pith. sign in

arxiv: 1907.11367 · v1 · pith:ON7HLXIQnew · submitted 2019-07-24 · 💻 cs.NI · cs.DC· cs.LG

Data Aggregation Techniques for Internet of Things

Sunny Sanyal This is my paper

Pith reviewed 2026-05-24 16:31 UTC · model grok-4.3

classification 💻 cs.NI cs.DCcs.LG
keywords data aggregationInternet of Thingsenergy efficient routinguncertain datamedical IoTfog computingclustering protocolprivacy preservation
0
0 comments X

The pith

Three frameworks address resource limits, uncertain data, and privacy in IoT aggregation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The dissertation designs data aggregation frameworks for massive IoT networks by tackling three challenges: limited power and computation on nodes, unreliable raw sensor readings, and issues of latency plus privacy in critical uses. It offers three separate methods, one using device clustering for energy-aware routing in both fixed and moving nodes, one applying processing techniques to raise the quality of uncertain data, and one using prediction to cut communication costs while safeguarding medical device information. These methods draw on optimization, learning algorithms, matrix theory, filtering, and fog or cloud resources. A reader would care because effective aggregation is required for IoT analytics to run reliably without exhausting device batteries or risking sensitive information.

Core claim

The dissertation establishes that three independent algorithmic frameworks, each built for a distinct IoT scenario and drawing on non-convex optimization, machine learning, stochastic matrix perturbation theory, federated filtering, fog computing, and cloud computing, can mitigate the stated open challenges of resource constraints, uncertain raw data, and network latency plus privacy for medical devices.

What carries the argument

Three independent novel approaches, each targeting one or more of the challenges through device-to-device clustering, data quality improvement, and prediction-based aggregation.

If this is right

  • Device clustering based on device-to-device links lowers power consumption for both stationary and mobile IoT nodes.
  • The scheme for handling uncertain data produces readings suitable for downstream decision making.
  • The prediction-based method for medical IoT reduces power loss from communication and addresses privacy concerns.
  • Fog and cloud resources combined with the listed algorithms enable the frameworks to operate at massive scale.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The three methods could be combined into a single system that handles all three challenges at once rather than separately.
  • Similar clustering and prediction ideas might extend to non-IoT sensor networks facing the same constraints.
  • Success would require testing how the frameworks interact with existing network protocols and standards.

Load-bearing premise

The three approaches using optimization, learning, matrix perturbation, filtering, and fog or cloud resources will successfully reduce the effects of resource limits, data uncertainty, and privacy or latency problems when applied to real large-scale IoT systems.

What would settle it

A field deployment of thousands of IoT nodes in which the proposed clustering fails to lower energy use, the quality scheme leaves data still unfit for decisions, or the medical prediction method does not reduce communication overhead or protect privacy.

Figures

Figures reproduced from arXiv: 1907.11367 by Sunny Sanyal.

Figure 1
Figure 1. Figure 1: 1 Massive IoT data Taxonomy............................................................................4 [PITH_FULL_IMAGE:figures/full_fig_p013_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: 1 Proposed IoT Analytics Architecture .............................................................20 [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: 2 Location of indoor temperature and humidity sensor deployment. (a) First floor [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: 6 True sensor data estimation error as a function of subspace reconstruction error. [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: 3 Normalized Tolerable Perturbation Error as a function of [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
Figure 1
Figure 1. Figure 1 [PITH_FULL_IMAGE:figures/full_fig_p022_1.png] view at source ↗
Figure 1
Figure 1. Figure 1: 2 Map of IoT domains and verticals As discussed IoT analytics have proven to bring value towards the society, and currently it is receiving considerable attention from both academia and industry. The growing interest in IoT analytics need its stakeholders to clearly understand the approaches involved in analytics, building blocks, technical requirements and open challenges. 1.2.1 IoT Analytics Architecture… view at source ↗
Figure 1
Figure 1. Figure 1: 3 Dominant IoT Analytics Architecture Typically, the analysis and decision making tasks of massive IoT analytics platform are performed inside the cloud servers in a centralized fashion. A centralized IoT analytic system (dominant) demands all the data at once to make inferences. 1.2.2 Taxonomy of IoT Analytics Extracting business value out of the raw IoT sensor data is anything but trivial. In order to le… view at source ↗
Figure 1
Figure 1. Figure 1: 4 Diffirent Categories of IoT Analytics based on use cases On the other hand, proactive analysis is slowly but steadily emerging as a new trend to generate and facilitate actionable insights from the massive IoT dataset. This category is further classified into stream analytics and real-time analytics. In the case of stream analytics, the incoming time series data is analyzed based on batches or streams; t… view at source ↗
Figure 1
Figure 1. Figure 1: 5 Proposed IoT Analytics Architecture 1.4.1 Objective The core objective of this dissertation is to design efficient data aggregation algorithms and frameworks for massive IoT networks in diverse scenarios to support the proper functioning of the entire IoT analytics sytem. To achieve this goal, the dissertation investigates data-driven approaches for massive IoT data aggregation that rely on methods based… view at source ↗
Figure 2
Figure 2. Figure 2: 1 Proposed IoT Analytics Architecture Following the standard procedure of data uploading in the cellular network, primarily the BS collects the cell-mode CQI (channel quality indicator) feedbacks from all IoT devices (nodes) willing to upload some data. Secondarily the BS collects the D2D mode CQI values and forms a D2D CQI matrix (DCM) [16]. Suppose each node 1,......, | | {} i in n n n N  is a part of … view at source ↗
Figure 2
Figure 2. Figure 2: 2 Logical flow of cluster formation scheme 2.5 Propsed Data Uploading Scheme As already mentioned the data size generated by the IoT devices are relatively small; thus it may not require a high CQI; taking advantage of this trait IoT devices form a multi-hop data routing from CM to the BS (sink). Moreover only the CH(s) participates in the multi-hop data transmission also known as intercluster data transmi… view at source ↗
Figure 2
Figure 2. Figure 2: 3 Logical flow of data uploading scheme 2.6 Implementation and Results Our proposed D2D based IoT data aggregation scheme is simulated on Matlab 2012. To present the competitive performance of the proposal with other related schemes, we [PITH_FULL_IMAGE:figures/full_fig_p044_2.png] view at source ↗
Figure 2
Figure 2. Figure 2: 4 Formation of simulated clusters based on co-relative mobility [PITH_FULL_IMAGE:figures/full_fig_p046_2.png] view at source ↗
Figure 2
Figure 2. Figure 2: 5 Energy efficiency vs Number of devices [PITH_FULL_IMAGE:figures/full_fig_p047_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: 1 Scenario of massive IoT data aggregation The selection scheme of cluster head among the other cluster members and the selection scheme of a next hop cluster head for the inter-cluster device to device communication can be found in chapter 1. 3.4 Background This section discusses some important prerequisites and notations ( [PITH_FULL_IMAGE:figures/full_fig_p054_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: 2 Location of indoor temperature and humidity sensor deployment. (a) First floor (b) Second floor [PITH_FULL_IMAGE:figures/full_fig_p058_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: 4 Fraction of Total link traffic variance captured by dominant subspace 3.5 Main approach As discussed in section 3.4.4 the chapter assumes that the raw IoT sensor data has an intrinsic true subspace which carries most of the dynamics of raw data and in section 3.4.1 we have defined the information content i.e the true data as a more reliable data with minimal or no uncertainties; also fit for data analyti… view at source ↗
Figure 3
Figure 3. Figure 3: 5 Subspace reconstruction error as a function of Gaussian noise variance. (a) Synthetic dataset. (b) Sensor dataset (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p068_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: 7 True sensor data estimation error asa function of subspace reconstruction error in the presence of outliers. Size of outliers between [-10, 10] with a density of 0.2 and all values are mean over 10 samples, where n=400 and k=10. (a) Synthetic dataset (b) Sensor Dataset (a) (b) [PITH_FULL_IMAGE:figures/full_fig_p069_3.png] view at source ↗
Figure 3
Figure 3. Figure 3: 9 Scalability, the packet size is 10 bytes 3.6.3 Discussion The baseline approach is fragile to high uncertainties due to the fact that the single value decomposition approach directly processes raw IoT sensor data to compute the optimal subspace basis. The empirical observations [31] [32] show that it rather generates random subspace basis in the presence of high uncertainties such as big outliers and mis… view at source ↗
Figure 4
Figure 4. Figure 4: 1 Federated Filtering Framework Some recent use cases of IoT based healthcare analytics such as [42] [43] also advocates centralized decision making, however, both of them lacks a theoretical formulation to ensure decision-making accuracy. 4.2.2 Prediction based IoT systems The literature [44] reports several prediction based approaches for reducing the communication overhead in sensor networks. The predic… view at source ↗
Figure 4
Figure 4. Figure 4: 2 Block diagram of an adaptive filter 1 0 ( ) i Y t P   (4.4) Where, 2 1 1 () M Yi j P Y j M    , and M is the number of iterations taken for training the LMS filter. 4.4.2 Perturbation Analysis at Fog Server The filter parameters play a key role in balancing the tradeoff between the desirable loss of decision accuracy (by allowing perturbation to Y ˆ ) and low communication overhead. This chapter us… view at source ↗
Figure 4
Figure 4. Figure 4: 3 Normalized Tolerable Perturbation Error as a function of  [PITH_FULL_IMAGE:figures/full_fig_p086_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: 5 Prediction performance of gyro sensor readings [PITH_FULL_IMAGE:figures/full_fig_p087_4.png] view at source ↗
Figure 4
Figure 4. Figure 4: 7 Energy efficiency as a function of number of devices 4.7 Chapter Conclusion This chapter considers open challenges concerning energy efficiency, privacy and latency for smart healthcare analytics. This chapter derives a theoretical upper bound on the eigenvalue perturbation and further formulates a relationship between the local quantization at IoMT devices with the global perturbation error at fog serve… view at source ↗
read the original abstract

The goal of this dissertation is to design efficient data aggregation frameworks for massive IoT networks in different scenarios to support the proper functioning of IoT analytics layer. This dissertation includes modern algorithmic frameworks such as non convex optimization, machine learning, stochastic matrix perturbation theory and federated filtering along with modern computing infrastructure such as fog computing and cloud computing. The development of such an ambitious design involves many open challenges, this proposal envisions three major open challenges for IoT data aggregation: first, severe resource constraints of IoT nodes due to limited power and computational ability, second, the highly uncertain (unreliable) raw IoT data is not fit for decisionmaking and third, network latency and privacy issue for critical applications. This dissertation presents three independent novel approaches for distinct scenarios to solve one or more aforementioned open challenges. The first approach focuses on energy efficient routing; discusses a clustering protocol based on device to device communication for both stationary and mobile IoT nodes. The second approach focuses on processing uncertain raw IoT data; presents an IoT data aggregation scheme to improve the quality of raw IoT data. Finally, the third approach focuses on power loss due to communication overhead and privacy issues for medical IoT devices (IoMT); describes a prediction based data aggregation framework for massive IoMT devices.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript is a dissertation proposal that identifies three open challenges in IoT data aggregation (resource constraints of nodes, uncertain raw data, and latency/privacy issues) and outlines three high-level intended approaches using non-convex optimization, machine learning, stochastic matrix perturbation theory, federated filtering, fog computing, and cloud computing. The approaches are described only at the level of goals: (1) a D2D-based clustering protocol for energy-efficient routing with stationary and mobile nodes, (2) an aggregation scheme to improve quality of uncertain IoT data, and (3) a prediction-based framework for medical IoT devices to reduce communication overhead and privacy risks.

Significance. The topic of scalable data aggregation for IoT analytics is relevant to the field. However, because the manuscript supplies no algorithms, derivations, analysis, or validation, no assessment of technical contribution or significance is possible from the current document.

major comments (1)
  1. [Abstract] Abstract: the central claim that the dissertation 'presents three independent novel approaches' is not supported by any content; the document only states that such approaches 'will be developed' without supplying any technical description, equations, pseudocode, or preliminary results.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed review of our dissertation proposal. We address the major comment below and note that the manuscript is a high-level proposal document rather than a completed technical paper.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the dissertation 'presents three independent novel approaches' is not supported by any content; the document only states that such approaches 'will be developed' without supplying any technical description, equations, pseudocode, or preliminary results.

    Authors: We agree that the manuscript is a dissertation proposal that identifies open challenges and outlines intended approaches at a conceptual level, without providing algorithms, derivations, or results. The phrasing 'presents three independent novel approaches' was meant to indicate the proposal of these research directions for the dissertation. We will revise the abstract (and related sections) to explicitly describe the document as a proposal outlining planned approaches to be developed, rather than implying completed technical contributions. revision: yes

Circularity Check

0 steps flagged

No circularity: proposal contains no derivations or equations

full rationale

This is a dissertation proposal that outlines three high-level intended approaches (energy-efficient routing via clustering, data quality improvement for uncertain IoT data, and prediction-based aggregation for IoMT) using techniques such as non-convex optimization and federated filtering. No equations, predictions, fitted parameters, self-citations of uniqueness theorems, or ansatzes are present in the provided text. The claims are prospective statements of intent rather than completed derivations that could reduce to their inputs by construction. The document is therefore self-contained against external benchmarks with no load-bearing steps to analyze.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are specified because the text is a high-level proposal without technical derivations or models.

pith-pipeline@v0.9.0 · 5751 in / 1134 out tokens · 28484 ms · 2026-05-24T16:31:46.762065+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 53 canonical work pages · 2 internal anchors

  1. [1]

    An Internet of Things blueprint for a smarter world,

    Nokia Strategic White Paper, “An Internet of Things blueprint for a smarter world,” 2016

    work page 2016

  2. [2]

    Available: https://innovate.ieee.org/innovation-spotlight-ieee-fueling-fourth-industrial-revolution/?LT=X PLHL_XPL_1.2019_LM_Innovation_Spotlight_4IR

    IEEE Xplore, IEEE is Fueling the Fourth Industrial Revolution [Online]. Available: https://innovate.ieee.org/innovation-spotlight-ieee-fueling-fourth-industrial-revolution/?LT=X PLHL_XPL_1.2019_LM_Innovation_Spotlight_4IR

    work page

  3. [3]

    Worldwide and Regional Internet of Things (IoT) 2014-2020 Forecast: A Virtuous Circle of Proven Value and Demand,

    D. Lund, C. Macgillivray, V. Turner, and M. Morales, “Worldwide and Regional Internet of Things (IoT) 2014-2020 Forecast: A Virtuous Circle of Proven Value and Demand,” 2014

    work page 2014

  4. [4]

    Gartner says 4.9 Billion Connected Things; Will be in use in 2015, Gartner 2014, [Online].Availble:https://www.gartner.com/en/newsroom/press-releases/2014-11-11-gartner-s ays-nearly-5-billion-connected-things-will-be-in-use-in-2015

    work page 2015

  5. [5]

    Algorithms, Big Data and the Importance of Smart Cities,

    Jimmy Daly, “Algorithms, Big Data and the Importance of Smart Cities,” StateTech Magazine, 2013. [Online]. Available: https://statetechmagazine.com/article/2013/02/algorithms-big-data-and-importance-smart-citie s-infographic

    work page 2013

  6. [6]

    When things matter: A survey on data-centric internet of things,

    N. J. G. Falkner, Q. Z. Sheng, S. Dustdar, A. V. Vasilakos, Y. Qin, and H. Wang, “When things matter: A survey on data-centric internet of things,” J. Netw. Comput. Appl., vol. 64, pp. 137–153, 2016

    work page 2016

  7. [7]

    Smart Home Ecosystem: IoT and Consumers,

    Parks Associates, “Smart Home Ecosystem: IoT and Consumers,” 2014

    work page 2014

  8. [8]

    Predicts 2015: The Internet of Things,

    Alfonso Velosa, James F. Hines and Hung LeHong, “Predicts 2015: The Internet of Things,” Gartner, 2014. [Online]. Available: https://www.gartner.com/doc/2952822/predicts--internet-things

    work page arXiv 2015

  9. [9]

    Intel and San Jose’s smart city will use real-time data to monitor air quality and more,

    Sarah Silbert, “Intel and San Jose’s smart city will use real-time data to monitor air quality and more,” Engadget, 2014. [Online]. Available: https://www.engadget.com/2014/06/11/intel-san-jose-smart-city/

    work page 2014

  10. [10]

    Parking made easy: Smarter parking project in Pisa kicks off

    Deutsche Telekom, “Parking made easy: Smarter parking project in Pisa kicks off”,

    work page

  11. [11]

    [Online]. Available: Thesis for Master’s Degree of CQUPT References 74 https://www.telekom.com/en/media/media-information/archive/parking-made-easy--smarter-p arking-project-in-pisa-kicks-off-361114

    work page

  12. [12]

    A Survey on Network Methodologies for Real-Time Analytics of Massive IoT Data and Open Research Issues,

    S. Verma, Y. Kawamoto, Z. M. Fadlullah, H. Nishiyama, and N. Kato, “A Survey on Network Methodologies for Real-Time Analytics of Massive IoT Data and Open Research Issues,” IEEE Communication Surveys and Tutorials, vol. 19, no. 3, pp. 1457–1477, 2017

    work page 2017

  13. [13]

    100,000 IoT Sensors Monitor a 1,400-Kilometer Canal in China,

    Tracy Staedter, “100,000 IoT Sensors Monitor a 1,400-Kilometer Canal in China,” IEEE Spectrum, 2018. [Online]. Available: https://spectrum.ieee.org/tech-talk/telecom/internet/a-massive-iot-sensor-network-keeps-watch -over-a-1400kilometer-canal

    work page 2018

  14. [14]

    The Internet of Trash: IoT Has a Looming E-Waste Problem

    Stacey Higginbotham, “The Internet of Trash: IoT Has a Looming E-Waste Problem”, IEEE Spectrum, 2018. [Online]. Available: https://spectrum.ieee.org/telecom/internet/the-internet-of-trash-iot-has-a-looming-ewaste-probl em

    work page 2018

  15. [15]

    Service Requirements for Machine-Type Communications (MTC),

    3GPP TS 22.368, “Service Requirements for Machine-Type Communications (MTC),” vol. 0, no. 14.0.0, pp. 1–26, 2014

    work page 2014

  16. [16]

    12, France, 2014

    3GPP,”TS 22.803,”Feasibility Study for Proximity Services (ProSe),” 3GPP: Sophia Antipolis Cedex, Rel. 12, France, 2014

    work page 2014

  17. [17]

    Energy efficient IoT data collection in smart cities exploiting D2D communications,

    A. Orsino, G. Araniti, L. Militano, J. Alonso-Zarate, A. Molinaro, and A. Iera, “Energy efficient IoT data collection in smart cities exploiting D2D communications,” Sensors (Switzerland), vol. 16, no. 6, pp. 1–19, 2016

    work page 2016

  18. [18]

    Multi-hop D2D networking and resource management scheme for M2M communications over LTE-A systems

    G. Rigazzi, F. Chiti, R. Fantacci, and C. Carlini, “Multi-hop D2D networking and resource management scheme for M2M communications over LTE-A systems”, in Proceedings of the 10th International Wireless Communications and Mobile Computing Conference, pp. 973–978, 2014

    work page 2014

  19. [19]

    Data aggregation for low power wireless devices,

    K. Schubert and N. Bambos, “Data aggregation for low power wireless devices,” in Proceedings of IEEE Military Communications Conference MILCOM, pp. 1-6, 2016. Thesis for Master’s Degree of CQUPT References 75

    work page 2016

  20. [20]

    Locating the nodes: cooperative localization in wireless sensor networks,

    A. O. Hero, R. L. Moses, N. Patwari, J. N. Ash, S. Kyperountas, and N. S. Correal, “Locating the nodes: cooperative localization in wireless sensor networks,” IEEE Signal Process. Mag., vol. 22, no. 4, pp. 54–69, 2005

    work page 2005

  21. [21]

    A distributed D-hop cluster formation for VANET,

    M. Azizian, S. Cherkaoui, and A. S. Hafid, “A distributed D-hop cluster formation for VANET,” in Proceedings of IEEE Wireless Communications Networking Conference WCNC, Doha, pp. 1–6, 2016

    work page 2016

  22. [22]

    Blind Calibration of Sensor Networks,

    L. Balzano and R. Nowak, “Blind Calibration of Sensor Networks,” in Proceedings of the 6th Int. Symp. Inf. Process. Sens. Networks, pp. 79–88, 2007

    work page 2007

  23. [23]

    Robust PCA and subspace tracking from incomplete observations using l0-surrogates,

    C. Hage and M. Kleinsteuber, “Robust PCA and subspace tracking from incomplete observations using l0-surrogates,” Comput. Stat., vol. 29, no. 3–4, pp. 467–487, 2014

    work page 2014

  24. [24]

    Challenges for Quality of Data in Smart Cities,

    P. Barnaghi, M. Bermudez-Edo, and R. Tönjes, “Challenges for Quality of Data in Smart Cities,” ACM Journal of Data and Information Quality, vol. 6, no. 2–3, pp. 1–4, 2015

    work page 2015

  25. [25]

    A review on approximate computing and some of the associated techniques for energy reduction in IOT,

    D. Marwaha and A. Sharma, “A review on approximate computing and some of the associated techniques for energy reduction in IOT,” in Proceedings of the 2nd International Conference on Inventive Systems and Control, pp. 319–323, 2018

    work page 2018

  26. [26]

    Low-Overhead and High-Precision Prediction Model for Content-Based Sensor Search in the Internet of Things,

    P. Zhang, Y. Liu, F. Wu, S. Liu, and B. Tang, “Low-Overhead and High-Precision Prediction Model for Content-Based Sensor Search in the Internet of Things,” IEEE Communication Letters, vol. 20, no. 4, pp. 720–723, 2016

    work page 2016

  27. [27]

    Adaptive cleaning for RFID data streams,

    S. R. Jeffery, U. C. Berkeley, M. J. Franklin, M. Garofalakis, and M. J. Franklin, “Adaptive cleaning for RFID data streams,” in Proceedings of the 32nd Int. Conf. Very large data bases, pp. 163–174, 2006

    work page 2006

  28. [28]

    Approximate Decision Making in Large-Scale Distributed Systems

    L. Huang, A. D. Joseph, and N. Taft, “Approximate Decision Making in Large-Scale Distributed Systems”

    work page

  29. [29]

    Communication-Efficient Online Detection of Network-Wide Anomalies,

    L. Huang et al., “Communication-Efficient Online Detection of Network-Wide Anomalies,” in Proceedings of IEEE INFOCOM, pp. 134–142, 2007

    work page 2007

  30. [30]

    Probabilistic inference over RFID streams in mobile environments,

    T. Tran, C. Sutton, R. Cocci, Y. Nie, Y. Diao, and P. Shenoy, “Probabilistic inference over RFID streams in mobile environments,” in Proceedings - International Conference on Data Engineering, pp. 1096–1107, 2009. Thesis for Master’s Degree of CQUPT References 76

    work page 2009

  31. [31]

    A novel fog computing enabled temporal data reduction scheme in iot systems,

    T. Yu, X. Wang, and A. Shami, “A novel fog computing enabled temporal data reduction scheme in iot systems,” in Proceedings of IEEE Global Communications Conference, pp. 1–5, 2018

    work page 2018

  32. [32]

    Robust PCA via outlier pursuit,

    H. Xu, C. Caramanis, and S. Sanghavi, “Robust PCA via outlier pursuit,” IEEE Trans. Inf. Theory, vol. 58, no. 5, pp. 3047–3064, 2012

    work page 2012

  33. [33]

    Thresholding based Efficient Outlier Robust PCA

    Y. Cherapanamjeri, P. Jain, and P. Netrapalli., "Thresholding based effcient outlier robust PCA", 2017 [Online]. Available: https://arxiv.org/abs/1702.05571

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  34. [34]

    Robust principal component analysis?

    E. J. Candes, X. Li and Y. Ma, “Robust principal component analysis?", Journal of ACM, vol. 58, no. 3, pp. 11:1-11:37, 2011

    work page 2011

  35. [35]

    Wang and Y

    X. Wang and Y. He, “Learning from uncertainty for big data: Future analytical challenges and strategies,'' IEEE Syst., Man, Cybern. Mag., vol. 2, no. 2, pp. 26-32, Apr. 2016

    work page 2016

  36. [36]

    Home M2M networks: Architectures, standards, and QoS improvement,

    Y. Zhang, R. Yu, S. Xie, W. Yao, Y. Xiao, and M. Guizani, “Home M2M networks: Architectures, standards, and QoS improvement,” IEEE Commun. Mag., vol. 49, no. 4, pp. 44–52, 2011

    work page 2011

  37. [37]

    Data driven prediction models of energy use of appliances in a low-energy house,

    L. M. Candanedo, V. Feldheim, and D. Deramaix, “Data driven prediction models of energy use of appliances in a low-energy house,” Energy Build., vol. 140, pp. 81–97, 2017

    work page 2017

  38. [38]

    Small, Low Power, 3-Axis Accelerometer (ADXL335),

    Analog Device, “Small, Low Power, 3-Axis Accelerometer (ADXL335),” 2009

    work page 2009

  39. [39]

    S. Boyd, N. Parikh, E. Chu, Eckstein, B. Peleato, and Jonathan, Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers, vol. 3, no. 1, 2014

    work page 2014

  40. [40]

    WHO | Global health workforce shortage to reach 12.9 million in coming decades,

    “WHO | Global health workforce shortage to reach 12.9 million in coming decades,” WHO, 2014 [Online]. Available: https://www.who.int/mediacentre/news/releases/2013/health-workforce-shortage/en/

    work page 2014

  41. [41]

    The Internet of Things for Health Care: A Comprehensive Survey,

    S. M. Riazul Islam, M. Humaun Kabir, M. Hossain, Daehan Kwak, and Kyung-Sup Kwak, “The Internet of Things for Health Care: A Comprehensive Survey,” IEEE Access, vol. 3, pp. 678–708, 2015. Thesis for Master’s Degree of CQUPT References 77

    work page 2015

  42. [42]

    Internet of Things for Smart Healthcare: Technologies, Challenges, and Opportunities,

    S. B. Baker, W. Xiang, and I. Atkinson, “Internet of Things for Smart Healthcare: Technologies, Challenges, and Opportunities,” IEEE Access, vol. 5, pp. 26521–26544, 2017

    work page 2017

  43. [43]

    Unobtrusive detection of body movements during sleep using Wi-Fi received signal strength with model adaptation technique,

    O. Ammae, J. Korpela, and T. Maekawa, “Unobtrusive detection of body movements during sleep using Wi-Fi received signal strength with model adaptation technique,” Futur. Gener. Comput. Syst., vol. 78, pp. 616–625, 2018

    work page 2018

  44. [44]

    An IoT-based computational framework for healthcare monitoring in mobile environments,

    H. Mora, D. Gil, R. M. Terol, J. Azorín, and J. Szymanski, “An IoT-based computational framework for healthcare monitoring in mobile environments,” Sensors (Switzerland), vol. 17, no. 10, 2017

    work page 2017

  45. [45]

    On the importance and feasibility of forecasting data in sensors

    G. M. Dias, B. Bellalta, and S. Oechsner, “On the importance and feasibility of forecasting data in sensors,” pp. 1–30, 2016 [Online]. Available: http://arxiv.org/abs/1604.01275

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  46. [46]

    A Survey about Prediction-Based Data Reduction in Wireless Sensor Networks,

    G. M. Dias, B. Bellalta, and S. Oechsner, “A Survey about Prediction-Based Data Reduction in Wireless Sensor Networks,” ACM Computing Surveys, vol. 49, no. 3, 2016

    work page 2016

  47. [47]

    An Adaptive Method for Data Reduction in the Internet of Things,

    Y. Fathy, P. Barnaghi, and R. Tafazolli, “An Adaptive Method for Data Reduction in the Internet of Things,” in Proceedings of IEEE 4th World Forum on Internet of Things (WF-IoT), pp. 7–13, 2018

    work page 2018

  48. [48]

    Communication-Efficient Learning of Deep Networks from Decentralized Data,

    H. B. Mcmahan and D. Ramage, “Communication-Efficient Learning of Deep Networks from Decentralized Data,” vol. 54, 2017 [Online]. Available: http://arxiv.org/abs/1602.05629

    work page arXiv 2017

  49. [49]

    Trading Computation for Communication: Distributed Stochastic Dual Coordinate Ascent,

    T. Yang, “Trading Computation for Communication: Distributed Stochastic Dual Coordinate Ascent,” Adv. Neural Inf. Process. Syst., pp. 1–9, 2013

    work page 2013

  50. [50]

    Haykin and S

    S. Haykin and S. Haykin, Adaptive filter theory. 1991

    work page 1991

  51. [51]

    Haykin and B

    S. Haykin and B. Widrow, Least-Mean-Square Adaptive Filters. 2005

    work page 2005

  52. [52]

    Bhatia, Review of matrix perturbation theory

    R. Bhatia, Review of matrix perturbation theory. by G.W. Stewart and Ji-guang Sun, vol. 160, no. C. 1992

    work page 1992

  53. [53]

    Distributed PCA and network anomaly detection,

    L. Huang, X. Nguyen, M. Garofalakis, M. Jordan, A. Joseph, and N. Taft, “Distributed PCA and network anomaly detection,” in Proceedings of Neaural Info. Proceesing conference, 2006. Thesis for Master’s Degree of CQUPT References 78 Thesis for Master’s Degree of CQUPT Ackno wledgement 79 Acknowledgement There are many people whom I wish to thank for their ...

    work page 2006