DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 16-30 are presented for examination.
Claims 16-30 rejected under 35 U.S.C. 112(b)
Claims 16- 30 are found ineligible under 35 USC 101.
Claims 16, 26-27 and 30 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019).
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Bolic, Miodrag. Architectures for efficient implementation of particle filters. State University of New York at Stony Brook, 2004.
Claim 28 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Thiyagalingam, Jeyarajan, Lykourgos Kekempanos, and Simon Maskell. "MapReduce particle filtering with exact resampling and deterministic runtime." EURASIP Journal on Advances in Signal Processing 2017.1 (2017).
Claim 29 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Doucet, Arnaud, and Stéphane Sénécal. "Fixed-lag sequential monte carlo." 2004 12th European Signal Processing Conference. IEEE, 2004.
Claims 18- 25 objected to as being dependent upon a rejected base claim.
This action is non-final rejection.
Priority
Acknowledgment is made for applicant’s foreign priority date for application GB2101274.5 filed on 01/29/20021. The foreign priority to application GB2101274.5 needs to be perfected (to include translation) to receive
priority to priority date of 01/29/2021.
Information Disclosure Statement
The IDS filed on 07/28/2023, 01/30/2024 and 01/31/2024 are reviewed and considered. See the attached documents.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 16-30 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The term “Lowest ranked”, “higher ranks” and “nearly sort” in claims 16 and 30 are a relative term which renders the claim indefinite. The term “lowest ranked” and “higher ranks” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. On claim 16, “moving the particles which are to be duplicated to the cores having lowest ranks to create gaps in the higher ranked cores using a rotational nearly sort process” it is not clear that what cores are considered to be higher and lower since there is no standard other than ranking the cores and “nearly sort” is a relative/subjective term so it also makes the above limitation unclear.
All dependent claims are either consisting of those relative terms or is dependent on the independent claim which recites a relative term, so claims 17- 29 are also rejected under the same rational.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent
therefor, subject to the conditions and requirements of this title.
Claims 16 -30 are rejected under 35 U.S.C. 101 because the claim invention recites a judicial exception, which is directed to judicial exception of an abstract idea, as it has not been integrated into practical application, and the claim further does not recite significantly more than the judicial exception.
Step 1: Yes, claims 16-29 are directed to a method, which is a process, so it is under a statutory category of the invention. Claims 30 is directed to a system, which is a machine, so it is under a statutory category of invention.
Step 2A: prong 1: Yes, claims 16- 30 recites abstract idea which falls under a mental process since a human mind can estimate the true state of a physical system using a manual load balancing using a pen and paper. Abstract ideas are bolded as shown below.
Regarding Claim 16:
implementing, a process using a plurality of statistically independent particles and the at least one measured parameter to estimate the true state of the physical system, distributed memory architecture has a plurality of cores each of which are ranked (under its broadest reasonable interpretation this limitation recites a mathematical based estimation based on collected data, and this can be done by a human mind by using a pencil and paper, so this limitation recites a mental process and the distributed memories cores are ranked but they is no specific way of ranking, or there is no specific rational to rank the cores so a human mind can make a judgment to rank the core based on the capacity/ manual).
determining the number of copies of particles required for each particle; (under a broadest reasonable interpretation this limitation recites a mathematical relationship since the number of copies are determined using a mathematical equation as it is recited on [0022]. organizing information and manipulating information through mathematical correlations, Digitech Image Techs., LLC v. Electronics for Imaging, Inc., 758 F.3d 1344, 1350, 111 USPQ2d 1717, 1721 (Fed. Cir. 2014). The patentee in Digitech claimed methods of generating first and second data by taking existing information, manipulating the data using mathematical functions, and organizing this information into a new form. The court explained that such claims were directed to an abstract idea because they described a process of organizing information through mathematical correlations, like Flook's method of calculating using a mathematical formula. 758 F.3d at 1350, 111 USPQ2d at 1721.).
redistributing copies of particles which are to be duplicated across the distributed memory architecture (under a broadest reasonable interpretation, this claim limitation recites a mental process. Since there is no specific rationality on how to differentiate a particle to be duplicated or not, a human can assign copies to different memory cores based on human judgment).
moving the particles which are to be duplicated to the cores having lowest ranks to create gaps in the higher ranked cores using a rotational nearly sort process (under a broadest reasonable interpretation, this limitation recites a mental process. The human mind can move some particles to lowest ranks of memory and leave a gap, as a manual load balancing. In this limitation, there is no rational to differentiate which core is considered as lowest and higher, so a human can basically make a judgment and move particles).
when the rotational nearly sort process is complete, filling the gaps with the required number of copies of the particles which are to be duplicated using a rotational split process; and (under a broadest reasonable interpretation, this limitation recites a mental process. In this limitation there is no specific way of determining the required number of copies of the particle, so a human can manually assign some particles to free core as a manual load balancing)
when the rotational split process is complete, filling the remaining gaps in each core using a sequential redistribute process; and (under a broadest reasonable interpretation, this limitation recites a mental process. As it recited above through manual load balancing, a human can assign some particles to free cores based on human judgment, since there is no criteria/rational which particle should go to which free core.
wherein after the redistributing, each of the plurality of cores only has copies of the particles which are to be duplicated and the total number of copies of each particle across the plurality of cores within the distributed memory architecture equals the determined number of copies ( under a broadest reasonable interpretation, this claim recites a mental process since a human mind can observe the manual load balancing to check if the total number of copies which are duplicated before and after distribution are the same).
Step 2A prong 2: No:
The above judicial exceptions do not recite additional elements that integrate the exceptions into a practical application of the exception because the claims do not have additional elements of a combination of additional elements that apply, rely on or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception.
Claims recite gathering data and outputting information which is insignificant extra solution activity. Adding insignificant extra-solution activity to the judicial exception, e.g., mere data gathering in conjunction with a law of nature or abstract idea such as a step of obtaining information about credit card transactions so that the information can be analyzed by an abstract mental process, as discussed in CyberSource V. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011) (see MPEP § 2106.05(g), and claims also recites data manipulation by “displaying” outputs - Selecting information, based on types of information and availability of information in a power-grid environment, for collection, analysis and display, Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1354-55, 119 USPQ2d 1739, 1742 (Fed. Cir. 2016); MPEP 2106.05(g).
The claim limitations which recite data gatherings are listed below.
Claim 16:
receiving, from at least one sensor(insignificant extra- solution activity – data gathering such as 'obtaining information'. See MPEP 2106.05(g).).
Step 2B: NO:
The claims do not recite additional elements which are significantly more than the abstract idea. As outlined above the claims merely used a computer as a tool to distribute load in a memory core to estimate the true state of a physical system. As of claim 16 and 30, additional element of a “sensor”, “a server comprising a distributed memory architecture, a sequential Monte Carlo (SMC)
process” are recited but the sensor is merely used as a tool to gather information and a server with a distributed architecture is also used as a tool to perform the abstract idea. While according to [0002], sequential Monte Carlo (SMC) is used to estimate a problem related to dynamic and statics model using gathered information is a WURC (insignificant extra solution activity-WURC, MPEP 2106.05(d).
Generally, based on the above analysis claim 16 found to be not eligible under 35 USC 10, while the dependent claims inherently recite abstract ideas from independent claim see if there are any significant more claim limitations exist for the dependent claims.
Regarding claim 17:
determining, for each particle to be duplicated, the number of particles in lower ranked cores which are not to be duplicated; and (under its broadest reasonable interpretation this, limitations recites a mental process of determining the number of particles on the core, so a human can determine through observation, evaluation and judgment the number of particles on the memory core, since in this limitation there is no any specific way of determining or any additional element).
obtaining, for each particle to be duplicated, an associated binary expression of the number of particles which are not to be duplicated (under a broadest reasonable interpretation, this limitation also recites a mental process of expressing number in a binary expression using a mathematical relationship, ii. a conversion between binary coded decimal and pure binary, Benson, 409 U.S. at 64, 175 USPQ at 674).
Regarding claim 18:
scanning, for each of the particles to be duplicated, a bit of the associated binary expression - insignificant extra-solution activity – WURC, MPEP 2106.05(d).
when the scanned bit is equal to one, rotating the associated particle to a lower ranked core; - it further definite the abstract idea of rotating based on criteria, so no new additional element is recited, which is significantly more.
repeating in sequence the scanning and rotating steps for each bit - insignificant extra-solution activity – WURC, MPEP 2106.05(d).
Regarding claim 19:
wherein the iterative process initially scans a least significant bit of each binary expression and finally scans a most significant bit of each binary expression - insignificant extra-solution activity – WURC, MPEP 2106.05(d).
Regarding claim 20:
computing a minimum shift value for each of the particles to be duplicated, wherein the minimum shift value is a value representing the minimum number of shifts each of the particles must take to securely distance itself from the particles in the lower ranked cores; and – is an abstract idea which can be done by a human mind using pen and paper since this limitation is performed using a mathematical equation as it is recited on [0017].
computing a maximum shift value for each of the particles to be duplicated, wherein the maximum shift value is a value representing the maximum number of shifts each of the particles may take to end up in a gap - is an abstract idea which can be done by a human mind using pen and paper since this limitation is performed using a mathematical equation as it is recited on [0017].
Regarding claim 21:
wherein computing the minimum shift value associated with a particle to be duplicated comprises summing for each lower ranked core one less than the number of copies which are required of each particle in the lower ranked cores – it further defines the abstract idea of computing a minimum shift value, so no new additional element is recited, which is significantly more.
Regarding claim 22:
wherein computing the maximum shift value associated with a particle to be duplicated comprises summing the minimum shift value with one less than the number of copies which are required of the associated particle- it further defines the abstract idea of computing a maximum shift value, so no new additional element is recited, which is significantly more.
Regarding claim 23:
obtaining, for each particle to be duplicated, an associated binary expression of the minimum shift value and the maximum shift value - under a broadest reasonable interpretation, this limitation also recites a mental process of expressing number in a binary expression using a mathematical relationship, ii. a conversion between binary coded decimal and pure binary, Benson, 409 U.S. at 64, 175 USPQ at 674.
Regarding claim 24:
scanning, for each of the particles to be duplicated, a bit of the associated binary expression of the minimum shift value and the maximum shift value; - insignificant extra-solution activity – WURC, MPEP 2106.05(d).
when both scanned bits are equal to one, rotating the associated particle to a higher ranked core by a number of positions corresponding to the scanned bits; - it further definite the abstract idea of rotating based on criteria, so no new additional element is recited, which is significantly more.
repeating in sequence the scanning and rotating steps for each bit- insignificant extra-solution activity – WURC, MPEP 2106.05(d).
Regarding claim 25:
when only one scanned bit is equal to zero, splitting the number of copies to be duplicated to determine excess duplicates; and – it further define abstract idea, so no new additional element is recited, which is significantly more.
rotating the excess duplicates to a higher ranked core by a number of positions corresponding the scanned bits -further define abstract idea, so no new additional element is recited, which is significantly more.
Regarding claim 26:
wherein filling the remaining gaps in each core using a sequential redistribute process comprises duplicating particles within the memory of each core – is further defines abstract idea of filling the gap, so no new additional element is recited, which is significantly more.
Regarding claim 27:
obtaining values for each of the particles; insignificant extra-solution activity – data gathering such as 'obtaining information'. See MPEP 2106.05(g).
computing weights for each of the particles, wherein the weight indicates the resemblance of the obtained values to the true state; - recites a mental process, which can be done by a human mind using pen and paper, so no new additional element is recited, which is significantly more.
normalising the computed weights - recites a mental process, which can be done by a human mind using pen and paper, so no new additional element is recited, which is significantly more.
determining the number of copies which are required based on the normalised weights; and after redistributing the particles, resetting the weights – it recites a mental process that can be done though observation, analysis, evaluation and judgment, so no new additional element is recited, which is significantly more.
Regarding claim 28:
determining the number of copies ncopies i required for each particle i is computed from:
PNG
media_image1.png
54
650
media_image1.png
Greyscale
… this limitation recites abstract idea since a mathematical equation is recited, so no new additional element is recited, which is significantly more.
Regarding claim 29:
wherein the SMC process is selected from a particle filter, a fixed-lag SMC sampler and an SMC sampler- this claim recites additional elements but the samplers are used as a tool to perform SMC process which is WURC, so no new additional element is recited, which is significantly more.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 16, 26- 17 and 30 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019).
As of claim 16, Varsi anticipates A method for estimating a true state of a physical system, the method comprising (Section 3.11 “particle filter”, Let Xt ∈ RM be the current state of the dynamic system that we want to estimate).
receiving, from at least one sensor, a measurement of at least one parameter within the physical system, wherein the at least one parameter is related to the true state of the physical system; and (Section 3.11 “particle filter”, Let Xt ∈ RM be the current state of the dynamic system that we want to estimate. At every time step t a new measurement Yt ∈ RD is collected…. Section 5.3, kth sensor with respect to the target).
implementing, on a server comprising a distributed memory architecture, (section 2, “distributed memory model”, Distributed memory architectures are a type of parallel system which are inherently different from shared memory architectures. In this environment, the memory is distributed over the cores, and each core can only directly access its own private memory). a sequential Monte Carlo (SMC) process using a plurality of statistically independent particles and the at least one measured parameter to estimate the true state of the physical system, (section 3.1 “sequential Monte Carlo methods”, SMC methods apply the Importance Sampling principle to make Bayesian inferences. The main idea consists of generating N statistically independent hypotheses called particles (or samples) at every given iteration t. The population of particles xt ∈ RN× M is sampled from a user-defined proposal distribution q(xt|xt−1) such that xt represents the pdf of the state of a dynamic model (in Particle Filters) or samples from a static target posterior distribution (in SMC Samplers1)). wherein the distributed memory architecture has a plurality of cores each of which are ranked, and wherein implementing the SMC method comprises (section 2, “distributed memory model”, Distributed memory architectures are a type of parallel system which are inherently different from shared memory architectures. In this environment, the memory is distributed over the cores, and each core can only directly access its own private memory …Section 4.3 “Parallel Nearly Sort “We call this algorithm Nearly Merge. Stage-by-stage, one core with MPI rank i is coupled with another core with MPI rank j.).
determining the number of copies of particles required for each particle; and (section 3.3 “Key components of Particle Filters and SMC Samplers”, The key idea of these algorithms is to process wt to generate an array of integers called ncopies ∈ ZN).
redistributing copies of particles which are to be duplicated across the distributed memory architecture, (section 3.3 “Key components of Particle Filters and SMC Samplers”, After that, it is necessary to perform a task called redistribute which duplicates xi t as many times as ncopiesi … Figure4: Nearly Sort Based Redistribute).
wherein the redistributing comprises:
moving the particles which are to be duplicated to the cores having lowest ranks to create gaps in the higher ranked cores using a rotational nearly sort process; and (section 4.2 “4.2 Nearly Sort: an alternative to single-core sorting”, when the rotational nearly sort process is complete, filling the gaps with the required number of copies of the particles which are to be duplicated using a rotational split process; and
PNG
media_image2.png
221
835
media_image2.png
Greyscale
Figure 4: Nearly Sort Based Redistribute).
when the rotational split process is complete, filling the remaining gaps in each core using a sequential redistribute process; and wherein after the redistributing, each of the plurality of cores only has copies of the particles which are to be duplicated and the total number of copies of each particle across the plurality of cores within the distributed memory architecture equals the determined number of copies.
PNG
media_image3.png
608
1030
media_image3.png
Greyscale
Examiner Note: after the rotation split process is completed, all core with 0 copies are removed and filled with copies of the particles which are to be duplicated as it is shown on table 5 - 5.3. After sequential redistribute, the total number of copies in each core is equal to the determined number of copies (for example on table 1, x 11, have 3 copies and after sequential distribution we have 1 x11 in table 5, and 2 x11 in table 5.1, so in total 3).
Claim 30 is in the same scope as that of claim 16, therefore claim 30 is also rejected under the same rational as claim 16.
As of claim 26, Varsi anticipates all the limitations of claim 16 and Varsi also anticipates wherein filling the remaining gaps in each core using a sequential redistribute process comprises duplicating particles within the memory of each core (Section 3.3.1 “Redistribute”, The same is described by Algorithm 5 in the appendix and referred to as Sequential Redistribute (S-R) in the rest of the paper. This routine simply iterates over ncopies, and, for the j−th element, it copies xj as many times as ncopiesj). Examiner note see Figure 4, Nearly Sort based redistribute, on claim 1, table 5 – 5.3
As of claim 27, Varsi anticipates all the limitations of claim 16 and Varsi also anticipates obtaining values for each of the particles; (Section 3.11 “particle filter”, Let Xt ∈ RM be the current state of the dynamic system that we want to estimate…. Algorithm 1 SIR Particle Filter, line 1 x0, w0 ←Initialization(), each particle is initially drawn from the prior distribution q(x0) = p(x0) and each weight is initialized to 1/N).
computing weights for each of the particles, wherein the weight indicates the resemblance of the obtained values to the true state (section 3.1, “Sequential Monte Carlo methods” Each particle xi t is then assigned to an unnormalized importance weight wi t = F(wi t−1,xi t,xi t−1), such that the array of weights wt ∈ RN provides information on which particle best describes the real state of interest).
normalising the computed weights (section 3.1, “Sequential Monte Carlo methods”
wt ∈RN represents the array of the normalised weights, each of them calculated as follows
PNG
media_image4.png
109
960
media_image4.png
Greyscale
).
determining the number of copies which are required based on the normalised weights; and (section 3.3 “Key components of Particle Filters and SMC Samplers”, The key idea of these algorithms is to process wt to generate an array of integers called ncopies ∈ ZN whose i-th element, ncopiesi, indicates how many times the i-th particle has to be duplicated. It is easy to infer that ncopies has the following property:
PNG
media_image5.png
58
413
media_image5.png
Greyscale
).
after redistributing the particles, resetting the weights (Section 3.3 “Key components of Particle Filters and SMC Sampler”, We note that the reset step (which sets all the weights to 1/N) after redistribute is trivially parallelized).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Bolic, Miodrag. Architectures for efficient implementation of particle filters. State University of New York at Stony Brook, 2004.
As of claim 17, Varsi anticipates all the limitations of claim 16 and Varsi also teaches wherein moving the particles using the rotational nearly sort process comprises: determining, for each particle to be duplicated, the number of particles in lower ranked cores which are not to be duplicated; (figure 4: Nearly Sort Based Redistribute, table 2
PNG
media_image2.png
221
835
media_image2.png
Greyscale
). Examiner note as it is shown on figure 4. The number of particles is sorted based by nearly sort and it sorts particles to be duplicated on the right.
Varsi does not explicitly teach obtaining, for each particle to be duplicated, an associated binary expression of the number of particles which are not to be duplicated.
While Bolic teaches obtaining, for each particle to be duplicated, an associated binary expression of the number of particles which are not to be duplicated(Page 42, section 4.2.1 “Proposed resampling scheme”, The second column represents the decimal values of the particle weights, and the third column their binary representation with two bits. The fourth column provides the replication factor, which is a decimal equivalent value of the K bits indicated in bold. According to the table, particle x(1) will be replicated twice, particle x(2) will be replicated once, and particles x(3) and x(4) will be eliminated.
PNG
media_image6.png
182
484
media_image6.png
Greyscale
).
Bolic is considered to be analogous to the claimed invention since it teaches Algorithms and architectures for particle filters. Therefore, it would be obvious for a person of ordinary skill in the art before the effective filing date to integrate Bolic’s teaching of expressing representing replication factor in binary expression into Varsi’s model to obtain an associated binary expression for the number of particles which are not to be duplicated.
The motivation would have been to create a low-complexity RR scheme for particle filters the proposed scheme uses a simple “particle-tagging” method to compensate for a possible error that can be caused by finite precision quantization in the resampling step of particle filtering. The scheme guarantees that the number of particles after resampling is always equal to the number of particles before resampling. The resulting scheme is suitable for high-speed physical realization when the number of particles is a power of two (Bolic, page 42, section 4.2).
Claim 28 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Thiyagalingam, Jeyarajan, Lykourgos Kekempanos, and Simon Maskell. "MapReduce particle filtering with exact resampling and deterministic runtime." EURASIP Journal on Advances in Signal Processing 2017.1 (2017).
As of claim 28, Varsi anticipates all the limitations of claim 27, but it doesn’t explicitly teach the limitations of claim 28.
While Thiyagalingam teaches wherein determining the number of copies ncopies i required for each particle i is computed from:
PNG
media_image1.png
54
650
media_image1.png
Greyscale
where cdf is the cumulative density function of the weights up to and including the ith particle and u is drawn from a uniform distribution with u-[0,1) (section 4.6, “Minimum variance resampling” As explained in Section 3.3, resampling involves deter mining the number of copies of each particle that are needed. We specifically describe minimum variance resampling, for which the number of copies of the ith particle is
PNG
media_image7.png
218
566
media_image7.png
Greyscale
).
Thiyagalingam is considered to be analogous to the claimed invention since it teaches filtering with exact resampling and deterministic runtime. Therefore, it would be obvious to try for a person of ordinary skill in the art before the effective filling date to generate an equation to find the number of copies using cumulative density function based on Thiyagalingam teaching Minimum variance resampling equation.
The motivation would have been to developed an improved parallel particle filtering algorithm and the implementation evaluated is limited by the communications overhead necessarily associated with giving each particle a unique key in the MapReduce framework: as a result, while it can achieve a speedup of threefold with 16 cores in a single node, with 512 cores spread across 28 nodes, we only achieve a speedup of approximately 1.4 (i.e., less) (Thiyagalingam, section 8, conclusion).
Claim 29 is rejected under 35 U.S.C. 103 as being unpatentable over Varsi, Alessandro, et al. "A single SMC sampler on MPI that outperforms a single MCMC sampler." arXiv preprint arXiv:1905.10252 (2019) in the view of Doucet, Arnaud, and Stéphane Sénécal. "Fixed-lag sequential monte carlo." 2004 12th European Signal Processing Conference. IEEE, 2004.
As of claim 29, Varsi anticipates all the limitations of claim 16, and Varsi also teaches’ the SMC process is selected from a particle filter (section 3.3.1 “particle filters” and an SMC sampler (section 3.1, The population of particles xt ∈ RN× M is sampled from a user-defined proposal distribution q(xt|xt−1) such that xt represents the pdf of the state of a dynamic model (in Particle Filters) or samples from a static target posterior distribution (in SMC 2 Samplers1)).
Varsi doesn’t teach SMC process selected from a fixed-lag SMC sampler.
While Doucet teaches SMC process selected from a fixed-lag SMC sampler (section 3, 3. fixed-lag sequential Monte Carlo).
Doucet is considered analogous to the claimed invention since it teaches a . fixed-lag sequential Monte Carlo. Therefore, it would be obvious for a person of ordinary skill in the art, before the effective filing date to integrate Doucet’s teaching of fixed-lag SMC sampler into Varsi model to process SMC from selected including fixed-lag SMC samplers.
The motivation would have been fixed-lag sequential Monte Carlo method is a cheaper and natural extension of standard SMC algorithms and can be used wherever these algorithms apply and in the standard case, one can only expect the algorithm developed to be efficient if it is possible to design some sensible approximation of the (fixed-lag) optimal importance distribution (Doucet, section 5, “Conclusion”).
Allowable Subject Matter
Claims 18- 25 objected to as being dependent upon a rejected base claim but would be allowable if rewritten in independent form including all of the limitations of the base claim, any intervening claims and if it overcomes claim rejections cited above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Brink; Kevin (US 10969752 B1, Date Published 2021-04-06), is similar to the claimed invention since it teaches estimating a state of a physical system and System states are predicted from prior estimates and then updated dependent upon the measurements and the state update weights to provide updated estimated states.
Uhlmann; Jeffrey K. (US 6092033 A, Date Published 2000-07-18) is similar to the claimed invention since it teaches one or more of the signals must encode some infonnation derived from measurements of the physical system, and every signal must encode sufficient information to obtain a mean estimate of the state of the physical system and an estimate of the covariance of that estimated state
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ABRHAM A. TAMIRU whose telephone number is (571)272-6987. The examiner can normally be reached Monday - Friday 8:00am - 5:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ryan Pitaro can be reached at 571 272 4071. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ABRHAM ALEHEGN TAMIRU/ Examiner, Art Unit 2188
/RYAN F PITARO/ Supervisory Patent Examiner, Art Unit 2188
Prosecution Insights