978-1-7281-8068-7/20/$31.00 ©2021 IEEE
Blockchain based smart marketplace for secure
internet bandwidth trading
Bello Musa Yakubu
Department of Computer Science
COMSATS University Islamabad
Ahamed Sani Kazaure
Department of Information
Technology, Federal University Dutse,
Muhammad Mahmoud Ahmad
Department of Computer science, Kano
University of Science and Technology
Majid Iqbal Khan
Department of Computer Science
COMSATS University Islamabad
Abdullahi Binta Sulaiman
Department of Technical and
Islamic University of Technology
Department of Computer Science
COMSATS University Islamabad
Abstract—Internet broadband usage is increasing
dramatically as the number of Internet of Things (IoT) devices
in smart homes that rely on Internet connectivity for effective
communication between them grows. As a result, many network
and Internet communication issues arise, including network
traffic congestion, insecurity, high connection rates, and excess
bandwidth waste. This work aims to create a secure and
decentralized smart marketplace (SMP) for a fair, transparent,
robust, and less expensive way of trading excess or idle traffic
from a smart home by directly paying the owner of the
bandwidth. The internet bandwidths may initially come from a
direct Tel-com product or from other users (smart homeowners)
who decide to sell their excess Internet bandwidth. The research
was carried out by integrating a given smart marketplace
scenario with blockchain technology using proof-of-authority
(PoA) consensus mechanism. The simulation results show that
the model is more reliable and appears to be an efficient solution
to dealing with excess bandwidth waste and offering safe excess
bandwidth trading within a smart city. Besides, the excess
bandwidth wastage is controlled at least by 75% by making it
available in the SMP for buyers.
Keywords— Excess Bandwidth, Channel utilization,
Blockchain, Smart contract, Gas, Smart marketplace
There is a chance that the number of internet users will rise
from 3.4 billion to 5 billion by 2025  and . This is due to
an increase in IoT impetus, which results in a rapid increase in
the number of bandwidth-hogging devices connecting to
networks . It is well known that IoT devices are mostly
resource constrained; as a result, many improvements are
made to them so that they can have good computing
capabilities and share large amounts of data. These
advancements are heavily reliant on consistent internet
connectivity. Based on this fact, some researchers propose that
an entirely new network model be established to
accommodate the current development, particularly the
unstoppable increase in bandwidth demand .
With this increasing growth, many problems arise and
therefore need to be addressed, such problems include:
• Excess bandwidth wastage: According to studies 
and , more than 90% of prepared internet bandwidth
remains unused, resulting in a reasonable amount of data
bandwidth being wasted. As previously stated, one of the
reasons for the increased demand for internet bandwidth is the
growing number of smart homes IoT devices that rely heavily
on internet connections for routine functionality. Similarly,
the functionalities of such devices may not always be
concurrent in a smart building. Most internet bandwidths are
purchased from telecommunications companies on a
subscription basis (daily, weekly, monthly, quarterly, or
yearly) and are forfeited if not consumed before the end of the
period. Although some companies provide a rollover service
that allows users to purchase a new subscription to roll over
unused bandwidth for a new period. This technique, however,
does not address the issue of bandwidth waste because the user
may not be able to consume all the bandwidth before the end
of the new period. As a result, there is a need to devise a
method of utilizing the unused bandwidth.
• Insecure bandwidth trading techniques and
environment: Many approaches – have attempted to
provide bandwidth trading mechanisms to utilize excess
bandwidth, but the majority of them are associated with
numerous security challenges and thus tend to be unreliable.
This is due to the fact that the trading environment or platform
they provide lacks sufficient security protocols to ensure the
safety of the participants, as many malicious nodes may
pretend to be part of the trading, which may result in the
breaching of legitimate participants' information or even
• Low network performance: This is another troubling
factor that most previous studies were unable to address,
particularly in spatially distant household environments 
and . Buildings in urban areas experienced significantly
poor network performance due to high demand for internet
bandwidth. It should also be noted that most buildings in rural
areas suffer from a severe lack of internet connectivity due to
a lack of good proximity to Tel-com base stations. This
necessitates the requirement for improved network
• High connection rates: Though some recent studies
tried addressing this problem using auction-based approaches,
yet the problem is not fully addressed , .
Several mitigation techniques, such as –, have been
proposed by various researchers, with the main goal of
providing fair and effective bandwidth trading. Some of them
also attempted to address high connectivity rate issues with
auction-based approaches. These approaches, however,
appear to be based on centralized architecture. Furthermore,
none of them have addressed the associated security
challenges such as non-repudiation, authentication, and
integrity, leaving the trading environment insecure and
vulnerable to a wide range of cyberattacks  and . Other
issues with these approaches include scalability, high
connection costs, flexibility, and user accessibility.
In this paper, we proposed a decentralized model using
blockchain technology to enable secure, fair, fast, and
affordable simple trade-based excess data bandwidth sharing
between smart buildings.
A Blockchain can be described as a distributed database or
ledger, which is free from any kind of mutilation. It’s made up
of blocks containing series of transactions. Each block
contains the hash of the block, some stored data, and the hash
of the previous block (except for the genesis block). The hash
of a block is unique and can be termed as signature or
fingerprint of that block. Therefore, changing anything in the
block will lead to change of the entire hash of the block. The
hash of the previous block is used to join the blocks together.
These characteristics makes it to become very secured as all
information entered cannot be easily changed .
The nature of the information in a block depend on the type
of the blockchain. In this work, we will consider working with
Ethereum blockchain, as it is one of the types of blockchain
that can be used as a private as well as public, unlike the
Bitcoin blockchain which is public.
B. Ethereum Blockchain
Ethereum blockchain was created by Vitalik Buterin in
2005 . It was a new evolution in blockchain technology in
a new platform different from Bitcoin. Though it supports
financial transactions just like Bitcoin, but it can also use in
non-financial transactions, e.g. change of user’s details,
balance inquiry, mini statement printing etc. Ether is the name
of the bit currency in Ethereum. Another difference between
Ethereum and Bitcoin is that, Ethereum supports smart
contracts (selves triggering program) running under a
specified consensus algorithm, while Bitcoin do not support
smart contracts. Similarly, Bitcoin is a public network while
Ethereum can be both public and private .
C. Consensus Algorithms
Computers nowadays are very fast and can calculate
hundreds of thousands of hashes per second. Therefore, using
hashes only is not enough to prevent tempering as you can
effectively tamper with a block and recalculate all the hashes
of other blocks to make your blockchain valid again. To
mitigate this, the blocks in a blockchain have to reach into
some understanding with each other before any changes can
be affected in the given block. This method is called consensus
algorithm. We have many different consensus algorithms, but
for the sake of this work, we will consider two of them, these
includes: Proof-of-Work (PoW) and Proof-of Authority (PoA)
I. Proof-of-Authority: in this consensus algorithm, a
participant node is selected based on its identity which must
be available to all other members of the network. Authority
(power) is given to the chosen one for verifying and allowing
any transaction to be added in the block. A compensation or
incentives is paid to this validator whenever a transaction
II. Proof-of-Work: in this consensus algorithm, a
transaction within each block are verified by miner nodes who
compete by solving mathematical puzzles known as Proof-of-
Work problem. Any miner that first solve the block problem
is given an incentive. Then all verified transactions will be
stored in the blockchain.
D. Smart Contracts
Blockchain is constantly evolving day by day. One of the
most recent developments is the creation of smart contracts.
These contracts are simple programs that are stored on the
blockchain and can be used to trigger an exchange of values
e.g. coins based on certain conditions. Smart contract are
programs that execute themselves when some specific
conditions are made. For example, an auction might
automatically transfer deeds of ownership to the highest
bidder after a certain time has elapsed, or father’s contract
might automatically send his son a set amount of money every
year on his birthday . Furthermore, each time smart
contract is executed, certain amount of gas is consumed to
enable transaction block mining.
E. Ethereum Gas
This is part of Ethereum blockchain token that is used by
the contract to pay miners for validating transactions. All
Ethereum blockchain needs gas to compute and execute their
smart contract and transactions. Enough gas must be paid to
the miners in order to ensure the smooth running of the
transactions, otherwise, the transactions will be terminated or
will not even run .
The proposed approach in addition employs a simple
exchange trading and bidding method in which each owner of
a smart building with excess data bandwidth displays the
amount of bandwidth he wishes to sell to other consumers in
need in the smart marketplace. This will help to reduce data
bandwidth waste while also allowing others to easily access
the internet without interruption in a secure and trusted
The major contributions of the paper are:
• A simple data bandwidth trading scheme was
proposed, specifically for areas of varying internet
• A decentralized smart marketplace that provide fair
and transparent transactions between all participants
using blockchain technology.
• A secure and trusted environment that provide high
network performance and cheaper connectivity rates.
The rest of the paper is organized as follows: Section II
explains the previous work carried out by other researchers in
bandwidth trading, data sharing and offloading. In section III,
we present our proposed model’s framework for secure
blockchain based internet bandwidth trading. In section IV we
explained our simulation setup and in section V we discussed
on the simulation results. The conclusion was discussed in
section VI, followed by references in section VII.
II. RELATED WORKS
Over the years, bandwidth marketing has been a hotly
debated topic that has piqued the interest of many scholars.
This is due to several issues associated with the technique,
including offloading, connectivity speed, internet
accessibility, financial gain, and others. These difficulties are
all related to the type of IoT devices we use. Previous research
indicates that most user devices, such as computing user
equipment (CUEs), are associated with computationally
intensive tasks, and as a result, they are unable to offload their
entire tasks to the cloud due to limited computing resources.
A bandwidth issuing approach where proposed by  to
enable a helper user equipment (HUEs) with idle processors
to assist these user devices in offloading their tasks.
In some cases, bandwidth trading is associated with an
explicit transaction between buyer and seller, which results in
an exchange or sharing of content. Authors in  proposed an
important method for allocating single source bandwidth in a
two-level hierarchical auction-based market. Their primary
goals were to reduce the seller-buyer direct transaction
difficulties and high computational and physical overhead,
particularly when dealing with a large and dispersed set of
customers. Other research studies ,  concentrate on
establishing bandwidth trading by offering incentives to
contribute P2P content sharing. The goal of their model was
to overcome high variability in download rates as well as
unfairness in the ratio of downloaded and uploaded data.
Randeep et al  proposed a model for revenue maximization
based on the linear programming relaxation approach. It is
also used to keep the cost of scheduling all jobs to a minimum.
Another model  was proposed to achieve most of the above-
mentioned goals. The primary goal of this approach is to
increase traffic, provide a lower connection rate, and solve all
bandwidth allocation issues. The greedy algorithm was used,
which solves offloading problems in polynomial time.
The studies have made significant contributions to the
research area of discussion, as many efforts have been made
to provide fair and effective bandwidth trading with an
emphasis on data offloading, data sharing, low connection
cost, and accessibility. In contrast, none of them have
addressed the current security challenges such as non-
repudiation, authentication, integrity, and so on. Similarly,
non-have provides a safe, equitable, low-cost, and robust
bandwidth transmission scheme. Second, most schemes are
not scalable and use centralized approaches, resulting in a
single point of failure.
In our paper, we proposed an approach using blockchain
features to provide a scalable, secure, robust, transparent,
cheap, reliable, and trusted excess internet bandwidth trading
environment in each smart city.
ROPOSED BLOCKCHAIN BASED SMART MARKETPLACE
FOR SECURE INTERNET BANDWIDTH TRADING
The proposed model comprises of a smart homes
connected to SMP via individual smart homes gateway device
(SHG) and the smart market gateway (SMG) respectively
using Ethereum blockchain. Furthermore, all gateways use a
unique Ethereum addresses (EA) to communicate in the
network. A participating smart home can be either bandwidth
seller (B.Seller) or bandwidth buyer (B.Buyer). The model
also uses Ethereum smart contract to carry out the bandwidth
trading in the smart marketplace (SMP) between any B.Seller
and the corresponding B.Buyer via the SMG. The B.Seller
might have initially purchased its bandwidth from other sellers
or from a Tel-com service provider (Telcom SP) as depicted
in Figure 1.
A. Case study scenario
A typical scenario consisting of a societal residential area
with multiple buildings was considered in this model. In such
cases, a local internet base station is installed so that residents
can connect to an internet bandwidth supply, either wirelessly
or wiredly. Each building has a home gateway that serves as a
gateway router or an access point to the local internet base
station bandwidth supply.
We assumed that each gateway can be configured to
monitor and set the amount of data package that is intended to
be used in the building based on the data consumption of
individual devices. Potential B.Sellers use their gateway to
connect to the SMP via the SMG and to set the amount of
excess bandwidth intended to sell in the SMP. With this, other
potential B.Buyers can use their gateways to access the SMP
and view the available stocks in the market as well as their
respective prices via the SMG. Buyers will be able to choose
the most affordable stock and make their payment in Ether
electronically via the smart contract.
Figure 1: Proposed blockchain based Smart marketplace for
secure internet bandwidth trading
In this model, both the buyer and the seller must be
subscribers of the same Telcom SP. Furthermore, the potential
B.Seller must notify his respective Telcom SP in order to
generate a key-value (K-value) for the amount of bandwidth
proposed to sell in the SMP prior to setting the stocks for sale.
The Telcom SP generates the K-value only if the seller has the
specified amount of bandwidth. When the K-value is
generated, it locks the specified amount of bandwidth stock.
Thus, unless the B.Seller withdraws from the SMP, he will be
unable to use the locked bandwidth again. Furthermore, the
B.Buyer uses the K-value to access the newly purchased
bandwidth amount from the respective Telcom SP following
successful trading completion.
Meanwhile, it was assumed that the SMG has a list of all
trusted Telcom SP to which both the seller and buyer must
subscribe. Furthermore, SMG verifies the potential seller's
claim of bandwidth ownership from the respective Telcom SP
before allowing it to participate in the SMP. Similarly, the
SMG notifies the respective Telcom SP whenever a payment
is made in order to remove the seller's complete ownership of
the said amount of bandwidth .
B. Registration and authentication
In the registration process, the SMG first registers each
home and maps it to the SMP. Any time a participant wants to
trade in the SMP, the smart contract authenticates it. Each
participant is verified by the system by ensuring that their EAs
are registered and legitimate. If the participant is legitimate,
the system will issue a public recognition event and privately
transmit an interaction token to the participant as Token =
(UID; Timestamp; SHG; SMG). The event is broadcast to all
The token contains the following data: Unique
Identification UID, SHG EA, SMG EA, and the Block
timestamp; the UID is created by hashing the SHG EA and
SMG EA using the keccak256 hashing algorithm feature built
into the Solidity language. Upon collecting the system's
recognition token, the requesting smart home can then use the
token information to authenticate itself before participating in
Following successful authentication, a standard secure
SSL interaction link is formed between the requesting smart
home and the other participating smart homes in the SMP for
the sharing of services, resources, and information.
C. Overall trading model
The proposed model is depicted in detail in Figure 1 and it
is described in Algorithm-1. It can be seen that it is made up
of three main stakeholders who have internet access to
Ethereum smart contracts, namely: B.Seller, B.Buyer, and
SMG. Both B.Seller and B.Buyer are individual buildings
within the smart city that participate in excess bandwidth
trading in this model. To participate in the trading, each
building has a separate Ethereum account. As a result, they all
have a direct connection to the blockchain, where individual
transactions are recorded and easily retrievable. The B.Seller
can acquire its own Internet bandwidth directly from the Tel-
com service providers or from other house owners (B.Sellers)
whom are willing to sell some of their bandwidth.
In this scenario, all legitimate B.Sellers will log in to the
SMP using their tokens in order to submit their excess
that they wish to sell out with their
. Therefore, they issue-out their bids
in form of (
) each time they join the SMP. Then any
legitimate B.Buyer that want to acquire a bandwidth will now
have to login to the SMP using their tokens in order to see all
available stocks. After making price comparatives, a B.Buyer
will select the stock that best suits his financial capacity and
then submit the corresponding price in ether through the smart
contract. After the confirmation of the transaction, the smart
contract will then notify the B.Seller about the commitment
made by the B.Buyer and request it to release the
corresponding bandwidth K-value before the fund will be
finally released to his Ethereum account. If the B.Seller fails
to release the required K-value after a given time frame δ, then
the fund will be returned to the respective B.Buyer and the
trading process canceled.
In view of this, market competition will be triggered as all
B.Sellers will have to provide best prices to their available
stocks because every B.Buyers will only buy a stock based on
their financial capacities. Secondly, security is provided as all
participant must be authenticated before they could join the
SMP, and once a participant is registered, its account address
will be recorded in the blockchain as a legitimate participant
Coming to network performance, transmission rate and
channel utilization are the key important parameter that help
in responding to any data traffic request. This technique
solemnly depends on the distance between the B.Seller and the
corresponding B.Buyer. Suppose the amount of traffic needed
by B.Buyer is given by
, then vector of the channel
utilization can be computed as:
where each pair () implies to all possible allocation of
B.Buyer to B.Seller , and is given as the number of
B.Sellers in the network. Therefore, channel utilization can be
is given as the channel utilization of B.Seller if
it is utilized in data traffic offloading of B.Buyer . Similarly,
its computed as the ratio of traffic request
and the highest
obtainable transmission rate of the wireless link capable of
Suppose represent the set of all B.Buyers willing to buy
some bandwidth in SMP and represent set of all the
corresponding B.Sellers in the SMP. Then
can be define as set of all B.Buyers within the radio range of a
given B.Seller. Similarly, the B.Seller whose stock was chosen
by a given B.Buyer is given as
. Then, !
is used as the variable that assign the B.Buyer to the
corresponding B.Seller whose stock was chosen in above.
Therefore, with these definitions above, we can drive the
• For any B.Buyer to be assigned to a given B.Seller
, this condition must hold:
This is to ensure that each B.Buyer is assigned to at
least one B.Seller which is in the same radio range.
This is to say that, B.Buyer must be assigned to the
particular B.Seller whose stock price is affordable
and chosen by the B.Buyer.
• Thus, we also introduce another constrain:
The above equation (5) and (6) must be satisfied
before any traffic request will be considered. This is
to ensure that the highest obtainable transmission
rate of the wireless link and available excess
bandwidth of a given B.Seller is not exceeded.
• The B.Buyer will submit his payment to the
particular chosen B.Seller via the smart contract
after all conditions are meet in the following form:
• Lastly, after the confirmation of the buyer’s
transaction, the seller submits the requested stock
(K-value) via the smart contract in the following
) -./) #0
In this section, we discuss the simulation setup of our
proposed model. In this work separate account are used to
create each B.Seller and B.Buyer in the Ethereum blockchain.
We create more than 100 different participant accounts
representing various community buildings. We group these
buildings in to clusters of 5, 10, 15, 20, 25 and 30 buildings to
represents different estates. We assume the distance between
one building to another in a given cluster to be at least 200
meters and from one cluster to another to be at least 500 meters
and at least one macro ISP base station (B.S) is installed
around a central site of these clusters as suggested in  and
. We assumed that any of the buildings can be a B.Buyer
or attend the position of B.Seller when it has excess bandwidth
The simulation was carried out on a python and solidity
using a meta mask Ethereum wallet. We run the simulation to
allow the pairs of participants to interact with each other based
on their clusters, and average transaction readings were taken
each time we had a complete selling and buying transactions.
To apply PoA, we use the rinkeby test net to fund the
transactions which is based on proof-of-authority consensus
architecture and for PoW, we use the ropsten test net which
was based on proof-of-work consensus architecture.
Similarly, the simulation was carried out through the
smart contract as described in section III-B and PoA
consensus. The performance of the model is then compared
with PoW consensus to ensure the effectiveness of our smart
contract. The channel utilization where computed based on
equation (2) in kilobytes per seconds (kbps) and the
connection rate was assumed to be a function of the
transaction cost. Thus, transaction cost was computed to
ensure that the model favors cheaper connection rate. The cost
where taken as the amount of gas that was paid to the validator
account, which run the program allowing them to put
transactions in blocks.
From figure 2, we can deduce that the cost of the
transaction drastically drops down with the increase in number
of B.Buyers for the both consensus algorithms. The PoA
forecast much better and affordable price as the number of
B.Buyers increases than that of PoW. This outcome is as result
of the ability of each B.Buyer to become a B.Seller when it has
an excess bandwidth, thereby providing high supply of stock
in the SMP. This will in turn force each B.Seller to lower the
price of their stock according to the law of supply and demand
. Similarly, figure 3 clearly shows that the channel
utilization increases with the increase of buyers. This is as
result of increased high number of B.Sellers in the SMP,
which in return decreases the rate of transmission which in
line with equation (2).
We also understand that equation (3) and (4) play an
important role in linking the participants together. This feature
helps in increasing their proximity. The constrains also allow
only those close participants (buildings in same cluster) to
partake in the SMP. This help in boosting the transmission
time and the quality of service (QoS) which result in
robustness of the system. Equation (5) and (6) provide the
constrains that prevent sending a request that is beyond the
bandwidth capabilities of a given B.Seller and also ensure that
the highest obtainable transmission rate of the wireless link is
not exceeded. This will increase the quality of the signal and
therefore remove all unnecessary traffic jam and delays.
Figure 2: Transaction Cost (PoW Vs. PoA)
In a nutshell, we can easily depict from figure 3 also that,
bandwidth is utilized as the number of the buyers increases.
This shows that, excess bandwidth wastage is controlled at
least by 75% by making it available in the SMP for buyers.
Similarly, security is provided as all participants must be
registered and authenticated. This will allow only trusted
members of the community to participate in the SMP.
Figure 3: Network performance based on channel utilization (PoW
In this work, we proposed a bandwidth SMP which utilizes the
features of blockchain architecture which in turn provide
authentication, transparency, and none-repudiation.
Simulation results shows that the proposed model can also
provide efficient and effective method of buying and selling
between all participants. It also provides a cheaper connection
rate. Similarly, by utilizing the PoA consensus algorithm,
simulation results show that the model demonstrate to be more
robust and tends to be effective approach of handling excess
bandwidth wastage and providing secure excess bandwidth
trading within a smart city.
We would like to express our appreciation to Ms. Surayya
Ado Bala of Kano University of Science and Technology
Wudil, Kano-Nigeria, and Mr. Bashir Yusuf of Kano State
Senior Secondary Management Board (KSSSMB) for sharing
their pearls of knowledge with us over the course of this
research. We are eternally grateful to everyone who
contributed directly or indirectly to the accomplishment of this
 BITCOIN PR BUZZ N, “PRIVATIX , THE WORLD ’ S FIRST
BLOCKCHAIN- BASED BANDWIDTH MARKETPLACE ,
REACHES ICO SOFT CAP .,” 2017. [Online]. Available:
bandwidth-marketplace-reaches-ico-soft-cap/. [Accessed: 29-
 T. Clancy, “This Blockchain Helps Users Sell Idle Bandwidth
and Bypass Internet Censorship,” 2017. [Online]. Available:
bandwidth-bypass-internet-censorship. [Accessed: 29-Apr-2019].
 R. O. B. Chamberlin, “IoT could demand a whole new ,
dedicated network — here ’ s why Existing networks can ’ t take
the strain of a growing IoT,” 2019. [Online]. Available:
new-dedicated-network-heres-why/. [Accessed: 29-Apr-2019].
 S. Paris, F. Martisnon, I. Filippini, and L. Clien, “A bandwidth
trading marketplace for mobile data offloading,” in Proceedings -
IEEE INFOCOM, 2013, pp. 430–434.
 R. Bhatia, J. Chuzhoy, A. Freund, and J. (Seffi) Naor,
“Algorithmic aspects of bandwidth trading,” ACM Trans.
Algorithms, vol. 3, no. 1, p. 10, 2007.
 K. Eger and U. Killat, “Bandwidth Trading in Unstructured P2P
Content Distribution Networks,” 2006, pp. 39–48.
 K. Eger and U. Killat, “Bandwidth Trading as Incentive.”
 M. Bitsaki, G. D. Stamoulis, and C. Courcoubetis, “Auction-
based bandwidth trading in a competitive hierarchical market,”
NGI 2005 - Next Gener. Internet Networks Traffic Eng., vol.
2005, pp. 372–379, 2005.
 S. Gupta and A. Lozano, “Computation-Bandwidth Trading for
Mobile Edge Computing,” 2019 16th IEEE Annu. Consum.
Commun. Netw. Conf. CCNC 2019, no. 1, pp. 1–6, 2019.
 PRIVATIX.IO, “Internet Broadband Marketplace powered by
VPN Network on Blockchain,” 2018.
 T. The and M. Data, “TOKENIZING AND LIBERATING THE
MOBILE DATA MARKET,” 2017.
 T. M. Fernández-Caramés and P. Fraga-Lamas, “A Review on
the Use of Blockchain for the Internet of Things,” IEEE Access,
vol. 6. pp. 32979–33001, 2018.
 V. Buterin, “Ethereum White Paper,” Etherum, 2014. [Online].
 M. Singh and S. Kim, “Branch based blockchain technology in
intelligent vehicle,” Comput. Networks, vol. 145, pp. 219–231,
 Y. Zhang, S. Kasahara, Y. Shen, X. Jiang, and J. Wan, “Smart
contract-based access control for the internet of things,” IEEE
Internet Things J., vol. 6, no. 2, pp. 1594–1605, 2019.
 Y.-C. Hu, T.-T. Lee, D. Chatzopoulos, and P. Hui, “Hierarchical
interactions between Ethereum smart contracts across Testnets,”
in CRYBLOCK ’18: 1st Workshop on Cryptocurrencies and
Blockchains for Distributed Systems, 2018, pp. 7–12.
 B. M. Yakubu, M. I. Khan, N. Javaid, and A. Khan, “Blockchain-
based secure multi-resource trading model for smart
marketplace,” Computing, 2021.
 T. B. Proposition, “The Law of Supply and Demand,” Math.
Scand., vol. 3, no. 1, pp. 2–4, 2000.