U.S. patent application number 13/107256 was filed with the patent office on 2011-09-01 for determining coverage of a wireless network.
This patent application is currently assigned to TROPOS NETWORKS, INC.. Invention is credited to Roman M. Arutyunov, Cyrus Behroozi.
Application Number | 20110211484 13/107256 |
Document ID | / |
Family ID | 39528935 |
Filed Date | 2011-09-01 |
United States Patent
Application |
20110211484 |
Kind Code |
A1 |
Behroozi; Cyrus ; et
al. |
September 1, 2011 |
Determining Coverage of a Wireless Network
Abstract
An apparatus and method of determining coverage of a wireless
network is disclosed. The method includes traveling to multiple
locations around access points of the wireless network, and for a
plurality of client applications, measuring a performance parameter
between a test client device and nodes of the wireless network, at
a plurality of the multiple locations.
Inventors: |
Behroozi; Cyrus; (Menlo
Park, CA) ; Arutyunov; Roman M.; (San Diego,
CA) |
Assignee: |
TROPOS NETWORKS, INC.
Sunnyvale
CA
|
Family ID: |
39528935 |
Appl. No.: |
13/107256 |
Filed: |
May 13, 2011 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11638274 |
Dec 13, 2006 |
7970394 |
|
|
13107256 |
|
|
|
|
Current U.S.
Class: |
370/252 |
Current CPC
Class: |
H04L 43/50 20130101 |
Class at
Publication: |
370/252 |
International
Class: |
H04L 12/26 20060101
H04L012/26 |
Claims
1. A method of determining coverage of a wireless network,
comprising: traveling to multiple locations in and around access
points of the wireless network; for a plurality of client
applications, measuring a performance metric between a test client
device and nodes of the wireless network, at a plurality of the
multiple locations.
2. The method of claim 1, wherein the performance metric comprises
a data throughput.
3. The method of claim 2, wherein characterizing the data
throughput at each of the plurality of locations comprises the test
client device transmitting a plurality of probe packets at a
predetermined rate, and counting a number of responses from the
nodes.
4. The method of claim 1, wherein measuring the performance metric
for the plurality of client applications comprises measuring the
performance metric for multiple types of data.
5. The method of claim 4, wherein the multiple types of data
comprises multiple packet sizes, multiple packet types and multiple
intervals.
6. The method of claim 5, further comprising measuring the
performance metric with a one of the multiple data types depending
upon the client application.
7. The method of claim 1, wherein measuring the performance metric
for the plurality of client applications comprises additionally
characterizing latency and jitter between the test client device
and the nodes at the plurality of locations.
8. The method of claim 1, wherein measuring the performance metric
for the plurality of client applications comprises measuring the
performance metric while transmitting data packets that mimic data
packets of each of the plurality of client applications.
9. The method of claim 1, further comprising measuring signal
strengths of nodes of the wireless network.
10. The method of claim 9, further comprising identifying potential
self-interference areas of the wireless network based on the
measured signal strengths, and the measured performance metric.
11. The method of claim 10, wherein the self-interference areas are
identified by identifying anomalies between the measured signal
strength of each node and a data throughput of each node.
12. The method of claim 1, wherein a single wireless link is
between the test client device and the node of the wireless
network.
13. The method of claim 1, wherein multiple wireless links are
between the test client device and the node of the wireless
network.
14. The method of claim 1, further comprising characterizing the
performance metric between the client device and a device that is
either a part of, or directly or indirectly connected to the
wireless network.
15. The method of claim 1, further comprising: sectoring an area of
the wireless network into grid points; associating each of the
locations with a grid point; designating grid points that
correspond with locations having a data throughput greater than a
threshold as passing grid points; calculating a percentage wireless
network coverage by comparing the passing grid points with a total
number of grid points in which the performance metric was
measured.
16. The method of claim 1, further comprising recording
identifications of nodes of the wireless network and reporting poor
and impaired nodes.
17. The method of claim 16, wherein poor nodes are identified as
nodes that have a lowest data throughput, or provide the least
coverage.
18. The method of claim 1, further comprising real-time displaying
of measured performance metrics.
19. The method of claim 18, wherein real-time displaying of the
measured performance metrics comprises showing the real-time
measured performance metrics on a map.
20. The method of claim 19, wherein a data throughput is indicated
on the map by providing different colors on the map for different
levels of data throughput.
Description
RELATED APPLICATIONS
[0001] This patent application is a continuation of U.S. patent
application Ser. No. 11/638,274, filed Dec. 13, 2006.
FIELD OF THE INVENTION
[0002] The invention relates generally to wireless communications.
More particularly, the invention relates to a method and apparatus
for determining coverage of a wireless network.
BACKGROUND OF THE INVENTION
[0003] Wireless mesh networks are gaining popularity because
wireless infrastructures are typically easier and less expensive to
deploy than wired networks. The wireless mesh networks typically
include wired gateways that are wirelessly connected to wireless
nodes, or wirelessly connected directly to client devices. Many
wireless nodes can collectively form a wireless mesh, in which
client devices can associate with any of the wireless nodes.
[0004] Wireless networks, however, can be more difficult to deploy
and maintain than wired networks. That is, wireless networks are
typically subjected to environmental influences that make operation
of the networks more problematic than wired networks. For example,
the wireless links of wireless networks can suffer from fading or
multi-path, which degrade the quality of transmission signals
traveling through the wireless links. Additionally, wireless
networks that include multiple access points can suffer from
self-interference.
[0005] During deployment of wireless networks, it is useful to have
information regarding fading, multi-path and self-interference to
allow for better deployment. Additionally, this information is
useful after deployment to allow determination of the quality of
possible network connections, and to provide information that can
be used to improve the wireless networks.
[0006] It is desirable to have a method and apparatus for
determining coverage a wireless network, allowing a system operator
to make strategic upgrades to the network. It is also desirable to
have measured network information available so that potential users
can determine a level of expected performance.
SUMMARY OF THE INVENTION
[0007] One embodiment of the invention includes a method of
determining coverage of a wireless network. The method includes
traveling to multiple locations around access points of the
wireless network, and for a plurality of client applications,
measuring a performance metric between a test client device and
nodes of the wireless network, at a plurality of the multiple
locations. The performance metric can be, for example, a
transmission data throughput, a signal jitter or latency.
[0008] Another embodiment includes a test client device that can be
used for characterizing a wireless network. The test client device
includes a means for traveling proximate to access nodes of the
wireless network. The test client device characterizes a
transmission data throughput between the wireless client test
device an access nodes of the wireless network, at multiple
locations, for a plurality of client applications.
[0009] Other aspects and advantages of the present invention will
become apparent from the following detailed description, taken in
conjunction with the accompanying drawings, illustrating by way of
example the principles of the invention.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] FIG. 1 shows a test device traveling around a wireless mesh
network measuring a performance parameter of the wireless mesh
network.
[0011] FIG. 2 shows the network of FIG. 1, having been sectored
into grids.
[0012] FIG. 3 is a map that shows levels of performance of the
wireless network as determined by the methods of characterizing the
wireless network.
[0013] FIG. 4 is a flow chart that includes steps of an embodiment
of a method for determining data throughput of a wireless
network.
DETAILED DESCRIPTION
[0014] As shown in the drawings for purposes of illustration, the
invention is embodied in methods for measuring a performance
parameter of a wireless network. The methods provide the
performance parameter measurements for different client
applications. The performance parameter measurements can be used by
network operators to improve performance of the wireless network.
Additionally, users and potential users can use results of the
performance parameter measurements to predict a level of expected
performance.
[0015] FIG. 1 shows a test device 110 traveling around a wireless
mesh network, and measuring a performance parameter (such as,
transmission data throughput) at various test locations (each
location depicted, for example, by an "X") of the wireless mesh
network. An exemplary wireless network includes a wireless mesh
network that includes gateways 120, 122, 124 and access nodes 130,
131, 132, 133, 134, 135, 136, 137. The gateways 120, 122, 124 can
be wired or wirelessly connected to a wired network 140. The wired
network 140 can be connected, for example, to the internet. The
test locations shown in FIG. 1 are merely an example of possible
test locations. Clearly, more, fewer, or different test locations
can be used for characterizing the data throughput of the wireless
mesh network. Generally, the denser the test locations are, the
better the characterization of the network.
[0016] Client devices (not shown) can wirelessly connect to any of
the gateways 120, 122, 124 and access nodes 130, 131, 132, 133,
134, 135, 136, 137, and therefore, obtain access to the wired
network 140. As previously described, the wireless links between
the gateways, access nodes and client devices, typically suffer
from some level of fading, multi-path and interference.
Characterizations of the wireless network can be useful in
determining whether new gateways and access nodes should be added
to the network, whether present gateways and access nodes should be
moved, or eliminated. Additionally, the characterizations can be
useful in predicting a level of expected performance.
[0017] The test client device 110 is used to characterize
performance of the wireless network. As shown, the test client
device 110 can be included within an automobile, for example, that
drives around and within the network while the test client device
characterizes network performance. The test client device 110 can
travel a path 160, for example, and test performance of the
wireless network at the various locations within the wireless mesh
network.
[0018] The test client device can obtain its location information
through a location determination service. The location
determination service can be, for example, a global positioning
system (GPS). As will be described, the test client device can
record its location at each point a network measurement is made.
The location information can be retrieved for display of
performance of the network.
[0019] The described methods of characterizing the wireless network
can provide a representation of the coverage of the wireless
network. Generally, the coverage of the wireless network can be
defined as performance (for example, data throughput) exceeding a
pre-determined threshold, and the coverage is the fraction of the
area of the network that exceeds the pre-determined threshold.
[0020] Characterizing Network Performance
[0021] One embodiment includes characterizing the performance of
the wireless network by measuring the data throughput at the
various locations within the network. The data throughput can
provide a better indication of the health of the network than just
transmission signal strength.
[0022] Testing throughput offers advantages over other performance
metrics that can be measured. For example, data throughput
measurements account for wireless network limitations such as
self-interference, which measuring signal strength does not
provide. Transmission signal impairments (such as multi-path and
fading) can be reflected in transmission data throughput test.
These signal impairments, however, may not be reflected in, for
example, a transmission signal strength measurement.
[0023] Transmission data throughput tests can be bi-directional
tests. That is, the throughput tests can include both the uplink
and the downlink between the test client device and the node of the
network being tested. Signal strength measurement tests are
typically a measurement of signal strength from the test node to
the test client device, and do not provide information of
transmission quality of the uplink between the test client device
and the test node.
[0024] The data packets of throughput tests can be configured to
mimic different client applications. Therefore, the performance
metric being measured can more accurately reflect the performance
of the wireless network for each of the client application types,
than a transmission signal strength measurement.
[0025] Additionally, transmission data throughput measurements are
sensitive to air-time congestion. That is, within a network having
many transmitting devices, the air-time available for transmission
can become occupied, and can therefore, affect performance of a
client connection to the network. Signal strength measurements
however, do not reflect the air-time congestion of the network like
data throughput measurements.
[0026] An embodiment for characterizing the data throughput
includes the test client device transmitting a plurality of probe
packets at a predetermined rate, and counting a number of responses
from the nodes. For an 802.11 wireless network, the test client
device transmits probe requests at a predetermined rate, and counts
probe responses.
[0027] An embodiment includes each of the nodes of the mesh network
being configuring with a common IP address. Each of the nodes may
have an individual IP address, but additionally have the common IP
address that can be used by the test client device for testing the
network. Therefore, the test client device does not have to know
the address of each node, improving mobility. Additionally, the
test client device does not have to reconfigure for testing of each
of the nodes.
[0028] Types of Client Applications
[0029] An embodiment includes determining the data throughput at
various locations of the wireless network for multiple types of
client applications. For example, client applications can include
data, voice, video, streaming data and low demanding applications,
such as, automatic meter reading.
[0030] The different types of client applications typically have
different data types. Therefore, if the wireless network is being
used for multiple types of client applications, it is useful to
measure the performance parameters, such as, the transmission data
throughput using the different data types of the different client
applications. Basically, the data packets can be configured to
mimic data packets of each of the plurality of client
applications.
[0031] The different data types can have, for example, different
data packet sizes. The size of a packet indicates the number of bit
included within the packet, which can vary depending upon the
client application. Additionally, the distribution of the data
packets can vary. That is, the number of bit within each packet can
vary from packet to packet during transmission. Additionally, the
data packets can be transmitted at different intervals. That is,
the time period between packets or the packets transmitted per
second can be adjusted. Different client applications can include
different transmission retry processes and age retries differently.
Therefore, measuring the performance parameters at different
intervals can be useful in characterizing the network for the
different client applications.
[0032] Network Interference Characterizations
[0033] An embodiment includes additionally measuring transmission
signal power throughout the wireless network. Based on anomalies
between the measured transmission signal power and the measured
transmission data throughput, areas of interference can be
identified. That is, generally higher power transmission signals
can provide a higher transmission data throughput because higher
power signals generally have a higher SNR. However, if the
transmission signal power is relatively higher, yet the
transmission data throughput is relatively lower, it can be assumed
that the transmission signal are being subjected to substantial
amounts of interference.
[0034] Characterizing Multiple Links
[0035] An embodiment includes characterizing the data throughput
for multiple links. That is, in addition to testing throughput
between the client test device and an access node the client device
is associated with, the throughput tests can be between the test
client device and any end-point device associated with the network,
or even outside of the network. As previously shown and described,
wireless mesh networks include many devices that can be wired or
wirelessly linked. An end-point device can be specified by an
address, such as, an IP address. The client test device can
exchange data with the end-point device according to the
transmission protocol of the client application being tested. In
one embodiment, the client test device sends requests to the
end-point device, and counts the responses from the end-point
device to determine an end-to-end data throughput value.
[0036] Presenting Results of the Characterization
[0037] An embodiment for reporting the results of the
characterization includes dividing (sectoring) an area around the
wireless network into grids. Performance of the wireless network
within each grid can be estimated based upon, for example, the
results of the throughput measurements for each of the client
application types. FIG. 2 shows the network of FIG. 1, having been
sectored into grids. A value of data throughput for each sector as
defined by the grids can be determined based on the data throughput
measurements by the test client device. As previously mentioned,
the data throughput is characterized at several locations. Based
upon the proximity of each of the locations with respect to the
areas of the grids, data throughput estimates can be made for each
of the grids.
[0038] As shown in FIG. 2, transmission data throughput
characterized, for example, at test locations 210, 212, 214, 216,
218 can be used to estimate the transmission data throughput in
each of the corresponding grid locations or grid points. The
accuracy of the characterizations is dependent upon the size of the
grids, and the number (density) of the test locations. A grid point
(or grid location) can include more than one throughput
characterization. A value for the grid point, can therefore, by
influenced by more than one characterization location.
[0039] The determination of the throughput at each grid location
can be determined by several throughput measurements. For one
embodiment, measurements that are clearly erroneous are filtered
(removed). The mean of the remaining measurements of a grid
location can be determined, or an average of the remaining
measurements can be used to estimate the throughput at each grid
location.
[0040] The other measure parameters (for example, latency and
jitter) can be estimated at each of the grid locations as well.
[0041] A coverage quality of the wireless network can be determined
(and therefore, represented) by designating grid points that
correspond with locations having a data throughput greater than a
threshold as passing grid points, and calculating a percentage
wireless network coverage by comparing the passing grid points with
a total number of grid points characterized. The resulting
percentage can be used to represent the coverage quality of the
network.
[0042] Additionally, nodes that provide poor performance and nodes
that are determined to be impaired, can be identified and recorded.
Nodes can be identified as poor performers if, for example, they
have a low transmission data throughput or are identified to have a
limited transmission range (poor coverage). An operator of the
network can use this information to improve the wireless network by
moving, adding or removing nodes within the wireless network.
[0043] Results of the data throughput characterizations can be
stored for future access and analysis. For example, the results at
each tested location can be stored at in a server that is connected
to the network. A network operator can gage the health of the
network by accessing the performance parameter measurements.
Additionally, potential future users can access the measured
parameters to determine a level a quality of connection to the
network the potential user could expect.
[0044] FIG. 3 is a map that shows levels of performance (as
determined by prior measurements of performance parameters) of the
wireless network as determined by the methods of measuring
performance parameters the wireless network. As shown, the path 360
traveled by the client test device 110 includes varying levels of
measured performance. That is, as shown in FIG. 3, the thickness of
the path 360 varies depending upon the value of the measured
transmission data throughput. The visual depiction on the map of
the value of the measured performance metric can be presented in
many different ways. For example, the map can include varying
colors in which each color represents a predetermined level of
measured performance metric. The map can include representations at
the points (locations) that measurements were actually made, or the
map can additionally fill-in between measurement locations with
values that are interpolated from between measured locations.
[0045] The map shown in FIG. 3 only shows the measured values over
the path 360 in which the measurements were made. However, as
described, the map can be filled in between measured values by, for
example, interpolating the measured values at the locations of the
path 360 in which actual measurements were made. It should be
understood that any type of visual depiction can be used as long as
different values of performance are distinguishable from each
other.
[0046] An embodiment includes a real-time display of measured
performance metrics. That is, the performance metrics are displayed
as they are being measured. The real-time performance metrics can
be displayed, for example, by showing the real-time measured
performance metrics on a map. The data throughput can be indicated
on the map by providing different colors on the map for different
levels of data throughput.
[0047] Storage and Accessing of the Results of the
Characterization
[0048] Various embodiments provide methods of storing the
characterized data throughput, and subsequently providing
characterizations for potential users of the network.
[0049] The results of the data throughput characterizations can be
stored in a server that potential users of the wireless network can
access to determine what level of network connection performance
they could expect.
[0050] FIG. 4 is a flow chart that includes steps of an embodiment
of a method for determining coverage of a wireless network. A first
step 410 includes traveling to multiple locations around access
points of the wireless network. Around the access points can be
defined as being within the range of the access point. Around the
access points can also be defined as within the range of the access
points, and just outside the range of the access points. A second
step 520 includes for a plurality of client applications, measuring
a performance metric between a test client device and nodes of the
wireless network, at a plurality of the multiple locations.
Exemplary performance metrics include transmission data throughput,
jitter and/or latency.
[0051] One embodiment for characterizing the data throughput at
each of the plurality of locations includes the test client device
transmitting a plurality of probe packets at a predetermined rate,
and counting a number of responses from the nodes. The latency can
be determined by timing response, and determining the timing
between the test client device and the node being tested. Jitter is
determined by the variation in the latency between multiple
responses.
[0052] Ping packets can be use for determining throughput of an
802.11 link between an access test client device and an access node
of a wireless mesh network. First, the test client device
intentionally corrupting ping packets by manipulating a checksum
(layer 3) of the ping packets while maintaining a proper CRC of the
ping packets. The test client sends the corrupted ping packets to
the access node. The test client device estimates a number of
packets received by the access node by counting how many layer 2
responses the test client device receives back from the access
node. The throughput is estimated by multiplying the number of
packets received by a size of the packets, and dividing a result by
a transmit time.
[0053] Ping Packets
[0054] TCP/IP protocols can facilitate in identification of network
problems. One of the most frequently used debugging tools invokes
ICMP echo request and echo reply messages. On many systems, the
command invoked to send ICMP echo requests is named "ping".
Sophisticated versions of ping send a series of ICMP echo requests,
capture responses, and provide statistics about datagram loss. The
ping allows specification of the length of the data being sent and
the interval between requests. Less sophisticated versions merely
send one ICMP echo request and await a reply.
[0055] The ping packets include a cyclic redundancy check (CRC) and
a checksum. The ping packets that have a corrupted checksum will
not generate a response at the second host. However, the layer two
CRC associated with 802.11 packets, do generate a response. The CRC
response only includes a very small number of bits, and therefore,
does not interfere with downlink throughput. The corrupted ping
packets do not generate a response.
[0056] The measurements can be made for the different client
applications can be measuring one or more of the measurement
parameters while setting the data packets of the transmissions
between the client test device and the node being tested according
to the protocols of each of the client applications. The can
include setting the packet size, packet type and packet interval.
The data packets during the measurement attempt to mimic the data
packets of the different client types.
[0057] Additional measurements, such as, signal strength can also
be made and used for determining areas of the wireless network that
are suffering from interference, such as, self-interference. These
areas can be identified by observing inconsistencies or anomalies
between the signal strength measurements and the performance metric
measurements.
[0058] The performance metrics can be measured for a single link
between the client test device and the test node, or for multiple
links (wired and wireless) between the client test device and test
node.
[0059] For presenting the measured performance parameters, an area
of the wireless network can be sectored into grid points. Each of
the locations is associated with a grid point. Grid points that
correspond with locations having a measured performance parameter
greater than a threshold can be designated as passing grid points.
A percentage of coverage of the wireless network is calculated by
comparing the passing grid points with a total number of grid
points in which the performance metric was measured. Additionally,
poor performing and impaired nodes are identifies and recorded.
Poor performing nodes can be identified as nodes that have a lowest
data throughput, or provide the least coverage.
[0060] The measured performance metrics can be stored for future
access and analysis. One embodiment includes generating a database
that includes the locations and the measured performance metric.
The locations can be either the actual test locations, or the
locations can be test locations and locations in which performance
metrics are calculated rather than measured.
[0061] The database can be accessed (for example, through the
internet) for determining how well the wireless network is working,
or to provide a potential user with an estimated expected
performance. That is, based on a requested location, the database
can be accessed, and a projected level data throughput can be
estimated at the requested location based on the locations and
measured performance metric of the database.
[0062] One method of providing the measured and calculated
performance metrics is to visually depict the values of the
performance metrics on a map of the area in which the wireless
network is located. For example, different levels of performance
can be depicted with different colors.
[0063] Although specific embodiments of the invention have been
described and illustrated, the invention is not to be limited to
the specific forms or arrangements of parts so described and
illustrated. The invention is limited only by the appended
claims.
* * * * *