Conference PaperPDF Available

BlocHIE: a BLOCkchain-based platform for Healthcare Information Exchange

Authors:
Conference Paper

BlocHIE: a BLOCkchain-based platform for Healthcare Information Exchange

Abstract and Figures

Nowadays, a great number of healthcare data are generated every day from both medical institutions and individuals. Healthcare information exchange (HIE) has been proved to benefit the medical industry remarkably. To store and share such large amount of healthcare data is important while challenging. In this paper, we propose BlocHIE, a Blockchain-based platform for healthcare information exchange. First, we analyze the different requirements for sharing healthcare data from different sources. Based on the analysis, we employ two loosely-coupled Blockchains to handle different kinds of healthcare data. Second, we combine off-chain storage and on-chain verification to satisfy the requirements of both privacy and authenticability. Third, we propose two fairness-based packing algorithms to improve the system throughput and the fairness among users jointly. To demonstrate the practicability and effectiveness of BlocHIE, we implement BlocHIE in a minimal-viable-product way and evaluate the proposed packing algorithms extensively.
Content may be subject to copyright.
BlocHIE: a BLOCkchain-based platform for
Healthcare Information Exchange
Shan Jiang, Jiannong Cao, Hanqing Wu, Yanni Yang, Mingyu Ma, Jianfei He
The Hong Kong Polytechnic University, Hong Kong, China
Huawei Technologies Co. Ltd., Shenzhen, China
{cssjiang,csjcao,cshwu,csynyang}@comp.polyu.edu.hk, derek.ma@connect.polyu.hk, jeffrey.he@huawei.com
Abstract—Nowadays, a great number of healthcare data are
generated every day from both medical institutions and individ-
uals. Healthcare information exchange (HIE) has been proved to
benefit the medical industry remarkably. To store and share such
large amount of healthcare data is important while challenging.
In this paper, we propose BlocHIE, a Blockchain-based platform
for healthcare information exchange. First, we analyze the
different requirements for sharing healthcare data from different
sources. Based on the analysis, we employ two loosely-coupled
Blockchains to handle different kinds of healthcare data. Second,
we combine off-chain storage and on-chain verification to satisfy
the requirements of both privacy and authenticability. Third,
we propose two fairness-based packing algorithms to improve
the system throughput and the fairness among users jointly.
To demonstrate the practicability and effectiveness of BlocHIE,
we implement BlocHIE in a minimal-viable-product way and
evaluate the proposed packing algorithms extensively.
I. INTRODUCTION
Healthcare has always been important to the society. Illness,
accidents, and emergencies do arise every day, and the incurred
ailments and diseases are supposed to be diagnosed, treated,
and managed. In recent years, healthcare information exchange
(HIE) among medical institutions has been proved to benefit
the medical industry a lot [1]. First, HIE can enhance the
understanding of each individual clinical trial. Second, the
researchers can get scientific insights by analyzing a bunch
of clinical trials. Third, the healthcare information interoper-
ability between clinical research enterprises strengthens their
collaborations.
Besides utilizing the data shared by the medical institutions,
daily data collection is also beneficial for personal health-
care. With the development of the Internet of things (IoT)
technology [2] [3], numerous personal healthcare data are
generated by the IoT devices every day [4]. The doctor can
take advantage of these data for precision medicine [5]. That
is, the doctor takes the individual variability in environment
and lifestyle into consideration when conducting disease treat-
ment or giving prevention advice. There is no doubt that the
data from individuals and various medical institutions benefits
healthcare. However, it is challenging to store and share such
large amount of data.
Early success in HIE arose from the field of cloud comput-
ing [6]. The idea to store the huge amount of data remotely
rather than locally is simple but effective. The cloud service
providers (CSPs) propose various schemes for reliable data
storage and efficient data processing. Then the stakeholders
choose a specific CSP by balancing various factors such as
cost and reliability. It has been a trend to resort to CSPs
when there are some data to be stored. The beneficiaries range
from patients, medical institutions, and research institutions
to big corporations. Therefore, the CSPs have been taking
great responsibilities to provide a controlled, cross-domain and
flexible HIE platform.
However, the CSPs have been struggling a lot to provide
data sharing service [7]. On the one hand, the cloud storage
market has been dominated by the largest CSPs such Google,
Dropbox, etc. They are unwilling to share their data with the
small/medium ones and between themselves due to market
competition. On the other hand, it is risky if the healthcare
data, which is highly private information, is exposed to the
malicious users unexpectedly. Fortunately, Blockchain tech-
nology, which starts at 2008 [8] and booms at 2014 [9],
provides great potential for HIE through its attractive features
such as security, privacy, decentralization, and immutability.
Blockchain technology has been successfully applied in
many areas. Bitcoin [8], as the first decentralized cryptocur-
rency, is also the first successful Blockchain application.
After the boom of cryptocurrencies, it comes to the era of
Blockchain 2.0 with the release of Ethereum [9]. During this
time, a lot of Blockchain-based systems are proposed for
the purpose of decentralization. The applications range from
transportation [10], e-government [11] to education [12].
When Blockchain technology meets HIE, there are only
few proposed systems [13][14] and they all suffer from the
following two problems. First, they only consider to store
and share the electronic medical records (EMRs) and ignore
the useful and numerous personal healthcare data (PHD).
The requirements to store and share the huge amount of
PHD are significantly different from storing and sharing the
EMRs, which brings new challenges in the aspect of system
throughput and fairness. Second, the existing systems directly
store the EMRs in the cloud environment with complicated
access control mechanism to prevent undesired data dissem-
ination. However, such system architecture heavily relies on
the security of the cloud environment.
To address the issues mentioned above, we propose
BlocHIE, a BLOCkchain-based platform for Healthcare In-
formation Exchange. In the system architecture, we use two
loosely-coupled Blockchain, namely EMR-Chain and PHD-
Chain to store EMRs and PHD separately. For the EMR-Chain,
49
2018 IEEE International Conference on Smart Computing
978-1-5386-4705-9/18/$31.00 ©2018 IEEE
DOI 10.1109/SMARTCOMP.2018.00073
we combine off-chain storage and on-chain verification to take
care of both privacy and authenticability, which also removes
the dependency on cloud services. For both of the EMR-Chain
and PHD-Chain, we propose two fairness-based transaction
packing algorithms to enhance the system throughput, and to
improve the fairness among the system users.
The contributions of this paper are as follows:
We analyzed the requirements for storing and sharing
EMRs and PHD. Based on the analysis, we propose to
use two loosely-coupled Blockchain, namely EMR-Chain
and PHD-Chain, as the system architecture. The EMR-
Chain stores EMRs from medical institutions while the
PHD-Chain serves the data from individuals. The usage
of multiple chains satisfies the different requirements of
storing and sharing different data.
We combine off-chain storage and on-chain verification
in the EMR-Chain, which fulfills the requirements of
privacy and authenticability, at the same time reduces the
storage overhead for the EMR-Chain.
We propose two fairness-based transaction packing algo-
rithms, namely FAIR-FIRST and TP&FAIR, for the EMR-
Chain and PHD-Chain respectively. The proposed algo-
rithms can enhance the system throughput and improve
the fairness among the users.
We implement BlocHIE in a minimal-viable-product way.
The implementation demonstrates the practicability of
BlocHIE. Moreover, we evaluate the packing algorithms
of FAIR-FIRST and TP&FAIR in terms of fairness and
throughput. The experimental result indicates that FAIR-
FIRST enhances fairness significantly and TP&FAIR im-
proves throughput remarkably while guaranteeing an ac-
ceptable fairness.
The rest of the paper is organized as follows. In Section II,
we describe the preliminaries towards developing BlocHIE.
Section III demonstrates the design of BlocHIE. In its subsec-
tions, three key novelty ranging from system architecture to
underlying algorithm are introduced. The system implemen-
tation and evaluation are showcased in Section IV. Finally,
Section V concludes the paper.
II. PRELIMINARIES
In this section, we formally describe the preliminaries
used in BlocHIE. We first introduce how Blockchain works,
the advantages of Blockchain, and how Blockchain benefits
BlocHIE in Subsection II-A. Then, we summarize the existing
distributed consensus algorithms and how BlocHIE is built
upon them in Subsection II-B.
A. Blockchain - Distributed Ledger Technology
A Blockchain is an append-only data structure, to store
a continuously growing list of transactions. A Blockchain is
replicated and maintained among the members of a network.
As a distributed ledger, Blockchain has two key features,
i.e., immutability and non-repudiability. The immutability is
achieved because it is computationally impossible to modify
any committed transaction in the Blockchain. The transactions
in a Blockchain are non-repudiable since they are replicated
by a large number of entities.
Size #Txs.
Header
Version
Previous Hash
Timestamp
Difficulty Target
Nonce
Merkle Root
H(*)
H12 = H(H1+H2)H34 = H(H3+H4)
H(H12+H34)
Transactions
Tx. 1 Tx. 2 Tx. 3 Tx. 4
H1H2H3H4
Block N-1 Block N
Size #Txs.
Header
Version
Previous Hash
Timestamp
Difficulty Target
Nonce
Merkle Root
Transactions
Tx. 1 Tx. 2 Tx. 3 ...
Fig. 1. Structure of a Traditional Blockchain
Traditionally, a Blockchain is a chain of blocks linked
and secured using cryptography. As shown in Fig. 1, each
block contains four components, namely block size, transac-
tion counter, block header, and transactions. The block size
and the transaction counter are the number of bytes of the
block and the number of transactions, respectively. The block
header contains six fields, namely version, previous block
hash, timestamp, difficulty target, nonce, and Merkle root. The
version is a version number to track the consensus protocol
upgrades, the timestamp is the approximate creation time of
the block, while the difficulty level and nonce are used for
proof-of-work consensus protocol. The Merkle root refers to
the hash of all the hashes of all the transactions. The previous
block hash is a reference to the hash of the previous block
along the chain. The hash value of a block, which is the
primary identifier of a block, is made by hashing the block
header twice through the SHA-256 hash function.
In our system BlocHIE, we take advantage of the im-
mutability and non-repudiability for HIE. On the one hand, the
feature of immutability is essential to prevent untrustworthy or
malicious modification on the healthcare records. On the other
hand, the healthcare records, as evidence to showcase the treat-
ment procedure between medical institutions and individuals,
should be non-repudiable. In addition, with the feature of non-
repudiability, an unnecessary disputation between the medical
institutions and individuals can be avoided.
B. Distributed Consensus
A Blockchain is replicated among the members of a net-
work, in which each member holds a replication of the
committed transactions and a pool of the submitted but uncom-
mitted transactions. Each member is responsible for packing
the transactions from the pool to the blocks to make them
committed. In order to make the Blockchain remain functional,
the members need to agree on a certain state of the Blockchain.
This procedure is accomplished by the underlying distributed
consensus algorithm.
50
Blockchain
BlockTransaction Pool
User
Header
Transactions
Size #Txs.
...
TxTx
Tx
Submit
Pack
Commit
Tx
Tx
Tx
Tx
Tx
Tx
Tx
Tx
Tx
Tx Tx
Tx Tx
Tx
Fig. 2. The procedure for commiting transactions
As shown in Fig. 2, it requires three steps for a transaction
to be committed. At first, the user has some raw transactions
(the red triangles) and wants to publish them. Then, the user
submits the raw transactions to the Blockchain network. Each
member in the Blockchain network receives the transactions
from the user and maintains a transaction pool. The transac-
tions in pool are called submitted transactions (the yellow and
the black ellipses). At this time, the members are supposed to
make consensus on the way to maintain the Blockchain based
on the transaction pool. The consensus consists of two steps,
namely packing and committing. At the packing stage, each
member selects some submitted transactions and puts them
into a block. The transactions that are packed into a block
but not yet committed are called packed transactions (the blue
rectangles), and the block containing packed transactions are
called uncommitted block. Finally, at the step of committing,
the members make efforts to get the uncommitted blocks
validated and committed. If a transaction is in a validated
block, it is said to be committed (the green tars).
As stated above, packing and committing are both required
for the distributed consesus. However, traditional Blockchain-
based systems, e.g., Bitcoin [8] and Ethereum [9], only focus
on the step of committing. Some of the applied committing
protocols include proof-of-work (PoW) [15], proof-of-stake
(PoS) [16], proof-of-burn [17], etc. In BlocHIE, we employs
PoW as one of the building blocks, which is introduced in
detail as follows. The PoW committing protocol is based
on some pre-defined puzzles that are difficult, i.e., costly
and time-consuming, to solve but easy to be verified. For
example in Bitcoin [8], the miner has to change the nonce
(as introduced in Fig. 1) constantly to make the hash value of
the block begin with a certain amount of zeros. It is difficult
to find such nonce while the validity is easy to be checked
once found. The cryptographic hash function used in Bitcoin
is twice SHA-256. Other hashing algorithms, including Scrypt
(used in Litecoin [18]) CryptoNight [19] (used in Monero
[20]), etc., are also employed.
III. BLOCHIE SYSTEM DESIGN
In this section, we outline the proposed platform BlocHIE,
a BLOCkchain-based platform for Healthcare Information
Exchange. We highlight the three key innovation points in the
following subsections.
A. System Architecture: Loosely-coupled EMR-Chain and
PHD-Chain
The system architecture of BlocHIE is presented in Fig. 3.
BlocHIE is envisioned for storing and sharing healthcare data
from medical institutions and individuals. There are mainly
three components in BlocHIE. The first component is the
Blockchain network. The Blockchain network is responsible
for storing and sharing the collected healthcare data. Anyone
who is willing to contribute to this platform can join the
network. The medical institutions, e.g., hospitals and clinics,
act as the second component. When there are new patients
in a hospital, their diagnostic records will be submitted to
the Blockchain network and shared with other hospitals and
clinics. The privacy issue will be discussed in subsection III-B.
The third component consists of all the individuals who are
willing to store and share their daily healthcare data. In a
smart home, numerous healthcare data are generated by the
IoT devices, e.g., smart watch, smart thermometer, and smart
sphygmomanometer. These devices can automatically submit
the generated data to the Blockchain network.
EMR-Chain PHD-Chain
Hospital Smart Home
Fig. 3. BlocHIE system architecture
Among the three components, there are two parties, i.e.,
medical institutions and individuals, who are submitting and
sharing healthcare data in BlocHIE. The reason why we
separate them is that there are different requirements to share
their data. For medical institutions, what they submit are
medical diagnostic report, medical examination report, etc.
These data are incredibly privacy-sensitive. Moreover, there
is a high demand to authenticate these data. For example, if
a patient receives some treatment in a medical hospital and
the medical diagnostic report is published with the signatures
from both the hospital and the patient, neither the hospital
nor the patient can deny the treatment. When it comes to
the data generated by the individuals, the primary concern
is the quantity. The amount of healthcare data generated by
each person is remarkable. Besides, the individuals compete to
publish their data for future healthcare usage. Consequently,
51
the key requirements to publish and share individuals’ data
are high throughput and substantial fairness. In the following
parts, we abbreviate the data generated by medical institutions
and individuals as critical data (EMR) and personal healthcare
data (PHD) respectively.
TABLE I
REQUIREMENTS TO PUBLISH AND SHARE HEALTHCARE DATA
Requirement EMR PHD
privacy high moderate
authenticability high no
throughput moderate high
latency moderate moderate
fairness moderate moderate
The requirements to publish and share EMRs and PHD are
summarized in Tab. I. As we can see from the summariza-
tion, their requirements are significantly different. Hence, we
propose to store and share EMR and PHD with two loosely-
coupled Blockchains, namely EMR-Chain and PHD-Chain.
Suppose that a person visits a hospital and some medical
diagnostic records are generated. If both the patient and the
hospital agree to publish the data, the data will be published to
the EMR-Chain with their signatures. Suppose that some daily
healthcare data are generated in a smart home, the data will
be published on PHD-Chain with the signature of the owner.
E
MR-Chain PHD-Chain
0x0012
0x0034 0x0056
0x0056
0x0078
Fig. 4. An individual can have multiple identities on EMR-Chain and PHD-
Chain
As shown in Fig. 4, EMR-Chain and PHD-Chain are
coupled because a single person can publish data on both of
the chains. However, they are only loosely coupled since the
identities of the same person can be different. An individual
knows the set of identities he/she owns. Indeed, the identity on
EMR-Chain can be interpreted as the unique record identifier,
while the identity on PHD-Chain can be treated as the unique
device identifier. When there is requirement to query the
healthcare data, the person can use the set of identities to
fetch data on both of the chains.
To conclude, we propose to use loosely-coupled EMR-
Chain and PHD-Chain to store and share EMR and PHD
respectively. The proposed system architecture can satisfy
different requirements of storing EMR and PHD concerning
privacy, authenticability, throughput, latency, and fairness.
B. Combining Off-chain Storage and On-chain Verification
In subsection III-A, we propose to use EMR-Chain to store
and share EMRs. As summarized in Tab. I, the key require-
ments of EMR are privacy and authenticability. However, in
existing Blockchain-based system, these two properties cannot
be guaranteed at the same time. Specifically, the whole data
is stored in existing Blockchain-based systems, which arouses
great privacy concern. To preserve privacy and authenticability
simultaneously, we propose to combine off-chain storage and
on-chain verification.
EMR-Chain
Transaction
Pool
Medical Record
Distributed
Database
Patient
Hospital
Timestamp
Medical Record Hash Value
Hospital Signature
Patient Signature
Keywords
Description
Fig. 5. The mechanism and structure of EMR-Chain
The process of publishing a piece of EMR is shown in
Fig. 5. When a medical record of a patient is generated at a
hospital, three copies of the medical record will be generated.
The first copy is stored in the database of the hospital, the
second copy is sent to the patient, and the third copy is
submitted to the Blockchain network. The first two copies are
identical and contain the full information of the EMR while
there is a vast difference between the third copy and the first
two. The full structure of the third copy is shown in Fig. 5. It
contains the timestamp, the hash value of the medical record,
the hospital signature, the patient signature, a set of keywords,
and extra description. The hash value of the medical record is
generated using some digest algorithm such as MD5.
Indeed, the third copy serves as a proof-of-existence copy
rather than a full copy. The advantages of such structure
are as follows. First, the detailed medical record is not
publicly accessible, which preserves the privacy of the pa-
tient. Second, EMR-Chain reduces the throughput requirement
significantly. The original medical records are large files of
several megabytes. If they are stored, it requires a very
high throughput of the system. Moreover, a single block can
even only contain a single record. In EMR-Chain, the hash
value, whose size is several kilobytes, is stored instead, which
reduces the throughput requirement. Third, authenticability is
preserved in EMR-Chain. The patient and the hospital can
compare their records in hand with the hash value along the
EMR-Chain to authenticate the medical record. It can prevent
the repudiation of the hospital and the patient. Finally, EMR-
Chain enhances the searchability. The keywords published
along with the medical record can be used for information
52
retrieval. For example, if the data of a certain kind of disease
is desired, the disease name be used for searching.
To conclude, the design concept of EMR-Chain is to
combine off-chain storage and on-chain verification. On the
one hand, the off-chain storage is achieved by storing in the
distributed databases of the hospitals. On the other hand, the
on-chain verification is achieved by including the hash value
of each medical record in the transaction.
C. Fairness-based Transaction Packing Algorithm
In subsection III-A, we propose to use PHD-Chain to store
and share data from individuals. As summarized in Tab. I,
the key requirement of PHD is throughput. However, exist-
ing Blockchain-based system cannot satisfy the throughput
requirement of sharing PHD. To this end, we propose two
fairness-based transaction packing algorithms. The proposed
algorithms can bring about not only high throughput but low
latency and moderate fairness as well.
To introduce the algorithm, we firstly define some termi-
nologies, i.e.,response time,waiting time, and fairness. Jain et.
al. introduced Jain’s fairness index to evaluate the fairness in
allocation of a resource to a set of users/devices [21]. Suppose
there are nusers sharing a network service and xito be the
throughput for the i-th user, then the Jain’s fairness index is
defined as J(x1,x
2,··· ,x
n)=(
n
i=1 xi)2/(n·n
i=1 x2
i).In
a Blockchain-based system, the fairness is defined in a similar
way.
Definition 1. Suppose a transaction xiis submitted at time
siand committed at time ei, then the response time tiof xi
is defined as:
ti=eisi(1)
Definition 2. Suppose there are ncommitted transactions
x1,x
2,··· ,x
nwith response time t1,t
2,··· ,t
n. The fairness
of the system is defined as:
J(x1,x
2,··· ,x
n)= (n
i=1 ti)2
n·n
i=1 t2
i
(2)
Definition 3. Suppose a submitted or packed transaction xi
is submitted at time si, then the waiting time wiof xiat time
tcis defined as:
wi=tcsi(3)
We assume that the submitting times of the transactions
are distinct. It is reasonable since there must be a slight
time difference between submitting two transactions. Even
if two transactions are submitted at the the same time, the
symmetry can be broken by comparing the transaction content
e.g., assume that the transaction with larger hash value is
submitted slightly later. According to Eq. 3, the waiting times
of transactions in pool are distinct as well.
In a Blockchain-based system, the number of transactions
inside a block should be bounded. A block will be of huge
size if too many transactions are included. When a huge
block is synchronized in the Blockchain network, the network
congestion will be very high. There are a lot of research on
setting the optimal block size [22][23]. In our system, we set
the maximum number of transactions inside a block to be an
adjustable parameter m.
When the Blockchain network wants to get some trans-
actions committed, each node in the Blockchain network is
supposed to follow the procedure illustrated in Fig. 2. That is,
the nodes are supposed to select some transactions from the
transaction pool first. Different transactions in the transaction
pool can have different waiting times. It is intuitive to pack as
many transactions as possible and to pack those transactions
that have the longest waiting times. On the one hand, it
can increase the throughput to pack as many transactions as
possible. On the other hand, it can enhance the fairness to pack
those transactions with the longest waiting times. To this end,
all the nodes will pack the transactions with top-mwaiting
times. However, it is indeed a waste of computing resources
for all the nodes to work on the same subset of transactions.
As a result, it is a problem to coordinate the nodes to pack
transactions in the Blockchain network to take both fairness
and throughput into consideration.
As above, we give the intuition to pack the transactions
with top-mwaiting times. Here, we formally prove that this
strategy achieves the maximum fairness.
Theorem 4. Given a setting of ntransactions x1,x
2,··· ,x
n
in pool with waiting time w1,w
2,··· ,w
nand mtransactions
are supposed to be packed, the strategy to pack the transac-
tions with top-mwaiting times achieves the maximum fairness.
Proof. Assume without loss of generality that the transactions
x1,x
2,··· ,x
nare sorted with decreasing waiting times, i.e.,
w1>w
2>··· >w
n. Let us notate the strategy to pack
the transactions with top-mwaiting times as MAX-PACK.
It is clear MAX-PACK pack the transactions in the order of
x1,x
2,··· ,x
n.
Assume by contradiction that there is another packing
algorithm OP-PACK that can achieve larger fairness. Let OP-
PACK packs the transactions in the order of σ1
2
n, where
σis a permutation other than (1,2,··· ,n). By definition, we
have the following property:
m
i=1
wi>
m
i=1
wσi(4)
2k<n
m,
km
i=1
wi
km
i=1
wσi(5)
n
i=1
wi=
n
i=1
wσi(6)
Assume the time to make a packed block to be committed to
be tp. Then the response times of the transactions using MAX-
PACK are w1+tp,w
2+tp,··· ,w
m+1 +2tp,··· ,w
n+n
m·tp.
The response times of the transactions using OP-PACK are
wσ1+tp,w
σ2+tp,··· ,w
σm+1 +2tp,··· ,w
σn+n
m·tp.To
53
this end, the fairness of MAX-PACK and OP-PACK and their
relationship are as follows:
JMaxPack =(n
i=1(wi+i
m·tp))2
n·n
i=1(wi+i
m·tp)2(7)
JOP Pack =(n
i=1(wσi+i
m·tp))2
n·n
i=1(wσi+i
m·tp)2(8)
JOP Pack >JMaxPack (9)
Since the algorithms are running on the same set of transac-
tions, we have n
i=1
w2
i=
n
i=1
w2
σi(10)
n
i=1
(wi+i
m·tp)=
n
i=1
(wσi+i
m·tp)(11)
According to Eq. 7,8,9,11, we have:
n
i=1
(wi+i
m·tp)2>n·
n
i=1
(wσi+i
m·tp)2(12)
Expand Eq. 12, we get:
n
i=1
w2
i+
n
i=1
(i
m·tp)2+2
n
i=1
(wi· i
m·tp)>
n
i=1
w2
σi+
n
i=1
(i
m·tp)2+2
n
i=1
(wσi· i
m·tp)
(13)
According to Eq. 10,13, we have:
n
i=1
(wi· i
m)>
n
i=1
(wσi· i
m)(14)
Adding Eq. 4 and all the inequations in Eq. 5, we have
n
m−1
i=1
im
j=1
wj>
n
m−1
i=1
im
j=1
wσj(15)
Adding the inquestions in Eq. 14 and Eq. 15, we have:
n
m
n
i=1
wi=
n
m−1
i=1
im
j=1
wj+
n
i=1
(wi· i
m)
>
n
m−1
i=1
im
j=1
wσj+
n
i=1
(wσi· i
m)
=n
m
n
i=1
wσi
(16)
Obviously, Eq. 16 is contradictory with Eq. 6. As a result,
the assumption does not hold and MAX-PACK achieves max-
imum fairness.
Similarly, we can get the corollary that the larger the sum of
the waiting times of the transactions is, the larger the fairness
is. To this end, we can get the strategies to pack transactions
with the largest, the 2-nd largest fairness and etc.. Suppose
there are knodes in the Blockchain network, then they can
coordinate to use the strategies with the largest, the 2-nd
largest, ···, and the k-th largest fairness to get the maximum
throughput and moderate fairness. However, there is a still a
gap towards finding the strategy with the k-th largest fairness.
The gap is the KTH-SUM problem defined as follows.
Definition 5. KTH-SUM: Given a set of npositive real
numbers X={x1,x
2,··· ,x
n}and a positive integer m<n,
there are n
mdistinct subsets of Xof size m. Among the n
m
subsets, find the one with the k-th largest sum.
Algorithm 1 An approaximate algorithm to find the subset of
size mwith k-th largest sum in a set Xof npositive real
numbers
aan array of size m
procedure APP-KTH-SUM(X, n, m, k)
for tar m(m+1)
2to do
prevk k
if DFS(n, m, k, 1,0,0,tar)then
Sort Xin decreasing order
return xa[1],x
a[2],··· ,x
a[m]
end if
if prevk =kthen
return FALSE
end if
end for
end procedure
procedure DFS(n, m, &k, d, p, sum, tar)Here, kis
passed by reference
if d=mthen
if tar sum > n then
return FALSE
end if
a[d]tar sum
kk1
if k=0then
return TRUE
end if
end if
for ip+1 to do
if sum +(2·i+md)·(md+1)
2>tar then
BREAK
end if
a[d]i
if DFS(n, m, k, d +1,i,sum+i, tar)then
return TRUE
end if
end for
return FALSE
end procedure
Actually, we do not need to solve the KTH-SUM problem
exactly. Instead, we only need to find an approximate solution.
Hence, we propose the algorithm APP-KTH-SUM as shown in
Alg. 1. The intuition of the algorithm is shown in Fig. 6. We
54
[ x1 x2 ... xm-2 xm-1 xm ]
[ x1 x2 ... xm-2 xm-1 xm+1 ]
[ x1 x2 ... xm-2 xm-1 xm+2 ][ x1 x2 ... xm-2 xm xm+1 ]
[ x1 x2 ... xm-2 xm-1 xm+3 ][ x1 x2 ... xm-2 xm xm+2 ][ x1 x2 ... xm-2 xm xm+2 ]
>
>>
>>> >
m(m+1)/2
m(m+1)/2 +1
m(m+1)/2 +2
m(m+1)/2 +3
.
.
.
.
.
.
.
.
.m(m+1)/2 +4
Index SumPacked Txs.
Fig. 6. Intuition for KTH-SUM problem
separate the subsets level by level. For a subset in a lower
level, there must be a subset which has larger subset sum
in the upper level. The level is determined by the sum of
the index. In Alg. 1, we enumerate the sum of the index
from the smallest possible one, i.e.,m(m+1)
2, to the infinity.
For each target sum of the index, we use depth-first-search
algorithm to find all the possible transaction combinations. For
a target sum of the index, if there is not a single transaction
combination whose index sum is the target, the procedure
returns false, which means k>n
m.Ifkreaches 0,it
means an approximate answer is found, and the corresponding
transaction combination is returned.
Algorithm 2 Throughput-first and fairness-first packing algo-
rithm running on node i
procedure TP&FAIR(X)
mthe maximum number of transactions in a block
XAPP-KTH-SUM(X, |X|,m,i)
return X
end procedure
procedure FAIR-FIRST(X)
mthe maximum number of transactions in a block
XAPP-KTH-SUM(X, |X|,m,1)
return X
end procedure
Based on the proposed APP-KTH-SUM algorithm, we fur-
ther propose two packing algorithms, i.e.,F
AIR-FIRST and
TP&FAIR, to coordinate the nodes in the Blockchain network.
The algorithms are shown in Alg. 2. FAIR-FIRST is used when
fairness is critical in the system while TP&FAIR sacrifices
a little fairness for higher throughput. The two packing al-
gorithms can be selected based on the required features of
the Blockchain-based system. In our system, PHD-Chain is
designed to use TP&FAIR for high throughput and moderate
fairness while EMR-Chain is designed to use FAIR-FIRST
since throughput is relatively less important. The intuition
for the FAIR-FIRST packing algorithm is to let all the nodes
work on the same transaction combination, which is of the
maximum fairness. The intuition for the TP&FAIR packing
algorithm is to let the nodes work on different transaction
combinations, while can achieve top fairness respectively.
IV. SYSTEM IMPLEMENTATION AND EVALUATION
To demonstrate the effectiveness and practicability of
BlocHIE, we implement BlocHIE in a minimal-viable-product
version. As shown in Fig. 7, the implementation is divided into
three layers, namely communication layer, Blockchain layer,
and GUI layer.
GUI
Layer
Blockchain
Layer
Communication
Layer
TP&FAIRFAIR-FIRST
PoW
EMR-Chain
I
PHD-Chain
Django Web Framework
gRPC-python
syntax= "proto2";
service Discovery {
rpc ExchangeNode(Node) returns (Node);
rpc Hello(Message) returns (Message);
}
service Synchronization{
rpc BlockFrom(Message) returns (Block);
rpc BlockTo(Block) returns (Message);
rpc ExchangeBlock(Block) returns (Block);
rpc TransactionTo(Transaction) returns (Message);
rpc TransactionFrom(Message) returns (Transaction);
}
message Transaction{
required bytes unixtime = 1;
required bytes body = 2;
required bytes txhash = 3;
required int32 type = 4;
required bytes txfrom = 5;
optional bytes txto = 6;
}
message Block{
required uint64 height = 1;
required bytes unixtime = 2;
required bytes previoushash = 3;
required bytes blockhash = 4;
required bytes difficulty = 5;
required bytes answer = 6;
repeated bytes txshash = 7;
required bytes miner = 8;
required int32 number = 9;
}
message Node{
required int32 number = 1;
repeated bytes ipport = 2;
}
message Message{
required bytes value = 1;
}
Fig. 7. Techniques for system implementation level by level
The bottom layer, i.e., communication layer, is imple-
mented using gRPC-python1. There are two services to support
Blockchain-based system, i.e., peer discovery service (“Dis-
covery”) and synchronization service (“Synchronization”) as
shown in Fig. 7. The “Discovery” service is used for dis-
covering the nodes inside the Blockchain network. When a
node is started, it will greet several static nodes (the same
as bootnodes in Ethereum) and exchange the connectivity
information with the static nodes. The block and transaction
synchronization is achieved by the “Synchronization” service,
which includes several remote procedure calls (RPCs) such
as “BlockFrom”, “BlockTo”, “BlockFrom”, “TransactionTo”,
and “TransactionFrom”.
At the middle layer, two Blockchains, i.e., EMR-Chain and
PHD-Chain are implemented. The EMR-Chain employs the
FAIR-FIRST transaction packing algorithm while the PHD-
Chain utilizes the TP&FAIR transaction packing algorithm.
1https://grpc.io/
55
For the block committing algorithm, both EMR-Chain and
PHD-Chain employs PoW.
Django web framework2is used in the top layer, i.e., the
GUI layer. It opens an HTTP port and presents HTML pages
using the port. In this way, the users can submit data fol-
lowing the HTTP protocol. When some data is submitted, we
invoke the methods on Blockchain layer to fulfill Blockchain
functions.
0.0
0.2
0.4
0.6
0.8
1.0
0
20
40
60
80
100
0.93
0.83
0.32
37.2
46 44.3
Fairness
Throughput
TP&FAIRFAIR-FIRST RANDOM
Fig. 8. Performance comparison of different packing algorithms
After implementation, we measure the performance of
BlocHIE with 8 nodes. Each node is serving as both server
and client, i.e., sending requests and packing transactions at
the same time. The frequency of sending requests of each node
is around 7 tx/s. Moreover, we set the number of transactions
inside a block, i.e.,m, to be 56, which is the approximate
transaction generating rate. We compare the performance of
TP&FAIR,FAIR-FIRST, and RANDOM concerning both fair-
ness and throughput. Here, the RANDOM packing algorithm
refers to the algorithm that randomly pick mtransactions
from pool. The result is shown in Fig. 8. We observe that in
terms of fairness, both FAIR-FIRST and TP&FAIR outperform
RANDOM significantly. Specifically, they achieve up to 2.9x
and 2.6x higher fairness than RANDOM respectively. From the
perspective of throughput, TP&FAIR achieves the maximum,
i.e., 46 tx/s, which improves FAIR-FIRST over 23.6%.
V. C ONCLUSION
In this paper, we propose BlocHIE, a Blockchain-based
platform for healthcare information exchange. We consider
two kinds of healthcare data, i.e., electronic medical records
and personal healthcare data, and analyzed the different re-
quirements to store and share them. Based on the analysis, we
architect BlocHIE on two loosely-coupled Blockchains, i.e.,
EMR-Chain for electronic medical records and PHD-Chain
for personal healthcare data. Inside EMR-Chain, we integrate
the techniques of off-chain storage and on-chain verification
to take good care of privacy and authenticability. Moreover,
we propose two transaction packing algorithms to enhance the
system throughput and the fairness among users. Finally, the
implementation and evaluation indicate the practicability and
effectiveness of BlocHIE.
2https://www.djangoproject.com/
ACKNOWLEDGMENTS
This work is supported by Huawei Technologies Co. Ltd.
with project code P15-0540 and RGC CRF with project
number CityU C1008-16G.
REFERENCES
[1] B. E. Dixon and C. M. Cusack, “Measuring the value of health
information exchange,” in Health Information Exchange. Elsevier,
2016, pp. 231–248.
[2] X. Liu, K. Li, G. Min, Y. Shen, A. X. Liu, and W. Qu, “Completely
pinpointing the missing rfid tags in a time-efficient way,” IEEE Trans-
actions on Computers, vol. 64, no. 1, pp. 87–96, 2015.
[3] S. Jiang, J. Cao, Y. Liu, J. Chen, and X. Liu, “Programming large-scale
multi-robot system with timing constraints,” in Computer Communica-
tion and Networks (ICCCN), 2016 25th International Conference on.
IEEE, 2016, pp. 1–9.
[4] S. R. Islam, D. Kwak, M. H. Kabir, M. Hossain, and K.-S. Kwak, “The
internet of things for health care: a comprehensive survey,” IEEE Access,
vol. 3, pp. 678–708, 2015.
[5] F. S. Collins and H. Varmus, “A new initiative on precision medicine,”
New England Journal of Medicine, vol. 372, no. 9, pp. 793–795, 2015.
[6] J. Zhou, Z. Cao, X. Dong, and X. Lin, “Tr-mabe: White-box traceable
and revocable multi-authority attribute-based encryption and its appli-
cations to multi-level privacy-preserving e-healthcare cloud computing
systems,” in INFOCOM. IEEE, 2015, pp. 2398–2406.
[7] N. Grozev and R. Buyya, “Inter-cloud architectures and application
brokering: taxonomy and survey,” Software: Practice and Experience,
vol. 44, no. 3, pp. 369–390, 2014.
[8] S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” 2008.
[9] G. Wood, “Ethereum: A secure decentralised generalised transaction
ledger,Ethereum Project Yellow Paper, vol. 151, 2014.
[10] A. Lei, H. Cruickshank, Y. Cao, P. Asuquo, C. P. A. Ogah, and
Z. Sun, “Blockchain-based dynamic key management for heterogeneous
intelligent transportation systems,” IEEE Internet of Things Journal,
vol. 4, no. 6, pp. 1832–1843, 2017.
[11] H. Hou, “The application of blockchain technology in e-government in
china,” in ICCCN. IEEE, 2017, pp. 1–4.
[12] M. Turkanovi´
c, M. H¨
olbl, K. Koˇ
siˇ
c, M. Heriˇ
cko, and A. Kamiˇ
sali´
c,
“Eductx: A blockchain-based higher education credit platform,” IEEE
Access, 2018.
[13] Q. Xia, E. B. Sifah, K. O. Asamoah, J. Gao, X. Du, and M. Guizani,
“Medshare: Trust-less medical data sharing among cloud service
providers via blockchain,” IEEE Access, vol. 5, pp. 14 757–14767, 2017.
[14] Z. Shae and J. J. Tsai, “On the design of a blockchain platform for
clinical trial and precision medicine,” in ICDCS. IEEE, 2017, pp.
1972–1980.
[15] M. Jakobsson and A. Juels, “Proofs of work and bread pudding proto-
cols,” in Secure Information Networks. Springer, 1999, pp. 258–272.
[16] “Nxt: The blockchain application platform via proof-of-stake,” https:
//nxtplatform.org/, accessed: 2018-01-28.
[17] “Slimcoin: A peer-to-peer crypto-currency with proof-of-burn,” https:
//slimcoin-project.github.io/, accessed: 2018-01-28.
[18] “Litecoin: Open source p2p digital currency,” https://litecoin.org/, ac-
cessed: 2018-01-26.
[19] “Cryptonight hash function,” https://cryptonote.org/cns/cns008.txt, ac-
cessed: 2018-01-28.
[20] “Monero [xmr]: a brand new uprising cryptocurrency which originates
from bitmonero,” http://dogecoin.com/, accessed: 2018-01-28.
[21] R. Jain, D.-M. Chiu, and W. R. Hawe, A quantitative measure of fairness
and discrimination for resource allocation in shared computer system.
Eastern Research Laboratory, Digital Equipment Corporation Hudson,
MA, 1984, vol. 38.
[22] K. Croman, C. Decker, I. Eyal, A. E. Gencer, A. Juels, A. Kosba,
A. Miller, P. Saxena, E. Shi, E. G. Sirer et al.,“On scaling decentralized
blockchains,” in International Conference on Financial Cryptography
and Data Security. Springer, 2016, pp. 106–125.
[23] A. Gervais, G. O. Karame, K. W¨
ust, V. Glykantzis, H. Ritzdorf,
and S. Capkun, “On the security and performance of proof of work
blockchains,” in CCS. ACM, 2016, pp. 3–16.
56
... It is additionally hard to track down broken hubs in this construction. A large portion of the IoT gadgets associate with the brought together data set and cloud network which builds the expense and computational necessities [21][22][23][25][26][27][28]. In the circulated blockchain innovation, hubs have least availability yet, the organization remains solid and safe. ...
Article
Full-text available
Block-chain is getting famous and one of the most widely recognized subjects can be thought of it has additionally changed the ways of life of many individuals in specific fields, in view of the effect on organizations and joins. Blockchain guarantees more dependable and fitting assets and it is exceptionally vital to remember that security and protection have a few deterrents as any innovation in this fields. The range of blockchain applications is extremely reached out into various regions in banking, wellbeing, car, the Internet of Things (IoT) and so forth. Numerous concentrates on focus on utilizing the block-chain information model in various executions. Motivated from these facts, in this paper, we present a systematic review of various blockchain-based solutions and their applicability in various Industry 4.0-based applications. In this paper, we attempt to portray block-chain innovation by talking about its model of information security and protection point of view with various agreement calculations, as well as issues and open doors in blockchains.
... The different edge devices need to share and synchronize stateful data to collaborate efficiently with applications in connected healthcare [67] and smart logistics [68]. While the topic of distributed data sharing has been studied in the context of traditional parallel and distributed systems and cloud, It is much more challenging to enable it in the context of edge computing. ...
Preprint
Full-text available
Edge computing has become a popular paradigm where services and applications are deployed at the network edge closer to the data sources. It provides applications with outstanding benefits, including reduced response latency and enhanced privacy protection. For emerging advanced applications, such as autonomous vehicles, industrial IoT, and metaverse, further research is needed. This is because such applications demand ultra-low latency, hyper-connectivity, and dynamic and reliable service provision, while existing approaches are inadequate to address the new challenges. Hence, we envision that the future edge computing is moving towards distributed intelligence, where heterogeneous edge nodes collaborate to provide services in large-scale and geo-distributed edge infrastructure. We thereby propose Edge-as-a-Service (EaaS) to enable distributed intelligence. EaaS jointly manages large-scale cross-node edge resources and facilitates edge autonomy, edge-to-edge collaboration, and resource elasticity. These features enable flexible deployment of services and ubiquitous computation and intelligence. We first give an overview of existing edge computing studies and discuss their limitations to articulate the motivation for proposing EaaS. Then, we describe the details of EaaS, including the physical architecture, proposed software framework, and benefits of EaaS. Various application scenarios, such as real-time video surveillance, smart building, and metaverse, are presented to illustrate the significance and potential of EaaS. Finally, we discuss several challenging issues of EaaS to inspire more research towards this new edge computing framework.
... As is the case with each new technology, it addresses certain problems while creating new ones, as is the case with quantum computing and security. These computers are capable of solving even the most difficult mathematical problems [42] [43]. While large-scale quantum computers remain undiscovered, some academics believe that developing new encryption methods for a postquantum world has come. ...
Article
Electronic health information exchange allows providers to healthcare providers and patients to securely access vital medical information of the patient and share it in electronically with other healthcare providers. The paper considers the electronic health system eHealth. The patient's medical records will be stored in a distributed ledger, and not on the server of the medical institution. So no one can change patient data or deny him access to his own data. The use of DAG is considered on the example of the transfer of medical patient data to the doctor. The process of creating and sending medical data to DAG is illustrated with a specific example.
Book
Full-text available
A key focus in recent years has been on sustainable development and promoting environmentally conscious practices. In today’s rapidly evolving technological world, it is important to consider how technology can be applied to solve problems across disciplines and fields in these areas. Further study is needed in order to understand how technology can be applied to sustainability and the best practices, considerations, and challenges that follow.
Article
Full-text available
Wireless body sensor networks (WBSNs) pose significant security and privacy risks. The Medical Server (MS) will only allow legitimate stakeholders access to confidential patient medical records when successful mutual authentication between all registered users and the MS has been confirmed using preset secret attributes. This paper proposes a novel approach to overcome the security and privacy problems in WBSNs by using CP-ABE and a consortium blockchain for key management and authentication. In this paper, a fixed-size session key is computed by utilizing several attribute base rules and AND/OR logic gate combinations. IEEE 802.15.6 is also used to transmit the encoded patient data from the register and legitimately deployed biosensor nodes on a patient’s body to the Base Station nearby (BS). This was done in part by leveraging consortium blockchains to construct partial blocks and then, transmit the encrypted partial blocks to MS via peer-to-peer networks, as well as aggregating critical physiological information. MS is now validating partial blocks with a hash function to ensure their integrity before converting them all into full blocks, which are subsequently mined and put on the blockchain effectively and ideally using a consensus mechanism. When sessions between MS and stakeholders are established, all legitimate consumers can view the secure medical records of a registered patient in a hospital using their predefined access structure.. The resource-constrained environment of WBSNs can benefit from enhanced data security and privacy by using CP-ABE in conjunction with the organization’s consensus to encrypt the patient’s critical features or attributes. Automated Validation of Internet Security Protocol and Applications (AVISPA) tool is used to verify the validity and correctness of the proposed authentication mechanism. The proposed scheme reduces transmission, processing and storage costs and energy usage by a significant margin when compared to current state-of-the-art alternatives. It is also worth noting that a multicriteria decision making (MCDM) approach known as Evaluation Based on Distance from Average Solution (EDAS) is employed to provide the ranking and determine which strategy is optimal across all of the domains involved.
Chapter
Deep learning is a subfield of machine learning that deals with artificial neural networks, which are algorithms encouraged by the structure and operation of the brain. Health practitioners and analysts are using deep learning to discover the latent opportunities in data and to support the healthcare industry in better way. Deep learning in healthcare enables physicians to correctly analyze any illness to properly manage it which results in better treatment decisions. Blockchain and Internet of things are also used as deep learning models in the healthcare system to manage the large healthcare data, such as electronic medical records. Mobile healthcare system is also a part of deep learning, which is used nowadays by health practitioners to manage the healthcare system by online mode. Therefore, there is a need to concentrate on the knowledge and use of certain methods in order to address the shortcomings of current mainstream approaches to human health monitoring. This chapter describes the role of deep learning in the patient care by using different models and managing the healthcare system effectively.
Article
The existing supply chain for the pharmaceutical industry is obsolete and lacks clear visibility over the entire system. Moreover, the circulation of counterfeit drugs in the market has increased over the years. According to the WHO report, around 10.5% of the medicinal drugs in lower / middle income countries are fake and such drugs may pose serious threats to public health, sometimes leading to death. Keeping these threats in mind, in this paper, we propose a blockchain-based model to track the movement of drugs from the industry to the patient and to minimize the chances of a drug being counterfeit. The reasons for using blockchain technology in our work include its immutability property and easy tracking of an entity in the blockchain. Through this proposed model, the manufacturer would be able to upload the details corresponding to a drug, after which it will be sent for approval to the Government. Thereafter, hospitals and pharmacies, based upon their requirements, can request the approved drugs. In the future, if a patient wants some medication, then he or she has to request it on the blockchain network. The request will be sent to the nearest hospital/pharmacy and thereafter, the patient can collect the medication. To implement this model, we have used Hyperledger fabric due to the presence of many auto-implemented features in it. Our implementation of the proposed blockchain based model highlights that the model can successfully detect any drug being counterfeit. This will be beneficial for the users getting affected with counterfeit drugs. Moreover, with the proposed model, we can also track the movement of the drug beginning from the manufacturer right up to the patient consuming that drug. Index Terms: Blockchain, Counterfeit Drugs, Drugs Tracking, Fake Medicines, Health Care.
Article
With the rapid development of blockchain, user’s daily financial bills and transfer records can be safely and completely stored in a transaction form. These transactions express commercial preferences for various scenarios and can be used to improve the quality of application services such as data analysis, data security, and recommendation systems. Therefore, there is a growing demand for processing diverse queries over blockchain transactions. In this paper, we formulate a new problem of blockchain top-k transaction path query (BCTkPQ) which returns first k transactions in a given query path according to the user-specified conditions. For processing BCTkPQ, we design a novel Collaborative Query Model (CQM) consisting of a set of collaborative peers and each of which contains three key components, i.e., parser, indexer and executor. First, in the parser, we propose a graph-based structure to denote the transaction paths. Then, in the indexer, we propose a two-level index to increase query and verification efficiency. Finally, in the executor, we present an optimized query algorithm for answering BCTkPQ securely and efficiently and further give a verification algorithm to guarantee the correctness of the query results. We conduct extensive experiments to illustrate the efficiency and effectiveness of our solution.
Chapter
In an age of rich data and information streams, the governments’ role with regard to new technology in terms of regulation is continuously changing. Novel areas for this era of big data must be considered when introducing new policies and issuing regulations. These areas cover privacy, security, retention, processing, ownership, and the integrity of information. In this paper, we have focused on the blockchain technology and its features such as distributed ledger and smart contracts in a regulation point of view. We have defined the current regulation landscape and initiatives across the globe and then highlighted some applications of the blockchain technology through regulation perspective in different sectors. We have emphasized some challenges faced by the blockchain technology and came up with some recommendations to tackle those challenges.KeywordsBlockchainDistributed ledgerPoliciesStandardsRegulations
Article
Full-text available
Blockchain technology enables the creation of a decentralized environment where transactions and data are not under the control of any third party organization. Any transaction ever completed is recorded in a public ledger in a verifiable and permanent way. Based on blockchain technology, we propose a global higher education credit platform, named EduCTX. This platform is based on the concept of the European Credit Transfer and Accumulation System (ECTS). It constitutes a globally trusted, decentralized higher education credit and grading system that can offer a globally unified viewpoint for students and higher education institutions (HEIs), as well as for other potential stakeholders such as companies, institutions, and organizations. As a proof of concept, we present a prototype implementation of the environment, based on the open-source Ark Blockchain Platform. Based on a globally distributed peer-to-peer network, EduCTX will process, manage and control ECTX tokens, which represent credits that students gain for completed courses such as ECTS. HEIs are the peers of the blockchain network. The platform is a first step towards a more transparent and technologically advanced form of higher education systems. The EduCTX platform represents the basis of the EduCTX initiative which anticipates that various HEIs would join forces in order to create a globally efficient, simplified and ubiquitous environment in order to avoid language and administrative barriers. Therefore we invite and encourage HEIs to join the EduCTX initiative and the EduCTX blockchain network.
Article
Full-text available
As modern vehicle and communication technologies advanced apace, people begin to believe that Intelligent Transportation System (ITS) would be achievable in one decade. ITS introduces information technology to the transportation infrastructures and aims to improve road safety and traffic efficiency. However, security is still a main concern in Vehicular Communication Systems (VCS). This can be addressed through secured group broadcast. Therefore, secure key management schemes are considered as a critical technique for network security. In this paper, we propose a framework for providing secure key management within the heterogeneous network. The security managers (SMs) play a key role in the framework by capturing the vehicle departure information, encapsulating block to transport keys and then executing rekeying to vehicles within the same security domain. The first part of this framework is a novel network topology based on a decentralised blockchain structure. The blockchain concept is proposed to simplify the distributed key management in heterogeneous VCS domains. The second part of the framework uses the dynamic transaction collection period to further reduce the key transfer time during vehicles handover. Extensive simulations and analysis show the effectiveness and efficiency of the proposed framework, in which the blockchain structure performs better in term of key transfer time than the structure with a central manager, while the dynamic scheme allows SMs to flexibly fit various traffic levels.
Article
Full-text available
The dissemination of patients’ medical records results in diverse risks to patients’ privacy as malicious activities on these records cause severe damage to the reputation, finances, etc. of all parties related directly or indirectly to the data. Current methods to effectively manage and protect medical records have been proved insufficient. In this paper we propose MeDShare, a system that addresses the issue of medical data sharing among medical big data custodians in a trust-less environment. The system is blockchainbased and provides data provenance, auditing and control for shared medical data in cloud repositories among big data entities. MeDShare monitors entities that access data for malicious use from a data custodians system. In MeDShare, data transitions and sharing from one entity to the other, along with all actions performed on the MeDShare system, are recorded in a tamper-proof manner. The design employs smart contracts and an access control mechanism to effectively track the behavior of the data and revoke access to offending entities on detection of violation of permissions on data. The performance of MeDShare is comparable to current cutting edge solutions to data sharing among cloud service providers. By implementing MeDShare, cloud service providers and other data guardians will be able to achieve data provenance and auditing while sharing medical data with entities such as research and medical institutions with minimal risk to data privacy.
Conference Paper
Full-text available
Proof of Work (PoW) powered blockchains currently account for more than 90% of the total market capitalization of existing digital cryptocurrencies. Although the security provisions of Bitcoin have been thoroughly analysed, the security guarantees of variant (forked) PoW blockchains (which were instantiated with different parameters) have not received much attention in the literature. This opens the question whether existing security analysis of Bitcoin's PoW applies to other implementations which have been instantiated with different consensus and/or network parameters. In this paper, we introduce a novel quantitative framework to analyse the security and performance implications of various consensus and network parameters of PoW blockchains. Based on our framework, we devise optimal adversarial strategies for double-spending and selfish mining while taking into account real world constraints such as network propagation, different block sizes, block generation intervals, information propagation mechanism, and the impact of eclipse attacks. Our framework therefore allows us to capture existing PoW-based deployments as well as PoW blockchain variants that are instantiated with different parameters, and to objectively compare the tradeoffs between their performance and security provisions.
Conference Paper
Full-text available
Recently years, research in multi-robot systems has attracted increasingly attentions. One important research topic is to design programming models that can facilitate the developers to programme large-scale multi-robot systems. However, existing works fail to manage the robots to perform tasks with real-time requirements. To address this issue, we propose a new programming model called RMR (Real-time Multi-Robot). RMR is a logic programming model with real-time support. On the basis of the logic programming paradigm, RMR allows the code for multi-robot system to be written from a global perspective, rather than managing a large collection of independent robots. Moreover, RMR allows developers to set timing constraints on the behaviors of an ensemble of robots, which is not implemented by state of the art. After designing RMR, we further develop a compiler and a runtime system for distributed execution of RMR programs. To evaluate the performance of RMR, we deploy it in a simulator and a test-bed, and then demonstrate RMR based on several applications. Our results indicate that RMR greatly facilitates implementing correct collaborative multi-robot applications.
Conference Paper
The increasing popularity of blockchain-based cryptocurrencies has made scalability a primary and urgent concern. We analyze how fundamental and circumstantial bottlenecks in Bitcoin limit the ability of its current peer-to-peer overlay network to support substantially higher throughputs and lower latencies. Our results suggest that reparameterization of block size and intervals should be viewed only as a first increment toward achieving next-generation, high-load blockchain protocols, and major advances will additionally require a basic rethinking of technical approaches. We offer a structured perspective on the design space for such approaches. Within this perspective, we enumerate and briefly discuss a number of recently proposed protocol ideas and offer several new ideas and open challenges.
Chapter
Demonstrating the effectiveness of health information exchange (HIE) in reducing costs, improving outcomes, and creating more efficient care delivery systems requires evaluation. To evaluate is to measure the significance or value that HIE brings to patient and population health. This chapter describes strategies and methods for evaluating HIE, principally from the perspective of a health system or other entity engaged in the implementation and use of HIE. The chapter begins with a review of the various types of evaluation, including research. Then the chapter outlines a strategy and the methods for the evaluation of HIE. Finally, the chapter provides guidance for how to disseminate outcomes from HIE development, implementation, adoption, and use. Only by evaluating HIE and sharing lessons can we expand the evidence base for HIE in support of adoption and sustainability.