication system according to a second frequency hopping
scheme, different from the first scheme. Both the first and the
second uplink messages are received and processed at the central
node.
Inventors: |
Mushkin; Mordechai; (Nirit,
IL) ; Zigdon; Shimon; (Netanya, IL) ; Ratner;
Miri; (Ramat Gan, IL) ; Harif; David; (Ramat
Gan, IL) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Tyco Fire & Security GmbH |
Neuhausen am Rheinfall |
|
CH |
|
|
Family ID: |
39944095 |
Appl. No.: |
14/529446 |
Filed: |
October 31, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12597945 |
Feb 19, 2010 |
8902933 |
|
|
PCT/IL2008/000581 |
Apr 30, 2008 |
|
|
|
14529446 |
|
|
|
|
60927506 |
May 2, 2007 |
|
|
|
Current U.S.
Class: |
370/350 |
Current CPC
Class: |
H04B 1/7136 20130101;
H04B 1/71635 20130101; H04B 1/713 20130101; H04W 56/001 20130101;
H04B 7/0808 20130101; H04L 5/14 20130101; H04B 1/7183 20130101;
H04W 72/085 20130101 |
Class at
Publication: |
370/350 |
International
Class: |
H04W 56/00 20060101
H04W056/00 |
Claims
1. A method for communication, comprising: transmitting periodic
status messages from a remote node to a central node in a wireless
communication system; in response to the periodic status messages,
transmitting acknowledgment messages from the central node to the
remote node, the acknowledgement messages comprising time-stamps;
and synchronizing the remote node with the central node using the
time-stamps.
2. The method according to claim 1, wherein synchronizing the
remote node comprises aligning a value of a local clock at the
remote node with a system clock maintained by the central node.
3. The method according to claim 2, wherein synchronizing the
remote node comprises aligning a frequency of the local clock with
an actual frequency of the system clock using the time-stamps.
4. The method according to claim 3, wherein the periodic status
messages are transmitted with a fixed interval between the status
messages, and wherein adjusting the frequency comprises computing a
frequency adjustment using the time-stamps and the fixed interval
without performing a division operation.
5. A wireless communication system, comprising: a remote node,
which is configured to transmit periodic status messages; and a
central node, which is configured to transmit acknowledgment
messages to the remote node in response to the periodic status
messages, the acknowledgement messages comprising time-stamps,
wherein the remote node is configured to synchronize with the
central node using the time-stamps.
6. The system according to claim 5, wherein the remote node
comprises a local clock, and is configured to align a value of the
local clock with a system clock maintained by the central node
using the time-stamps.
7. The system according to claim 6, wherein the remote node is
configured to align a frequency of the local clock with an actual
frequency of the system clock using the time-stamps.
8. The method according to claim 7, wherein the periodic status
messages are transmitted with a fixed interval between the status
messages, and wherein the remote node is configured to compute a
frequency adjustment using the time-stamps and the fixed interval
without performing a division operation.
Description
CROSS REFERENCE TO RELATED APPLICATIONS
[0001] This application is a Division of U.S. application Ser. No.
12/597,945, filed on Apr. 30, 2008, which is a National Stage
Application of PCT/IL2008/000581, filed on Apr. 30, 2008, which
claims priority to U.S. Provisional Application 60/927,506, filed
on May 2, 2007, all of which are incorporated herein by reference
in their entirety.
[0002] This application is related to U.S. application Ser. No.
______, filed on an even date herewith with attorney docket number
0309.0001US2, now U.S. Patent Publication No. ______, U.S.
application Ser. No. ______, filed on an even date herewith with
attorney docket number 0309.0001US4, now U.S. Patent Publication
No. ______, and U.S. application Ser. No. ______, filed on an even
date herewith with attorney docket number 0309.0001US5, now U.S.
Patent Publication No. ______.
FIELD AND BACKGROUND OF THE INVENTION
[0003] The present invention relates to wireless communication
systems. Some wireless communication systems, such as event
reporting systems, use synchronized, time-slotted,
frequency-hopping schemes and switched antenna diversity
mechanisms
SUMMARY OF THE INVENTION
[0004] There is provided, in accordance with an embodiment of the
present invention, a method for communication, including: [0005]
transmitting a first uplink message from a first remote node to a
central node in a wireless communication system according to a
first frequency hopping scheme; [0006] transmitting a second uplink
message from a second remote node to the central node in the
wireless communication system according to a second frequency
hopping scheme, different from the first scheme; and [0007]
receiving and processing both the first and the second uplink
messages at the central node.
[0008] In a disclosed embodiment, the second frequency hopping
scheme is a non-synchronized frequency hopping scheme, and
receiving the second uplink message includes scanning a receiver of
the central node over the frequencies used in the second frequency
hopping scheme in order to detect the second uplink message. The
second frequency hopping scheme may be used by remote nodes in the
wireless communication system that are not synchronized with the
central node.
[0009] In some embodiments, the first frequency hopping scheme is a
synchronized, time-slotted, frequency hopping scheme. The method
may include synchronizing the first remote node with the central
node prior to transmission of the first uplink message. A second
remote node may transmit the second uplink message using the second
frequency hopping scheme in order to join the wireless
communication system and become synchronized with the central
node.
[0010] Additionally or alternatively, the second remote node
transmits the second uplink message using the second frequency
hopping scheme in order to re-synchronize with the central node
after having lost synchronization or in order to deliver an
operational message to the central node after having lost
synchronization. The second remote node may transmit the second
uplink message using the second frequency hopping scheme in order
to re-synchronize with the central node after having lost
synchronization in addition to delivering the operational
message.
[0011] Optionally, the second frequency hopping scheme is used by
remote nodes in the wireless communication system that include a
one-way radio transmitter for delivering operational messages.
[0012] Typically, receiving and processing both the first and the
second uplink messages includes implementing both the synchronized
and the non-synchronized frequency hopping schemes together in a
receiver of the central node. In some embodiments, implementing
both the schemes includes defining a time-slot that includes a
system frequency window (SFW) for receiving the first uplink
message and a scanning window (SCW) for receiving the second uplink
message. Receiving the first uplink message may include tuning the
receiver of the central node during the SFW to a current frequency
value of a system frequency-hopping function, and upon receiving a
valid preamble, remaining tuned to the current frequency value
until the entire first uplink message has been received.
Additionally or alternatively, receiving the second uplink message
includes performing a fast frequency scan during the SCW, and upon
detecting a valid preamble at a given frequency, remaining tuned to
the given frequency until the entire second uplink message has been
received.
[0013] In some embodiments the method may also include defining an
antenna switching function of the central node, the function
specifying respective time slots in which each of a plurality of
antennas of the central node is to receive signals, [0014] wherein
transmitting the first uplink message includes selecting, at the
first remote node, an antenna of the central node as a favored
antenna for receiving transmissions from the first remote node at
the central node, and transmitting the first uplink message in a
time-slot selected responsively to the antenna switching function
so that the message will be received by the favored antenna at the
central node.
[0015] Additionally or alternatively, the method may include
selecting, for each first antenna of the central node, a respective
favored second antenna among two or more second antennas of the
first remote node for transmitting signals that will be received at
the central node via the first antenna, and transmitting the first
uplink message in a specified time slot via the favored second
antenna with respect to the first antenna specified by the antenna
switching function for the specified time slot.
[0016] In some embodiments, transmitting the first uplink message
includes forwarding an uplink message received by the first remote
node which is configured to operate as a repeater.
[0017] In a disclosed embodiment, receiving and processing the
uplink messages includes monitoring events detected by the remote
nodes, and the method includes issuing an alarm in response to one
or more of the detected events.
[0018] There is also provided, in accordance with an embodiment of
the present invention, a wireless communication system, including:
[0019] a plurality of remote nodes, which are configured to operate
in accordance with first and second different frequency hopping
schemes; and [0020] a central node, which is configured to receive
and process both a first uplink message transmitted by a first
remote node in the wireless communication system according to the
first frequency hopping scheme and a second uplink message
transmitted by a second remote node in the wireless communication
system according to the second frequency hopping scheme.
[0021] There is additionally provided, in accordance with an
embodiment of the present invention, apparatus for wireless
communication, including: [0022] a radio transceiver, which is
configured to transmit uplink messages to a central node in a
wireless communication system; and [0023] a processor, which is
coupled to drive the radio transceiver to transmit a first uplink
message to the central node according to a first frequency hopping
scheme and to transmit a second uplink message to the central node
according to a second frequency hopping scheme, which is different
from the first frequency hopping scheme.
[0024] There is further provided, in accordance with an embodiment
of the present invention, apparatus for wireless communication,
including: [0025] a radio transceiver, which is configured to
receive uplink messages transmitted by remote nodes in a wireless
communication system in accordance with first and second different
frequency hopping schemes; and [0026] a processor, which is coupled
to process both a first uplink message transmitted by a first
remote node in the wireless communication system according to the
first frequency hopping scheme and a second uplink message
transmitted by a second remote node in the wireless communication
system according to the second frequency hopping scheme.
[0027] There is moreover provided, in accordance with an embodiment
of the present invention, a method for communication, including:
[0028] transmitting periodic status messages from a remote node to
a central node in a wireless communication system; [0029] in
response to the periodic status messages, transmitting
acknowledgment messages from the central node to the remote node,
the acknowledgement messages including time-stamps; and [0030]
synchronizing the remote node with the central node using the
time-stamps.
[0031] In a disclosed embodiment, synchronizing the remote node
includes aligning a value of a local clock at the remote node with
a system clock maintained by the central node. Additionally or
alternatively, synchronizing the remote node includes aligning a
frequency of the local clock with an actual frequency of the system
clock using the time-stamps. In one embodiment, the periodic status
messages are transmitted with a fixed interval between the status
messages, and adjusting the frequency includes computing a
frequency adjustment using the time-stamps and the fixed interval
without performing a division operation.
[0032] There is additionally provided, in accordance with an
embodiment of the present invention, a wireless communication
system, including: [0033] a remote node, which is configured to
transmit periodic status messages; and [0034] a central node, which
is configured to transmit acknowledgment messages to the remote
node in response to the periodic status messages, the
acknowledgement messages including time-stamps, [0035] wherein the
remote node is configured to synchronize with the central node
using the time-stamps.
[0036] There is furthermore provided, in accordance with an
embodiment of the present invention, a method for communication,
including: [0037] transmitting downlink messages in a
time-division-duplexing (TDD) wireless communication system using a
first frequency hopping function; and [0038] transmitting uplink
messages in the TDD wireless communication system using a second
frequency hopping function, which is synchronized with but
different from the first frequency hopping function.
[0039] The first and second frequency hopping functions may be
mutually orthogonal or mutually pseudo-orthogonal.
[0040] Typically, transmitting the downlink messages includes
transmitting the downlink messages from a central node in the
wireless communication system during predetermined downlink
windows, and the method includes receiving the downlink messages at
receivers of remote nodes in the wireless communication system,
wherein the remote nodes actuate the receivers only during the
downlink windows. In a disclosed embodiment, when the central node
does not have any outstanding downlink messages for transmission,
the uplink messages are received at the central node during the
downlink windows. Receiving the downlink messages at the remote
nodes may have priority over transmitting the uplink messages.
[0041] In one embodiment, transmitting the downlink messages
includes transmitting at least some of the downlink messages from
the central node to one or more repeaters in the wireless
communication system in a manner identical to transmitting the
downlink messages from the central node to the remote nodes.
Additionally or alternatively, transmitting the downlink messages
includes transmitting at least some of the downlink messages from
the central node to one or more repeaters in the wireless
communication system in a manner identical to transmitting of the
uplink messages from the remote nodes to the central node, and the
method may include receiving the at least some of the downlink
messages from the central node at the repeater in a manner
identical to receiving the downlink messages from the central node
at a remote node. Alternatively, the method includes receiving the
at least some of the downlink messages from the central node at the
repeater in a manner identical to receiving the uplink messages
from the remote nodes at the repeater.
[0042] There is also provided, in accordance with an embodiment of
the present invention, a wireless communication system, including:
[0043] a central node, which is configured to transmit downlink
messages in a time-division-duplexing (TDD) wireless communication
scheme using a first frequency hopping function; and [0044] one or
more remote nodes, which are configured to receive the downlink
messages and to transmit uplink messages to the central node using
a second frequency hopping function, which is synchronized with but
different from the first frequency hopping function.
[0045] There is also provided, in accordance with an embodiment of
the present invention, a method for communication, including:
[0046] defining an antenna switching function of a first node in a
wireless communication system, the function specifying respective
time slots in which each of a plurality of antennas of the first
node is to receive signals; [0047] selecting, at a second node in
the wireless communication system, an antenna of the first node as
a favored antenna for receiving transmissions from the second node;
and [0048] transmitting a message from the second node to the first
node in a time slot selected responsively to the antenna switching
function so that the message will be received by the favored
antenna at the first node.
[0049] Typically, selecting the antenna includes evaluating
respective scores of the antennas of the first node.
[0050] In some embodiments, evaluating the respective scores
includes evaluating a history of successful receptions via each of
the antennas.
[0051] Additionally or alternatively, evaluating the respective
scores includes evaluating a quality of a respective propagation
channel between the second node and each of the antennas.
[0052] In some embodiments, evaluating the quality of the
respective propagation channel includes evaluating signal quality
parameters measured by the second node when receiving an
acknowledgment transmitted via each of the antennas.
[0053] Additionally or alternatively, evaluating the quality of the
respective propagation channel includes measuring signal quality
parameters at the first node when receiving transmissions from the
second node, and incorporating a value of the measured signal
quality parameters within the ACK reply for use in computing the
respective scores at the second node.
[0054] Commonly, the signal quality parameters evaluated at the
first node and/or the second node include a received signal
level.
[0055] There is also provided, in accordance with an embodiment of
the present invention, a method for communication, including:
[0056] defining an antenna switching function of a first node in a
wireless communication system, the first node having a plurality of
first antennas, the function specifying respective time slots in
which each of the first antennas is to receive signals; [0057]
selecting, for each first antenna among the plurality of the first
antennas, a respective favored second antenna among two or more
second antennas of a second node in the wireless communication
system for transmitting signals that will be received via the first
antenna; and [0058] transmitting a message from the second node to
the first node in a specified time slot via the favored second
antenna with respect to the first antenna specified by the antenna
switching function for the specified time slot.
[0059] In some embodiments, transmitting the message includes
forwarding an uplink message.
[0060] There is also provided, in accordance with an embodiment of
the present invention, a wireless communication system, including:
[0061] a first node, including a plurality of antennas and having a
predefined antenna switching function, which specifies respective
time slots in which each of the plurality of the antennas is to
receive signals; and [0062] a second node, which is configured to
select one of the antennas of the first node as a favored antenna
for receiving transmissions from the second node, and to transmit a
message to the first node in a time slot selected responsively to
the antenna switching function so that the message will be received
by the favored antenna at the first node.
[0063] There is also provided, in accordance with an embodiment of
the present invention, a method for communication, including:
[0064] a first node, including a plurality of first antennas and
having a predefined antenna switching function, which specifies
respective time slots in which each of the first antennas is to
receive signals; and [0065] a second node, which includes two or
more second antennas, and which is configured to select, for each
first antenna among the plurality of the first antennas, a
respective favored second antenna among the two or more second
antennas for transmitting signals that will be received via the
first antenna, and to transmit a message to the first node in a
specified time slot via the favored second antenna with respect to
the first antenna specified by the antenna switching function for
the specified time slot.
[0066] In some embodiments, the second node includes a repeater,
which is configured to forward the message from a remote node.
BRIEF DESCRIPTION OF THE DRAWINGS
[0067] The present invention will be more fully understood from the
following detailed description of the embodiments thereof, taken
together with the drawings in which:
[0068] FIG. 1 presents a block diagram of a wireless event
monitoring system, in accordance with an embodiment of the present
invention;
[0069] FIG. 2 presents an example of a time-slotted scheme, in
accordance with an embodiment of the present invention;
[0070] FIGS. 3(a) and 3(b) are timing diagrams that illustrate a
power-save scheme, in accordance with an embodiment of the present
invention;
[0071] FIG. 4 presents a block diagram of a node in a wireless
communication system, in accordance with an embodiment of the
present invention;
[0072] FIG. 5 is a timing diagram that illustrate active
synchronization with receiver fast frequency scan, in accordance
with an embodiment of the present invention; and
[0073] FIGS. 6(a), 6(b) and 6(c) are timing diagrams that
illustrates concurrently employment of access schemes, in
accordance with an embodiment of the present invention.
[0074] FIG. 7 is a timing diagram that illustrates an intra-message
switched antenna diversity scheme, in accordance with an embodiment
of the present invention;
[0075] FIG. 8 a timing diagram that presents an example of an
antenna switching function, in accordance with an embodiment of the
present invention;
DETAILED DESCRIPTION OF EMBODIMENTS
[0076] The embodiments that are described herein below relate to
wireless communication systems, and more specifically to
synchronized, time-slotted, frequency-hopping wireless
communication systems. Such systems commonly require methods for
initial synchronization, maintenance of synchronization,
re-synchronization in cases of synchronization loss, high
reliability of message delivery, low latency of message delivery,
message delivery during periods of synchronization loss, antenna
diversity, low current consumption of battery-powered nodes, low
cost and simple implementation. An example of a synchronized,
time-slotted, frequency-hopping wireless communication system in
which these embodiments could be applied is a wireless event
reporting system, which serves in the description that follows as a
platform for describing the features of these embodiments and the
manner in which they meet the above requirements. This event
reporting system is described solely by way of example, and the
principles of the present invention may similarly be applied in
wireless communication systems of other types.
[0077] FIG. 1 presents a block diagram of a wireless event
reporting system, for example a wireless alarm system, in
accordance with an embodiment of the present invention. The system
comprises a central unit 100 and one or more distributed monitoring
devices 200. The system might also comprise one or more signaling
devices 300, one or more human interface devices 400 and one or
more repeaters 500. Throughout the present document, the components
of the system are also referred to as nodes, the central unit and
the repeaters are referred to as central nodes and all other nodes
are referred to as remote nodes.
[0078] The monitoring devices detect events of interest and report
them to the central unit. The monitoring devices might, for
example, be any of the various detectors of an alarm system, such
as motion detectors, glass break detectors, magnetic contact
detectors, smoke/fire detects, gas detectors, flood detects, panic
buttons, human health monitors and similar devices.
[0079] The signaling devices produce signals according to commands
received from the central unit. For example, a signaling device
might be an audio signaling device such as a siren or a visual
signaling device such as a warning light. The signaling device
might also be a communication device for sending messages from the
event monitoring system to other systems, for example a device that
sends messages to a remote location over the telephone line.
[0080] The human interface devices provide the system user with a
remote interface to the central unit. For example, a human
interface device might be a remote keypad.
[0081] Communication between the nodes of the wireless event
monitoring system is usually based on a TDD (time division
duplexing) scheme, meaning that at a given moment the radio
transceiver is able either to transmit or to receive. The TDD
scheme is selected since it is more economical than the alternative
FDD (frequency division duplexing) scheme, but the methods
described herein below may alternatively be adapted for use in FDD
systems.
[0082] Communication between the nodes of the wireless event
monitoring system comprises uplink messages, transmitted from the
remote nodes to the control unit, directly or via one or more
repeaters, and downlink messages, transmitted from the control unit
to the remote nodes, directly or via one or more repeaters. Both
types of messages are usually relatively short. For example, the
length of a typical message transmitted in an alarm system is
typically on the order of ten to fifty bytes of data.
[0083] Uplink traffic comprises event-reporting messages and might
also comprise periodic status messages. An event-reporting message
is sent from a monitoring device to report an event detected by the
device. Usually, event-reporting messages are infrequent. For
example, in an alarm system the average rate of event reporting
messages is typically on the order of ten to fifty messages per
day. On the other hand, those messages need to be delivered with
high reliability and at short latency. For example, typical
requirements in the context of an alarm system are probability of
missed events less than about 10.sup.-5 and latency on the order of
about one or two seconds.
[0084] Status messages are sent periodically from a remote node to
indicate the status of the sending node. Periodic reception of
status messages from a remote node also provides an indication of
the quality of the wireless link from the remote node to the
control unit or the repeater. For example, the European standard EN
50131-5-3 for wireless alarm systems specifies that in a grade 3
system the status RF link should be monitored at least once per 100
seconds, which calls for periodic status messages at intervals of
less than 100 seconds between successive messages.
[0085] Downlink traffic comprises messages transmitted from the
control unit to the signaling devices and the human interfaces
devices, and might also comprise messages transmitted from the
control unit to the monitoring devices. Messages to the signaling
devices are usually infrequent and need to be delivered with high
reliability and short latency. Typical requirements in the context
of alarm systems are similar to the requirements for event
reporting messages.
[0086] Communication between the control unit and a given remote
node might be either direct, or via a repeater or a sequence of
repeaters, depending on system deployment and on the quality of the
wireless links between the remote node and the central unit and the
between the remote node and the repeaters.
[0087] The central nodes of an event monitoring system are
typically mains powered, while the remote nodes are typically
battery powered. Battery size and battery life-time are usually
important factors in wireless event monitoring systems, and
therefore one of the important considerations in devising a
communication scheme for a wireless event monitoring system is to
minimize the current consumption of the remote nodes.
[0088] Since a primary function of the distributed monitoring
devices is to send report messages, those devices might in
principle be fitted with one-way radio transmitters rather than
with two-way radio transceivers. Throughout the present document,
devices fitted with one-way radio transmitters are referred to as
one-way nodes and devices fitted with two-way radio transceivers
are referred to as two-way nodes. Although one-way nodes might be
more economical, two-way nodes enable much better performances in
various aspects, some of which are described below. An embodiment
of the present invention that is described hereinbelow relates to
systems in which all the remote nodes are two-way nodes. The
principles of the present invention, however, may also be applied
in systems in which some of the monitoring devices are one-way
nodes.
[0089] One of the advantages of fitting the remote nodes of the
wireless event monitoring system with a two-way radio transceiver
is the ability to utilize an automatic repeat request (ARQ)
mechanism for the uplink traffic. According to the ARQ mechanism,
when an uplink message transmitted by a given remote node is
successfully received by a central node, the central node replies
with an acknowledgement (ACK) message. If the remote node does not
receive an ACK, the remote node retransmits the message one or more
times, until an ACK is received, or until some limit for the number
of retransmissions is reached. An alternative mechanism, in case
ARQ cannot be utilized, is blind repetition, wherein every message
is transmitted several times. The advantages of ARQ over blind
repetition are increased reliability and reduced current
consumption. Another advantage of ARQ over blind repetition is the
lower occupancy of the wireless media, which implies lower
probability of collision.
[0090] Another advantage of two-way communication is the ability to
maintain synchronization. Synchronization means that every node in
the wireless network is fitted with a local clock. The local clock
of the central unit is referred to as the system clock, and all
other nodes keep their local clocks synchronized with the system
clock. The mechanism of synchronization is based on time stamps
transmitted by the central nodes and received by the other nodes.
Although wireless event monitoring system might, in principle, be
non-synchronized, synchronization enables much better performances
in various aspects, some of which are described below.
[0091] One of the advantages of synchronizing a wireless event
monitoring system is the ability to employ a time-slotted access
scheme for the uplink traffic. According to the time-slotted access
scheme, the time axis is divided into time-slots, and the duration
of a time-slot is sufficient to accommodate a typical uplink
message following by an ACK reply. The advantage of the slotted
access scheme is reduced probability of collision, which implies
higher reliability, lower latency and lower current
consumption.
[0092] An example of a time-slotted scheme for the uplink of a
wireless alarm system is presented in FIG. 2. According to this
example, the time access is divided into super-frames of fixed
duration, for example 100 seconds, each super frame is divided into
frames of fixed duration, for example one second, and each frame is
divided into time-slots of fixed duration, for example 40 ms. Each
remote node in the system is assigned a respective frame, which
restricts the number of remote nodes in this specific example to
100. The first time-slot, or the first few time-slots, in each
given frame, are dedicated for transmission of uplink periodic
status messages from the remote node associated with the given
frame, and the rest of the time-slots within each frame are used
for transmission of uplink random access messages, such as event
reporting messages. This scheme eliminates potential collision of a
given periodic status message with other periodic status messages,
because each node transmits the periodic status messages in a
different frame. This scheme also eliminates potential collision of
a given periodic status message with random access messages,
because random access messages are not transmitted in the first
time-slot, or time-slots, of the frame that are dedicated for
periodic status messages. The probability of collision between
random access messages is low, due to the low frequency of those
events, and in case such collision does occur, it is recovered by
the ARQ mechanism, with random back-off.
[0093] Another advantage of synchronizing a wireless event
monitoring system is the ability to employ a power save scheme for
the downlink. According to the power save scheme, the receiver of
the remote node spends most of the time in sleep mode, also
referred to as stand-by mode or power-save mode. When in sleep
mode, most of the functions of the receiver are disabled, thus
keeping the current consumption as low as possible. The local clock
of the remote node is retained operative in sleep mode too, in
order to maintain synchronization with the system. The local clock
is also utilized to wake up the receiver for the downlink windows,
as described below.
[0094] According to the power save scheme, downlink messages are
transmitted at determined points in time, referred to as downlink
windows. It is advantageous, although not necessary, to align the
downlink window structure with the uplink time-slot structure. A
power-save scheme, aligned with the uplink time-slot structure, is
presented in FIGS. 3(a) and 3(b), wherein FIG. 3(a) illustrates the
case of no downlink transmission and FIG. 3(b) illustrates the case
of a downlink transmission. The downlink window, denoted in the
Figures as DW, is located at the beginning of the time-slot.
[0095] According to the power save scheme, the receiver of the
remote node is woken up at the beginning of the downlink window and
is kept open for a short time-period, referred to as the snapshot
window, trying to detect a valid preamble transmitted by a central
node. The snapshot window is denoted in the Figures as SNPW. If a
valid preamble is detected during the snapshot window, the receiver
remains open, detects the rest of the message and then returns to
sleep mode, as illustrated in FIG. 3(b). The preamble is denoted in
the Figure as PR. Otherwise, the receiver returns to sleep mode
after the snapshot window, as illustrated in FIG. 3(a).
[0096] The interval between downlink windows might depend on the
node type. For example, a typical arrangement for a wireless alarm
system would be to schedule downlink windows for the signaling
devices on the order of once per second, while scheduling downlink
windows for monitoring devices on the order of once per minute.
[0097] An advantage of the power-save mode for a wireless event
monitoring system is the great reduction in current consumption. A
limitation of the power save mode is that a downlink message can be
repeated, either according to an ARQ scheme, or according to a
blind repetition scheme, only at the next downlink window.
Therefore, when power-save mode is employed, it is desirable to
protect the downlink messages, as much as possible, from being
interfered with by other messages. A method for providing such
protection is described further hereinbelow.
[0098] A block diagram of a node in a wireless communication
system, for example a wireless event monitoring system, is
presented in FIG. 4. The node comprises one or more antennas 1000,
a radio transceiver 2000, a processor 3000, a clock source 4000 and
a power supply 5000. In case the node comprises more than one
antenna, the node usually comprises also an antenna selecting
switch 6000. The node might comprise other functions 7000 too,
depending on the type of the node. For example monitoring devices
200 comprise sensing functions, signaling devices 300 comprise
signaling functions, and human interfacing devices 400 comprise
human interfacing functions.
[0099] In wireless alarm systems the central nodes are typically
fitted with two antennas and employ receive antenna diversity on
the uplink and transmit antenna diversity on the downlink. The
monitoring devices, which are typically much lower-cost devices,
may be fitted with one antenna only.
[0100] The simplified functional blocks shown in FIG. 4 do not
necessarily reflect the actual components that are used in
constructing the node. For example, the implementation of the radio
transceiver might require auxiliary components such as a low noise
amplifier (LNA), power amplifier, frequency source and SAW filter.
The processing function, as another example, can be implemented by
a single processor or by several processors, for example, one
processor to handle the communication related tasks and another
processor(s) to handle other function(s). Alternatively, the radio
transceiver and the processing function may be integrated into one
component, such as a Texas Instruments CC1110 System-on-Chip (SoC)
device. The clock source, for example, can be implemented as an
internal function of the processor or as a separate component that
is external to the processor.
[0101] Wireless event monitoring systems commonly operate over
unlicensed frequency bands. Some examples of unlicensed bands are
the 433.29 to 434.79 MHz band, the 868 to 870 MHz band, the 902 to
928 MHz band, the 2.4 to 2.5 GHz band and the 5.725 to 5.875 GHz
band. For wireless communication systems that operate in unlicensed
frequency bands, it is advantageous to utilize a spread-spectrum
scheme, such as frequency-hopping or direct sequence, in order to
reduce the probability of co-channel interference with other
systems operating in the same band. In some regulatory domains,
spread-spectrum is mandatory for this very reason. An embodiment of
the present invention addresses systems employing frequency-hopping
schemes, and more specifically, systems employing slow
frequency-hopping scheme. The term "slow frequency hopping" means
that the frequency-hopping rate is substantially lower than the
communication symbol rate. In the description below, the term
"frequency hopping" always refers to slow frequency hopping.
[0102] For a synchronized wireless communication system that
employs frequency hopping, it is advantageous to utilize a
synchronized frequency-hopping scheme. Synchronized
frequency-hopping means that the radio channel used for
communication at a given moment in time is a deterministic function
of the system clock, and thus it is known to all synchronized nodes
in the system. This function is referred to in the present document
as the system frequency-hopping function. The range of the system
frequency-hopping function might be the entire set of radio
channels available for the system, or a subset of that set. The
advantage of a synchronized frequency-hopping scheme is that the
receiver knows in advance on which radio channel a potential
message might be expected, and tunes to this radio channel.
[0103] An alternative to a synchronized frequency hopping is
non-synchronized frequency hopping, wherein the frequency-hopping
function used by a potential transmitter is not known to the
receiver, for example since they do not share the same clock. An
inherent difficulty with non-synchronized frequency hopping is that
the receiver has no prior knowledge of the radio channel used by
the transmitter. One method to solve this difficulty is by receiver
fast frequency scan. According to this method, the transmitted
message begins with a long preamble, and the receiver continuously
scans all the potential radio channels. At each radio channel, the
receiver stays for a time-period sufficient to detect the existence
of a valid preamble. If a valid preamble is detected on a given
radio channel, the receiver stays at that channel and detects the
rest of the message. Otherwise the receive switches to the next
radio channel.
[0104] One disadvantage of non-synchronized frequency hopping,
compared to synchronized frequency hopping, is that the transmitted
messages are necessarily longer, due to the long preamble, and the
current consumption of the transmitting nodes is therefore higher.
Another disadvantage is that the longer messages imply higher
occupancy of the wireless media and higher collision
probability.
[0105] For a synchronized communication system employing
frequency-hopping and time-slotted access, it is advantageous to
synchronize the frequency hopping with the slotted access, meaning
that the frequency-hopping function is changed at the beginning of
a time-slot. This sort of synchronization is use in an embodiment
of the present invention.
[0106] In order to join a synchronized frequency-hopping wireless
communication system, the joining node (such as a remote node in
the system of FIG. 1 that has not yet synchronized with a central
node) needs to become synchronized with the system. This process is
referred to as initial synchronization. In order to become
synchronized, the joining node needs to adjust its local clock to
system clock, and to obtain the information required for
calculating the system frequency-hopping function. One method for
initial synchronization with a synchronized frequency-hopping
wireless communication system is active synchronization with
receiver fast frequency scan.
[0107] The method of active synchronization with receiver fast
frequency scan is illustrated in FIG. 5. According to this method,
the joining node selects a radio frequency (denoted in the figure
by f(m) and a moment in time for transmitting a probe message. The
radio channel f(m) is selected from a predetermined set of radio
channels, which might be the entire set of radio channels available
for the system, or a subset of that set. The selection of the radio
channel and the transmission moment are performed locally, since
the joining node is not yet synchronized with the system. The probe
message comprises a long preamble, which is used by the receiver
for fast frequency scan. The central nodes continuously scan all
the radio channels (denoted by f(1), f(2), . . . in the figure)
within the set of channels allocated for non-synchronized
transmissions, staying in each radio channel for the minimum
duration required to detect the existence of a valid preamble. If a
valid preamble is detected on some given radio channel, the
receiver stays on the given radio channel and detects the entire
message. If the message is successfully detected, the central node
transmits a response message on radio channel f(m), or on another
radio channel known to both sides. The response message comprises a
time-stamp and other information needed for synchronization, for
example the parameters of the frequency-hopping function. Note that
the response message might be a standard ACK message, provided that
the standard ACK message comprises all the information needed for
initial synchronization. The response message might also be a
special ACK message with additional information, or a special probe
response message.
[0108] A potential alternative to active synchronization is passive
synchronization, wherein the central node periodically transmits
beacon messages, and the joining node employs fast frequency scan
until a beacon message is detected.
[0109] Active synchronization with fast frequency scan is an
advantageous method for wireless event monitoring systems, compared
to passive synchronization, some of the advantages being a shorter
joining process and reduced current consumption. On the other hand,
the scheme of active synchronization with fast frequency scan
apparently conflicts with the scheme of time-slotted synchronized
frequency hopping, which is an advantageous scheme for the
steady-state operation of the event monitoring system. The conflict
is caused by the fact that the latter scheme dictates that the
receiver reside on one given radio channel during each time-slot,
while the former scheme dictates that the receiver scan
continuously over all radio channels. In order to resolve this
conflict, in an embodiment of the present invention, the scheme of
time-slotted, synchronized frequency-hopping is utilized by the
synchronized remote nodes, the scheme of active synchronization
with receiver fast frequency scan is utilized by the
non-synchronized remote nodes, such as joining nodes, and the
central node receiver employs both schemes concurrently.
[0110] The method of concurrent employment of the two schemes is
illustrated in FIGS. 6(a) through 6(c), wherein FIG. 6(a)
illustrates the case of no remote node transmission, FIG. 6(b)
illustrates the case of transmission by a synchronized remote node,
and FIG. 6(c) illustrates the case of transmission by a
non-synchronized remote node. Each time-slot (TS) comprises a
system frequency window (SFW) and a scanning window (SCW). When
transmitting an uplink messages, a synchronized remote node always
starts transmitting at the beginning of the SFW. More specifically,
the remote node starts the transmission a short time after the
beginning of the SFW, in order to compensate for potential
misalignment between the local clock of the remote node and the
system clock. The uplink message transmitted by the remote node
comprises a short preamble, denoted in the figure as pr, followed
by the content of the message, denoted in the figure as data. The
uplink message is transmitted at the current value of the system
frequency-hopping function, denoted in the figure by fs(n)), to
indicate that the frequency is a function of the time-slot number
n. The SFW is long enough to accommodate a minimal detectable
preamble, plus the tolerance dictated by the potential misalignment
of the clock at the remote node.
[0111] During the SFW, the receiver of the central node stays at
fs(n) and attempts to detect a valid preamble. If a valid preamble
is detected during the SFW, the receiver of the central node stays
at fs(n) and detects the rest of the message. If the message is
detected successfully, and the message requires acknowledgment or
some other immediate reply, the central node replies at fs(n), or
at another frequency known to both sides. If a valid preamble is
not detected during the SFW, the receiver of the central node
starts a fast frequency scan over the radio channels assigned for
non-synchronized transmissions, denoted in the figure as fu(i)). If
a valid preamble is detected at some radio channel, denoted by
fu(x), during the SCW, the receiver stays at fu(x) and attempts to
detect the entire message. If the message is successfully detected,
and the message requires acknowledgment or other reply, the central
node transmits the reply at fu(x), or at some other radio frequency
known to both sides. If a valid preamble is not detected during the
SCW, the receiver stops the fast frequency hopping during the SFW
of the next time-slot, and continues the fast frequency scan at the
SCW of the next time-slot.
[0112] One of the advantages of the method of concurrent employment
of the two schemes lies in the fact that the receiver employs the
fast frequency-scanning scheme during the idle periods of the
time-slotted scheme. Thus, the employment of one scheme does affect
the efficiency of the other scheme.
[0113] When joining a synchronized wireless communication system,
the joining node compares the current value of the system clock, as
expressed in the time stamp incorporated in the reply message
received from the central node, with the current value of it local
clock, and aligns its local clock with the system clock. Assuming
hypothetically that the system clock and the local clock have
exactly the same frequency, the local clock will theoretically
remain identical to the system clock for an unlimited period of
time. In practice, however, there is usually a slight difference
between the frequencies of the local clocks, which causes the
difference between their values to increase gradually. Therefore,
for a synchronized node to maintain synchronization, the node needs
to receive messages comprising time-stamps and to align its local
clock according to those time-stamps, wherein the maximum time
interval between received time-stamps is inversely proportional to
the maximum frequency difference between the local clocks and to
the maximum allowable synchronization error. Consider for example
the case in which the accuracy of the local clocks is 50 parts per
million (ppm) and the maximum acceptable synchronization error is 1
ms. In this case the maximum difference between the frequencies of
the system clock and the local clock might reach 100 ppm and
therefore the maximum interval between received time-stamps needs
to be below 10 seconds.
[0114] In order to increase the synchronization accuracy, or
increase the allowable gap between time-stamped messages, the
frequency of the local clock of the remote node may be adjusted
relative to that of the system clock, using a method such as the
following: Let the value of the local clock of the remote node when
receiving a first time-stamp be t(1), and let the value of the
first time-stamp be x(1). When receiving the first time-stamp, the
remote node sets its local clock to the value x(1)+d, wherein d is
the estimated delay between the creation of the time stamp and the
reception of the time-stamped message. The remote node also saves
the value x(1) for future reference. Now let the value of the local
clock of the remote node when receiving a second time-stamp be
t(2), and let the value of the second time-stamp be x(2). When
receiving the second time-stamp, the remote node sets its local
clock to the value x(2)+d and saves the value x(2) for future
reference. The remote node also calculates the value
f=(t(2)-x(2))/(x(1)-x(2)), which is an estimation of the frequency
different between the local clock and the system clock. The remote
node then adjusts the frequency of its local clock by af, wherein a
is some positive factor, lower than or equal to unity.
[0115] The difference in frequency between local clocks is due to
long-term reasons, such as the inherent variation between the
devices and effect of aging, and to short-term reasons such as
temperature variations. Due to the short-term reasons, frequency
adjustment should also be performed regularly, for example at each
reception of a time-stamped message.
[0116] As seen above, the process of calculating the frequency
difference f involves division, an operation that might be rather
complex for the ultra-low-power and ultra-low-cost processor of the
remote node. In an embodiment described hereinbelow, the division
may might be avoided by using the periodic nature of the status
messages.
[0117] As explained above, in order to maintain synchronization a
remote node typically needs to receive regularly time-stamped
messages. A common method for ensuring regular reception of
time-stamped messages is to have the central nodes periodically
transmit beacon messages that are time-stamped, and to have the
remote nodes receive those beacon messages. One disadvantage of
this method is that when a remote node fails to receive a beacon
message, it must wait for the next beacon message, which increases
the risk of losing synchronization. Another disadvantage of this
method for a wireless event monitoring system is the extra current
consumption of the remote node, due to the need to wake up to
receive the beacon messages.
[0118] A method of synchronization of a wireless mesh network
without using dedicated beacon messages is described in a white
paper entitled: "Technical Overview of Time Synchronized Mesh
Protocol (TSMP)" by Dust Networks, 30695 Huntwood Avenue, Hayward,
Calif. 94544, USA (document number: 025-0003-01, last revised: Jun.
20, 2006). This method may be used in conjunction with the
embodiments of the present invention that are described
hereinabove.
[0119] In an embodiment of the present invention, a wireless event
monitoring system may be synchronized without using dedicated
beacon. This embodiment uses a method which is based on the regular
nature of periodic status messages sent by the remote nodes of the
event monitoring system. According to this method, the ACK messages
transmitted in reply to the periodic status messages comprise
time-stamp fields, and synchronization is based on those time-stamp
fields. One advantage of this method is that no extra current
consumption is required for the reception of those time-stamped
messages. Another advantage of this method is that the ARQ
mechanism ensures periodic reception of time-stamped messages,
because when the ACK reply fails to be received by the remote node,
the status message is retransmitted several times, until an ACK is
successfully received. As a result, the probability of losing
synchronization due to failure to receive a time-stamp message is
very low. Another advantage is due to the periodic nature of the
periodic status messages. Since the interval between status
messages is fixed (for example 100 seconds), the interval between
successive time-stamps is also fixed, and therefore the frequency
difference f=[t(2)-x(2)]/[x(1)-x(2)] can be approximated by
f=[t(2)-x(2)]k, where k is a constant, thus avoiding the division
operation mentioned above. In summary, this method results in high
accuracy and good reliability of the synchronization mechanism,
along with relaxed requirements on the stability of the local
clock.
[0120] A remote node of a synchronized wireless event monitoring
system might lose its synchronization with the system, for example,
due to failure in receiving messages for a period of time long
enough to cause the difference between the system clock and the
local clock to exceed the maximum tolerance. In such cases, the
node needs to re-synchronize to the system, and the process of
re-synchronization is similar to that of initial synchronization.
According to an embodiment of the present invention,
re-synchronization is performed according to the scheme of active
synchronization with receiver fast frequency scanning. According to
this embodiment, a node that has lost synchronization with the
system transmits re-synchronization probe messages according to the
scheme of receiver fast frequency scan, as presented in FIG. 5, and
those messages are received and replied to by the central node
according to the scheme presented in FIG. 6(c).
[0121] A remote node of a synchronized wireless event monitoring
system that has lost synchronization with the system might fail to
re-synchronize with the system for a long time. One reason for such
a situation might be a persistent interference source located in
the vicinity of a given remote node. If the given remote node is a
monitoring device, it would be advantageous for the node to be able
to transmit operative messages, such as event reporting messages
and periodic status messages, although the node is not synchronized
with the system. According to an embodiment of the present
invention, the scheme of receiver fast frequency scan is utilized
also by nodes that have lost synchronization in order to transmit
operational messages, such as periodic status messages and event
reporting messages. According to this embodiment, a node that has
lost synchronization with the system transmits the operational
message according to the scheme of receiver fast frequency scan, as
presented in FIG. 5, and the message is received and acknowledged
by the central node according to the scheme presented in FIG. 6(c).
If the ACK reply for the operational message is received by the
non-synchronized node, the node uses the time-stamp field in the
ACK message to re-synchronize to the system. Otherwise the
non-synchronized node repeats the operational message several
times, until an ACK is received or until some re-transmission limit
is exceeded. Thus, the operational message serves both for delivery
of information from the remote node to the central unit, and for
re-synchronization.
[0122] As explained above, a wireless event monitoring system in
which all remote nodes are two-way nodes performs better than a
system in which the monitoring devices are one-way nodes. Yet, in
some situations, it might turn out to be advantageous to allow for
the incorporation of some one-way nodes along with the two-way
nodes. According to an embodiment of the present invention, one-way
nodes might be incorporated into the system, wherein the one-way
nodes transmit their operational messages according to the scheme
of receiver fast frequency scan, as presented in FIG. 5, and the
messages are received by the central node according to the scheme
presented in FIG. 6(c), wherein the ACK is superfluous.
[0123] An embodiment of the present invention addresses the
power-save scheme employed for the downlink. According to the
power-save scheme, as explained above, downlink messages are
transmitted at determined points in time, referred to as downlink
windows. An advantage of the power-save mode for a wireless event
monitoring system is the great reduction in current consumption. A
limitation of the power-save mode is that a downlink message can be
repeated only at the next downlink window. Therefore, it is
desirable to protect the downlink messages from being interfered
with by uplink messages. This feature is especially important for
the downlink messages sent to the signaling devices, because those
messages require high reliability and short latency.
[0124] A straightforward method for protecting the downlink
messages from being interfered with by uplink messages is to
identify the uplink time-slots that overlap with the downlink
windows and to avoid the utilization of those time-slots for
transmission of uplink messages. The disadvantage of this
straightforward method is that it reduces the number of available
time-slots for uplink transmission, thus implying an increase of
the collision probability between uplink messages.
[0125] An alternative method for protecting the downlink messages
from being interfered with by uplink messages is provided by an
embodiment of the present invention. According to this method, the
radio channel used for transmission of the downlink messages is
determined not by the function used for the uplink transmissions,
which is referred to as the uplink frequency-hopping function, but
rather by a different function referred to as the downlink
frequency-hopping function.
[0126] It would be desirable, as far as is permitted by
regulations, to have the two frequency-hopping functions orthogonal
to each other, wherein the term orthogonal means that the two
functions never collide, thus achieving full protection of the
downlink messages from being interfered with by uplink messages. In
cases in which regulations do not permit employing orthogonal
frequency hopping functions, for example in FCC regulations,
pseudo-orthogonal functions can be utilized. The term
pseudo-orthogonal means that each frequency hopping function is a
different pseudo-random function. Since the two functions are
different, the probability of collision is about 1/n, wherein n is
the number of radio channels in the range of the frequency hopping
function.
[0127] As explained above, the damage caused by failure in
receiving a downlink messages is higher than the damage caused by
failure in receiving an uplink message, because retransmission of
an uplink message can take place at one of the near time-slots,
while retransmission of a downlink message can take place only at
the next downlink window. Therefore, in order to protect the
downlink messages, these messages have priority over the uplink
messages. This priority is achieved by implementing the following
algorithms at the transmitter side and at the receiver side.
[0128] At the transmitter side, the algorithm for transmitting of
downlink messages is as follows: At each downlink window, if the
central node does not have any outstanding downlink message, the
central node receiver carries on with the reception of potential
uplink messages, using the uplink frequency hopping function for
reception of potential messages from synchronized nodes, and, if
applicable, employing fast frequency scan for reception of
potential messages from un-synchronized nodes. On the other hand,
if the central node does have an outstanding downlink message, the
central node transmits the message in the downlink window, using
the downlink frequency hopping function. Thus, transmission of a
downlink message has priority over reception of a potential uplink
message. Yet, since downlink messages are infrequent, this priority
is applied infrequently and has only a minor effect on the
reception of uplink messages.
[0129] At the receiver side, the algorithm for reception of
downlink messages is as follows: In each downlink window applicable
to a given remote node, the node tunes its receiver to the current
value of the downlink frequency hopping function and attempts to
receive a potential downlink message. Reception of downlink
messages has priority over transmission of uplink messages, meaning
that a given remote node never transmits an uplink message during a
downlink window applicable for the given node. Yet, a remote node
can transmit uplink messages during downlink messages applicable to
other remote nodes. In order to appreciate the advantage of this
arrangement, consider an example of a wireless alarm system, in
which the rate of the downlink windows of the monitoring devices is
much lower than that of the signaling devices, for example once per
minute for the monitoring devices versus once per second for the
signaling devices. In this example, the monitoring devices are not
prevented from transmitting uplink messages during the downlink
windows of the signaling devices, and the probability that an
uplink message transmitted during those downlink windows will be
lost due to a concurrent downlink message is low, since downlink
messages are typically infrequent.
[0130] Downlink messages are transmitted to the remote nodes either
directly or via one or more repeaters, which implies that there
should be a mechanism for transmitting downlink messages from a
central device to a repeater, wherein the term "central device"
refers to the central node and to any of the other repeaters. This
mechanism can be implemented according to the following two
methods:
[0131] The first method for transmitting downlink messages from a
central device to a repeater is similar to the method for
transmitting downlink messages from a central device to a remote
node. The algorithm employed by the central node is identical to
the algorithm for transmitting downlink messages to remote nodes.
(Actually, a message transmitted by the central node might be
simultaneously addressed to more than one remote node and/or
repeater.) The algorithm employed by the repeater is similar to the
algorithm employed by a remote node. At each downlink window, the
repeater stops its activity in receiving potential uplink messages,
sets its receiver to the current value of the downlink frequency
hopping function, and attempts to receive a potential downlink
message.
[0132] The second method for transmitting downlink messages from a
central node to a repeater is similar to the method for
transmitting uplink messages from a remote node to a central node.
The algorithm employed by the receiving repeater is identical to
the algorithm for receiving uplink messages. Actually, the repeater
employs the same scheme for receiving both uplink messages from
remote nodes and downlink messages from other central nodes. The
algorithm employed by the transmitting central node is similar to
the algorithm employed by a remote node for transmitting an uplink
message. When transmitting a downlink message to a repeater, the
central node is necessarily unable to receive uplink messages, but
since downlink messages are infrequent, this fact has minor effect
on the probability of missing an uplink message.
[0133] One advantages of the second method is the better
reliability and shorter latency of the hop between the central node
and the repeater, because the time gap between retransmissions is
much shorter.
[0134] Another advantage of the second method is that it avoids the
main drawback of the first method, which is the fact that during
the downlink windows the repeater has to switch to a different
radio channel and is therefore not able to receive potential uplink
messages.
[0135] In a wireless communication link that is subject to
multipath propagation, the propagation channel between the
transmitting antenna and the receiving antenna can vary
significantly as a result of relatively small displacements of the
antenna. Furthermore, the propagation channel between the
transmitting antenna and the receiving antenna might also vary
significantly as a result of the relative orientation of the two
antennas, due to polarization. Therefore, it is a common practice
in wireless communication systems to utilize antenna diversity, by
fitting the transmitter or the receiver, or both, with more than
one antenna (usually two). The antennas are located at some
distance from one another, and might also have different
orientations (usually perpendicular to one another). A simple and
common method for antenna diversity is switched antenna diversity,
which can be employed by the transmitter and/or by the receiver.
According to this method, when transmitting or receiving via a
given antenna results in a significantly better propagation channel
than via the other antenna, the better antenna is selected for
transmission or reception, respectively.
[0136] FIG. 7 is a timing diagram that schematically illustrates a
method of intra-message antenna diversity that may be implemented
by a central node of a wireless event monitoring system, in
accordance with an embodiment of the present invention. In this
example, the central node uses intra-message antenna diversity for
the reception of a synchronized uplink messages, using the method
illustrated in FIGS. 6(a) and 6(b). According to FIG. 7, the SFW
comprises three sub-windows, denoted in the Figure as SW(1), SW(2)
and SW(3). During SW(1) the central node receiver is switched to
the first antenna and measures some quality parameter of the signal
received via the first antenna, such as the received signal level.
During SW(2) the central node receiver performs the same operation
for the second antenna. During SW(3) the central node receiver is
switched to the antenna with the higher quality parameter, and
attempts to detect a valid preamble. The subsequent operation of
the central node receiver depends on the result of the preamble
detection process, as illustrated in FIGS. 6(a) and 6(b). In a
similar manner, the method of intra-message antenna diversity can
also be applied for the reception of non-synchronized uplink
messages in accordance with FIGS. 6(a) and 6(c), wherein each of
the time-periods denoted in the Figures by fu(i) comprises
sub-windows SW(1), SW(2) and SW(3) with the same functionality.
[0137] The disadvantages of this method of intra-message antenna
diversity is the excessive length of the preamble that it requires.
In this method the preamble needs to accommodate SW(1) and SW(2) in
addition to SW(3), which is required regardless of the antenna
diversity function. This disadvantage of the of intra-message
antenna diversity method is especially significant for a wireless
event monitoring system, since for such systems, the longer
preamble means (a) higher current consumption at the remote nodes
and (b) longer time-slots, which implies fewer time-slots per
frame. An alternative embodiment, described hereinbelow, provides a
method of transmitter-selected receiver antenna diversity that
requires no excess preamble length.
[0138] FIG. 8 is a timing diagram that schematically illustrates a
method of transmitter-selected receiver antenna diversity that
operates in accordance with the uplink time-slotted scheme of FIG.
2. According to this embodiment, each time-slot (TS) within a frame
is associated with a given antenna, and the antenna associated with
a given time-slot is utilized by the receiver of the central node
to receive synchronized uplink messages transmitted during the
given time-slot. The association between time-slot number and
antenna number is a fixed function, referred to as the antenna
switching function, that is also known to the remote nodes. In the
example presented FIG. 8, the antenna switching function associates
the odd time-slots with the first antenna (denoted in the Figure as
ANT1) and the even time-slots with the second antenna (ANT2). FIG.
8 also shows that the first two time-slots in each frame are
reserved for periodic status messages, while the rest of the
time-slots are available for random access.
[0139] Since the antenna switching function is known to the remote
nodes, each remote node that has an outstanding uplink message is
able to select the antenna of the central node that will receive
the message by selecting the time-slot in which the message is
transmitted accordingly to the antenna switching function. In the
case of a periodic status message, the remote node is able to
select one of the two time-slots reserved for periodic status
messages, and in the case of random access, the remote node is able
to select one of the time-slots available for random access. In
case of retransmission, the time-slot for retransmission is
selected according to the random back-off mechanism and according
to the antenna selection mechanism.
[0140] In order to select the antenna of the central node for
reception of an outstanding message, each remote node calculates
the score of each of the antennas of the central node. Any suitable
method of calculation may be used. For example the antenna used for
the last successful uplink transmission might have a high score and
the other antenna(s) a low score. Calculating the score can also
involve more elaborate statistics, for example, the number of
successful receptions via each antenna during the last x successful
receptions or the last y time units.
[0141] Calculating the score can also depend on link quality
parameters associated with each successful transmission. The link
quality parameters can be signal quality parameters measured by the
remote node when receiving the ACK reply. A simple, useful and
commonly-used signal quality parameter is the received signal
level, which is inversely proportional to the channel attenuation,
assuming a fixed or known transmit signal level.
[0142] Alternatively or additionally, the link quality parameters
can be signal quality parameters measured by the central node when
receiving the uplink message, provided that these parameters are
incorporated in the ACK reply. For example, the signal quality
parameters can comprise the received signal level, which is
commonly incorporated in the ACK reply in order to facilitate
transmission power control. It should be noted that in
bi-directional communication links employing the TDD scheme, which
utilize the same radio frequency for both directions, the
propagation channel between the antennas is reciprocal, and
therefore similar scoring applies to both directions of the
link.
[0143] Communication systems usually employ ARQ, as described
above. According to this method, when a message is not replied by
an ACK, the message is retransmitted according to some random
back-off mechanism. When transmitter-selected receiver antenna
diversity is employed, the time-slot for retransmission is selected
according to the random back-off mechanism and according to the
antenna selection mechanism. The design of the antenna selection
mechanism for retransmission depends on various parameters. For
example, if the temporal variations of the propagation channels are
expected to be slow, a suitable policy would be to retransmit
several times to the same antenna before switching to the other
antenna. The advantage of such a policy is that it minimizes the
average number of retransmissions. But on the other hand, such a
policy increases the worst-case time for message delivery.
Therefore, in cases in which worst-case delivery time must be
short, or in cases in which the temporal variations of the
propagation channel are quick, a more suitable policy would be to
switch to the other antenna at each retransmission.
[0144] Uplink antenna diversity has been discussed above in the
context of a single-antenna transmitter and a multi-antenna
receiver, which is the common case for event monitoring systems
since monitoring devices 200 are typically single-antenna nodes.
However, cases of uplink antenna diversity between a multi-antenna
transmitter and a multi-antenna receiver may also be applicable.
For example, signaling device 300 and human interface device 400 of
a wireless event monitoring system are typically bigger and more
expensive, and may be fitted with more than one antenna. Another
example are repeaters 500 of the wireless event monitoring system,
which are usually fitted with two antennas.
[0145] In case of a multi-antenna uplink transmitter, two methods
can be used. The first method is to select one of the two antennas
of the uplink transmitter as the transmitting antenna, and to
employ the same method as in the case of a single-antenna uplink
transmitter. The second method is to select the transmit antenna
according to the receive antenna. According to the second method,
the transmitter calculates the score of each of its antennas with
respect to each antenna of the receiver. When transmitting a
message at a given time, the transmitter selects for transmission
the antenna with the best score with respect to the antenna
utilized by the receiver at the given time.
[0146] An advantage of the method of selecting the transmit antenna
according to the receive antenna is that the transmitter is usually
able to select the transmission time regardless of the antenna
switching function utilized by the receiver, because for each
receive antenna there is usually at least one transmit antenna such
that the propagation channel between the transmit and receive
antennas is not subject to severe degradation due to multipath or
polarization. The freedom in selecting the transmission time may,
in many cases, be valuable. Consider, for example, a wireless event
monitoring system in which the communication between the central
unit and the remote nodes is either direct or via at most one
repeater, and in which the central unit and the repeaters are
dual-antenna units. In this system it is advantageous to reserve in
each frame one or more time-slots for forwarding periodic status
messages by a potential repeater, thus avoiding any interference
between forwarded periodic messages and other uplink messages. For
example, the first two time-slots in each frame can be reserved for
transmission of periodic status messages, and the next one or two
time-slots might be reserved for forwarding those messages by a
potential repeater, when applicable. Now, the number of time-slots
reserved for potential forwarding of periodic status messages
depends on the antenna diversity method employed by the repeater.
If the first method is employed, two time-slots need to be
reserved, whereas if the second method is used, one time-slot is
sufficient.
[0147] Antenna diversity is applicable to the downlink messages,
too. For single-antenna remote nodes, the central node should
transmit the message multiple times, once via each antenna, and
each transmission should take place in different downlink window.
The mapping between the transmit antenna and the downlink window is
a deterministic function, which is also known to the remote nodes.
The wake-up policy employed by the remote nodes might be (a) to
wake up in the downlink windows corresponding to each transmit
antenna or (b) to wake up only in the downlink window of the best
transmit antenna. The first policy can be employed when the
temporal variations of the propagation channels are expected to be
slow, the potential increase in latency can be tolerated, and the
current consumption requirements are strict. The second policy can
be employed when the temporal variations of the propagation
channels are expected to be quicker, the potential increase in
latency cannot be tolerated, and the current consumption
requirements are more relaxed.
[0148] It will be appreciated that the embodiments described above
are cited by way of example, and that the present invention is not
limited to what has been particularly shown and described
hereinabove. Rather, the scope of the present invention includes
both combinations and sub-combinations of the various features
described hereinabove, as well as variations and modifications
thereof which would occur to persons skilled in the art upon
reading the foregoing description and which are not disclosed in
the prior art.
* * * * *