The Anatomy of an Infrastructure for Digital Underground Mining

Sreekant Sreedharan

1,*

, Muthu Ramachandran

2,†

Soma Ghosh

3,‡

and Suraj Prakash

1,§

TEXMiN Foundation, IIT-ISM Dhanbad, 3

Floor CRE Building, Dhanbad, India

Department of Computer Science, University of Southampton, Southampton, U.K.

Engineering and Architecture AI/ML, JPMorgan Chase & Co., Bangalore, India

Keywords: Architecture, Industrial IoT, Mining4.0, Edge Computing, Distributed Computing, Smart Infrastructure,

Private Cloud Computing, Real-Time Operating Systems, Wireless Mesh Networking, Wearable Computing,

Distributed Automation.

Abstract: Over 41.6 billion IoT devices are expected to come online by 2025, collectively capable of generating 80

zettabytes (ZB) of data. Despite the relentless progress in adoption of smart devices occurring around all us -

in our smart homes, our wearable devices, our smart cities and workplaces -, progress in the adoption of smart

technology in the mining sector has languished. Yet the mining sector powers our energy systems and makes

components for smart devices possible, while employing over 4.5 million people world-wide in some of the

most extreme & hostile environments. This paper presents the design of a prototype industrial IoT platform

for large-scale industrial automation of conventional mines.

1 INTRODUCTION

Mines in India are located in some of the country’s

most remote and inhospitable areas. In these sites,

poor digital connectivity leads to operational

deficiencies on a wide range of issues, such as

dangerous working conditions, loss and pilferage of

products and low productivity. Situations like these

prevail in mining operations across most low-income

and developing countries in Latin America, Africa

and Asia. To say that mining is a critical sector is an

understatement. Large emerging countries still rely

on coal mining to fulfil their energy needs - over

49.7% of India’s electricity demands come from coal.

Moreover, mined materials are also needed to

construct roads and buildings, build automobiles, and

even make computers and satellites that power the

modern economy - mining powers our modern

civilization. However, mining is a dangerous business

- the global mining sector collectively accounts for an

estimated loss of 15,000 lives each year. More

recently, several active working groups and

committee set up by sovereign governments are

https://www.linkedin.com/in/ssreedharan/

†

https://www.linkedin.com/in/muthuuk/

‡

https://www.linkedin.com/in/soma-ghosh-kohli/

https://www.linkedin.com/in/surajprakash1/

accelerating the digitization of mines to make them

more productive, sustainable and safer to operate. As

a result, there is considerable pressure on the mining

industry from the government in most emerging

countries to accelerate the adoption of digitized

mining practices. These initiatives collectively fall

under the encompassing umbrella theme – advancing

‘Connected Mining’ (or Mining 4.0) adoption.

The 'Connected Mining' market is not entirely

new – it is estimated to be worth USD 12.7 billion in

2022, growing annually by 13.3%. But, our current

research focuses on digitizing the largely

underdeveloped underground coal mine operation in

emerging countries. In India, these mines number a

total of 273 individual sites. Typically, a large mining

company manages upwards of twenty such

geographically distributed remote mining sites. A

typical site may have 3-10 levels (or seams), typically

covering an underground area of 1-4 sq. km. One way

to conceptualize such a site is to consider that a

mining facility sits atop several floors of the mining

zone. Each floor is stacked one on top of the other like

slices of bread, and they can cover an area as large as

a small rural town.

218

Sreedharan, S., Ramachandran, M., Ghosh, S. and Prakash, S.

The Anatomy of an Infrastructure for Digital Underground Mining.

DOI: 10.5220/0011982700003482

In Proceedings of the 8th International Conference on Internet of Things, Big Data and Security (IoTBDS 2023), pages 218-225

ISBN: 978-989-758-643-9; ISSN: 2184-4976

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

We aim to provide intelligent automation and

connectivity to these remote underground mining

sites.

2 MOTIVATION

Over 81% (2.53 million sq.km.) of India's vast

mineral reserves remain untapped. Why? - Poor

network connectivity in mines limits automated

mining adoption in the industry. Moreover, the

conventional, labour-intensive mining practice

warrants that work in the sector is punishing and

dangerous. It also consequently leaves vast portions

of active mining sites unexplored and untapped.

The main reason is that no technology existed that

balanced - cost-effective deployment and

adaptability to the dynamic nature of mining

operations while operating through virtually

impenetrable bedrock. Consequently, MineNet, a

path-breaking subterranean mesh networking

technology, was developed by researchers at IIT-

ISM as a prerequisite connectivity fabric to enable

cost-effective alternatives to connected mining

solutions in the industry. The infrastructure detailed

in this paper is the first generation of a

comprehensive platform for autonomy in mining

that emerged out of the MineNet initiative, under

development at IIT-ISM.

Figure 1: Capability Matrix of Mining 4.0.

The initial goal of this project is to improve

mine safety and sustainability in remote Indian

mines using recent advances in industrial

automation and communication technology.

Longer-term, we intend to leverage the inherent

capabilities of emerging technologies like private

5G networks, AI/ML & robotics to advance mining

practices, thus ushering in a new phase of

accelerated adoption of digital technologies to

augment the conventional approaches. The concept

diagram in Figure 1 landscapes the capabilities the

platform is eventually expected to support. The

current release focuses on integrating advanced

industrial IoT – Mining 4.0 – to enable real-time

surveillance and tracking systems of equipment and

personnel operating on a remote mining facility.

While doing so, we assume we are also laying a

foundation for open innovation on unexplored areas

for advancement in autonomous mining - including

visualization, robotic & drone-assisted systems in

the future.

2.1 Concept Description

The presented system consists of two functional parts:

an automation platform for large-scale deployment of

autonomous systems and a wearable application

platform to provide miners with real-time information

about their environment. The project’s initial focus is

to address immediate challenges in mining safety and

operational efficiency. Future platform generations

are expected to evolve around learning models around

accrued operational data and sensor telemetry, thus

opening as yet unexplored avenues into AI-assisted

use cases.

2.1.1 Distributed Automation

The system’s core is a centralized orchestration

platform that serves as the network operations nerve

center for the entire site. Complementing the core

system is a rapidly deployable off-grid mesh

communications infrastructure designed to enable,

among other things - collaborative mapping, texting,

and emergency beaconing. The infrastructure is

expected to permeate every nook and corner of the

mine, providing ubiquitous connectivity for all

miners. A family of field apparatuses complements

the infrastructure to facilitate telemonitoring and

teleoperation.

2.1.2 Wearable Computing

Central to this infrastructure is a wearable router to be

carried by the field staff, which creates a mobile

'wireless bubble' hotspot around each individual.

These bubbles are interconnected through a network

of router devices forming a highly-resilient, robust,

wireless spinal cord that spans the entire area of off-

grid workspace. The infrastructure is a built around a

novel long-range, low-SWaP (size, weight & power)

short-burst wireless radio technology designed to

network large off-grid areas (such as a UG coal

mine), that are usually inaccessible through

conventional telecommunication technology.

The Anatomy of an Infrastructure for Digital Underground Mining

219

Beyond its original conception for interpersonal

communication, the proposed mesh serves both

people and smart devices (sensors & controllers) for

applications including: location tracking, telemetry

acquisition, search & rescue, surveillance & patrol

planning and alerts.

2.1.3 Thing Computing

The aforementioned use cases are only a few

exemplars conceived by our research team. By

opening up the platform by delivering software

development kits & hardware toolkits to the broader

research academic community we are also

concurrently developing several useful applications

of digital mining that harness sensor networks,

robotics, drones & predictive modelling to augment

conventional practices in existing mining sites.

3 RELATED WORK

The adoption of contemporary industrial automation

technology has reached an inflection point wherein,

driven by the accelerated adoption and promise of

AI/ML; the industry is transitioning to emergent

models for large-scale automation. As such, although

a large corpus of academic literature exists on

industrial automation platforms, only a few address

the challenges of large-scale distributed automation

from a general-purpose, full-stack platform

perspective (Chehida, 2022). Such generalization

unintentionally ignores the practical element of

usability in edge case conditions - like mine safety.

Others explore theoretical paradigms -

Ramachandran (2021) has proposed a software

engineering framework for IoT and CPS, which can

also be adapted for industrial IoT (IIoT) and Wireless

Sensor Networks (WSN).

3.1 Automation Platforms

Recognizing that as industrial IoT has matured, more

recent approaches to the problem take a domain-

centric approach at its core leading to highly

specialized solutions in healthcare (Said, 2021), smart

cities (Meiling, 2018) and farming (Fruhner, 2019).

Consequently, our design takes a human-centric

approach to mining applications - putting the miner at

the heart of the problem to explore novel ways to

ensure his safety and alleviate the drudgery of his

occupation while he operates in extraordinarily

hostile and inhospitable environments. In doing so,

we have designed a full-stack, turnkey platform

purpose-built for the mining sector.

3.2 Communication Technologies

Underground coal mining operations fall under one of

4 categories: room-and-pillar, longwall, short-wall,

and thick-seam. They differ in operational

characteristics, but in all cases, personnel and

machinery operate in tight and constrained

compartments deep under layers of bedrock accessed

through portals: drifts, slopes and shafts. Typically,

vertical shafts may interconnect compartments with

specific functional roles.

The proposed communication system that would

operate in this environment would be a wireless, ad-

hoc network for both interpersonal communication &

telemetry in the future (eg: ground control,

ventilation, haulage, drainage, power supply,

lighting, and communication. These goals can only be

achieved by using devices that are lightweight,

durable, long-lasting and inexpensive. Moreover, the

network must be decentralized for resilience and easy

deployment and to support the mobility of mining

personnel. Lastly, the network needs to cover as large

an area of the underground mining network as well.

Ramanathan (2005) explores the challenges of

implementing such a network.

The ideal technology for these requirements is a

radio device with a long range and the ability to multi-

hop or mesh a network of such devices into a single

self-organizing ad-hoc network. It must also be

lightweight enough to be wearable. Existting

solutions built around off-the-shelf technology (Wi-

Fi, LTE, 5G, ZigBee, Bluetooth, LoRaWAN) fall

short on one or more the following requirement

criteria: low cost, decentralized, wearable, long-

range, mesh network, low-cost, light-weight and

requiring no public commercial infrastructure.

Our wireless router technology utilizes a

proprietary WIFI-over-radio (WFoR), multi-hop,

mesh-networking technology built over long-range

radio (1-4 km), making it possible for anyone to

create a reliable, off-grid, peer-to-peer, ad-hoc,

communication network at will (Ramanathan, 2018).

This proprietary radio technology makes it suitable

for short burst radio communication for inter-

personal texting, emergency beacons and transfer of

critical data.

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

220

4 SYSTEM ARCHITECTURE

4.1 Design Principles

Industrial IoT has entered its next phase – large-scale

industrial automation. Over 7 million micro-devices

are projected to come online by 2030 in the coal

sector in India alone as the mining sector begins to

digitize proactively. This transformational number is

only the tip of the iceberg when it includes other

mining sectors (mineral mines or open cast mines)

and the prospect of global sectoral transformations

across Latin America, Africa or other Asian

countries. Moreover, such micro-devices for

industrial applications will include the entire gamut

of emerging IoT applications today: networking

equipment (like routers and hubs), sensors, smart

controllers and wearable devices, and as such, the

ideal solution for the industry must also foment

solution convergence, and therefore it must have the

following characteristics:

• Modelled for adoption of turnkey industrial

automation technology that can be deployed in

stages to enable effective cost control on capital

investment.

• Solutions that augment and co-exist with

existing operational and regulatory practices, to

minimize transformational downtime.

• Reliable and fault-tolerant systems that are

resilient to failures that occur often in such

dangerous environment.

• Turnkey platforms that operate entirely off-grid

while still adopting the advantages of

conventional technology paradigms in cloud

computing.

• Open platforms with established industry

standards that can evolve and co-opt, over time

thus allowing easy integration through multi-

vendor solution offerings.

• Human-centric solutions that factor in safety

and ease of adoption at the heart of its design so

as to accelerate the adoption of technologies

substantially.

In order to deliver a scalable solution that

addresses all of the aforementioned constraints, we

have adopted the following requirements and

Figure 2: Conceptual Design of Platform.

The Anatomy of an Infrastructure for Digital Underground Mining

221

principles as design guidelines.

• A simple wearable device work by each miner

and serves as the functional user interface for the

miner regardless of where he is. The device

must have a built-in application platform that

allows its capabilities to be extended

boundlessly and a deployment platform to

provision and administer it from a central

location.

• An on-premises, private-cloud platform that

provides all the conventional services that we

assume standard on mainstream cloud

platforms, including: storage services, compute

services, databases and messaging systems.

• A low-cost, unified, hybrid (wired+wireless)

internetworking fabric providing ubiquitous

connectivity over the entire mining area. It is

both resilient to failures (e.g., outages,

accidents, calamities) while also able to deliver

a range of service capabilities to support: large

data transfers for sensor telemetry, real-time

messaging for interpersonal communication and

system-wide machine-to-machine control

messaging.

• A low-code IoT platform that operates on

commercial off-the-shelf (COTS) hardware to

accelerate the prototyping and development of

data acquisition (sensors) & teleoperation

(controllers) solutions.

• A plug-n-play network architecture based on

REST-based web services, wherein open-source

and third-party vendor solutions can be

integrated quickly.

The presented platform has evolved through

several iterations of the design. Consequently, our

current approach to achieving the previously

mentioned design goals has culminated in developing

a distributed automation solution that integrates field

apparatus, devices, communications and software

applications. The system is also designed to support

the orchestration of all autonomous mining activities

from a centralized network operations center.

4.2 System Overview

Delivering on the requirements above resulted in the

development of two disparate yet complementary,

interoperating platforms:

• Ceres Cortex Cloud Platform: A private-

cloud infrastructure that serves as the central

orchestration & administration platform for all

system services and IoT devices. It includes

cloud platform services, web applications

serving the console, grid-wide centralized

messaging and storage and database services.

• Ceres IoT Platform: An open-source operating

system for automating large-scale, distributed,

mesh-based IIoT infrastructure. Running on any

conventional embedded platform as a host

operating system, it provides a low-code,

managed environment for rapid edge device

development. It also serves as a unified interface

into a seamless distributed internetworking

fabric to allow distributed automation & edge

computation capabilities.

Conceptually, the system operates as the brain

integrated with appendages through a distributed

nervous system as detailed in Figure 2.

4.2.1 Cloud Platform

Cortex is a complete, on-premise, private IoT Cloud

infrastructure that allows you to customize and

deploy our entire suite of solutions on a secure,

private network, fully uncoupled from the Internet.

Cortex is a turnkey solution providing industrial

orchestration capabilities for a disconnected remote

industrial site located in places with unreliable or no

internet connectivity. It can also be deployed in

locations where cloud connectivity is impossible or

undesirable (e.g., for security reasons). In either of

these environments, Cortex ensures that all

equipment deployed within an industrial site inter-

operate seamlessly without needing a public cloud

infrastructure (like AWS, Azure, or GCP) or a local

area network (LAN). Cortex achieves this by

cooperating with the suite of networking solutions

that create a secure, wireless radio perimeter around

the site for all Ceres-compatible devices. The

capabilities of Cortex allow it to:

• Provide an off-grid cloud infrastructure for

remote and offline sites, where computational

tasks and data needs to be located on-premise

either for security or performance reasons.

• Provide advanced reporting and control

capabilities of all autonomous operations

running on field apparatus across the entire site.

• Act as a fault tolerant, centralized exchange for

all inter-process communication occurring

between devices deployed at disparate zones in

the subterranean mine working areas.

• Serve as a central repository and administrative

console for device profiles that define the

behaviour of individual IoT devices deployed at

the site.

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

222

• Provide API services through REST-based

programming interfaces for integration with

third-party software and tools.

(Sreedharan, 2019)

Figure 3: Ceres IoT Architecture.

4.2.2 IoT Platform

Ceres IoT is an operating system for off-grid &

occasionally-connected industrial equipment. Ceres

is purpose-built to address the delivery of AI-enabled

industrial automation at a massive scale. It allows the

central administration, provisioning & analysis of

equipment in mines. At scale, equipment powered by

Ceres can be connected over a geographically

distributed network using wired, wireless or mobile

networks.

Ceres IoT is designed to operate in highly

constrained and hostile physical environments. It is,

therefore, ideally suited for deployment in a farm

where computers are expected to work with minimal

power, often on batteries, and are exposed to harsh

environmental conditions like dust, rain and heat. The

capabilities of Ceres allow it to:

 Augment existing industrial infrastructure with

intelligence deployed either on the cloud or on-

site

 Facilitate rapid reconfiguration and upgrade of

autonomous equipment on a large scale using

existing mobile or wireless networks.

 Dramatically reduce the cost of industrial

automation deployment & maintenance by

leveraging wireless interconnects, 5G networks

& enhanced capabilities of modern low-cost

embedded devices

 Allow a heterogeneous set of sensors &

switches from disparate vendors to interoperate

over a shared network

 Facilitate the acquisition of large amounts of

real-time sensor measurements for further

analysis and monitoring.

 Allow big data, machine learning, and AI to

power the industrial transformation of mining

using data sciences.

 Synchronize execution of distributed tasks

based on a universal clock, scheduled by a fleet

of connected devices, thus enabling large-scale

orchestration.

(Sreedharan, 2022)

An important lesson is that standardizing the

entire network architecture around a common

portable, managed operating-system environment

allows us to simultaneously simplify the deployment

use cases and accelerate solution development of field

apparatus (like sensors and controllers). It also

enables a versatile communication fabric that spans

the entire industrial site, with assured interoperability

across devices. This design approach allowed us to

offload a substantial amount of the computing

capability to the edge, reducing the throughput

demands on the network while also making the

platforms highly resilient to zonal outages and

accidents. Figure 4 outlines the Ceres architecture.

Although not a comprehensive list of current

solutions – and it will undoubtedly continue to

evolve, the following classes of devices are

implemented entirely using Ceres IoT in the current

generation of the platform.

• Wireless Sensor Networks: A growing family

of sensor devices is either delivered or under

development on-premises. The sensors include

air quality & toxic gas sensors, micro-

seismometers, environmental sensors, activity

monitors and incendiary sensors.

• Networking Components: A family of low-

cost, industrial-grade gateways & hubs provide

the main communication backbone that

integrates a wide range of technologies like

smart switches or environmental sensors using

WIFI-over-radio, over long distances. The

current generation of solution includes a suite of

routers built using LoRa-based, wireless

LPWAN (low-power, wide-area network)

technologies. These routers make deploying and

integrating Wi-Fi-capable Ceres-compatible

devices easier over long distances.

• Wearable Pagers: A smart wearable, called

‘pagers’, operating on a personal area network

(PAN) that interfaces with an indigenously

developed LPWAN-based wireless mesh

networking technology that bridges over a wide

range of conventional backbone network (5G,

SPE or RS485) to provide ubiquitous tracking

and communication capability for miners.

The Anatomy of an Infrastructure for Digital Underground Mining

223

4.3 Communication

Implementing wireless technology in underground

mines is challenge (Ranjan, 2013). The fundamental

limitation is that the transmissibility of radio is

limited in any form or rock. The amount of energy

powering the transceiver. To circumvent these

limitations, we developed a novel variation of a

mobile ad-hoc network (MANET) (Ramanathan,

2019) that mimics the behavior of swarm insects in

colonies – in particular, ant colonies

The internetworking platform is built atop

MineNet, a path-breaking mesh networking for

ubiquitous underground mine connectivity. The

networking technology behind MineNet functions by

segregating the entire network into three logical

subnets. A Layer-2, LPWAN radio & serial transport

ensure interoperability across the subnets. Traffic is

moderated using a variation of the OSPF routing

protocol (Baccelli, 2010), with proprietary extensions

optimized for scalable, decentralized, short-burst

communications over long-range radio possible.

 Central Network: Located on the surface,

usually within an office environment, it serves as

the brain of the network. Components of the

entire mesh are provisioned & administered

through applications running on a central server

setup with a Private IIoT Cloud Infrastructure

software. One or more Gateways provide a

Layer-2 (MAC) abstraction whereby every

single device on the network may be addressed

and accessed over LPWAN radio and Modbus.

The gateways expose those devices over

HTTP/REST services via Wi-Fi, thus providing

a universal interface to program the radio mesh

network over LAN.



Zonal Network: A fleet of fixed and portable

routers strategically placed within range

proximity are interspersed across the entire

underground mine environment to cover shafts,

compartments and other areas of activity.

Collectively these zonal routers form the spinal

cord of the network. They function as moderators

for inter-zonal traffic while also brokering

administrative & control signals from the central

network to devices connected to the mesh.

 Personal Area Network: A wearable router,

usually attached to a belt buckle, creates a

'wireless bubble' hotspot around the miner. This

device is paired with a Wi-Fi or BLE-compatible

hand-held device (iPhone/Android phone or,

optionally, a custom-built touch screen

computer) or sensor. It provides Layer-3 protocol

abstractions for a wide range of applications

(e.g.: communication, telemetry, emergency

beaconing), on the hand-held device.

4.3.1 Network Topology

The physical network architecture closely shadows

the logical network design detailed above. As such,

the physical network is also broken down into three

separate interoperating zones. The transport layer of

the networking stack ensures that packets generated

in one zone are efficiently routed to appropriate

destinations. The following summary outlines the

operating characteristics of each site.

• Spinal Network: Computing devices in the

Central Network are connected to routers in the

Zonal Network through several standard wired

protocols. The solution deploys either an ethernet

network or serial communication cables to allow

reliable connectivity from the surface entrance to

the active areas of the mines. The network winds

through the mining tunnels to locations within

meters of any mining activity. This wired

network forms a spine that can extend up to 1.5

km from the entrance to areas deep in the seam.

• Regional Network: Mobile wireless routers

called extenders are placed at strategic locations

to extend the spinal network's reach by up to 200

meters at each hop from the spinal periphery. The

wireless network allows mining activity to

evolve organically into new working areas.

• Team Network: Wearable routers interact with

extenders wirelessly to enable short-spurt

communication to the surface. These pagers

interoperate (peer-to-peer), forming a swarm

cloud of networking devices that ensures reliable

communication with the regional network.

4.3.2 Network Stack

The networking stack follows a layered approach

drawing inspiration from the implementation of

TCP/IP protocol suite and SLIP. In Figure 5, we detail

the layers of the networking stack. Layering ensures

that the network can provide a wide range of services

including time synchronization, remote procedure

calls, file transfer and data acquisition. Considering

the dynamic nature of the mining operations and

supporting unforeseen use cases in the future, the

network is designed to be programmable through an

API drawing on key design principles of a software

defined network (SDN).

IoTBDS 2023 - 8th International Conference on Internet of Things, Big Data and Security

224

Figure 4: Layers of the Networking Stack.

4.4 Applications

Although the platform continues to evolve and

expand, the current suite of solutions includes the

following components:

 Network Operation Centre: A centralized web

application nerve centre of the operation.

 Safety & Surveillance: Wearable solution to

track miner location and movement.

 Teleoperation: A family of smart controllers to

schedule and control electrical equipment and

flow valves.

 Telemonitoring: A family of sensors tracking

environmental and equipment parameters.

5 CONCLUSION

Adoption of advanced industrial IoT in mining has

lagged in emerging countries due to lack of practical

solution alternatives. In this paper, we demonstrate

how an integrated platform approach to large-scale

automation, purpose-built for the mining sector, can

accelerate technology adoption. More specifically, in

the mining industry it promises to augment the

conventional capabilities of tracking, surveillance

and activity monitoring to augment them with

emerging capabilities in smart wearables, industrial

automation, drones & robotics, as a new era of

autonomy in industries driven by artificial

intelligence emerges in the horizon.

REFERENCES

Baccelli, Emmanuel & Cordero, Juan & Jacquet, Philippe.

(2010). OSPF over Multi-Hop Ad Hoc Wireless

Communications. International Journal of Computer

Networks & Communications. 2. 10.5121/ijcnc.20

10.2503.

Bbesa, & Mulenga, Sunday & Mazimba, C. (2018). Mines

Safety and Accident Communication System for

underground mines.

Chehida, Salim & Bensalem, Saddek & Conzon, Davide &

Ferrera, Enrico & Tao, Xu. (2022). BRAIN-IoT

Architecture and Platform for Building IoT Systems. 67-

77. 10.5220/0011086000003194.

Dusia, Ayush & Ramanathan, Ram & Ramanathan, Warren

& Servaes, Christophe & Sethi, Adarshpal. (2019).

VINE: Zero-Control-Packet Routing for Ultra-Low-

Capacity Mobile Ad Hoc Networks. 521-526.

10.1109/MILCOM47813.2019.9020768.

Fruhner, Maik & Iggena, Thorben & Kraatz, Franz &

Nordemann, Frank & Tapken, Heiko & Tönjes, Ralf.

(2019). OPeRAte: An IoT Approach towards

Collaborative, Manufacturer-independent Farming 4.0.

165-176. 10.5220/0007760101650176.

Meiling, Sebastian & Purnomo, Dorothea & Shiraishi, Julia-

Ann & Fischer, Michael & Schmidt, Thomas. (2018).

MONICA in Hamburg: Towards Large-Scale IoT

Deployments in a Smart City.

Ramachandran, M (2021) SEF4CPSIoT: Software

Engineering Framework for Cyber-Physical and IoT

Systems, International Journal of Hyperconnectivity and

the Internet of Things (IJHIoT), 5(1), DOI: 10.4018/

IJHIoT.2021010101

Ramanathan, Ram & Servaes, Christophe & Ramanathan,

Warren & Dusia, Ayush & Sethi, Adarshpal. (2019).

Long-Range Short-Burst Mobile Mesh Networking:

Architecture and Evaluation. 1-2. 10.1109/SAHCN.20

19.8824803.

Ramanathan, Ram & Servaes, Christophe & Ramanathan,

Warren. (2018). ECHO: Efficient Zero-Control Network-

Wide Broadcast for Mobile Multi-Hop Wireless

Networks. 1-6. 10.1109/MILCOM.2018.8599 737.

Ramanathan, Ram. (2005). Challenges: a radically new

architecture for next generation mobile ad hoc networks.

132-139. 10.1145/1080829.1080843.

Ranjan, Alok & Sahu, Himanshu. (2013). Communications

Challenges in Underground Mines. Search & Research.

V. 23-29.

Said, Omar & Tolba, Amr. (2021). Design and Evaluation of

Large-Scale IoT-Enabled Healthcare Architecture.

Applied Sciences. 11. 3623. 10.3390/app11083623.

Sreedharan, S. (2022). Ceres OS. ExoCortex Inc. https://

www.exocortex.systems/product/technology

Sreedharan, S. (2019). Bloom Cerebrum. ExoCortex Inc.

https://www.exocortex.systems/product/bloom-

suite/bloom-cerebrum

Disclaimer: This paper is not a product of JPMorgan Chase & Co. (JPMC) or its affiliates. Neither JPMC nor any of its

affiliates makes any explicit or implied representation or warranty regarding the contents of this paper, nor accepts any

liability in connection with this paper, including, without limitation, with respect to the completeness, accuracy, or reliability

of the information contained herein, whether the contents of the paper violates or misappropriates the intellectual property of

another, and the potential legal, compliance, tax, or accounting effects thereof.

The Anatomy of an Infrastructure for Digital Underground Mining

225