Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant’s arguments, see page 10, lines 14-17, filed 9/16/2025, with respect to claim 21 have been fully considered and are persuasive. The 35 U.S.C. § 101 rejection of claim 21 has been withdrawn.
Applicant’s arguments, see page 11, lines 1-4 , filed 9/16/2025, with respect to claim 11 have been fully considered and are persuasive. The 35 U.S.C. § 112 rejection of claim 11 has been withdrawn.
Applicant's arguments filed 9/16/2025, regarding the 35 U.S.C. § 103 rejection of the independent Claims 1, 19, and 21, have been fully considered but they are not persuasive. The applicant argues:
Independent Claims 1, 19, and 21 recite, inter alia: "selecting the new transmission patterns for the IoT devices comprising selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs."
…Sadek does not teach or suggest Applicant's above-emphasized features. Sadek's aligning is between the access point and the access terminals. This is shown in, e.g., Sadek's FIGS. 5 and 6, which show the pattern between the "access terminal DRX cycle" and the "access point CSAT cycle." In other words, Sadek's access point aligns between access terminals and itself, not between access terminals and devices other than the access point. Accordingly, Sadek does not describe selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs.
The examiner calls attention to the prior art of SADEK et al. (US 20150296560 A1). SADEK writes, “FIG. 1 illustrates an example wireless communication system including an Access Point (AP) in communication with an Access Terminal (AT). Unless otherwise noted, the terms “access terminal” and “access point” are not intended to be specific or limited to any particular Radio Access Technology (RAT). In general, access terminals may be any wireless communication device allowing a user to communicate over a communications network (e.g., a mobile phone, router, personal computer, server, entertainment device, Internet of Things (IOT)/Internet of Everything (IOE) capable device, in-vehicle communication device, etc.), and may be alternatively referred to in different RAT environments as a User Device (UD), a Mobile Station (MS), a Subscriber Station (STA), a User Equipment (UE), etc. Similarly, an access point may operate according to one or several RATs in communicating with access terminals depending on the network in which the access point is deployed, and may be alternatively referred to as a Base Station (BS), a Network Node, a NodeB, an evolved NodeB (eNB), etc. Such an access point may correspond to a small cell access point, for example. “Small cells” generally refer to a class of low-powered access points that may include or be otherwise referred to as femto cells, pico cells, micro cells, Wi-Fi APs, other small coverage area APs, etc. Small cells may be deployed to supplement macro cell coverage, which may cover a few blocks within a neighborhood or several square miles in a rural environment, thereby leading to improved signaling, incremental capacity growth, richer user experience, and so on” (paragraph 0024).
SADEK further explains, “While the foregoing disclosure shows various illustrative aspects, it should be noted that various changes and modifications may be made to the illustrated examples without departing from the scope defined by the appended claims. The present disclosure is not intended to be limited to the specifically illustrated examples alone. For example, unless otherwise noted, the functions, steps, and/or actions of the method claims in accordance with the aspects of the disclosure described herein need not be performed in any particular order. Furthermore, although certain aspects may be described or claimed in the singular, the plural is contemplated unless limitation to the singular is explicitly stated” (paragraph 0065).
Based on the teachings of SADEK and the citations above, the examiner finds the applicant’s arguments regarding the 35 U.S.C. § 103 rejections of independent claims 1, 19, and 21 not persuasive. Therefore, the rejections of the independent claims 1, 19, and 21 remain, and, likewise, the rejection of the dependent claims also remain as described below.
Claim Rejections - 35 USC § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 1, 17-19, and 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over SONG et al. (CN 107 820 321 A, hereinafter, "SONG") in view of CHU (US 10856228 B1, hereinafter, "CHU"), KAUR et al. (US 20170142741 A1, hereinafter, "KAUR"), and SADEK (US 20150296560 A1, hereinafter, "SADEK").
Regarding claim 19, SONG teaches a node in a communications network for scheduling
transmissions of a plurality of Internet of Things (IoT) devices to network resources in the
communications network (figure 1; paragraph 62; paragraph 74), the node comprising:
obtain transmission patterns of transmissions from the IoT devices;
SONG writes, “Step 1. For a narrowband Internet of Things community, when communicating between
each Internet of Things device and the base station, the base station monitors and collects the request
access information of each Internet of Things device in real time; The requested access information
includes: the length of the data packet to be transmitted by the IoT device, the maximum tolerance limit
of the device to time delay, and the power consumption value of the device when transmitting data”
(paragraph 14-15). SONG states the base station monitors and collects request access information of
each IoT device in real time. Therefore, SONG indicates the transmission patterns are obtained from the
IoT devices.
cluster the IoT devices into clusters based on the obtained transmission patterns;
SONG writes, “Step 3: The base station uses the K-means algorithm to iteratively group the IoT devices
requesting access” (paragraph 17). SONG continues, “Step 302, read the requested access information
of each IoT device, and sequentially calculate the energy consumption required by each IoT device to
transmit unit data, and store it in the array D” (paragraph 21). SONG concludes, “Step 4, when the
grouping of the IoT devices is completed, allocate free network resource blocks to each IoT device group” (paragraph 37). SONG specifies the bases station uses an algorithm to group the IoT devices,
reads the requested access information, such as the transmission patterns, of each IoT device, calculate energy consumption, and store it in the array. After the grouping has finished, resource blocks are
allocated to each IoT device group.
SONG fails to explicitly disclose information regarding, “a memory comprising instruction data
representing a set of instructions;”, “and a processor configured to communicate with the memory
and to execute the set of instructions,”, “the set of instructions, when executed by the processor,
causing the processor to:”, “and schedule transmissions of IoT devices in different clusters to
different network resources;”, “obtain transmission patterns for one or more User Equipments (UEs) that are making transmissions using the network resources;”, “select new transmission patterns for the IoT devices, the new transmission patterns being predicted to result in clustering with increased synchronisation of the transmission patterns of the IoT devices in the clusters,”, “selecting the new transmission patterns for the IoT devices comprising selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs;”, and “and repeat the clustering and scheduling for the new transmission patterns.”
However, in analogous art, CHU teaches a memory comprising instruction data representing a
set of instructions;
CHU writes, “Various embodiments and components disclosed herein are configured to be at least
partially operated and/or implemented by processor-executable instructions stored on one or more
transitory or non-transitory processor-readable media” (column 11, lines 60-64).
and a processor configured to communicate with the memory and to execute the set of
instructions,
CHU writes, “Various embodiments and components disclosed herein are configured to be at least
partially operated and/or implemented by processor-executable instructions stored on one or more
transitory or non-transitory processor-readable media” (column 11, lines 60-64).
the set of instructions, when executed by the processor, causing the processor to:
CHU writes, “Various embodiments and components disclosed herein are configured to be at least
partially operated and/or implemented by processor-executable instructions stored on one or more
transitory or non-transitory processor-readable media” (column 11, lines 60-64).
and schedule transmissions of IoT devices in different clusters to different network resources;
CHU writes, “In some implementations, the [lower power (LP)] device may request a scheduled start
time for transmission, an interval between two scheduled transmissions, and the AP may accordingly
group the LP devices based on the requested schedules of transmissions. For example, if LP devices A, B
and C are intended receivers for the first transmission, while LP devices D, E and F are intended receivers
for the second transmission, the AP may group devices A, B and C as members of the same group, and
devices D, E and F as members of a different group” (column 8, lines 10-19). CHU adds, “In a WLAN
environment, an LP device may constantly turn off its Wi-Fi module to save power when the LP device is
not in use. To maintain the connection between the AP and the LP device, duty-cycle protocols are
sometimes defined to synchronize the AP and the LP device...” (column 4, lines 30-34). CHU continues,
“For example, to increase batter[y] life, the LP device attempts to sleep more...” (column 4, lines 38-39).
CHU states the AP may accordingly group the lower power (LP) devices based on the requested
schedules of transmission. CHU specifies the LP device may constantly turn off its Wi-Fi module to save
power when the device is not in use, though, to maintain the connection between the device and the
AP, the duty-cycle protocols are sometimes defined to synchronize the AP and the LP device. CHU points
out that to increase battery life, the LP device attempts to sleep more.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG to include aspects described by CHU that “relates to managing lower power (LP) devices in a wireless communication system, and specifically, to transmitting a wake-up request (WUR) frame to LP devices in a wireless communication system.” CHU provides the motivation for modification stating, “A type of the wake-up data frame is determined from a type field in the wake-up data frame. Whether the wake-up radio packet is intended for the lower power wireless device based on information corresponding to one or more intended lower power wireless devices from an identifier field in the wake-up data frame is then determined. In response to determining that the wake-up radio packet is intended for the lower power wireless device, the primary wireless receiver is activated for data transmission from the wireless access point” (column 3, lines 28-37). CHU indicates through the disclosed method that battery life can be extended.
SONG and CHU fail to explicitly disclose information regarding, “obtain transmission patterns for one or more User Equipments (UEs) that are making transmissions using the network resources;”, “select new transmission patterns for the IoT devices, the new transmission patterns being predicted to result in clustering with increased synchronisation of the transmission patterns of the IoT devices in the clusters,”, “selecting the new transmission patterns for the IoT devices comprising selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs;”, and “and repeat the clustering and scheduling for the new transmission patterns.”
However, in analogous art, KAUR teaches obtain transmission patterns for one or more User Equipments (UEs) that are making transmissions using the network resources;
KAUR writes, “A D2D enabled WTRU may perform a resource selection procedure to determine
resources for D2D transmission and/or reception. The D2D enabled WTRU may determine the D2D
resource to be a physical resource block (PRB) in time and frequency for D2D operations. A D2D enabled
WTRU may be configured with one or more D2D resources, such as a D2D resource pool. A D2D enabled
WTRU may select, based on a D2D resource selection, a pattern of transmission and reception
opportunities from the resource pool in a defined duration. The pattern may include a D2D transmission
pattern and/or a D2D reception pattern” (paragraph 0121).
select new transmission patterns for the IoT devices, the new transmission patterns being predicted to result in clustering with increased synchronisation of the transmission patterns of the IoT devices in the clusters,
KAUR writes, “A WTRU may be configured to determine when to use a pattern (e.g., pool) based on
detecting triggers. A WTRU may receive an explicit indication to use a transmission pattern (e.g.,
transmission pool) or a reception pattern (e.g., reception pool) according to one or more criteria. For
example, when the synchronization message (e.g., the synchronization message of the selected
synchronization source or relay) includes a pattern, and/or the WTRU may explicitly receive a pattern
from a base station, an eNB or from another controlling entity. When a WTRU receives an explicit
indication, the WTRU may initiate the use of the pattern and/or associated pool” (paragraph 0231).
KAUR adds, “The WTRU may, for example, start/stop relaying synchronization based on certain
conditions. An example might be that the WTRU may select a synchronization source based on certain
conditions or the WTRU may select a resource pool based on the same or other conditions” (paragraph
0085). KAUR mentions, “The WTRU may include the resource pool it expects to use for transmission
and/or reception and/or is using in the cluster configuration message, for example, for the
synchronization message” (paragraph 0395). KAUR states that a WTRU may receive an explicit indication
to use a transmission pattern according to one or more criteria. The WTRU may explicitly receive a
pattern from a base station, an eNB or from another controlling entity. KAUR specifies the WTRU may
select a synchronization source based on certain conditions or the WTRU may select a resource pool
based on the same or other conditions. KAUR also points out the WTRU may include the resource pool it
expects to use for transmission and/or reception and/or is using in the cluster configuration message.
and repeat the clustering and scheduling for the new transmission patterns.
KAUR writes, “In particular, FIG. 1F illustrates an example of a distributed synchronization scheme where
a serving base station (S-BS) may broadcast an indication of one or more pools of D2D resources to one
or more WTRUs for discovery and communication within its own cell (e.g., cell 1), for reception and/or
transmission. A receiver disposed in a WTRU may determine a time and/or frequency reference, for
example, such that the receiver window and frequency correction may be aligned when detecting D2D
channels and/or the transceiver timing and parameters may be aligned when transmitting D2D
channels. As shown in FIG. 1F, example embodiments described below provide information to WTRUs
that are exposed to D2D channels transmitted by neighbor WTRUs camping on different cells (e.g., N-BS
in Cell No. 2), PLMNs (not shown) and clusters (e.g., SCH Synchronization Cluster No. 1). WTRUs camping
on neighbor cells, PLMNs, and/or clusters may signal their own D2D resource pool along their
synchronization reference, in order to allow detection of the D2D communication resources” (figure 1F;
paragraph 0084). KAUR indicates that a serving base station (S-BS) may broadcast an indication of one or
more pools of D2D resources to one or more WTRUs for discovery and communication within its own
cell (e.g., cell 1), for reception and/or transmission, and the transceiver timing and parameters may be
aligned when transmitting D2D channels. Thus, KAUR implies that after new transmission patterns are
introduced cluster and scheduling will repeat.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG and CHU to include aspects described by KAUR that is “related to power control when changing coverage state. The WTRU may be configured to determine the D2D power control mode, for example, based on the D2D WTRU coverage situation.” KAUR provides the motivation for modification stating, “With respect to supporting public safety communications, there may be a need to harmonize the radio access technology across jurisdictions and lower the cost of radio-access technology to public safety (PS) officials.” (paragraph 0003). KAUR adds, "D2D communications may be available for commercial uses, for example utility companies or the like. For example, D2D communication techniques may be used for enhancing communications in areas that may have poor coverage from network infrastructure. Commercial and social users may request D2D communications, for example, to assure the consistency of their user experience. For example, a user may request D2D communications to improve the user's reachability and/or mobility" (paragraph 0004).
SONG, CHU, and KAUR fail to explicitly disclose information regarding, “selecting the new transmission patterns for the IoT devices comprising selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs;”
However, in analogous art, SADEK teaches selecting the new transmission patterns for the IoT devices comprising selecting new transmission patterns for the IoT devices so as to increase a measure of overlap between the transmission patterns of the IoT devices and the transmission patterns of the one or more UEs;
SADEK writes, “For example, the access point may align the DRX pattern of one or more high-traffic
access terminals (e.g., an access terminal served in the 5 GHz band, such as on a frequency in the range
of 5.15 GHz to 5.725 GHz, which is generally associated with high traffic) with the CSAT TDM
communication pattern to maximize or at least increase the overlap between the DRX ON period and
the CSAT ON (activated) period, thereby increasing transmission opportunities and overall throughput
for the access terminals” (paragraph 0045). SADEK specifies that the access point may align the DRX
pattern with the CSAT TDM communication pattern to maximize or at least increase the overlap
between the DRX ON period and the CSAT ON period to increase transmission opportunities and overall
throughput for the access terminals.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG, CHU, and KAUR to include aspects described by SADEK that “relate generally to telecommunications, and more particularly to co-existence between wireless Radio Access Technologies (RATs) and the like.” SADEK provides the motivation for modification stating, “In cellular networks, “macro cell” access points provide connectivity and coverage to a large number of users over a certain geographical area. A macro network deployment is carefully planned, designed, and implemented to offer good coverage over the geographical region. To improve indoor or other specific geographic coverage, such as for residential homes and office buildings, additional “small cell,” typically low-power access points have recently begun to be deployed to supplement conventional macro networks. Small cell access points may also provide incremental capacity growth, richer user experience, and so on” (paragraph 0004). SADEK adds, "Small cells may be deployed to supplement macro cell coverage, which may cover a few blocks within a neighborhood or several square miles in a rural environment, thereby leading to improved signaling, incremental capacity growth, richer user experience, and so on" (paragraph 0024).
Regarding claim 17, SONG, CHU, KAUR, and SADEK teach the method as in claim 1,
Additionally, CHU teaches wherein clustering the IoT devices based on the obtained
transmission patterns comprises clustering the IoT devices based on one or more from a group
consisting of: performance patterns; network access patterns; coverage areas; data rates; and
criticality of the IoT devices.
CHU writes, “FIG. 1 is a block diagram 100 illustrating an example WUR data frame defined by the basic
service sets (BSS) coloring, according to some embodiments described herein. In 802.11 WLANs, a
service set is defined as a group of wireless network devices that are operating with the same
networking parameters. The service sets are arranged hierarchically, basic service sets (BSS) are defined
as units of devices operating with the same medium access characteristics (i.e. radio frequency,
modulation scheme etc.)” (column 4, lines 64-67; column 5, lines 1-5). CHU explains that a service set is
defined as a group of wireless network devices that are operating with the same networking
parameters.
Regarding claim 18, SONG, CHU, KAUR, and SADEK teach the method as in claim 1,
Additionally, CHU teaches wherein the increased periods of inactivity comprise one or both of
more frequent periods of sleep mode and longer periods of sleep mode.
CHU writes, “In some implementations, the [lower power (LP)] device may request a scheduled start
time for transmission, an interval between two scheduled transmissions, and the AP may accordingly
group the LP devices based on the requested schedules of transmissions. For example, if LP devices A, B
and C are intended receivers for the first transmission, while LP devices D, E and F are intended receivers
for the second transmission, the AP may group devices A, B and C as members of the same group, and
devices D, E and F as members of a different group” (column 8, lines 10-19). CHU adds, “In a WLAN
environment, an LP device may constantly turn off its Wi-Fi module to save power when the LP device is
not in use. To maintain the connection between the AP and the LP device, duty-cycle protocols are
sometimes defined to synchronize the AP and the LP device...” (column 4, lines 30-34). CHU continues,
“For example, to increase batter[y] life, the LP device attempts to sleep more...” (column 4, lines 38-39).
CHU states the AP may accordingly group the lower power (LP) devices based on the requested
schedules of transmission. CHU specifies the LP device may constantly turn off its Wi-Fi module to save
power when the device is not in use, though, to maintain the connection between the device and the
AP, the duty-cycle protocols are sometimes defined to synchronize the AP and the LP device. CHU points
out that to increase battery life, the LP device attempts to sleep more.
Claims 1 and 21 are method and memory claims corresponding to the apparatus claim 19 that
has already been rejected above. The applicant’s attention is directed to the rejection of claim 19.
Claims 1 and 21 are rejected under the same rational as claim 19.
Claim(s) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable over SONG, CHU, KAUR, and SADEK as applied to claim 1 above, and further in view of LE HOUEROU et al. (US 20110103248 A1, hereinafter, "LE HOUEROU").
Regarding claim 3, SONG, CHU, KAUR, and SADEK teach the method as in claim 1,
SONG, CHU, KAUR, and SADEK fail to explicitly disclose information regarding, “wherein the new
transmission patterns comprise perturbations of the obtained transmission patterns and wherein the
perturbations are based on predetermined flexibilities associated with the transmission patterns of
the IoT devices.”
However, in analogous art, LE HOUEROU teaches wherein the new transmission patterns comprise perturbations of the obtained transmission patterns and wherein the perturbations are based on predetermined flexibilities associated with the transmission patterns of the IoT devices.
LE HOUEROU writes, “According to one advantageous characteristic, the method for managing
comprises steps of: determining, for at least one transmission configuration, a number of second sender
devices for which said rate or rates of estimated disturbance are below said predefined threshold;
selecting a configuration from among said plurality of transmission configurations as a function of the
determined number or numbers of second sender devices for which the estimated rate or rates
of disturbance are below said predefined threshold; configuring the first sender device and a receiver
device according to the selected transmission configuration” (paragraphs 0049-0052). LE HOUEROU
states that for at least one transmission configuration, determining the rate of estimated disturbance
are below said predefined threshold. Therefore, LE HOUEROU indicates that if the new transmission
patterns comprise perturbations, that at least one new transmission contain perturbations below
said predefined threshold.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG, CHU, KAUR, and SADEK to include aspects described by LE HOUEROU that “relates to a technique for managing communications in a wireless communications network. Such a technique is aimed at setting up simultaneous communications in the network.” LE HOUEROU provides the motivation for modification stating, “the use of a narrow reception angle 520 either increases the power of the radio signal at input of the receiver device (positive gain), thus increasing the distance of transmission or, for equal distances, increases the signal-to-noise ratio (SNR). The radio signal reception quality is thereby improved and the error rate of the transmission channel is reduced” (paragraph 0121).
Claim(s) 5-6, 13, and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over SONG, CHU, KAUR, and SADEK as applied to claim 1 above, and further in view of ABDALA et al. (US 20190141146 A1, hereinafter, "ABDALA").
Regarding claim 5, SONG, CHU, KAUR, and SADEK teach the method as in claim 1,
SONG, CHU, KAUR, and SADEK fail to explicitly disclose information regarding, “wherein selecting new transmission patterns for the IoT devices comprises:”, “using a model trained using a machine learning process to select the new transmission patterns,”, and “wherein the model takes as input transmission parameters of the IoT devices and outputs the new transmission patterns for the IoT devices, based on the transmission parameters.”
However, in analogous art, ABDALA teaches wherein selecting new transmission patterns for
the IoT devices comprises:
ABDALA writes, “In the illustrative embodiment, the system uses the trained machine-learning-model to
predict an effective transmission pattern for a new communication” (paragraph 0021).
using a model trained using a machine learning process to select the new transmission
patterns,
ABDALA writes, “In the illustrative embodiment, the system uses the trained machine-learning-model to
predict an effective transmission pattern for a new communication” (paragraph 0021).
wherein the model takes as input transmission parameters of the IoT devices and outputs the
new transmission patterns for the IoT devices, based on the transmission parameters.
ABDALA writes, “If the various communications are sufficiently similar, the system can then use these
effective transmission patterns as training data in order to train a machine-learning-model to predict an
effective transmission pattern for a new communication not yet transmitted to the user. The system can
determine if communications are sufficiently similar by analyzing one or more attributes of the
communications for a preset combination of common characteristics, such as type, length, or keywords”
(paragraph 0019). ABDALA continues, “The system can then use those effective transmission patterns to
train the machine-learning-model...” (paragraph 0020). ABDALA adds, “In the illustrative embodiment,
the system uses the trained machine-learning-model to predict an effective transmission pattern for a
new communication” (paragraph 0021). ABDALA concludes, “In block 310, the processing device 202
trains one or more machine-learning-models 120 using the training data. Training a machine-learning-
model 120 can involve tuning weights for nodes in the machine-learning model in order to transform the
machine-learning-model 120 from an untrained state into a trained state. Once trained, the machine-
learning-model(s) 120 can receive an input (e.g., a piece of content or characteristics of a piece of
content) and provide a transmission pattern as output” (paragraph 0047). ABDALA specifies that
transmission patterns may be used as training data in order to train a machine-learning-model to predict
an effective transmission pattern for a new communication not yet transmitted to the user. That the
system can determine if communications are sufficiently similar by analyzing one or more attributes of
the communications for a preset combination of common characteristics, and the system will use those
effective transmission patterns to train the machine-learning-model. ABDALA explains the system uses
the trained machine-learning-model to predict an effective transmission pattern for a new
communication. ABDALA indicates training a machine-learning-model can involve tuning weights for
nodes in the machine-learning model in order to transform the machine-learning-model from an
untrained state into a trained state. Once trained, the machine-learning-model(s) can receive an input
(e.g., a piece of content or characteristics of a piece of content) and provide a transmission pattern as
output.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG, CHU, KAUR, and SADEK to include aspects described by ABDALA that “relates generally to managing data transmissions. More specifically, but not by way of limitation, this disclosure relates to managing data transmissions based on a user's digital footprint.” ABDALA provides the motivation for modification stating, “...by using a user's digital footprint to determine an optimal transmission pattern at which to transmit data to the user. For example, a server can receive data from multiple sources about the user's Internet activities in order to develop a digital footprint for the user.” (paragraph 0014). ABDALA continues, "The server can use these insights to optimize transmission of similar datasets to the user's devices in the future (e.g., rather than transmitting the datasets indiscriminately). This may prevent the unnecessary consumption of computing resources" (paragraph 0014).
Regarding claim 6, SONG, CHU, KAUR, SADEK, and ABDALA teach the method as in claim 5,
Additionally, ABDALA teaches wherein the model is trained to output transmission patterns for
the IoT devices that optimise a number of clusters one or both of obtained when clustering the IoT
devices and that are predicted to satisfy performance requirements of the IoT devices.
ABDALA writes, “The system can then use those effective transmission patterns to train the machine-
learning-model... (paragraph 0020). In the illustrative embodiment, the system uses the trained
machine-learning-model to predict an effective transmission pattern for a new communication
(paragraph 0021). For example, the server 112 can use classifiers (e.g., Naïve bias classifiers) to classify
content into various groups, whereby each content group has common characteristics. The server 112
then generates separate training data from each content group. The training data can include
relationships between (i) the common characteristics defining a content group and (ii) the effective
transmission patterns corresponding to the content in the content group. The server 112 can then use
the training data for each content group to train separate machine-learning-models 120 tailored to the
respective content groups (paragraph 0029). ABDALA indicates the system will use effective
transmission patterns to train the machine-learning-model. ABDALA explains the system uses the
trained machine-learning-model to predict an effective transmission pattern for a new communication.
ABDALA specifies that classifiers may be used to classify content into various groups with common
characteristics, and that separate training data may be generated from each content group. Where the
training data may be used for each content group to train separate machine-learning-models tailored to
the respective content groups.
Regarding claim 13, SONG, CHU, KAUR, SADEK, and ABDALA teach the method as in claim 5,
Additionally, ABDALA teaches wherein the model takes as input transmission parameters
comprising one or more from a group consisting of: the obtained transmission patterns; flexibility
associated with the obtained transmission patterns obtained; service level agreements; and locations;
of the IoT devices.
ABDALA writes, “If the various communications are sufficiently similar, the system can then use these
effective transmission patterns as training data in order to train a machine-learning-model to predict an
effective transmission pattern for a new communication not yet transmitted to the user. The system can
determine if communications are sufficiently similar by analyzing one or more attributes of the
communications for a preset combination of common characteristics, such as type, length, or keywords”
(paragraph 0019). ABDALA continues, “The system can then use those effective transmission patterns to
train the machine-learning-model...” (paragraph 0020). ABDALA adds, “In the illustrative embodiment,
the system uses the trained machine-learning-model to predict an effective transmission pattern for a
new communication” (paragraph 0021). ABDALA concludes, “In block 310, the processing device 202
trains one or more machine-learning-models 120 using the training data. Training a machine-learning-
model 120 can involve tuning weights for nodes in the machine-learning model in order to transform the
machine-learning-model 120 from an untrained state into a trained state. Once trained, the machine-
learning-model(s) 120 can receive an input (e.g., a piece of content or characteristics of a piece of
content) and provide a transmission pattern as output” (paragraph 0047). ABDALA specifies that
transmission patterns may be used as training data in order to train a machine-learning-model to predict
an effective transmission pattern for a new communication not yet transmitted to the user. That the
system can determine if communications are sufficiently similar by analyzing one or more attributes of
the communications for a preset combination of common characteristics, and the system will use those
effective transmission patterns to train the machine-learning-model. ABDALA explains the system uses
the trained machine-learning-model to predict an effective transmission pattern for a new
communication. ABDALA indicates training a machine-learning-model can involve tuning weights for
nodes in the machine-learning model in order to transform the machine-learning-model from an
untrained state into a trained state. Once trained, the machine-learning-model(s) can receive an input
(e.g., a piece of content or characteristics of a piece of content) and provide a transmission pattern as
output.
Regarding claim 16, SONG, CHU, KAUR, SADEK, and ABDALA teach the method as in claim 5,
Additionally, ABDALA teaches wherein the model comprises a neural network model.
ABDALA writes, “Examples of machine-learning-models 120 can include neural networks, support vector
machines, and classifiers” (paragraph 0029).
Claim(s) 7-12 and 14-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over SONG, CHU, KAUR, SADEK, and ABDALA as applied to claim 5 above, and further in view of PANTELIDOU et al. (US 20220295324 A1, hereinafter, "PANTELIDOU").
Regarding claim 7, SONG, CHU, KAUR, SADEK, and ABDALA teach the method as in claim 5, further comprising:
Additionally, ABDALA teaches providing a first feedback to the model based on the clusters;
ABDALA writes, “In block 308, the processing device 202 generates training data that includes a
relationship between (i) one or more characteristics of the first content, and (ii) the first transmission
pattern. For example, the processing device 202 can build a dataset that includes a relationship between
a combination of keywords in the first content and the first transmission pattern (e.g., the number of
times and/or frequency at which the first content was transmitted). The dataset can serve as the
training data” (paragraph 0045). ABDALA states the processing device generates training data that
includes a relationship between (i) one or more characteristics of the first content, and (ii) the first
transmission pattern. ABDALA already specified that classifiers may be used to classify content into
various groups with common characteristics, and that separate training data may be generated from
each content group. Therefore, ABDALA indicates a first feedback to the model based on the clusters.
SONG, CHU, KAUR, SADEK, and ABDALA fail to explicitly disclose information regarding, “and re-training the model, using the first feedback, to output transmission patterns that increase synchronization of the transmissions scheduled on each network resource.”
However, in analogous art, PANTELIDOU teaches and re-training the model, using the first
feedback, to output transmission patterns that increase synchronisation of the transmissions
scheduled on each network resource.
PANTELIDOU writes, “After receiving 1550, 1552 the new filter sets, the gNB 1510 can use them
for retraining 1554 its ML optimization algorithms. ML algorithms may compute a cost or reward which
may determine how well the ML algorithm performs” (paragraph 0145). PANTELIDOU adds, “If the cost
or reward is farther from the optimal cost or reward, more measurements may be requested, and the
ML optimization problems may be retrained” (paragraph 0142). PANTELIDOU indicates that after
receiving new filter sets the gNB can use them for training its ML optimization algorithms. The
algorithms may compute a cost or reward which may determine how well the ML algorithm performs. If
the cost or reward is farther from optimal, more measurements may be requested, and the ML
optimization problem retrained. Here, optimization can be inferred as synchronization of the
transmissions.
It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to modify the method and invention of SONG, CHU, KAUR, SADEK, and ABDALA to include aspects described by PANTELIDOU that “relate to radio access network data collection, e.g. minimization of drive test data collection, and machine learning based optimization of networks.” PANTELIDOU provides the motivation for modification stating, “A radio access network (RAN) optimization algorithm may comprise an algorithm for optimizing and/or improving operation, performance and/or one or more functions of a RAN. RAN optimization may comprise, for example, increasing or decreasing a priority of a service. RAN optimization, targeting end-user perception improvement, comprises e.g. capacity and coverage optimization, load sharing, load balancing, random access channel (RACH) optimization and energy saving. These functions may be optimized by SON algorithms. A radio access network optimization algorithm may be implemented with, for example, a machine learning technology” (paragraph 0071).
Regarding claim 8, SONG, CHU, KAUR, SADEK, ABDALA, and PANTELIDOU teach the method as in claim 7,
Additionally, ABDALA teaches wherein the first feedback comprises a measure of energy usage
associated with the network resources performing the scheduled transmissions.
ABDALA writes, “In block 308, the processing device 202 generates training data that includes a
relationship between (i) one or more characteristics of the first content, and (ii) the first transmission
pattern. For example, the processing device 202 can build a dataset that includes a relationship between
a combination of keywords in the first content and the first transmission pattern (e.g., the number of
times and/or frequency at which the first content was transmitted). The dataset can serve as the
training data” (paragraph 0045). ABDALA states the processing device generates training data that
includes a relationship between (i) one or more characteristics of the first content, and (ii) the first
transmission pattern. ABDALA already specified that classifiers may be used to classify content into
various groups with common characteristics, and that separate training data may be generated from
each content group. Therefore, ABDALA indicates a first feedback to the model based on the clusters.
Training data and characteristics of the first content may include energy usage.
Regarding claim 9, SONG, CHU, KAUR, SADEK, ABDALA, and PANTELIDOU teach the method as in claim 8,
Additionally, CHU teaches wherein the measure of energy usage comprises a measure of an
energy saving associated with the periods of inactivity of the network resources.
CHU writes, “In some implementations, the [lower power (LP)] device may request a scheduled start
time for transmission, an interval between two scheduled transmissions, and the AP may accordingly
group the LP devices based on the requested schedules of transmissions. For example, if LP devices A, B
and C are intended receivers for the first transmission, while LP devices D, E and F are intended receivers
for the second transmission, the AP may group devices A, B and C as members of the same group, and
devices D, E and F as members of a different group” (column 8, lines 10-19). CHU adds, “In a WLAN
environment, an LP device may constantly turn off its Wi-Fi module to save power when the LP device is
not in use. To maintain the connection between the AP and the LP device, duty-cycle protocols are
sometimes defined to synchronize the AP and the LP device...” (column 4, lines 30-34). CHU continues,
“For example, to increase batter[y] life, the LP device attempts to sleep more...” (column 4, lines 38-39).
CHU states the AP may accordingly group the lower power (LP) devices based on the requested
schedules of transmission. CHU specifies the LP device may constantly turn off its Wi-Fi module to save
power when the device is not in use, though, to maintain the connection between the device and the
AP, the duty-cycle protocols are sometimes defined to synchronize the AP and the LP device. CHU points
out that to increase battery life, the LP device attempts to sleep more.
Regarding claim 10, SONG, CHU, KAUR, SADEK, ABDALA, and PANTELIDOU teach the method as in claim 7, further comprising:
Additionally, KAUR teaches negotiating with the IoT devices to determine whether the new
transmission patterns satisfy performance requirements of the IoT devices.
KAUR writes, “A WTRU may be configured to determine when to use a pattern (e.g., pool) based on
detecting triggers. A WTRU may receive an explicit indication to use a transmission pattern (e.g.,
transmission pool) or a reception pattern (e.g., reception pool) according to one or more criteria. For
example, when the synchronization message (e.g., the synchronization message of the selected
synchronization source or relay) includes a pattern, and/or the WTRU may explicitly receive a pattern
from a base station, an eNB or from another controlling entity. When a WTRU receives an explicit
indication, the WTRU may initiate the use of the pattern and/or associated pool” (paragraph 0231).
KAUR adds, “The WTRU may, for example, start/stop relaying synchronization based on certain
conditions. An example might be that the WTRU may select a synchronization source based on certain
conditions or the WTRU may select a resource pool based on the same or other conditions” (paragraph
0085). KAUR mentions, “The WTRU may include the resource pool it expects to use for transmission
and/or reception and/or is using in the cluster configuration message, for example, for the
synchronization message” (paragraph 0395). KAUR states that a WTRU may receive an explicit indication
to use a transmission pattern according to one or more criteria. The WTRU may explicitly receive a
pattern from a base station, an eNB or from another controlling entity. KAUR specifies the WTRU may
select a synchronization source based on certain conditions or the WTRU may select a resource pool
based on the same or other conditions. KAUR also points out the WTRU may include the resource pool it
expects to use for transmission and/or reception and/or is using in the cluster configuration message.
Regarding claim 11, SONG, CHU, KAUR, SADEK, ABDALA, and PANTELIDOU teach the method as in claim 10, further comprising:
Additionally, ABDALA teaches providing a second feedback to the model based on an outcome of the negotiating;
ABDALA writes, “In block 312, the processing device 202 provides second content as input into the one
or more machine-learning-models 120. The second content may be different from the first content. In
response, the machine-learning-model(s) 120 can output a second transmission pattern for transmitting
the second content to one or more of the user devices 102a-b. The second transmission pattern can be
an effective transmission pattern related to the second content” (paragraph 0048). ABDALA indicates a
second content as input to the one or more machine-learning-models. The second content may be
different from the first content. The second content may include an outcome of the negotiating.
Additionally, PANTELIDOU teaches and re-training the model, using the second feedback, to output transmission patterns for the IoT devices that satisfy the performance requirements of the IoT devices.
PANTELIDOU writes, “After receiving 1550, 1552 the new filter sets, the gNB 1510 can use them
for retraining 1554 its ML optimization algorithms. ML algorithms may compute a cost or reward which
may determine how well the ML algorithm performs” (paragraph 0145). PANTELIDOU adds, “If the cost
or reward is farther from the optimal cost or reward, more measurements may be requested, and the
ML optimization problems may be retrained” (paragraph 0142). PANTELIDOU indicates that after
receiving new filter sets the gNB can use them for training its ML optimization algorithms. The
algorithms may compute a cost or reward which may determine how well the ML algorithm performs. If
the cost or reward is farther from optimal, more measurements may be requested, and the ML
optimization problem retrained. Here, optimization can be inferred as synchronization of the
transmissions.
Regarding claim 12, SONG, CHU, KAUR, SADEK, ABDALA, and PANTELIDOU teach the method as in claim 7, further comprising:
Additionally, ABDALA teaches repeating the clustering and the scheduling using the re-trained
model.
To implement the effective transmission pattern for the new content, the server 112a can transmit data
to one or more other servers 112b-c, whereby the data is configured to cause a content-transmission
schedule 122 corresponding to the unknown content to be adjusted. The data can cause the content -
transmission schedule 122 to be adjusted such that the server(s) 112b-c deliver the unknown content to
one or more of the user device(s) 102a-b in accordance with the effective transmission pattern
predicted by the machine-learning-model(s) 120. In some examples, the data includes commands,
program code, or other instructions configured to cause the one or more other servers 112b-c to
automatically adjust the content-transmission schedule 122 (e.g., to match the predicted effective-
transmission pattern) (paragraph 0032). For example, the server 112 can use classifiers (e.g., Naïve bias
classifiers) to classify content into various groups, whereby each content group has common
characteristics. The server 112 then generates separate training data from each content group. The
training data can include relationships between (i) the common characteristics defining a content group
and (ii) the effective transmission patterns corresponding to the content in the content group. The
server 112 can then use the training data for each content group to train separate machine-learning-
models 120 tailored to the respective content groups (paragraph 0029). ABDALA indicates the effective
transmission pattern for the new content, the server can transmit data to one or more other servers,