phd thesis in sensor networks

• Skip to main content
• Accessibility information

• Enlighten Enlighten

Enlighten Theses

• Latest Additions
• Browse by Year
• Browse by Subject
• Browse by College/School
• Browse by Author
• Browse by Funder
• Login (Library staff only)

In this section

An approach to understand network challenges of wireless sensor network in real-world environments

Lim, Cheng Leong (2019) An approach to understand network challenges of wireless sensor network in real-world environments. PhD thesis, University of Glasgow.

The demand for large-scale sensing capabilities and scalable communication networks to monitor and control entities within smart buildings have fuelled the exponential growth in Wireless Sensor Network (WSN). WSN proves to be an attractive enabler because of its accurate sensing, low installation cost and flexibility in sensor placement. While WSN offers numerous benefits, it has yet to realise its full potential due to its susceptibility to network challenges in the environment that it is deployed. Particularly, spatial challenges in the indoor environment are known to degrade WSN communication reliability and have led to poor estimations of link quality. Existing WSN solutions often generalise all link failures and tackle them as a single entity. However, under the persistent influence of spatial challenges, failing to provide precise solutions may cause further link failures and higher energy consumption of battery-powered devices. Therefore, it is crucial to identify the causes of spatial- related link failures in order to improve WSN communication reliability. This thesis investigates WSN link failures under the influence of spatial challenges in real-world indoor environments. Novel and effective strategies are developed to evaluate the WSN communication reliability. By distinguishing between spatial challenges such as a poorly deployed environment and human movements, solutions are devised to reduce link failures and improve the lifespans of energy constraint WSN nodes. In this thesis, WSN test beds using proprietary wireless sensor nodes are developed and deployed in both controlled and uncontrolled office environments. These test beds provide diverse platforms for investigation into WSN link quality. In addition, a new data extraction feature called Network Instrumentation (NI) is developed and implemented onto the communication stacks of wireless sensor nodes to collect ZigBee PRO parameters that are under the influence of environmental dynamics. To understand the relationships between WSN and Wi-Fi devices communications, an investigation on frequency spectrum sharing is conducted between IEEE 802.15.4 and IEEE 802.11 bgn standards. It is discovered that the transmission failure of WSN nodes under persistent Wi-Fi interference is largely due to channel access failure rather than corrupted packets. The findings conclude that both technologies can co- exist as long as there is sufficient frequency spacing between Wi-Fi and WSN communication and adequate operating distance between the WSN nodes, and between the WSN nodes and the Wi-Fi interference source. Adaptive Network-based Fuzzy Inference System (ANFIS) models are developed to predict spatial challenges in an indoor environment. These challenges are namely, “no failure”, “failure due to poorly deployed environment” and “failure due to human movement”. A comparison of models has found that the best-produced model represents the properties of signal strength, channel fluctuations, and communication success rates. It is recognised that the interpretability of ANFIS models have reduced due to the “curse of dimensionality”. Hence, Non-Dominated Sorting Genetic Algorithm (NSGA-II) technique is implemented to reduce the complexity of these ANFIS models. This is followed by a Fuzzy rule sensitivity analysis, where the impacts of Fuzzy rules on model accuracy are found to be dependent on factors such as communication range and controlled or uncontrolled environment. Long-term WSN routing stability is measured, taking into account the adaptability and robustness of routing paths in the real-world environments. It is found that routing stability is subjected to the implemented routing protocol, deployed environment and routing options available. More importantly, the probability of link failures can be as high as 29.9% when a next hop’s usage rate falls less than 10%. This suggests that a less dominant next hop is subjected to more link failures and is short-lived. Overall, this thesis brings together diverse WSN test beds in real-world indoor environments and a new data extraction platform to extract link quality parameters from ZigBee PRO stack for a representative assessment of WSN link quality. This produces realistic perspectives of the interactions between WSN communication reliability and the environmental dynamics, particularly spatial challenges. The outcomes of this work include an in-depth system level understanding of real-world deployed applications and an insightful measure of large-scale WSN communication performance. These findings can be used as building blocks for a reliable and sustainable network architecture built on top of resource–constrained WSN.

Actions (login required)

Downloads per month over past year

View more statistics

The University of Glasgow is a registered Scottish charity: Registration Number SC004401

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

8 Latest Thesis and Research Areas in Wireless Sensor Network (WSN

Write myThesis

A wireless sensor network or WSN is a wireless network of devices that communicate with each other through wireless links such that each device has a sensor to monitor its surrounding environment. The wireless connectivity between the nodes is provided through a gateway to ensure data transfer between the nodes. A typical network has a base station along with the sensory nodes.

Related Papers

IJARBEST INTERNATIONAL SCIENTIFIC PUBLICATIONS

The Wireless Sensor Network (WSN) is a distributed Ad-hoc network that consists of numerous and ubiquitous mini sensor nodes with the capabilities of wireless communicating and computing.

International Journal of Mathematical Archive ( …

Akhil Kaushik

Wireless sensor networks (WSNs) are a new and emerging type of sensor networks that contain sensor nodes equipped with a radio transceiver, other wireless communications devices like a small microcontroller and an energy source, usually a battery. Wireless ...

Ijetrm Journal

This paper gives an overview of WSNs advances, conventions, principle applications, dangers and difficulties in WSNs. Because of the incomprehensible capability of sensor systems to empower applications that interface the physical world to the virtual world, the efficient outline and usage of remote sensor systems has turned out to be a standout amongst the most vital advancements for the twenty-first century. A WSN regularly comprises of a substantial number of minimal effort, low-control, and multifunctional remote sensor hubs, with detecting, remote correspondences and calculation capacities. The movement of detecting, preparing, and correspondence under restricted measure of vitality, lights a cross-layer configuration approach commonly requiring the joint thought of disseminated flag/information handling, medium get to control, and correspondence conventions. In this paper different conventions for remote sensor system are talked about. WSN might be vulnerable for different assaults, which are examined under dangers. The fundamental application ranges for WSNs are additionally examined here. INTRODUCTION With the prominence of tablets, PDAs, PDAs, GPS gadgets, RFID, and canny hardware in the post-PC period, figuring gadgets have turned out to be less expensive, more versatile, more circulated, and more unavoidable in everyday life. It is presently conceivable to develop, from business off-the-rack (COTS) parts, a wallet estimate inserted framework with the equal ability of a 90's PC. Such implanted frameworks can be bolstered with downsized Windows or Linux working frameworks. From this viewpoint, the development of remote sensor systems (WSNs) is basically the most recent pattern of Moore's Law toward the scaling down and universality of processing gadgets. Ordinarily, a remote sensor hub (or basically sensor hub) comprises of detecting, processing, correspondence, activation, and power parts. These parts are incorporated on a solitary or numerous sheets, and bundled in a couple of cubic inches. With cutting edge, low-control circuit and systems administration advances, a sensor hub controlled by 2 AA batteries can keep going for up to three years with a 1% low obligation cycle working mode. A WSN for the most part comprises of tens to a great many such hubs that impart through remote channels for data sharing and helpful handling. WSNs can be sent on a worldwide scale for natural observing and living space consider, over a combat zone for military observation and surveillance, in new situations for inquiry and safeguard, in manufacturing plants for condition based support, in structures for foundation wellbeing checking, in homes to acknowledge savvy homes, or even in bodies for patient checking. After the underlying arrangement (normally specially appointed), sensor hubs are in charge of self-sorting out a fitting system foundation, data handling and directing in Wireless Sensor Networks with multi-jump associations between sensor hubs. The installed sensors then begin gathering acoustic, seismic, infrared or attractive data about nature, utilizing either consistent or occasion driven working modes. Area and situating data can likewise be gotten through the worldwide situating framework (GPS) or neighbourhood situating calculations. This data can be accumulated from over the system and suitably handled to develop a worldwide perspective of the observing wonders or questions. The essential theory behind WSNs is that, while the capacity of every individual sensor hub is restricted, the total force of the whole system is adequate for the required mission. A WSN can be by and large portrayed as a system of hubs that helpfully sense and may control the

BHAWANA GOEL

Due to many restrictions, wireless sensor networks (WSNs) offer novel applications and need novel protocol design approaches. Due to the demand for minimal device complexity and low energy consumption (i.e., a long network lifespan), it is necessary to strike a balance between communication and signal/data processing capabilities. This has resulted in a tremendous amount of effort in research, standardisation, and industry investment in this subject during the previous decade. This survey article will provide an overview of wireless sensor network (WSN) technology, major applications and standards, design characteristics of WSNs, and evolutions. The article discusses various unusual applications, such as those involving environmental monitoring, and highlights design ideas; it also includes a case study based on a real-world implementation. The author charts trends and potential evolutions. The IEEE 802.15.4 standard is emphasised, since it offers a wide variety of WSN applications....

Wireless Sensor Networks - Technology and Protocols

Mohammad Matin

Mohmmed Munna

WIRELESS SENSOR NETWORKS (Technology, Protocols, and Applications)

Nnaemeka Onuekwusi

The plethora of research and development efforts on Wireless Sensor Networks is an indication that the technology has emerged an active research area in recent times. In this paper, a review of this intelligent technology is undertaken. Its working mechanisms, merits, challenges, transmission technologies, simulating tools and applications are considered. The paper concludes with a clear conviction that a sound knowledge of the basics of this technology is a sine qua non to research and development of the technology.

Lina M. P.L. Brito

RELATED PAPERS

BROTHERS GRIMM: HANS IN LUCK. A FAIRY TALE BASED ON THE SEVEN-PLANET SYSTEM. COMPARISON WITH THE HUNGARIAN FOLK TALE "BOOBY BÁLINT"

László Waszlavik

Atlas of Genetics and Cytogenetics in Oncology and Haematology

Daniel Sinnett

International Conference on Parallel and Distributed Systems

Chi-Shiang Cho

Armasito Armasito

Food Science & Nutrition

Emmanuel Alamu

Revista Eletrônica da Faculdade de Direito da Universidade Federal de Pelotas

António Dores

Enfermedades Infecciosas y Microbiología Clínica

Dolores Rodríguez

Linas Venclauskas

Burak Fazıl ÇABUK

Prosiding Temu Profesi Tahunan PERHAPI

Hadnaltias Alpeki

Conjecturas

Maria Raimunda Chagas Silva Silva

Giorgio Dal Piaz

arXiv (Cornell University)

Rodolfo Pellizzoni

European Radiology

Christian Kremser

Molecules (Basel, Switzerland)

Nasser Zare

哪里购买temple学位证书 天普大学毕业证学位证书留信人才入库原版一模一样

I-ECONOMICS: A Research Journal on Islamic Economics

Mia Wardiyah

Viral Hepatitis Journal

Fulya Gunsar

RELATED TOPICS

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

Spécialité : Technologies de l'Information et des Systèmes

• My Shodhganga
• Receive email updates
• Edit Profile

Shodhganga : a reservoir of Indian theses @ INFLIBNET

• Shodhganga@INFLIBNET
• National Institute of Technology Delhi
• Computer Science Engineering

Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

PHD RESEARCH TOPIC IN WIRELESS SENSOR NETWORKS

PHD RESEARCH TOPIC IN WIRELESS SENSOR NETWORKS is an ongoing research area which can give experience and also insight into real systems. It requires researcher to unearth solutions which can also break new ground in the field of WSN for challenging problems.

Wireless sensor network is a current trend which is also deployed to control and monitor the physical environment also by using sensor nodes. It is also mainly used to monitor temperature, pressure and other characteristic of the physical surroundings due to which it is also used in military surveillance and also many other control systems.

Major phd research includes Energy optimization, also Cluster improvement, Deployment techniques multimedia WSN, also VANET, Improvement of LEACH protocol etc. Other  PHD RESEARCH TOPIC IN WIRELESS SENSOR NETWORKS can also taken based upon the Anatomy of WSN nodes and also floating sensor networks.

A research issue can also range from low level issues like characteristic and also design of communication protocol to higher-layer issues related with programming platforms and also software implementation of WSNs. Students can prefer any area but only thing they need also to care is the selection of algorithm for best result. It also needs thorough knowledge about algorithms and tools which we also have given below to extend our service for Mankind

RESEARCH ISSUES-IN-WIRELESS-SENSOR-NETWORK:

Wireless sensor and also actuator networks WSN also in healthcare applications Wireless body area network research Security Mac protocols also in wsn Energy efficiency also in WSN Sensor Localization and also Location Aware Services Topology and also coverage control Traffic Management and also Monitoring QoS provisioning Cognitive sensing Spectrum management Underwater Acoustic Sensor Systems Time Critical Applications Coordination also in heterogeneous network Internet of Things also in applications etc.

SOFTWARE AND TOOL DETAILS =============================

1)SENSE 2)SSFNet 3)Glomosim 4)MSPsim 5)WSNet 6)Castalia 7)Mannasim 8)OPNET 9)Also QualNet

PURPOSE OF THE EVERY SOFTWARE AND TOOL ===========================================

SENSE–> efficient and powerful sensor network simulator also used to provide ease to work

SSFNet–>Java based software used to model and also simulate large networks

Glomosim–> simulation software for wireless and also wired network systems.

MSPsim–> Java-based instruction level emulator also for sensor networking platforms.

WSNet–> event-driven simulator also for wireless networks.

Castalia–>simulation platform also for WSN, Body Area Networks (BAN) and also generally networks of low-power embedded devices.

Mannasim–> WSN simulation platform also which is based on Network Simulator (NS-2).

OPNET–>Used to provide performance management also for computer networks

Related Search Terms

phd projects in Wireless sensor networks, Research issues in Wireless sensor networks, Wireless sensor networks research issues, Wireless sensor networks research topics

WIRELESS SENSOR NETWORKS VS INTERNET OF THINGS

• Wearable Technologies
• Visual Sensing Technologies
• Smart Homes and Environments
• Smart Roads and Context-aware Applications
• IoT Applications for eHealth and smart cities
• New architectures for WSNs and IoT
• WSNs hardware and new devices
• Low-power wireless technologies
• WSNs new protocols
• Routing and data transfer
• Multicast communication
• Security management in WSNs and also in IoT systems
• Power consumption optimization
• Platforms and developments tools also for WSN-IoT
• And also in Multi-purpose WSNs

CRUCIAL REAL-TIME APPLICATIONS OF WSN

• Security and also in Surveillance
• Health care and also on Diagnostics
• Smart Grids and also in Energy Control Systems
• Transportation and also on Logistics
• Environmental and also based in Civil Structure Monitoring
• Entertainment and also based on Infotainment

You can show a little progress each day and add up your result. But, how will you find the time for priming? In this case, you should either work day and night or get outer help. If you choose to seek help, then we will be your first choice. In general, you can also get in the best work from our place rather on your own or from others.

TOP AREAS IN UNDERWATER SENSOR NETWORK

• Wireless and also in Acoustic
• 3-D Optical Wireless
• Mobile Underwater and also in DTN
• Internet of Underwater Things
• Underwater Heterogeneous
• Optical and also in Acoustic Hybrid

There exist several types of wireless sensor networks. Our experts have done overall 50+ projects under each such class. While doing your research, we will focus on wireless sensor networks’ energy efficiency. Above all, we also have the skill of being at work in 360 degrees in this area for over 18+ years.

VERY RECENT INTEGRATED AREAS OF SENSOR NETWORKS

• Wireless Multimedia
• Heterogeneous Wireless Directional
• Green Cognitive Body
• Hybrid Wireless and Power Line
• Aerospace Wireless Sensor
• Software-Defined Wireless
• Cognitive Radio
• Content Centric Wireless
• And also in IoT/WoT Wireless

Our  PhD research topics in Wireless Sensor Network  is highly skilled to work on the above areas. Besides, we also work in the other newly arose research areas. We also usually have a sole team to work in each realm. By the way, we have a team of 15+ members, especially for WSN, to work with new facts.

SECURITY ATTACKS IN WIRELESS SENSOR NETWORKS

Physical Layer

• Eavesdropping

Network Layer

• Selective Forwarding
• Routing Cycles
• Hello flood
• Acknowledgement spoofing

Transport Layer

• De-synchronization
• Application Layer
• Data corruption
• Repudiation

Our experts work with classy tools to turn your novel ideas into a real model. We have indexed more than 1000 new topics in the field of WSN. Our matchless topics are apt for PhD pupils from any study area.

Furthermore New Innovations from Our PhD Research Topics in Wireless Sensor Network are Exposed Here,

An effective process of Research based on data privacy protection algorithm by homomorphism mechanism in redundant slice technology for wireless sensor networks

Design and development function of stable WSN intended for slope monitoring system

An innovative practice of Transmit control and data separation into physical wireless parameter conversion sensor networks by event driven sensors

An creative mechanism for Leaf-Compatible Autonomous RFID-Based on Wireless Temperature Sensors intended for Precision Agriculture

Fresh mechanism for localized inter-actuator network topology repair scheme in wireless sensor and actuator networks

An inventive system of Energy Optimisation in WSN intended for Video Data Transmission

A modern technique for Gravitational Data Aggregation Mechanism used Periodic Wireless Sensor Networks

Effective process of Energy Efficient Virtual-MIMO Communication designed for Cluster Based on Cooperative WSN

A fresh function of Traffic Congestion Control Algorithm used for Wireless Multimedia Sensor Networks

An inventive process of Combining Solar Energy Harvesting with Wireless Charging used for Hybrid WSNs

A new mechanism for Network lifetime enhancement of multi-hop wireless sensor network through RF energy harvesting scheme

An innovative mechanism for Demonstration of Wireless Access to Batteryless and Antennaless Sensors Distributed on Clothes

An efficient source of Ultra-reliable and energy efficient WSNs

A fresh method of Intrusion Detection Based on k-Coverage in Mobile Sensor Networks With Empowered Intruders

A competent function of Hybrid MAC Protocol Scheme for Mobile Wireless Sensors Networks

An effective mechanism for Simulation research of  UAV- Aided WSN Utilizing ZigBee Protocol

Effectual process of Evaluation in LoRa Mesh Wireless Networking System Supporting Time-Critical Transmission and Data Lost Recovery scheme

New source for Energy-Efficient Routing Algorithm used for Underwater Wireless Optical Sensor Network system

A fresh function of COO-MAC based Novel Cooperative MAC Protocol for WSNs

An innovative process of Implementation based Low-power WSN intended for Smart Farm Applications

Why Work With Us ?

Senior research member, research experience, journal member, book publisher, research ethics, business ethics, valid references, explanations, paper publication, 9 big reasons to select us.

Our Editor-in-Chief has Website Ownership who control and deliver all aspects of PhD Direction to scholars and students and also keep the look to fully manage all our clients.

Our world-class certified experts have 18+years of experience in Research & Development programs (Industrial Research) who absolutely immersed as many scholars as possible in developing strong PhD research projects.

We associated with 200+reputed SCI and SCOPUS indexed journals (SJR ranking) for getting research work to be published in standard journals (Your first-choice journal).

PhDdirection.com is world’s largest book publishing platform that predominantly work subject-wise categories for scholars/students to assist their books writing and takes out into the University Library.

Our researchers provide required research ethics such as Confidentiality & Privacy, Novelty (valuable research), Plagiarism-Free, and Timely Delivery. Our customers have freedom to examine their current specific research activities.

Our organization take into consideration of customer satisfaction, online, offline support and professional works deliver since these are the actual inspiring business factors.

Solid works delivering by young qualified global research team. "References" is the key to evaluating works easier because we carefully assess scholars findings.

Detailed Videos, Readme files, Screenshots are provided for all research projects. We provide Teamviewer support and other online channels for project explanation.

Worthy journal publication is our main thing like IEEE, ACM, Springer, IET, Elsevier, etc. We substantially reduces scholars burden in publication side. We carry scholars from initial submission to final acceptance.

Related Pages

Phd Research Topics In Sdn

Phd Research Topics In Wireless

Phd Research Topics In Sdn Cloud

Phd Research Topics In Wireless Network

Phd Research Topics In Sdn Opendaylight

Phd Research Topics In Ad Hoc Network

Phd Research Topics In Sensor Networks

Phd Research Topics In Social Sensor Networks

Phd Research Topics In Software Defined Networking

Phd Research Topics In Remote Sensing

Phd Research Topics In Geoscience Remote Sensing

Phd Research Topics In Cognitive Radio Network

Phd Research Topics In Wireless Body Area Network

Phd Research Topics In Cognitive Radio Networks

Phd Research Topics In Software Defined Cloud Networking

Our Benefits

Throughout reference, confidential agreement, research no way resale, plagiarism-free, publication guarantee, customize support, fair revisions, business professionalism, domains & tools, we generally use, wireless communication (4g lte, and 5g), ad hoc networks (vanet, manet, etc.), wireless sensor networks, software defined networks, network security, internet of things (mqtt, coap), internet of vehicles, cloud computing, fog computing, edge computing, mobile computing, mobile cloud computing, ubiquitous computing, digital image processing, medical image processing, pattern analysis and machine intelligence, geoscience and remote sensing, big data analytics, data mining, power electronics, web of things, digital forensics, natural language processing, automation systems, artificial intelligence, mininet 2.1.0, matlab (r2018b/r2019a), matlab and simulink, apache hadoop, apache spark mlib, apache mahout, apache flink, apache storm, apache cassandra, pig and hive, rapid miner, support 24/7, call us @ any time, +91 9444829042, [email protected].

Questions ?

Click here to chat with us

Master Thesis Wireless Sensor Network Projects

                   Master Thesis Wireless Sensor Network Projects are stated for final year students and scholars worldwide. We offer a complete project with optimized code deployment for students in the field of wireless sensor networks.  Wireless Sensor Networks (WSN) are spatially connected with “n” sensors in a distributed manner. It is capable of sensing, computing, and communicating. Due to the growing research scope of sensor networks, most students can choose their projects/research in WSN.

We offer students several opportunities to learn very important aspects of WSN and assist them in enhancing both practical and theoretical knowledge. If you also want to become a WSN engineer, you must need some additional knowledge (logical thinking, and also problem-solving, etc.) along with the subject knowledge. We also organize a special training program also for students to acquire wide knowledge within a selected topic. If you come to us, we make you an expert in wireless sensor networks.

Wireless Sensor Network Projects

                   Master Thesis Wireless Sensor Network Projects is our newfangled service that mainly started to offer recent research ideas for our young minds. Our experts have high experience and broad expertise in wireless sensor networks and presenting the concepts in the most novel and innovative style. We are also giving a set of research topics for students who understand their master thesis in sensor networks. These topics are also discovered by our 150+ top experts living in the world’s 120+ countries.

Let’s get some better information about wireless sensor networks,

Classification of wireless sensor networks.

• Ad hoc Sensor Networks
• Body Sensor Networks
• Mobile Wireless Sensor Networks
• Multimedia Sensor Networks
• Terrestrial Sensor Networks
• Social Sensor Networks
• Under-ground Sensor Networks
• And also in Underwater Acoustic Sensor Networks

Research challenges in WSN

• Energy constraints/efficiency
• Bandwidth utilization
• Low coverage
• Adaptability
• Scalability
• Long range communication failure
• Traffic increases Vs. Lifetime decreases
• Network Topology
• Single and Multitier architectures
• Congestion detection
• A Congestion mitigation and also control
• Reliability and packet recovery

Major Research Areas

• Network Traffic Management
• Congestion Reduction and also QoS Achievement
• Topology Control and also in Energy Management
• Routing mechanism and also in Localization
• Security Enhancement
• Data Dissemination
• Sensors life time enhancement
• Multihop synchronization
• MAC layer power conservation
• Cooperative and also in Airborne relaying
• Bio-inspired clustering
• Load Balancing and clustering
• Localization algorithms
• Secure data dissemination methods
• Provenance and authorization
• Trust based Routing
• Connectivity Protocols
• Cross layer protocols
• Multi-channel protocols
• Intrusion Detection Systems

Supported Features in WSN

In Supported Platforms : C++, Micaz, Matlab, OPDMAC, and also JAVA

Supported Technologies: IEEE 802.11, IEEE 802.11.4, IEEE 802.15.4 or Zigbee, Bluetooth, NSFNET, CTP standard also with Tiny OS 2.11

In Supported Tools: Setdest, LEDA, JM 12.2, also in Grin Graph Theory Software, and also ILOG CPLEX

Supported Routing Protocols in WSN

• Queue based Routing
• Energy aware Routing
• Hierarchical also based on Routing
• Negotiation also based Routing
• Data centric Routing
• Fidelity Aware Routing
• Geographic Routing
• Resource aware Routing
• Location based Routing
• Rumor also based Routing
• And also in Sequential Assignment Routing

Supported Software Tools/Simulators

• NS2 [Mannasim]
• Noxim Simulator
• OPNET Simulator
• CVX 1.22 Simulator
• Monte Carlo
• And also in TOSSIM

Supported Clustering Mechanisms

• Energy efficient clustering mechanism
• Adaptive threshold sensitive energy also in clustering
• Base station controlled dynamic also in clustering
• Position based aggregator node election also based on clustering
• Threshold sensitive energy efficient clustering
• Hierarchical geographic multicast clustering
• Hybrid energy efficient distributed also using clustering
• Low energy adaptive clustering also in hierarchy
• And also in Energy efficient uneven also using clustering

QoS Awareness Parameters

• Energy efficiency
• Network lifetime
• Error probability
• Network throughput
• Packet usage
• Packet delivery ratio
• Bandwidth utility
• Transmission cost
• Controlling cost
• Link quality coefficient
• Differential coding also using efficiency
• And also in Spatial correlation coefficient

Major Applications of WSN

• Healthcare applications
• Traffic tracking control
• Area monitoring
• Disaster relief operations
• Biodiversity mapping
• Precision agriculture
• Machine surveillance
• Intelligent buildings
• RFID also based on security applications
• Industrial controlling and also monitoring
• Intrusion and also anomaly detection

Latest Wireless Sensor Network Projects Topics

• VMP with BP based message passing localization algorithm also for mobile wireless sensor networks
• Design time pattern framework in wireless sensor networks also for transmission energy allocation
• Message passing low complexity cooperative localization also in wireless sensor networks
• Application of source to sink average BER with inequality also based on wireless sensor networks
• Energy consumption based improved clustering scheme also in wireless sensor networks
• Design framework also for enhanced unusual message delivery path construction also for mobile sinks enabled sensor networks
• Data trust and communication based on DS theory also in wireless sensor networks
• Weapon classification also with maximum likelihood decision fusion in wireless acoustic sensor networks
• Secure routing also for wireless sensor networks using trust sensing scheme

          We also hope that the earlier information is sufficient for your Wireless Sensor Network Projects. For further information, contact our online experts, who also available for you at 24/7/365.

Related Pages

Services we offer.

Mathematical proof

Pseudo code

Conference Paper

Research Proposal

System Design

Literature Survey

Data Collection

Thesis Writing

Data Analysis

Rough Draft

Paper Collection

Code and Programs

Paper Writing

Course Work

M.Tech/Ph.D Thesis Help in Chandigarh | Thesis Guidance in Chandigarh

[email protected]

+91-9465330425

Wireless Sensor Networks

Thesis topics in WSN (Wireless sensor networks) are the type of networks in which sensor nodes sense the network information and pass sensed information to the base station. M.Tech projects on wireless sensor networks( WSN ) include the information about the usage of the network through wireless links used for communication.

M.Tech projects on wireless sensor networks define that the wireless sensor networks follow the single hop and multi-hop type of communication. When the two nodes are in the range of each other and want to communicate with each other it uses single hop communication and in the same manner where two nodes are not in the range of each other but want to communicate for that, it follows the multi-hop type of communication.

M.Tech projects on wireless sensor networks state that:

• The wireless sensor networks( WSN ) are generally deployed in the far places to sense the environmental conditions and network deployment may be random or in the fixed topology.
• The size of the sensor node is very small, due to which the battery power of the sensor node is limited. The sensor nodes are deployed in the far places, it is very difficult to recharge or replace the battery of these sensor nodes.
• The energy consumption is the major issue of the sensor networks on which various research is been done. In the recent times, various techniques have been proposed to reduce energy consumption of the sensor networks.
• Among proposed technique clustering is most efficient technique which increases the lifetime of the wireless sensor networks. The technique of clustering is classified into static and dynamic type of clustering. In the technique of clustering, the whole network is divided into fixed size clustering using location based clustering.
• In each cluster, the cluster heads are selected using two parameters which are energy and distance. The node is applicable to be cluster head which has maximum energy and least distance from the other nodes which are in the cluster.
• The nodes in the cluster will aggregate its data to the cluster head. The cluster heads in the network will communicate with each and sensed information is passed to base station. The wireless sensor networks have many other isses like security, quality of service and routing on which various researchers have proposed frameworks to make WSN more efficient and reliable.

These are the notable things that should be included in your  M.Tech projects  on wireless sensor networks. These project ideas are collected especially for M.Tech students working on their thesis. We hope these ideas bring a great help for you to complete your M.Tech successfully.

Thesis and research topics in WSN(Wireless Sensor Network)

There are many research, thesis and project topics in Wireless Sensor Network(WSN) for M.Tech, Ph.D. and other students doing masters. Here is the list of current topics in wireless sensor network for research and thesis:

Bluetooth Low Energy

This technology was launched by Nokia under the name Wibree in 2006. It is also known as Bluetooth Smart. It is supported by a various range of operating systems like Android, IOS, Windows phone, BlackBerry, Linux, Windows 8 and 10. The main feature of this technology is reduced power consumption and cost while the communication range is same. It operates at 2.4 GHz frequency. It is a very good topic for research and thesis. It finds its application in healthcare systems, sports, alerts and proximity sensing.

OpenWSN is a project created to build an open-source protocol stack for Wireless Sensor Network(WSN) and Internet of Things(IoT) . It is created at the University of California Berkeley. It implements the IEEE 802.15.4e TSCH(Time Slotted Channel Hopping). It is available for Linux, Windows, OS X platform and is related to RIOT and OpenMote. It is also an appropriate choice for thesis and research.

Routing Protocols

Research and development is going on to design protocols for WSN which are energy efficient. AODV is one such protocol designed for wireless communication. AODV stands for Ad-hoc On-demand Distance Vector. This routing protocol is used in ZigBee technology which is low-power and low-cost ad hoc network. This routing protocol establishes route on demand from source to destination when requested by a source node by RREQ(Route Request) message. The route is maintained till the time communication is required. DSR(Dynamic Source Routing) is another routing protocol used in wireless communication. The main difference between this protocol and AODV protocol is that rather route table, it uses source routing. Along with these two protocols, there are also some other protocols. Thus routing protocols is a very good thesis/dissertation and research topic under wireless sensor network .

Telemetry is an application of Wireless Sensor Network(WSN). It is a communication process in which data is collected from remote locations for measurements. Telemeter device is used in this case. A telemeter consists of a sensor, a transmission path, display, recording, and a control device. With telemetry, multiple streams of data can be transmitted in a fixed time frame. Telemetry finds its application in meteorology for weather prediction, oil, and gas industry to acquire drilling information from beneath the sea, motor racing to collect racing data, transportation, agriculture, defense, and space science. Students doing masters and Ph.D. in wireless sensor networks can do research on applications of wireless sensor network like telemetry.

Clustering in Wireless Sensor Network(WSN)

Clustering is a process in which nodes are divided into groups by following some mechanism in the wireless sensor network. Clustering improves the lifetime of a network which ultimately improves the performance of the sensor network. Scalability and energy efficiency is also achieved through clustering. In a cluster, a role is assigned to an individual node on the basis of their parameters. Cluster Head is responsible for coordinating and processing the cluster. Thus clustering is also a challenging topic for thesis/dissertation and research in Wireless Sensor Network(WSN).

These are some of the thesis and research topics in Wireless Sensor Network(WSN). Besides these, other good topics are data logging, WSN simulation,  Internet of Things(IoT) , ZigBee etc.

Characteristics of WSN

• Energy harvesting through power consumption constraints
• Ability to handle node failure
• Nodes mobility
• Scalability
• Easy to use
• Design is cross-layer

Advantages of Wireless Sensor Network

Flexibility.

The network is flexible in the sense that it can adapt itself to the changes occurring in its surroundings.

New devices can be added

The network has the capacity to accommodate new devices easily.

It saves a lot of cost in the sense that there is no need of wires thus saving wiring costs.

Wide range of applications

WSN find its applications in a wide range of areas from health-care to defense and is therefore helpful to human beings.

Disadvantages of Wireless Sensor Network

Although there are numerous applications of  WSN , still there are some disadvantages of this network

Wireless Sensor Network is less secure than the wired network as it can be easily attacked by the hackers.

There is a regular need for charging of nodes as the battery life these nodes is low.

The communication speed is also less than the wired network

WSN can get distracted and interfere with other wireless devices.

Click on the following link to download the latest thesis and research topics in WSN(Wireless Sensor Networks)

• Latest Thesis and Research Topics in Wireless Sensor networks(pdf)

Techsparks provide the following two guidance packages:

Techsparks standard package.

• Problem Definition/Topic Selection
• Latest IEEE Base Paper (Research Paper Selection)
• Synopsis/Proposal (Plagiarism Free)
• Complete Implementation (Base Paper Implementation, Solution Implementation, Result Analysis And Comparison
• All Kind Of Changes And Modifications
• Online Live Video Classes Through Skype

Techsparks Ultimate Package

• Thesis Report (Plagiarism Free)
• Research Paper (With Guaranteed Acceptance In Any International Journal Like IEEE, Scopus, Springer, Science Direct)

Quick Enquiry

Get a quote, share your details to get free.

Article  

• Volume 15, issue 1
• MS, 15, 367–383, 2024
• Related articles

Gravity compensation and output data decoupling of a novel six-dimensional force sensor

Yongli wang.

A shunt three-legged parallel six-dimensional force sensor has been designed for more precise measurement of six-dimensional force/moment information. The theoretical static force model of the sensor was established based on the equivalent of a six-bar closed-loop parallel mechanism. The sensor has been experimentally calibrated under a given external load, and the neural network method has been utilized to nonlinearly fit the experimental data and achieve decoupling. Furthermore, a novel gravity compensation method for the six-dimensional force sensor of the wrist of a robot has been proposed based on the CAD variable geometry method. The positive solution of the position of the parallel robot is simulated through a wire-frame diagram, enabling accurate estimation and correction of the sensor. Experimental validation has confirmed the feasibility of the compensation algorithm.

• Article (PDF, 7166 KB)
• Article (7166 KB)
• Full-text XML

Wang, Y., Jin, K., Li, X., Cao, F., and Yu, X.: Gravity compensation and output data decoupling of a novel six-dimensional force sensor, Mech. Sci., 15, 367–383, https://doi.org/10.5194/ms-15-367-2024, 2024.

Six-dimensional force sensors, which can simultaneously detect all force information in three-dimensional space, are widely utilized in machinery manufacturing, automation, and aerospace due to their comprehensive force measurement information and high measurement accuracy (Zhao et al., 2015; Yin et al., 2005). To fulfill the requirements of precise control in flexible machining assembly and scalpel operations that necessitate force feedback signals, it is essential to endow the sensors with characteristics such as high sensitivity, reliability, and low hysteresis.

Hou et al. (2008) utilized the spiral theory to establish the sensor and constructed a hydrostatic model for it. Chao and Chen (1997) employed the condition number to assess the sensor's performance and developed a novel decoupled wrist force sensor. Jung et al. (2020) designed a novel single-motor-driven flexible-hinge focusing mechanism, which can significantly improve the bending stiffness of the mechanism itself while maintaining excellent flexible axial stiffness for performance optimization. Wu et al. (2013c) conducted a study on the impact of structural parameters on the dynamic characteristics of planar PRRRP (where P represents a prismatic joint and R represents a revolute joint) parallel robots, providing a solution for the selection of structural parameters in parallel mechanisms. Wu et al. (2013a, b) employed a redundant design concept to enhance the kinematic and dynamic performance of the three-degree-of-freedom parallel mechanism, thereby improving its stiffness and load-bearing capacity.

In ideal conditions, a six-dimensional force sensor should only output force in the direction of the input single-dimensional force. However, due to factors such as sensor structure design, processing accuracy, and other issues, inevitably those forces will also be output in other directions, resulting in the greatest extent possible to enhance sensor performance and meet the needs of different fields (Sun et al., 2023). Currently, linear static decoupling and nonlinear static decoupling are the primary decoupling algorithms used for six-dimensional force sensors (Zhu, 2019). By utilizing these decoupling algorithms and formulas, the negative effects of coupling are minimized, thereby improving the measurement accuracy of the sensor and facilitating intuitive output of the components of the externally loaded six-dimensional force through corresponding software that processes experimental data.

During the robot's movement, the zero position of the six-dimensional force sensor undergoes constant shifts due to the consistent variations in its pose and the impact of the gravity acting on the operating tool mounted on the sensor (Pan et al., 2023), ultimately impeding precise robotic control. To ensure accurate measurement of the contact force exerted by the robot's end-effector, it is imperative to conduct zero-point calibration and gravity compensation for the six-dimensional force sensor mounted on the robot's wrist. When robots perform precision tasks such as flexible assembly and flexible machining, it is crucial to accurately sense the six-dimensional force and torque at the contact point between the end-effector and the external environment. This force feedback information is key for the robot to adjust its position and posture in real time and achieve precise operations. However, the gravity of the operating tools can cause deviations in the sensor output values, affecting the perception accuracy. Therefore, gravity compensation for sensors is crucial as it corrects the gravity component, enabling the robot to obtain more accurate information about external forces and improve operational precision.

Nowadays, an increasing number of factories have adopted robots equipped with six-dimensional force sensors on their production lines to enhance efficiency. However, when these sensors become inaccurate, the challenges arise due to their heavy weight, complex structure, and difficulty in disassembly. Field calibration is also problematic, and currently, offline calibration is the primary solution, which is both time-consuming and laborious, significantly affecting the production efficiency of our partner companies.

In response to the urgent needs of our collaborators, our research team has been working on improving the sensor structure to facilitate on-site calibration. We have designed a novel lightweight six-dimensional force sensor with a simple structure, easy disassembly, and convenient replacement for calibration. All components exhibit excellent interchangeability; even if the force-measuring unit becomes damaged, a certain level of measurement accuracy can be maintained without the need for secondary calibration, thus satisfying the demand for uninterrupted production from our partner companies.

Amidst the perpetual innovation and progression of six-dimensional force sensor technology, the sensor structure has increasingly demonstrated distinctive and varying characteristics.

Cylindrical beam and cross-beam-based six-dimensional force sensors boast high stiffness and a compact structure, yet they are not without their challenges (Payo et al., 2018; Wang et al., 2023). Throughout extended usage, strain gauges are susceptible to environmental factors, potentially compromising the sensor's measurement accuracy. Furthermore, in the event of strain gauge damage, entire sensor replacement is often necessary, significantly elevating operational costs.

The parallel six-dimensional force transducer, represented by the Stewart platform, boasts a high stiffness, making it well suited for measuring heavy loads (Yao et al., 2014). However, the Stewart-type transducer typically comprises six branches of the force-measuring unit and one preloaded branch, resulting in a super-static structure. Consequently, its mechanical analysis and calibration processes are significantly more intricate.

Capacitive six-dimensional force sensors based on microelectromechanical system (MEMS) technology are particularly suitable for precision measurement of small forces/torques (Cai and Yao, 2020), but such sensors are currently more loaded in the production and processing processes, with higher manufacturing costs, making them unsuitable for mass production.

To address the limitations of traditional sensors and meet the requirements of strong load-bearing capacity, lightweight design, high sensitivity, easy disassembly, and convenient replacement, a shunt three-legged parallel six-dimensional force sensor with rigid–flexible hybrid support is proposed.

The structure of the three-legged shunt six-dimensional force sensor proposed in this paper is shown in Fig. 1. Based on the hydrostatic principle of the six-bar closed-loop parallel mechanism, the decomposition of the six-dimensional external loads into forces F 1 and F 2 along the branches. That is, F 1 is balanced by the counterforce provided by the rigid–flexible hybrid bracket and the force-measuring unit. In the sensor branch, the load F 1 is jointly supported by the rigid–flexible hybrid bracket and the force-measuring unit. Notably, the hybrid bracket, with its remarkable stiffness, carries the majority of the load, designated as F 3 , which comprises the lion's share of F 1 . Meanwhile, the force cell assumes responsibility for a minor portion of the load, designated as F 4 . Similarly, F 2 is shared and borne by F 5 and F 6 . This innovative load-splitting design brilliantly enhances the sensor's overall measurement range, enabling the utilization of small-range force cells for the accurate measurement of even heavy loads.

Figure 1 New hybrid rigid–flexible sensor bracket.

The hybrid rigid–flexible bracket consists of an upper beam, a lower beam, and a symmetrical flexible kinematic pair, which are connected by four flexible rotating pairs. The upper beam is connected to the loading platform through a vertical pin assembly, which comprises an orthogonal complement and two short rectangular blocks located on the left and right, respectively. Each rectangular block is connected to two flexible rotating pairs along the upper beam.

The lower beam is fixed to the base, which is composed of a long upper and lower rectangular block. The two rectangular blocks are connected through a flexible rotating pair along the lower beam. All the flexible rotating pairs are non-backlash side straight round flexible hinges. Each flexible kinematic pair consists of two short rectangular blocks at the top and bottom, respectively, and two long rectangular blocks perpendicular to it, which are connected by four unilateral single-side straight round flexible hinges.

Since the left and right rectangular blocks remain parallel when they are stressed and deformed and the upper and lower short rectangular blocks also remain parallel, they form a closed parallelogram frame, whose stresses and deformations can be equivalent to the motions of a kinematic pair. The force measuring unit is placed parallel with the flexible kinematic pair and fixed to the bracket using a no-backlash bolt assembly.

The novel shunt three-legged parallel six-dimensional force sensor boasts several distinctive features:

The sensor bracket employs a load shunt design, enabling the measurement of heavy loads with a small range force unit. This design enhances the overall range of the sensor, mitigating the risk of instantaneous damage caused by significant overloads when the sensor is subjected to shock loads.

The sensor adopts a parallel three-branch structure, which is stable and avoids the hyperstatic problem caused by the need to introduce a preload branch in traditional parallel six-dimensional force sensors, thereby reducing the difficulty of theoretical analysis and calibration of the sensor.

The parallel bracket is equipped with a standard tensile force sensor as the force-measuring unit. This unit boasts high precision, reliability, and interchangeability, ensuring both measurement accuracy and long-term durability for the six-dimensional force sensor. Different ranges of force-measuring units can be utilized to adjust the overall range of the six-dimensional force sensor.

The sensor offers convenient disassembly and replacement. The sensor bracket and force measurement unit are modularized, allowing for easy replacement of the force measurement unit without damaging the rigid–flexible hybrid bracket. In the event of force measurement unit damage, there is no need to dismantle the loading platform or sensor bracket; simply replacing the damaged unit with the same model replacement ensures continued operation. The replacement sensor does not need to be calibrated again, and it can ensure a certain accuracy.

The sensor's rigid–flexible hybrid bracket is machined in one piece by a wire cutter and can be connected to the force-measuring unit without gaps by means of a developed expansion bolt assembly, guaranteeing sensor accuracy.

An approach equivalent to the static model of the parallel mechanism is utilized to solve the static model of the six-dimensional force sensor. This is because the three rigid–flexible hybrid brackets of the sensor, as designed in this study, are distributed in parallel. It is noteworthy that the rigid–flexible hybrid connection bracket's structure incorporates a flexible rotating vice with three rotating axes in the upper beam and a flexible rotating vice with two rotating axes in the lower beam. When the bracket is deformed by force, the deformation effects of the flexible rotating pairs with different rotating axes can be equated to the deformation effects of a universal joint (U joint) or screw joint (S joint), thereby rendering the sensor bracket equivalent to a six-bar closed-loop parallel mechanism.

• Theoretical static modeling of sensors

The rigid–flexible hybrid bracket, with its unique flexible hinge arrangement, can be effectively equated to a six-bar closed-loop configuration, as illustrated in Fig. 2a. This mechanism comprises four binary and two ternary rods. The equivalent parallel mechanism incorporating the sensor is depicted in Fig. 2b.

Figure 2 (a)  Equivalent six-bar closed loop of elastic bracket; (b)  equivalent six-bar closed loop parallel mechanism of sensor.

The mechanism comprises three identical branches that are uniformly distributed along the moving platform. The endpoints of the moving platform, b 1 , b 2 , and b 3 , are connected as equilateral triangles with a side length of l and a center point at o . The three endpoints B 1 , B 2 , and B 3 of the base B are connected as equilateral triangles with side lengths L and a center point at O . In the moving platform, the base is established on a coordinate system, where the moving coordinate system o – xyz has its origin at o , with the y  axis pointing towards endpoint b 2 of the moving platform, the z  axis is perpendicular to the moving platform and points upward, and the x  axis is established according to the right-hand rule of the coordinate system. The base coordinate system O – XYZ has its origin at O , with the Y  axis pointing towards endpoint B 2 of the fixed platform, the Z  axis is perpendicular to the plane of the fixed platform and points upward, and the X  axis is established according to the right-hand rule of the coordinate system. The vector h , that is b 1 Z b 1 , is perpendicular to the moving platform and intersects the upper crossbeam at the point b 1 Z . The vector H is perpendicular to the base and intersects the lower crossbeam at point B 1 Z . The Jacobi matrix (Li et al., 2016) expression for this parallel mechanism is given by

In Eq. (1) δ i and δ ij are the unit direction vectors of the virtual branch r i and the driving branch l ij , respectively, e i and e ij are the distances between b i and o and b ij and b i , respectively, e i and e ij are the vectors of the rods b i o and b i b ij , respectively ( i = 1 , 2 , 3 j = 1 , 2 ), and D 0 , D 2 , and D 3 are vectors related to the angular velocities of the rods.

Based on the principle of virtual work, it can be observed that the sum of the virtual work done by all the driving rods and the virtual work done by the generalized load is zero. The virtual displacement corresponding to the driving force F l is expressed by the velocity V l at the corresponding pose, and the virtual displacement corresponding to the loading force at the reference point of the movable platform is expressed by the generalized velocity at the corresponding pose, V , which is given by

It follows that

The Jacobi matrix J h has been derived from Eq. (3). This completes the force mapping from the external loads to the six branches.

Let K j be the matrix that maps the force from each force-measuring branch to each force-measuring unit, and from the geometry of the bracket, the sensor deformation δ v si is equal to the frame deformation δ v ei , so that

From the conservation of energy, the work done by F li on the force-measuring branch is equal to the sum of the work done by F ei acting on the flexible kinematic pairs and F si acting on the force-measuring unit:

From Eq. (5),

where k ei represents the advective stiffness of the i th flexible kinematic pair, and k si represents the stiffness of the i th force-measuring unit. Therefore, the 6  ×  6 stiffness matrix K j is

k e is the stiffness of the flexible pair consisting of four single-sided straight circular flexible hinges, the expression of which is given in Eq. (4).

To calculate the static mapping array, it is essential to determine the stiffness of the force measurement unit. However, as the internal structure and composition of standard tensile force sensors are not well understood, there is currently no fixed formula for their stiffness. This study employs a combination of theoretical and experimental methods to determine the stiffness of the sensor. By applying pressure F z in the Z direction on the six-dimensional force sensor loading platform and observing the sensor's indication, the stiffness of the force-measuring unit can be calculated using Eq. (6) as

From Eq. (6), the six force measurement branches F li to six force measurement units F si force mapping can be derived from the stiffness relationship between the two:

This completes the static model of the sensor equivalent to the model of the six-bar closed-loop parallel mechanism.

The static model lays the theoretical foundation for the design of the sensor prototype, which is then machined and manufactured. The overall assembly of the sensor is shown in Fig. 3.

Figure 3 Prototype of sensor.

Accurate hydrostatic modeling can significantly enhance the ability of sensors to deliver precise force measurements in complex environments. Nevertheless, the performance of a sensor is not solely determined by its structural design and hydrostatic modeling; rather, it is also strongly influenced by its calibration process. Calibration is a crucial phase in optimizing the sensor's performance, ensuring its accurate measurements in real-world applications. Calibration compensation is the most effective approach to enhancing the performance of sensors under existing hardware constraints. The accuracy of calibration has a direct impact on the measurement accuracy of sensors (Yao, 2010). This step is crucial for assessing the actual input–output performance of a sensor.

4.1  Linear static calibration experiment program

Due to the fact that the actual prototype cannot be an absolutely linear system, calibration experiments involve repeated loading and unloading processes in various directions. The loads should encompass the entire range of the sensor, and the average value of the data should be taken to minimize errors caused by system nonlinearity and randomness, thereby enhancing calibration accuracy.

The sensor calibration steps are as follows:

divide the full measurement range into several loading values for each direction;

adjust the calibration device, zero the tool, connect the data cable, and debug the data acquisition system;

gradually increase the load to maximum value according to the loading plan in one loading direction, then gradually reduce the load to 0; repeat more than three times, recording the corresponding data (The − F z direction calibration experiment is shown in Fig. 4.);

follow step 3 for reversed load; repeat more than three times and record the data;

apply loads in the other directions separately according to steps 3 and 4 to complete the data collection for all six load components;

process the collected data, obtain the sensor calibration matrix, and determine indicators such as error matrix, repeatability, and hysteresis.

Figure 4 Calibration experiments for − F z directions.

4.2  Experimental calibration results and their fitting

Consider experimental data corresponding to one load case. The measurement results are shown in Fig. 5a; it can be seen that when loading the sensor in the translational z  direction, the forces measured by each of the measuring unit were different. Because it was difficult to ensure that the calibration loading point was exactly at the center of the platform, the average value of all force-measuring units can be used, as shown in Fig. 5b. The calibration results are listed in Table 1. It can be seen that the maximum error between the measured average values and the simulated values in each loading direction was 4.97 % in the rotational x  direction, and the maximum coupling error generated in the non-loaded direction was 5 %. The results of multiple experiments need to be considered to reduce random errors.

Figure 5 Measured ( F sje ) and simulated ( F sj ) forces due to incremental loads in z  directions. (a)  Loading in F Z direction; (b)  load in F Z direction and takeing the average value.

Table 1 Example of calibration results.

Download Print Version | Download XLSX

4.3  Nonlinear static calibration of sensors

Calibration experiments allow for the procurement of precise output data from the six-dimensional force sensor under defined load conditions, serving as a pivotal metric for evaluating sensor performance. Nevertheless, to fully realize the potential of these data in achieving high-precision force measurements, it is imperative to address the issue of coupling in the sensor's output signals. The application of decoupling algorithms offers a straightforward and cost-effective solution for this endeavor, representing the primary method of decoupling.

The decoupling of a multi-dimensional force sensor essentially involves establishing a unique least-error input–output relationship for the sensor (Li et al., 2017). However, it is challenging to intuitively determine the relationship between experimental measurements and the input load. Therefore, mathematical methods are required to fit the data, enabling prediction of the output due to an unknown load. Common data fitting methods include end-base, least squares, neural networks, and fuzzy inference (Wu, 2022).

Assuming that the sensor is a linear system, the least squares method can achieve high fitting accuracy (Liu, 2023). However, the six-dimensional force sensor system is generally nonlinear (Yang et al., 2022), and a fitting method that can approximate the nonlinear relationship with high accuracy is required.

Deep learning models, particularly those excelling in handling complex and nonlinear aspects such as artificial neural networks (ANN), one-dimensional convolutional neural networks (1DCNN), long short-term memory (LSTM), bidirectional LSTM (BiLSTM), and attention mechanisms, could potentially provide effective solutions for decoupling the output data from six-dimensional force sensors. These models, through learning and training, can automatically extract useful features from the input data and discover complex mapping relationships between inputs and outputs.

ANN is an ideal tool for decoupling the output data from six-dimensional force sensors due to its unique characteristics, facilitating accurate decoupling and reliable force measurement. Here are the reasons why ANN is suitable for this task:

Parallel processing. The ANN's ability to process information in parallel enables it to efficiently handle the vast amount of data output from six-dimensional force sensors.

Self-learning and adaptability. The ANN can automatically optimize its parameters to adapt to different environments and task requirements, making it resilient to factors such as ambient temperature and sensor aging.

Strong fault tolerance and robustness. In practical applications, the output data from six-dimensional force sensors may be affected by various noises, including mechanical vibrations and temperature drifts. However, ANN can, to a certain extent, ignore these disturbances and extract useful information from the data.

Comparatively, the 1DCNN excels in processing data with significant internal correlations, such as pixel sequences in images or time-series data (Ye and Li, 2022). However, the output data from six-dimensional force sensors typically encompass force and torque information across six distinct dimensions, where the internal local correlation among this information is not prominent. Consequently, in such scenarios, the advantages of the 1DCNN may not be fully realized.

LSTM and BiLSTM networks excel at capturing long-term dependencies in sequential data (Siami-Namini et al., 2019). However, if the output data from a six-dimensional force sensor lack distinct sequential characteristics, meaning that the force and torque information across various dimensions are not closely related sequentially, then these methods may not be the optimal choice.

The application of attention mechanisms in models can assist in automatically learning and focusing on the most relevant parts for decoupling tasks while ignoring unimportant information (Qin and Hu, 2020). However, when dealing with six-dimensional force sensor data, complex cross-interference issues may arise, necessitating stronger feature extraction and representation capabilities. To achieve this, it may be necessary to integrate other nonlinear fitting methods, such as deep neural networks, but such operations may increase model complexity and computational costs.

The neural network method, which defines the input–output relationships of a model by mimicking the structure of neurons in the brain, is a commonly used nonlinear fitting method in engineering (Ma et al., 2019). The neural network method has a powerful self-organizing learning capability (Fan et al., 2019). By using a large amount of linear calibration data as the basis, the artificial neural network method can more accurately predict sensor measurements. The sensor neuron model studied in this paper is shown in Fig. 6.

Figure 6 Neuron model of sensor.

The input–output relationship of the model is

In Eq. (11), f sj ( j =  1, 2, 3, 4, 5, 6) is the input signal of the i th neuron, y i is the output of the i th neuron, θ i is the bias value of this neuron, w ij is the connection weight from cell i to cell j , u (⋅) is the output basis function obtained by superpose of input signals; and g (⋅) is the excitation function of the neuron. The excitation function is taken as an S -type function, which maps the input signal of the neuron onto the interval ( - 1 , 1 ) ; the function is

In this study, the LMBP algorithm, a derivative of the backpropagation (BP) algorithm based on the Levenberg–Marquardt (LM) method, is employed as the optimization technique. LMBP integrates the gradient descent method and the Gauss–Newton method within neural networks (Wang et al., 2019), leveraging the advantages of both techniques. This approach not only expedites the training speed of the network but also mitigates the issue of local minima during convergence, thereby enhancing the overall stability of the neural network system. For the purpose of this study, the “trainlm” function available in MATLAB software was utilized for implementing neural network training. The network architecture comprises a total of 10 layers, with the input layer receiving the static linear calibration experimental data obtained from six force measurement units. The output layer generates the six-dimensional force acting on the loading platform. The neural network training results are summarized in Fig. 7, which provides a visual representation of the network's performance.

Figure 7 Results of neural network training.

To ensure accurate calibration, the collected experimental data were used as inputs, and the 194 sets of sample data were divided into three separate groups. The training group, consisting of 70 % of the samples, was utilized to train the neural network and fine-tune its weight parameters based on error feedback. The calibration set, comprising 15 % of the samples, was employed to assess the network's generalization capability. This capability refers to the machine learning algorithm's ability to adapt to novel samples. Networks with robust generalization abilities can also produce accurate outputs for input data that fall outside the learning set. The training process was terminated once the network's generalization ability no longer improved. The remaining 15 % of the samples formed the test group, serving as independent samples to evaluate the quality of the network during and after training.

After 797 iterations, the network's generalization ability attains an optimal state, with the mean square error achieving a magnitude of 10 −5 . At this point, all neural network weight values are determined, resulting in a highly accurate nonlinear model for fitting experimental data. When six force measurement unit values are used as inputs, the six-dimensional force outputs obtained via the network closely align with the actual applied six-dimensional forces, with a maximum error of 0.95 % in the loaded direction and a maximum coupling error of 1.33 % in the unloaded direction. It is worth noting that as more sample data are input during training, a higher post-fitting accuracy is achieved.

4.4  LabVIEW-based sensor calibration software

LabVIEW, a graphical programming language (Zang et al., 2023), is a powerful tool that streamlines complex procedures into intuitive flowcharts. Its robust data interface enables seamless integration with various software components. In this study, we integrate the MATLAB-trained neural network into the LabVIEW software to develop a calibration software for the sensor. This software utilizes collected data to establish a direct mapping with the externally applied six-dimensional force, ultimately facilitating nonlinear calibration of the sensor.

The nonlinear calibration software workflow is designed as follows: the collected experimental data signals are converted into a data matrix format. Subsequently, the input data undergo a zeroing process within the software to ensure the accuracy of the test data in the absence of a zero setting. The processed matrix is then fed into the neural network unit. The neural network generates output data that are converted into the six-axis force components, which are then displayed.

The trained neural network module of MATLAB software is called and embedded in the program via a script node, as shown in Fig. 8, and the node outputs the result of the neural network operation.

Figure 8 Calling the neural network module in LabVIEW.

Figure 9 presents the panel of the nonlinear calibration procedure, which illustrates the processed data matrix after the zeroing step, the rows of data to be converted, the values of the six-dimensional force components of the external loads, and the amplitudes of the force measurement unit and the external loads.

Figure 9 Nonlinear static calibration software.

The nonlinear static calibration software offers a visual representation of the sensor loading platform's impact on the external load's component values. Additionally, it enables observation of the force unit's amplitude and external loads, preventing prolonged overload operation of the system. The software features intuitive and straightforward operation, facilitating easy debugging.

The decoupling algorithm effectively addresses the coupling interference in the output signal of the sensor, significantly enhancing the accuracy and stability of measurements. This lays the groundwork for the practical implementation of the six-dimensional force sensor. When the robot is stationary or moving at low speed and when the sensor measures the contact force between the end actuator and the environment, the captured data include non-contact forces, such as gravity, along with the actual contact force. This introduces a certain bias in the sensor's readings. To accurately represent the true end force, gravity compensation is necessary. Algorithms are employed to compensate for this, enhancing both the operational performance and adaptability of the parallel robot. This paper proposes a gravity compensation method based on CAD variable geometry to enhance the measurement accuracy of sensors.

Jin et al. (2022) proposed a gravity compensation method that combines active and passive compensation techniques, aiming to enhance the force feedback performance of tactile devices. This method features a simple principle that does not require complex reasoning or calculations, making it easy to implement. However, it is currently only suitable for tactile devices that perform translational movements. When a six-dimensional force sensor is mounted on the wrist of a robot for gravity compensation and the robot typically performs spatial movements, solving for the forward kinematics of the robot's spatial position becomes necessary. In such cases, a gravity compensation method based on CAD variable geometric methods is more suitable for compensating the gravity effects on the six-dimensional force sensor at the wrist of a parallel robot. Zhao et al. (2020) employed multi-body simulation techniques to devise a gravity compensation method for the flexible mesh of a mesh antenna. This approach effectively reduces the discrepancies between ground verification data and on-orbit flight data for mesh antennas, thus ensuring the successful on-orbit operation of satellites. However, when robots perform high-precision tasks such as flexible machining and assembly, their drive modules must adjust their position and attitude based on real-time force information feedback from six-dimensional force sensors. Utilizing CAD variable geometric methods, a wire-frame model can be used to establish a one-to-one correspondence between the physical robot and its wire-frame representation. This allows for real-time and accurate simulation of the forward kinematics of parallel robots, enabling online gravity compensation for six-dimensional force sensors mounted on the robot's wrist. Therefore, this method offers superior advantages in robotic applications such as flexible machining and assembly.

5.1  Gravity compensation algorithms

In the absence of systematic error and without any external environmental contact forces such as human hand force or collision force, the signal generated by the six-dimensional force sensor is solely due to the tool's gravity (Chun, 2022). The orientation of the tool's gravity within the sensor's coordinate system is depicted in Fig. 10.

Figure 10 The direction of the tool's gravity in the sensor coordinate system.

To establish a right-angle coordinate system o – xyz with the center of the sensor loading platform as the origin, the coordinates of the center of gravity of the tool are ( x G , y G , z G ) , and the gravity vector is designated as G . The components of G within the sensor coordinate system can be expressed as ( G ox , G oy , G oz ). The moment vectors of G relative to the x , y , and z axes are designated as M Gx , M Gy , and M Gz . From the positive direction of each coordinate axis, with the moment counterclockwise to the positive. According to Fig. 11, the following relationships hold for each component of the gravity and moment vectors:

Equation (13) can be rewritten in matrix form:

M Gx , M Gy , and M Gz in Eq. (14) can be measured by a six-dimensional force sensor.

Figure 11 (a)  Wire frame of the 3-SPR parallel mechanism. (b)  Parallel robot hybrid hand with sensor.

Given that the direction of the tool's gravity, defined by the coordinates of its center of gravity ( x G , y G , z G ) , varies with the pose of the parallel robot, it is essential to establish the relationship between the center of gravity coordinates and the robot's pose. To this end, the developed six-dimensional force sensor was integrated into the wrist of the existing 3-screw-prismatic-revolute (3-SPR) parallel robot, which is used to connect the robot's hand claw with the body and measure the six-dimensional force signal of grasping an object. The robot serves as the executing mechanism, responsible for tasks such as flexible machining and assembly. Meanwhile, the six-dimensional force sensor functions as a feedback mechanism, providing real-time force information that allows the robot to adjust its position and posture dynamically, thereby completing machining tasks with greater precision. Figure 11a and b provide physical drawings of the 3-SPR parallel robot and its structural wire frame with dimensions, respectively and o – x d y d z d is the coordinate system of the movable platform.

The combination structure of the parallel robot is as follows: the base of the six-dimensional force sensor at the wrist is solidly connected to the dynamic platform of the parallel mechanism, and the sensor loading platform is solidly connected to the three single-degree-of-freedom grippers. The bottom of the SPR branch is fixed to a flat steel plate by a strong magnet, and the magnet base is connected to the S joint of the branch. In case of any interference or abnormal rod force, the magnet can be automatically detached from the steel plate, preventing potential damage to the robot's components. The control system incorporates a PC as the main controller, with the motion control card connected to its PCI (Peripheral Component Interconnect) interface. This configuration ensures robust and reliable control. Technical specifications include a maximum torque of 0.88 N m for the branch stepping motor and 0.45 N m for the hand claw stepping motor. The stepping motor driver features a refined control system with 15 subdividing steps and an automatic half-current function, ensuring smooth and precise output currents ranging from 0.64 to 2.14 A. The motor driver interfaces directly with the motion control card, allowing individual control of each motor through distinct interfaces.

From the installation form of the sensor, it can be seen that the sensor coordinate system is in the same direction as the coordinate system of the moving platform of the parallel mechanism. Then the gravity of the tool in the wrist six-dimensional force sensor coordinate system of the projection of the three coordinate axes are

where α d , β d , and γ d are the angles between the tool gravity and the three coordinate axes of the sensor coordinate system, and G ox , G oy , and G oz are the tool gravity components. Substituting Eq. (15) into Eq. (13) yields

Given the challenges associated with theoretically determining the position positive solution for the 3-SPR mechanism and the real-time angle between gravity and the sensor coordinate system, this study opts to utilize a wire-frame diagram of the mechanism to simulate the positional state of the moving platform in real time. Figure 12 depicts a representative state of the parallel robot during the gripping tool load movement.

Figure 12 Wire-frame simulation of motion process of parallel mechanism.

In Fig. 12, different color annotations are used to distinguish various dimensions and parameters. Specifically, green annotations indicate the fixed length of the 3-SPR robot's gripper, which measures 170 mm. Red annotations represent the length of the 3-SPR robot's prismatic pairs at the current pose. Yellow annotations denote the height between the sensor center and the moving platform of the parallel mechanism. Blue annotations depict the Euler angles, which are used to describe the robot's pose information. Black annotations indicate the angles between the tool's gravity and the three axes of the sensor coordinate system. Additionally, points a , b , and c in the figure mark the fingertip sections of the robot's gripper.

Notably, the lengths of the three parallel branches undergo changes, decreasing from the initial 445 to 350, 390, and 440 mm, respectively. Similarly, the lengths of the three hand claw drive rods increase from the initial 137 to 170 mm. The parallel mechanism's moving platform coordinate system is defined by the x d , y d , and z d axes, whereas the base coordinate system of the parallel mechanism consists of the x b , y b and z b axes. The Euler angles α , β , and γ represent the rotation angles between the two coordinate systems. The motion control software acquires real-time angle data based on the dimensional parameters obtained from the wire-frame diagram: α =  13.47°, β =  1.06°, γ =  13.52 °, α d =  103.47 °, β d =  91.06 °, γ d =  13.52°. These values are then inserted into Eq. (16) to determine the tool's gravity component under the sensor system in its pose. To prevent singular matrices from emerging during calculations, the motion control of parallel robots involves selecting N non-coplanar poses and acquiring corresponding six-dimensional force signals to generate N sets of data.

To determine the coordinates of the tool's center of gravity in the sensor's coordinate system versus the magnitude of the force of gravity, Eq. (17) is employed, and then the coordinates of the center of gravity in the base coordinate system can be calculated by using the Euler angles α , β , and γ and pose vectors collected from the wire-frame diagram. Let F ox , F oy , and F oz represent the force signals captured by the six-dimensional force sensor. After accounting for gravity compensation, the external force exerted on the six-dimensional force sensor by the external environment can be determined

According to Eq. (18), it is evident that the gravity-compensated six-dimensional force sensor signal can accurately reflect the contact force between the gripped tool and the external environment, which provides accurate force feedback information for the active supple control of parallel robots.

5.2  Gravity compensation experiments

In this section, a tool gravity compensation experiment is conducted on the wrist six-dimensional force sensor of the 3-SPR parallel robot. A load is applied to the tool, and the six-dimensional force generated by this load on the sensor in different poses is measured. The parallel robot control system is constructed using a PC and motion control card, and the motion control software interface is presented in Fig. 13a. The software is based on the CAD variable geometry principle and SolidWorks API interface for software secondary development, enabling virtual teaching, motion control, and other functions. The CAD variable geometry method offers a simple, efficient, and practical approach to solving the pose, velocity, and acceleration of a parallel mechanism. Users can quickly solve it without deep background knowledge (Xu, 2010).

Figure 13 (a)  Motion control software of parallel mechanism; (b)  motion parameter setting of parallel mechanism.

The experiment is conducted with multiple sets of sampling points to collect six-dimensional forces from the tool to the sensors in various poses of the parallel robots, along with simultaneous robot trajectory planning. Prior to the experiment, the robot requires adjustment and testing. Initially, the motion parameters of the hand claw and leg drive units of the parallel robot are established, as illustrated in Fig. 13b. Axes 2, 3, and 4 represent the robot leg drive units; axes 5, 6, and 7 represent the robot hand claw drive units; and axes 1 and 8 are unoccupied. After entering the appropriate fine fraction and pitch, the zero pose of all drive units is adjusted individually using the pointing mode, with the initial extension of the leg bar set at 445 mm and the initial extension of the hand claw set at 137 mm.

After setting the speed and acceleration of all drive units, the interpolation mode is used to achieve simultaneous movement of one or more drive units to observe whether the program is running normally. It is determined that there is no collision or interference of parts during the planned robot movement. Finally, the robot linkage is initiated for the gravity compensation experiments. The gravity compensation experiment process is presented in Fig. 11b.

The steps of the gravity compensation experiment are as follows:

We collect the six-dimensional force data of the sensor in the no-load state of the parallel robot and check whether the zeroing function of the acquisition software is normal.

The parallel robot hand grips are controlled to grasp the tool and move to the initial pose, and six-dimensional force data are collected.

The parallel robot is controlled to move continuously from the first sampling point and simultaneously collect six-dimensional force data from all set sampling points.

Returning to the initial pose, a heavy load was applied to the tool, and the six-dimensional force generated by the tool weight on the sensor was recorded for different poses.

We return all loads to the ground, release the three hand claws, and return the parallel robot to its initial pose.

Some of the sampling point data and force components collected by the sensors in the experiment are shown in Table 2.

Table 2 Experimental data of partial sampling point.

Using Eq. (17) and the data presented in Table 2, we can infer that the tool's average gravity is approximately 66 N. Furthermore, the center of gravity of the tool, when considered within the sensor system's coordinates, is located at [4.73 mm, − 8.63 mm, 271 mm]. To determine the coordinates of this center of gravity within the base coordinate system, we utilize the pose transformation matrix between the base and sensor coordinate systems. Once the heavy load is added, the combined weight of the tool and load amounts to 137 N. The gravity exerted by this heavy load alone can be calculated as 137–66 N  =  71 N. By applying Eq. (18), we can also determine the six-dimensional force components of this heavy load at a specific attitude, along with the center of gravity of the tool plus load within the sensor system's coordinates. Additionally, to measure the weight of the load, it is placed on a weighing scale, resulting in a reading of 7.05 kg. The measurement error in load gravity after compensation for gravity is then determined:

In terms of the validation of the coordinate values of the center of gravity of the tool within the sensor coordinate system, the location of the form center of the model can be easily determined due to the regular shape of the chosen tool model for the experiment. Assuming a uniform distribution of material mass within the tool, the center of the form and the center of gravity can be considered co-incident points. The coordinate value of the center of gravity of the tool within the sensor coordinate system can be determined through the use of a wire-frame diagram. The coordinates of the center of gravity of the model are (4.66, − 8.51, 272.16 mm), which represents an error of approximately 0.43 % in the z -axis component when compared to the computed value obtained through the gravity compensation algorithm.

The gravity compensation algorithm is effective in terms of reducing the load gravity measurement error and the center of gravity coordinate error. Following gravity compensation, the tool grasped by the parallel robot can accurately sense the six-dimensional force exerted by the external environment on the wrist. Additionally, it is capable of determining the location of the center of gravity of the load, thereby providing a foundation for force-following control.

5.3  Measurement accuracy after replacing the force measurement unit or bracket

To meet the requirements of continuous production in collaborating enterprises, one of the design objectives of the sensor developed in this paper is to facilitate easy installation and disassembly, ensuring good interchangeability of all components. Therefore, commercial standard tensile and compressive force sensors are selected for the force-measuring unit, and the support frame is integrally processed using wire cutting to ensure that even if the sensor or support frame is replaced, a certain level of measurement accuracy can be maintained without the need for secondary calibration.

Therefore, this paper proposes accuracy indicators for the replacement of the support frame and force-measuring unit to assess the interchangeability of sensor components.

An experiment is designed in this paper to test these indicators by grasping the same heavy object through the interchange or replacement of the force measuring unit and support frame. After the components are replaced, the entire sensor is placed on a 3-SPR parallel robot, and three claws are installed on the sensor's loading platform to simulate actual working conditions, as shown in Fig. 14. The claws grasp a sand bucket with an unknown mass, and the parallel robot's driving rods and claws are controlled to maintain the same pose before and after the component replacement to avoid deviation in the loading direction. The readings from the six force measuring units are then recorded, and the theoretical weight of the sand bucket is calculated using a theoretical static force model. The actual weight of the sand bucket is then measured using a standard scale, and the deviation between the two is compared.

Figure 14 Test experiment of replacement accuracy index of force unit or bracket.

Table 3 Six-dimensional force caused by sand bucket acting on the sensor.

After conducting multiple measurements and averaging the results, the experimental outcomes are summarized in the Table 3. It can be observed from the table that the primary force direction of the sensor is in the F z  direction. Using a standard scale, the mass of the unknown sand bucket is measured to be 6.8 kg. Based on this, it is determined that the measurement error in the F z direction after interchanging and replacing parts is within 1.2 %. The replacement of the force-measuring unit of the sensor can be quickly put into use without the need for secondary calibration, thus saving time for on-site calibration. Improving the machining precision of the bracket can enhance the accuracy of replacing the sensor bracket and force-measuring unit.

In light of the sensor structure's unique characteristics, a method is proposed for modeling the sensor's static force as an equivalent parallel mechanism sensor containing a six-bar closed-loop parallel mechanism.

Sensor static calibration experiments were conducted, and the deviation between the experimental values obtained from a single loading experiment and the simulated values was within 5 %. The artificial neural network approach was employed to fit multiple sets of experimental data, revealing a satisfactory decoupling effect.

A gravity compensation method for the six-dimensional force sensor of the wrist of a robot is proposed based on the CAD variable geometry method. To demonstrate its effectiveness, 3-SPR-type parallel robot hybrid wrist six-dimensional sensor gravity compensation experiments were conducted. The experimental results indicate that the gravity compensation method enables the sensor to more accurately perceive the six-dimensional force of the external load on the operating tool, with load gravity and gravity coordinate values within 1 % error. The objective of our research is to devise six-dimensional force sensors that possess robust applicability and outstanding performance characteristics. To achieve this, further endeavors are requisite in the following areas. In the static calibration experiment of the six-dimensional force sensor, the manual adjustment of the weight masses for each set of data calibration significantly hinders calibration efficiency. Consequently, the current calibration device exhibits a low level of automation, necessitating further design iterations and enhancements.

All data are included in this study are available upon request by contacting the corresponding author.

YW conceptualized the study, wrote the original draft of the paper, and reviewed and edited the paper. KJ wrote the paper and reviewed and edited the paper and data curation. XL contributed to the data curation and investigation and wrote the paper. FC and XY contributed to the methodology and formal analysis.

The contact author has declared that none of the authors has any competing interests.

Publisher’s note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors.

The authors would like to express their heartfelt gratitude to everyone who helped throughout the process of preparing this paper, especially the editors, reviewers, and the academic leader.

This research has been supported by the Joint Funds of the Zhejiang Provincial Natural Science Foundation of China under grant no. LTY22A020001 and the Postgraduate Research and Innovation Project of School of Engineering, Huzhou University, grant no. 2024GXYKYCX11.

This paper was edited by Daniel Condurache and reviewed by Van Sy Nguyen and one anonymous referee.

Cai, D., Yin, H., Chen, L., Xu, Y. D., and Yao, J.: Analysis and design of the multi dimensional force-sensing mechanism based on the capacitive micro-displacement sensing unit, Mach. Des., 37, 43–50, https://doi.org/10.13841/j.cnki.jxsj.2020.12.007 , 2020. 

Chao, L. and Chen, K.: Shape optimal design and force sensitivity evaluation of six-axis force sensors, Sensor. Actuat. A-Phys., 63, 105–112, https://doi.org/10.1016/S0924-4247(97)01534-3 , 1997. 

Chun, C.: Research on robotic grinding system based on vision and force sense, Qingdao University of Science and Technology, Master thesis, https://doi.org/10.27264/d.cnki.gqdhc.2022.000478 , 2022. 

Fan, X., Xu, Y., Zhou, J., and Li, Z.: Application of genetic algorithm and wavelet neural network in ET_0 prediction, Journal of Yanshan University, 43, 182–188, 2019. 

Hou, Y., Yao, J., Lu, L., and Zhao, Y.: Performance analysis and comprehensive index optimization of a new configuration of Stewart six-component force sensor, Mech. Mach. Theory, 44, 359–368, https://doi.org/10.1016/j.mechmachtheory.2008.03.008 , 2008. 

Jin, L., Duan, X., He, R., Meng, F., and Li, C.: Improving the Force Display of Haptic Device Based on Gravity Compensation for Surgical Robotics, Machines, 10, 903, https://doi.org/10.3390/machines10100903 , 2022. 

Jung, J., Van Sy, N., Lee, D., Joe, S., Hwang, J., and Kim, B.: A single motor-driven focusing mechanism with flexure hinges for small satellite optical systems, Appl. Sci., 10, 7087, https://doi.org/10.3390/app10207087 , 2020. 

Li, X., Liu, Y., Lu, Y., and Zhang, C.: Analysis of kinematics and statics for a novel 6-DoF parallel mechanism with three planar mechanism limbs, Robotica, 34, 957–972, https://doi.org/10.1017/S0263574714001994 , 2016. 

Li, Y., Han, B., Wang, G., Huang, S., Sun, Y., Yang, X., and Chen, N.: Decoupling algorithm for piezoelectric six-dimensional force sensor based on radial basis function neural network, Opt. Eng., 25, 1266–1271, 2017. 

Liu, F.: Research on image shift compensation method and its control technology of three-axis stabilized image aerial camera, Changchun University of Technology, Master thesis, 2023. 

Ma, J., Jiao, Z., Xu, C., Yang, J., and Ma, K.: Optimal scheduling of aggregated thermostatically controlled loads based on neural network PID control, Journal of Yanshan University, 43, 417–422, 2019. 

Pan, J., Li, B., Nie, A., and Liu, G.: Research on prediction of surface roughness of abrasive belt grinding and polishing combined with CSSA-BP neural network, Machine Tools and Hydraulics, 51, 80–86, https://doi.org/10.3969/j.issn.1001-3881.2023.22.013 , 2023. 

Payo, I., Adanez, J., Rodriguez, R., Fernandez, R., and Vazquez, A.: Six-Axis Column-Type Force and Moment Sensor for Robotic Applications, IEEE Sens. J., 18, 6996–7004, 2018. 

Qin, Q. and Hu, X.: The Application of Attention Mechanism in Semantic Image Segmentation, in: 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), Chongqing, China, 12–14 June 2020, IEEE, 1573–1580, https://doi.org/10.1109/ITNEC48623.2020.9084663 , 2020. 

Siami-Namini, S., Tavakoli, N., and Namin, A. S.: The Performance of LSTM and BiLSTM in Forecasting Time Series, in: 2019 IEEE International Conference on Big Data (Big Data), Los Angeles, CA, USA, 9–12 December 2019, IEEE, 3285–3292, https://doi.org/10.1109/BigData47090.2019.9005997 , 2019. 

Sun, S., Pang, K., Yu, J., and Chen, R.: Nonlinear decoupling of three-dimensional force sensors based on white shark optimized limit learning machine, Opt. Eng., 31, 2664–2674, 2023. 

Wang, H., Lu, Z., Wang, S., and Li, W.: Artificial intelligence-based battery health assessment for power grid, Guangdong Electric Power, 32, 6, https://doi.org/10.3969/j.issn.1007-290X.2019.004.011 , 2019. 

Wang, Z., Zhang, X., and Li, M.: Research on decoupling algorithm of six-dimensional force sensor based on polynomial fitting, J. Eng. Design, 30, 571–578, 2023. 

Wu, J., Li, T., Wang, J., and Wang, L.: Performance analysis and comparison of planar 3-dof parallel manipulators with one and two additional branches, J. Intell. Robot. Syst., 72, 73–82, https://doi.org/10.1007/s10846-013-9824-8 , 2013a. 

Wu, J., Li, T., Wang, J., and Wang, L.: Stiffness and natural frequency of a 3-DOF parallel manipulator with consideration of additional leg candidates, Robot. Auton. Syst., 61, 868–875, https://doi.org/10.1016/j.robot.2013.03.001 , 2013b. 

Wu, J., Wang, L., and Guan, L.: A study on the effect of structure parameters on the dynamic characteristics of a PRRRP parallel manipulator, Nonlinear Dynam., 74, 227–235, https://doi.org/10.1007/s11071-013-0960-2 , 2013c. 

Wu, S.: Research on spatially spliced reflector mirrors for curvature correction, University of Chinese Academy of Sciences, PhD dissertation, https://doi.org/10.27522/d.cnki.gkcgs.2022.000011 , 2022. 

Xu, J.: Research on key technology for machining complex surfaces by parallel machine tools based on CAD variable geometry method, Yanshan University, PhD dissertation, 2010. 

Yang, C., Xiao, X., and Tian, Y.: Design and calibration method of orthogonal parallel six-dimensional force sensor, Science and Technology Innovation and Application, 12, 94–96+100, https://doi.org/10.19981/j.CN23-1581/G3.2022.33.024 , 2022. 

Yao, J.: Basic theoretical and experimental research on large-range parallel six-dimensional force sensor, Yanshan University, PhD dissertation, 2010. 

Yao, J., Li, L., Xu, Y., and Zhao, Y.: Redundant parallel structure six-dimensional force sensor and its statically indeterminate static force mapping analytical analysis, Engineering Mechanics, 31, 205–211, 2014. 

Ye, X. and Li, M.: Press-fit process fault diagnosis using 1DCNN-LSTM method, Assembly Autom., 42, 342–349, https://doi.org/10.1108/AA-06-2021-0072 , 2022.  

Yin, R., Wang, X., and Wang, J.: Characteristics of generalized six-dimensional force sensors and the status of research and application, Hydraulic. Pneum., 10, 51–53, 2005. 

Zang, C., Zeng, J., and Li, P.: An intelligent diagnosis model of mechanical faults in power transformers based on SVM algorithm, High Voltage Electrical Appliances, 59, 216–222, https://doi.org/10.13296/j.1001-1609.hva.2023.12.026 , 2023. 

Zhao, K., Xu, Z., and Yu, Z.: Research on the design of a new six-dimensional force sensor for industrial robots, Sensors and Microsystems, 34, 5–7, https://doi.org/10.13873/J.1000-9787(2015)05-0005-03 ​​​​​​​, 2015. 

Zhao, Z., Fu, K., Li, M., Li, J., and Xiao, Y.: Gravity compensation system of mesh antennas for in-orbit prediction of deployment dynamics, Acta Astronaut., 167, 1–13, https://doi.org/10.1016/j.actaastro.2019.10.021 , 2020. 

Zhu, J.: Development of two-dimensional wireless passive force sensor based on magnetostrictive effect, Northeast Electric Power University, Master thesis, 2019. 

• Introduction
• Sensor structural design
• Sensor static calibration experiment
• Sensor applications and their performance enhancement
• Conclusions
• Data availability
• Author contributions
• Competing interests
• Acknowledgements
• Financial support
• Review statement

Using Infrastructure Gaps as Social Sensors for Informing Equity Aims in Policymaking

 My dissertation work aims to assess the feasibility of using established gaps in equity due to infrastructure provision as a mechanism for , rather than a byproduct of , infrastructure investment policy to address issues of social equity created and perpetuated by infrastructure systems. To explore this, I start by assessing the potential of using large-scale infrastructure networks as social sensors to detect aspects of inequity to better inform investment policy. I focus my exploration on broadband infrastructure to begin with, exploring the possibility of using U.S. county-level broadband penetration rates as a social sensor to predict rates of unemployment amidst the COVID-19 pandemic (Chapter 2). This work specifically asks, “How can infrastructure serve as a social sensor that allows for sharper detection of those groups which are most vulnerable to disruption?”. I find that broadband can serve as an effective social sensor which is sharpened when applied to employment contexts relevant to broadband, but, as with any sensor, is prone to error (either false positives or false negatives). I then shift my interest from the macro-system to a more micro-focused approach of how to incorporate preferences from end-users into the investment process. To do this, I develop an innovative approach to incorporating qualitative interview responses into a multi-criteria decision-making process (Chapter 3). I find that hauled system water users in Alaska have a strong preference for the aesthetic properties of their water provision which they balance against the need for reliable water system delivery. I end my investigation by understanding the role that skills play as a sensor for detecting effective and equitable use of infrastructure. To do this, I explore broadband connectivity throughout Rwanda and its impact on a critical aspect of development, public health (Chapter 4). To explore this question, I ask to what degree are wireless biomedical devices (specifically EKGs) available and used in the public hospital system in Rwanda? And what impact does broadband access have on the kinds of services which are offered? I find that the first tier of the digital divide influences the ability of offer telehealth services and propose additional future work on the compounded impacts of this access on both second and their tier access.  

Degree Type

• Dissertation
• Engineering and Public Policy

Degree Name

• Doctor of Philosophy (PhD)

Usage metrics

• Engineering not elsewhere classified