U.S. patent application number 11/328567 was filed with the patent office on 2007-07-12 for system and method for clustering wireless devices in a wireless network.
Invention is credited to Amit Jain.
Application Number | 20070160016 11/328567 |
Document ID | / |
Family ID | 38232665 |
Filed Date | 2007-07-12 |
United States Patent
Application |
20070160016 |
Kind Code |
A1 |
Jain; Amit |
July 12, 2007 |
System and method for clustering wireless devices in a wireless
network
Abstract
Described are a system and method for clustering wireless
devices in a wireless network. The system comprises a wireless
access point and a plurality of wireless computing units grouped
into a cluster as a function of a predetermined parameter. The
cluster includes a cluster head unit and at least one cluster
member unit. The at least one cluster member unit utilizes a first
power level when wirelessly communicating with the cluster head
unit, and the cluster head unit utilizes a second power level when
communicating with the AP.
Inventors: |
Jain; Amit; (Bangalore,
IN) |
Correspondence
Address: |
INGRASSIA FISHER & LORENZ, P.C.
7150 E. CAMELBACK, STE. 325
SCOTTSDALE
AZ
85251
US
|
Family ID: |
38232665 |
Appl. No.: |
11/328567 |
Filed: |
January 9, 2006 |
Current U.S.
Class: |
370/338 |
Current CPC
Class: |
H04W 52/283 20130101;
H04W 52/367 20130101; H04W 84/20 20130101 |
Class at
Publication: |
370/338 |
International
Class: |
H04Q 7/24 20060101
H04Q007/24 |
Claims
1. A system, comprising: a wireless access point ("AP"); and a
plurality of wireless computing units grouped into a cluster as a
function of a predetermined parameter, the cluster including a
cluster head unit and at least one cluster member unit, wherein the
at least one cluster member unit utilizes a first power level when
wirelessly communicating with the cluster head unit, the cluster
head unit utilizing a second power level when communicating with
the AP.
2. The system according to claim 1, wherein each of the wireless
computing units includes at least one of a laser-based scanner, an
image-based scanner, an RFID reader, a laptop, a cell phone, a PDA
and a network interface card.
3. The system according to claim 1, wherein the second power level
is greater than the first power level.
4. The system according to claim 3, wherein the first power level
is approximately 2 mW.
5. The system according to claim 1, wherein the predetermined
parameter includes at least one of a distance between the cluster
head unit and the at least one cluster member unit, an internal
load level, a battery level, a hardware configuration and a radio
frequency range between the wireless computing units.
6. The system according to claim 1, wherein the cluster head unit
is the selected from the wireless computing units as a function of
a respective amount of data to transmit by each of the wireless
computing units.
7. The system according to claim 6, wherein the cluster head unit
is the wireless computing unit with a greatest amount of data to
transmit than the remaining wireless computing units of the
cluster.
8. The system according to claim 6, wherein the cluster member unit
acts as the cluster head unit after a predetermined time period
expires, the time period corresponding to a period of time
necessary to transmit corresponding data by each of the at least
one cluster member unit.
9. The system according to claim 5, wherein the first power level
is determined as a function of the predetermined parameter.
10. The system according to claim 1, wherein the cluster head unit
at least one of aggregates data transmissions from the at least one
cluster member unit to the AP and disaggregates data transmissions
from the AP to the at least one cluster member unit.
11. The system according to claim 10, wherein the cluster head unit
performs the at least one of the aggregation and the disaggregation
as a function of at least one of a type of the data transmissions,
a clustering efficiency and a power-saving efficiency.
12. The system according to claim 11, wherein the cluster head unit
refrains from the at least one of the aggregation and the
disaggregation when the type is at least one of voice and video
data transmissions.
13. A method, comprising: determining a location of each of a
plurality of wireless computing units; selecting at least two units
of the plurality as a function of a first predetermined parameter;
creating a cluster of the at least two units, the cluster including
a cluster head unit and at least one cluster member unit;
designating one of the at least two units as the cluster head unit
as a function of a second predetermined parameter; and conducting
wireless communications between the cluster head unit and the at
least one cluster member unit at a predetermined power level.
14. The method according to claim 13, wherein each of the wireless
computing units includes at least one of a laser-based scanner, an
image-based scanner, an RFID reader, a laptop, a cell phone, a PDA
and a network interface card.
15. The method according to claim 13, further comprising:
conducting wireless communications between the cluster head unit
and an access point at a further predetermined power level, the
further predetermined power level being greater than the
predetermined power level.
16. The method according to claim 13, wherein the predetermined
power level is approximately 2 mW.
17. The method according to claim 13, wherein the first
predetermined parameter includes at least one of a geographical
range of the wireless computing units relative to each other, an
internal load level, a battery level, a hardware configuration and
a radio frequency range between the wireless computing units.
18. The method according to claim 13, wherein the second
predetermined parameter is a respective amount of data to transmit
by each of the at two units.
19. The method according to claim 18, wherein the cluster head unit
is the wireless computing unit with a greatest amount of data to
transmit than the remaining wireless computing units of the
cluster.
20. The method according to claim 18, further comprising: when a
predetermined time period expires, designating one of the at least
one cluster member unit as a next cluster head unit, the time
period corresponding to a period of time necessary to transmit
corresponding data by each of the at least one cluster member
unit.
21. The method according to claim 13, wherein the conducting step
includes the following substeps: aggregating, by the cluster head
unit, data transmissions from the at least one cluster member unit
into a single transmission; and transmitting the single
transmission to the AP.
22. The method according to claim 13, wherein the conducting step
includes the following substeps: disaggregating, by the cluster
head unit, a data transmission from the AP to the at least one
cluster member unit into multiple transmissions; and transmitting
each of the multiple transmissions to the corresponding cluster
member unit.
23. The method according to claim 21, wherein the aggregating step
is performed as a function of at least one of a type of the data
transmissions, a clustering efficiency and a power-saving
efficiency.
24. The method according to claim 23, wherein, when the type is at
least one of voice and video data transmission, bypassing the
aggregating step by the cluster head unit.
25. An arrangement, comprising: a processor selecting at least two
wireless computing units from a plurality of wireless computing
units as a function of a first predetermined parameter, the
processor creating a cluster of the at least two units, the cluster
including a cluster head unit and at least one cluster member unit,
the processor designating one of the at least two units as the
cluster head unit as a function of a second predetermined
parameter, the cluster head unit conducting wireless communications
with the at least one cluster member unit at a first power level;
and a communications arrangement conducting wireless communications
with the cluster head unit at a second power level.
26. The arrangement according to claim 25, wherein the arrangement
includes at least one of an access point and a switch.
27. An arrangement, comprising: a locationing means determining a
location of each of a plurality of wireless computing units; a
clustering means selecting at least two units of the plurality as a
function of a first predetermined parameter, the clustering means
creating a cluster of the at least two units, the cluster including
a cluster head unit and at least one cluster member unit, the
clustering means designating one of the at least two units as the
cluster head unit as a function of a second predetermined
parameter; and a communication means conducting wireless
communications with the cluster head unit at a predetermined power
level.
Description
FIELD OF INVENTION
[0001] The invention generally relates to clustering wireless
computing devices in a wireless network.
BACKGROUND INFORMATION
[0002] A conventional wireless network may operate in one of two
distinct modes: an infrastructure mode and an ad hoc mode. In the
infrastructure mode, a mobile unit ("MU") transmits wireless
signals to other MUs via an access point ("AP"). The MU utilizes a
maximum power level when transmitting to the AP, regardless of the
proximity to the AP, consuming a significant amount of battery
power and potentially causing interference with communications
between one of the other MUs and a further AP. In the ad hoc mode,
the MU communicates directly with another MU, i.e., without use of
the AP. However, the ad hoc mode cannot typically support a large
number of MUs, and these MUs, while in the ad hoc mode, cannot
bridge to the wireless network (e.g., a WLAN) or the Internet,
limiting functionality. Therefore, there is a need for an improved
network architecture.
SUMMARY OF THE INVENTION
[0003] The present invention relates to a system and method for
clustering wireless devices in a wireless network. The system
comprises a wireless access point and a plurality of wireless
computing units grouped into a cluster as a function of a
predetermined parameter. The cluster includes a cluster head unit
and at least one cluster member unit. The at least one cluster
member unit utilizes a first power level when wirelessly
communicating with the cluster head unit, and the cluster head unit
utilizes a second power level when communicating with the AP.
BRIEF DESCRIPTION OF DRAWINGS
[0004] FIG. 1 shows an exemplary embodiment of a system according
to the present invention; and
[0005] FIG. 2 shows an exemplary embodiment of a method according
to the present invention.
DETAILED DESCRIPTION
[0006] The present invention may be further understood with
reference to the following description and the appended drawings,
wherein like elements are provided with the same reference
numerals. The present invention describes a system and method for
clustering wireless devices in a wireless network. In particular,
the present invention relates to an improved wireless network
architecture that, for example, uses location information of
wireless computing units to form clusters thereof. As will be
understood by the following description, the present invention may
be utilized to limit power consumption by the wireless devices and
improve throughput in the wireless network.
[0007] FIG. 1 shows an exemplary embodiment of a system 100
according to the present invention. The system 100 may include a
communications network 10 (e.g., a wired/wireless LAN, WAN or the
Internet) having at least one access point/port ("AP"), such as an
AP 30 and/or an AP 35, providing access thereto. The system 100 may
further include a server 20 in communication with the network 10.
The network 10 may include a plurality of interconnected computing
devices such as, for example, servers, hubs, routers, switches,
etc.
[0008] The system 100 may include a plurality of wireless computing
devices (e.g., mobile units ("MUs")), such as MUs 41-45 and 51-54.
Each MU may include, for example, a laser/image-based scanner, an
RFID reader/tag, a cell phone, a PDA, a network interface card,
etc. The MUs 41-54 may conduct wireless communications over network
10 via an AP (e.g., AP 30). When the MUs 41-54 utilize the AP 30 to
communicate over the network 10, the MUs transmit packets at a
first power level (e.g., a maximum power level) regardless of the
proximity to the AP 30. Each MU may also communicate directly with
another MU without use of the AP 30. During this direct
communication between MUs and/or while the MUs are clustered, the
transmitting MU may utilize a second power level. These
methodologies of communication will be described further below.
[0009] According to exemplary embodiments of the present invention,
two or more MUs may be grouped into a cluster for improved
communications between the cluster and an AP. In one exemplary
embodiment, the MUs in the cluster may be in an infrastructure
mode, communicating with each other via the AP. However, the MUs
may also be in an ad-hoc mode with a cluster-head MU (described
below) for communication therebetween when, for example, the MUs
are unable to communicate directly with the AP or the MUs are
within a predetermined communicable range of each other or the
cluster-head MU. That is, the MUs in the clusters may send first
data (e.g., high priority data--e.g., voice) directly to the AP
while sending second data (e.g., low priority) to the cluster-head
MU which aggregates data from a plurality of MUs (including itself)
before forwarding the data to the AP, as will be explained below.
The cluster may be formed based on a plurality of factors such as
geographical proximity of the MUs to one another and/or the AP, an
internal load level, a battery level, a hardware configuration, RF
range between MUs, etc. For example, the MUs 41-45 may be grouped
into a cluster 40 and the MUs 51-54 may be grouped into a cluster
50. In a preferred embodiment, each cluster includes one
cluster-head MU and at least one cluster-member MU. For example,
the cluster 40 may include the cluster-head MU 41 and the
cluster-member MUs 42-45. Similarly, the MU 51 may be the
cluster-head of the cluster 50 and the MUs 51-54 are the
cluster-member MUs.
[0010] In one exemplary embodiment, a particular MU (e.g., the MU
41) may be designated as the cluster head MU as a function of, for
example, an amount of data the MU is going to transmit. For
example, the MU 41 may have a largest amount of data (e.g., as
measured in bytes) to transmit. Thus, the MU 41 may be selected as
an initial cluster head MU for the cluster 40. In another exemplary
embodiment, the MU 41 may function as the cluster head for a
predetermined time proportional to the amount of data it is going
to transmit. For example, if a total data to be transmitted by the
MUs 41-45 equaled 100 bytes, and the MU 41 had 50 bytes to
transmit, the MU 41 may be designated as the cluster head MU for
50% of the time the cluster 40 is intact. When the predetermined
time expires, the MU with a second largest amount of data to
transmit may be designated as a subsequent cluster head MU, and so
on, until each MU functions as the cluster head MU.
[0011] Inter-cluster communication, i.e., between the cluster head
MU and the AP, may utilize a conventional IEEE 802.11x protocol
which may be the same as or different from the protocol used for
intra-cluster communication. The cluster head MU (e.g., MU 41) may
communicate with the AP (e.g., AP 30) at the first power level
(e.g., maximum power). Each cluster head MU (e.g., the MUs 41 and
51) associated with the AP (e.g., AP 30) may utilize a CSMA/CA
mechanism when communicating therewith.
[0012] Intra-cluster communication, i.e., between the
cluster-member MUs and/or between the cluster-member MUs and the
cluster head MU, may occur using a conventional IEEE 802.11x
protocol. For example, each MU in the cluster may utilize a CSMA/CA
mechanism to limit congestion and interference within the cluster.
Furthermore, intra-cluster communication (e.g., between the cluster
member MU and the cluster head MU) may occur at the second power
level. In one exemplary embodiment, the second power level is no
more than about 2 mW. In another exemplary embodiment, the second
power level is variable as a function of a size of the cluster. For
example, as the number of MUs in the cluster increases and/or the
geographic distance or RF range between MUs increases, the second
power level may increase.
[0013] When one of the cluster member MUs has data to transmit out
of the cluster, e.g., to the AP 30, the network 10, etc., the
cluster head MU may function as an aggregation point for data from
the cluster member MUs. For example, when the MU 42 has a packet to
transmit to the server 20, the packet is first sent to the MU 41.
Depending on a type of the data in the packet (e.g., data, voice,
video, etc.), the MU 41 may aggregate the packet with one or more
packets previously received/stored by the MU 41 which has not been
transmitted to the AP 30. That is, the MU 41 may include the packet
from the MU 42 with packets from one or more other MUs in the
cluster 40, including a packet from the MU 41. The aggregated
packet may then be transmitted to the AP 30. Disaggregation of an
aggregated packet by the MU 41 may occur in a similar manner. That
is, the MU 41 may receive an aggregated packet from the AP 30,
divide the aggregated packet into individual packets intended for
each recipient MU and distribute the individual packets to the
corresponding MU(s).
[0014] As noted above, both aggregation and disaggregation may be
sensitive to the type of data being transmitted. For example, when
data which is sensitive to latency (e.g., VoIP packets) or marked
as an emergency transmission is received by the MU 41, it may
transmit the data without dis/aggregation. In this instance, there
may be a plurality of rounds of dis/aggregation executed at the
cluster head MU, as will be further described below. Whether to
utilize the aggregation may be determined as a function of one or
more predetermined factors, e.g., clustering efficiency, power
saving, etc. The clustering efficiency may include any
inter-cluster interference. For example, clusters may be located
adjacent to each other so that even low power communications may
degrade the clustering efficiency. This may be accounted for when
determining the cluster efficiency.
[0015] Similarly, in an instance where the data is larger than a
fragmentation threshold, fragmentation may be performed for data
communicated between the cluster head MU and the AP and between the
cluster head MU and the cluster member MUs.
[0016] Those of skill in the art will understand that the cluster
may include two or more cluster-head MUs, and/or the cluster may
further be subdivided into a plurality of sub-clusters with each
sub-cluster having a structure similar to one of the cluster.
[0017] As one of ordinary skill in the art will understand,
membership in the cluster may change dynamically depending on, for
example, proximity of the MUs relative to each other. As shown in
FIG. 1, the MU 45 is the cluster-member MU of the cluster 40.
However, if the MU 45 changes location (e.g., in a direction
towards the MUs 51-54), then the MU 45 may become a cluster-member
of the cluster 50.
[0018] FIG. 2 shows an exemplary method 200 of communication
according to the present invention. The method 200 is described
with reference to the system 100 in FIG. 1. Those skilled in the
art will understand that other systems having varying
configurations, for example, different numbers of networks, APs,
and MUs may also be used to implement the exemplary method.
[0019] In step 201, locations and/or relative locations of the MUs
which are in communication with the network 10 may be determined
using any real-time locationing algorithm or any other methods
known to those of ordinary skill in the art. For example, a
Received Signal Strength Indication ("RSSI") may be measured by one
or more APs to determine the locations of the MUs. That is, the
server 20 may generate and transmit a signal via one or more APs
(e.g., APs 30 and/or 35). Each MU may then transmit a response
signal to the APs 30 and/or 35 which forward it to the server 20.
The server 20 may then measure the RSSI for the response signals
and compare it with predetermined geographically marked locations
or points (e.g., within the network 10) to determine the relative
locations of the MUs (e.g., the MUs 41-45 and 51-54).
Alternatively, or in combination, a Time Difference Of Arrival
("TDOA") method may be utilized to determine the relative locations
of the MUs.
[0020] In step 203, the MUs are formed into one or more clusters
based on a plurality of factors as mentioned herein. In the
exemplary embodiment, the MUs form the cluster as a function of
their relative locations to each other. That is, when the MUs 41-45
are within a predetermined distance and/or RF range of each other,
they form the cluster 40. When the cluster 40 is formed, each of
the MUs 41-45 are cluster member MUs.
[0021] Also, in step 203, a cluster-head MU is selected from among
the cluster member MUs as a function of, for example, an amount of
data each MU has stored for transmission at a preselected time
(e.g., at formation of the cluster). For example, the MU 41 may be
selected as the cluster-head MU of the cluster 40, because the MU
41 has more data to transmit than the MUs 42-45.
[0022] As one of ordinary skill in the art will understand, steps
201 and 203 may be repeated at any predetermined time interval,
upon request by a user or upon occurrence of a predetermined event.
For example, if the MU 41 terminates membership in the cluster 40
(e.g., powers down/works offline, leaves RF range of MUs 42-45,
etc.), the membership in the cluster 40 may be reassessed and a new
cluster head may be selected. In a preferred embodiment, the new
cluster head is the MU with a second largest amount of data to
transmit after the MU 41. Additionally, as described above, the
cluster head designation may be reassigned so that each MU in the
cluster functions as the cluster head for a time proportional to an
amount of data for transmission (e.g., the more data to transmit,
the more time functioning as the cluster head). Therefore, the MUs
included in a cluster and/or the cluster-head MU may change over
time.
[0023] In step 205, one or more cluster-member MUs and/or the
cluster head MU, generates a packet for transmission to the AP. For
example, the cluster-member MU 43 generates a packet for
transmission to the AP 30. In step 207, the cluster-member MU 43
transmits the packet to the cluster-head MU 41. As described above,
the MUs 41-45 may utilize the second power level for intra-cluster
communications. Thus, the cluster-member MU 43 may transmit its
packet to the cluster head MU 41 at the second power level. As
described above, in an exemplary embodiment, the second power level
is no more than 2 mW; however, the second power level may vary with
a geographic distance and/or RF range covered by the cluster.
[0024] In step 209, the cluster-head MU 41 transmits the packet
from the MU cluster-member 43 to the AP 30. As described above, the
cluster head MU may utilize the first power level (e.g., maximum
power) when transmitting to the AP. In an exemplary embodiment, the
MU 41 may have received or be waiting to receive packets from the
other cluster-member MUs (e.g., MUs 42, 44, 45) for transmission to
the AP. Thus, the MU 41 may aggregate the packets from itself and
the MUs in the cluster into fewer (e.g., one) packet(s) prior to
transmission to the AP 30. The aggregate packet may be transmitted
to the AP 30. Those of skill in the art will understand that
transmission of the aggregated packet may result in decreased
congestion and/or interference in the network 10 and less overhead.
Thus, in the example described above, the packet from the MU 43 may
be transmitted to the AP 30 by itself or as part of an aggregate
packet which contains a plurality of packets.
[0025] According to the present invention, the cluster head MU may
suspend aggregation as a function of, for example, a type of
packet. For example, if the packet from the MU 43 is a voice packet
(e.g., VoIP) or an emergency/urgent packet, the MU 41 may transmit
the voice packet to the AP 30 by itself. Alternatively, the MU 41
may receive the voice packet during aggregation. Thus, when the MU
41 receives the voice packet, it may aggregate the voice packet
into the aggregated packet and immediately transmit the aggregated
packet to the AP 30 without waiting for data packets from other
cluster-member MUs. The MU 41 may then resume aggregation upon
receipt of packets from the cluster member MUs.
[0026] In step 211, the cluster-head MU 41 receives a response
signal from the AP 30. The response signal may include packets for
one or more of the cluster member MUs and/or the cluster head MU.
The cluster-head MU 41 may determine whether the response signal
includes response packets for one or more of the MUs 41-45 by
disaggregating the response signal. Those of skill in the art will
understand that disaggregation in downstream communication may be
optional. Also, the AP 30 may transmit directly to the cluster
member MUs.
[0027] In step 213, the response packet(s) are then transmitted to
the corresponding MU(s) (e.g., the MU 43). Those of skill in the
art will understand that the above-description with respect to
aggregation as a function of packet type applies similarly to
disaggregation. That is, the MU 41 may receive a plurality of
response signals from the AP 30 before disaggregating and
distributing the response packets to the corresponding MUs.
However, in one exemplary embodiment, when at least one of the
response packets includes voice and/or emergency data, the MU 41
may transmit that response packet to the corresponding MU without
waiting for further response signals from the AP 30.
[0028] In another exemplary embodiment of the present invention, an
efficacy factor may be computed for each cluster in the system 100
as a function of RF proximity of MUs for a given power level. For
example, a histogram may be generated for the MUs in the system 100
based on an amount of data transmitted by each MU. Each cluster,
with a data byte count in a predefined range, may have a different
efficacy factor. The efficacy factor of each cluster may be
utilized to determine a benefit in throughput and power consumption
provided by utilizing the present invention. Thus, any algorithm
granting permission to a cluster to transmit uplink data (e.g.,
cluster head to AP) may utilize the efficacy factor.
[0029] The present invention provides several advantages both at
the network level and the MU level. By allowing simultaneous or
substantially simultaneous intra-cluster communication of adjacent
clusters, the throughput of the network may increase without a
concomitant increase in interference and congestion. Additionally,
by allowing intra-cluster communication at a lower power level than
inter-cluster communication, the MUs in the cluster may consume
less power and extend a life of their power sources (e.g.,
batteries).
[0030] The present invention has been described with reference to
an embodiment having the MUs 41-45 and 51-54, the network 10, and
the APs 30 and 35. One skilled in the art would understand that the
present invention may also be successfully implemented, for
example, for any number of MUs, APs, and/or a plurality of the
networks 10. Accordingly, various modifications and changes may be
made to the embodiments without departing from the broadest spirit
and scope of the present invention as set forth in the claims that
follow. The specification and drawings are accordingly to be
regarded in an illustrative rather than restrictive sense.
* * * * *