
Citation: Benjamin Hersh, Amin Mirkouei, John Sessions, Behnaz Rezaie, Yaqi You. A review and future directions on enhancing sustainability benefits across food-energy-water systems: the potential role of biochar-derived products[J]. AIMS Environmental Science, 2019, 6(5): 379-416. doi: 10.3934/environsci.2019.5.379
[1] | Jin Wang, Liping Wang, Ruiqing Wang . MFFLR-DDoS: An encrypted LR-DDoS attack detection method based on multi-granularity feature fusions in SDN. Mathematical Biosciences and Engineering, 2024, 21(3): 4187-4209. doi: 10.3934/mbe.2024185 |
[2] | Yanling An, Shaohai Hu, Shuaiqi Liu, Bing Li . BiTCAN: An emotion recognition network based on saliency in brain cognition. Mathematical Biosciences and Engineering, 2023, 20(12): 21537-21562. doi: 10.3934/mbe.2023953 |
[3] | Francesca Sapuppo, Elena Umana, Mattia Frasca, Manuela La Rosa, David Shannahoff-Khalsa, Luigi Fortuna, Maide Bucolo . Complex spatio-temporal features in meg data. Mathematical Biosciences and Engineering, 2006, 3(4): 697-716. doi: 10.3934/mbe.2006.3.697 |
[4] | Dandan Fan, Dawei Li, Fangzheng Cheng, Guanghua Fu . Effects of congestion charging and subsidy policy on vehicle flow and revenue with user heterogeneity. Mathematical Biosciences and Engineering, 2023, 20(7): 12820-12842. doi: 10.3934/mbe.2023572 |
[5] | Shuo Zhang, Yonghao Ren, Jing Wang, Bo Song, Runzhi Li, Yuming Xu . GSTCNet: Gated spatio-temporal correlation network for stroke mortality prediction. Mathematical Biosciences and Engineering, 2022, 19(10): 9966-9982. doi: 10.3934/mbe.2022465 |
[6] | Chuanqin Zheng, Qingshuang Zhuang, Shu-Juan Peng . Efficient motion capture data recovery via relationship-aggregated graph network and temporal pattern reasoning. Mathematical Biosciences and Engineering, 2023, 20(6): 11313-11327. doi: 10.3934/mbe.2023501 |
[7] | Sunghyun Yu, Yoojae Won . A survey of methods for encrypted network traffic fingerprinting. Mathematical Biosciences and Engineering, 2023, 20(2): 2183-2202. doi: 10.3934/mbe.2023101 |
[8] | H. Y. Swathi, G. Shivakumar . Audio-visual multi-modality driven hybrid feature learning model for crowd analysis and classification. Mathematical Biosciences and Engineering, 2023, 20(7): 12529-12561. doi: 10.3934/mbe.2023558 |
[9] | Sangeeta Kumari, Sidharth Menon, Abhirami K . Dynamical system of quokka population depicting Fennecaphobia by Vulpes vulpes. Mathematical Biosciences and Engineering, 2025, 22(6): 1342-1363. doi: 10.3934/mbe.2025050 |
[10] | Xihe Qiu, Xiaoyu Tan, Chenghao Wang, Shaotao Chen, Bin Du, Jingjing Huang . A long short-temory relation network for real-time prediction of patient-specific ventilator parameters. Mathematical Biosciences and Engineering, 2023, 20(8): 14756-14776. doi: 10.3934/mbe.2023660 |
Network traffic comprised of data encapsulated in network packets belonging to a variety of Internet-based applications. With the profusion of encrypted Internet applications, network administrators always have the need to monitor some specific network protocols or applications [1]. Traffic identification is fundamental to network traffic management, which helps the network operators to analyze the network traffic and identify the specific applications and protocols. In addition, this topic has received increased attention due to application prioritization for network Quality of Service (QoS) and traffic engineering [2,3], network management [4] and security protection [5].
With the increasing number of encrypted protocols and obfuscation tools, the accurate identification of network traffic has become a significant challenge in current security industry [6]. The encryption protocols such as secure socket layer/transport layer security (SSL/TLS) are used to intertwine the network traffic flows and thus lose their unique characteristics. Hence the traditional TCP/UDP port numbers based and deep packet inspection (DPI) classification approaches no longer fulfill efficient recognition due to non-standard port and encryption [7]. To overcome the failure of these traditional network traffic classification approaches, the feature-based traffic classification methods show high accuracy in the accurate identification of encrypted network traffic and further characterization with specific applications and protocols.
Virtual private network (VPN) provide secure communication for remote users to access geographically and content-based locked services, which is a locked tunnel between the remote user device and the Internet, having the capability for avoiding the user traffic from spoofing, sniffing, and censorship. The identification of the VPN-tunneled traffic is a quite challenging task due to its packet-level encryption. VPNs are generally created on the application layer, network layer, or data link layer. The standard protocols used for the mentioned layers include SSL/TLS, IPSec, and L2TP [8]. Due to the implementation of these encryption protocols, VPN traffic analysis still remains many difficult issues. Classification of tunneled traffic depends on the ultimate purpose, such as categorized them according to the specific applications, (e.g., Facebook, YouTube, Skype, GTalk, Primus Softphone, QQ). Furthermore, characterization into specific application type (e.g., Chat, Streaming, File Transfer, Voice call) as many of these mentioned applications support multiple services. Still, the VPN traffic analysis is in the early stages and needs innovative identification techniques.
In the last few years, Voice over IP (VoIP) protocols grows to be the most important services for individuals and companies for making phone calls between VoIP end-users over the Internet due to its dramatic functionality over the traditional telephone network and cost-effectiveness. VoIP is also an important communication form for online meetings and education. VPN tunnels are used to ensure the security of sensitive communication over suspicious Internet infrastructure. The VoIP communication through VPN tunnels hides its content by applying encryption to circumvent firewalls and Network Address Translation (NAT) restrictions. Therefore, a competent classification engine is required to differentiate encrypted VPN traffic according to application type. It is an essential concern for enterprises to assure the appropriate consumption of bandwidth to client applications.
The followings are the main contributions of this paper:
(1) This paper aims to provide the Flow Spatio-Temporal Features (FSTFs) for distinguishing VoIP flows from Non-VoIP ones in VPN traffic. FSTFs are mainly composed of temporal components, because the rate-adaptive techniques widely are used in VoIP resolutions and products. Meanwhile, FSTFs are more suitable for traffic inspection near the enterprise network boundary where the flow temporal signals are more distinct than those captured at the inspection nodes far away from the VoIP users.
(2) The proposed identification method validated via six well-known machine learning algorithms including decision trees (C4.5, Random Forest), K-Nearest Neighbors, bagging and boosting (via C4.5), and neural network classifier (MLP). The algorithms applied to both 10-Fold Cross-Validation and training-testing technique. According to the consideration of the practical implementation efficiency demand, only traditional machine-learning-based classifers are used and tested in this paper.
The structure of the remaining paper is divided into several sections. The related work is given in Section 2. The main design of the proposed scheme is described in Section 3. This section defined the preprocessing steps such as dataset description, flow generation, feature set selection, and generation of final datasets. Furthermore, the learning algorithms and the accessing predictive ability used for the experimental results are explained briefly. Finally, the experimental results are evaluated and discussed at the end of this section. The conclusion is forecasted, and future work is discussed in Section 4.
In this section, we first give a brief review for the methods of network traffic classification. Next, the machine learning-based network traffic classification approaches are discussed from a historical view. Last, we also survey VoIP traffic classification methods that have been proposed so far.
Existing network traffic classification methods can be categorized into three types: Port-based classification, payload-based classification, and statistic-based classification.
In the early stage of network traffic classification, the port-based technique used to be the most popular and fastest. Port-based classification uses port number in the TCP/UDP header of the IP packet to classify the network traffic without any information storage. Some port numbers reserved for privileged services were designated as 'well-known' port numbers as defined by Internet Assigned Numbers Authority (IANA) [9]. This method has several drawbacks which makes it unreliable. Some applications have registered port numbers (e.g., SSH and SMTP uses port 22 and 25 respectively), however some modern applications associated with non-standard ports, or dynamically allocated port numbers. For instance, online gaming, peer to peer (P2P) applications use random port numbers. A P2P varies the ports, and the client might use TCP port 80 to generate traffic that would appear as HTTP. Due to these allocations of random port numbers to different applications makes the detection very hard to classify network traffic. However, this type of classification is still applicable for specific applications that use their default assigned port numbers, especially for the case where accuracy is not the point of interest (e.g., network traffic monitoring).
Payload-based traffic classification generally looking for distinctive application signatures in the payload of an IP packet. Thus with knowledge of these signatures, it is easy to identify individual protocols [7]. Payload-based method inspects the entire packet payload against a set of known protocol signature to classify the packet flow. This technique has a high classification accuracy of approximately 100% for unencrypted traffic [10]. However, payload-based classification has some drawbacks. Firstly, this method fails to classify encrypted packets because the unique patterns, regular expressions, and strings are invisible in encrypted network flow. Secondly, as this method inspect each packet of the entire traffic flow to classify them, therefore the computational cost is much higher.
In order to maintain client privacy, the accessibility to the payload of the transmitted IP packets is prevented by imposing some legal restrictions such as encrypting payload information. Thus statistical classification method introduced to overcome the challenges for encrypted traffic analysis by using payload-independent statistical network flow attributes that are unique for different applications [11]. This method relies on packet-based or flow-based parameters and does not need to access the packet payload information. Frequently statistical parameters used by researchers are packet-based features (e.g., packet length, packet intervals, packet directions, etc.) and flow-based features (e.g., flow packets per second, flow bytes per second, flow duration, inter-arrival time, etc.) [12]. Therefore, the advantage of this method is that they can be applied to encrypted or tunneled flow, as they have no concern with the actual content of the packet. Machine learning techniques are employed to use these unique statistical attributes to characterize the encrypted network traffic into different applications. In the past, many works have been done on the application of machine learning in the field of network traffic classification [13]. The machine learning techniques used for statistical classification are further categorized as an unsupervised, supervised, and semi-supervised learning technique.
Unsupervised learning technique infers hidden patterns from unlabeled network traffic flows and groups them into a set of clusters with similar properties. The distance measuring approaches define the similarity between two data points (e.g., Manhattan, and Euclidean distance). The k-means and DBSCAN clustering are the most frequently used unsupervised learning techniques. These techniques do not need any labeled instances and training phase. It can accomplish the classification task for similar network traffic categories generated by distinct protocols. Some of the contributed work in the field of network traffic classification based on unsupervised learning are [14,15,16,17].
Supervised learning technique used for the classification of the network traffic flows where the class labels are known. A set of pre-labeled instances with statistical features are grouped to build a classification model during the training phase. Further, an unseen network traffic flows are predicted by the trained model developed in the training phase. The ultimate goal of supervised learning is to learn a function that maps an input features to output class. Mathematically, it can be shown as [13]
f(mi,nj)=(m1,n1),(m2,n2),…,(ml,nk). | (2.1) |
the f(mi, nj) is the dataset, where mi are the input attributes corresponding ith instances, and nj is its output class labels. Wide range of the well-known supervised learning techniques such as Naïve Bayes, decision trees, Support Vector Machine (SVM), Genetic Programming (GP), Multi-Objective Genetic Algorithm (MOGA) [18,19,20,21,22,23,24,25] are applied for the identification of P2P traffic, VoIP services and especially encrypted traffic classification with the accuracy more than 90%.
Semi-supervised learning technique falls in between supervised and unsupervised learning. The dataset contains both a small amount of labeled and a large number of unlabeled flows that are fed into a clustering algorithm to build the classification model [26]. Many authors used semi-supervised learning models for the classification of different protocols and applications [26,27,28].
This section provides a brief overview of the recent contributions in the field of traffic classification using machine learning. Most of the recent researchers used machine learning as a key methodology with the flow and packet-based features to replace traditional classification methods (port-based or payload-based) to classify encrypted traffic efficiently. Machine-learning-based classification is specialized in encrypted and tunneled traffic with acceptable computational complexity and accuracy respecting user's privacy. The statistic clustering, machine learning technique, and some other heuristic approaches have been applied to identify the network traffic and specific applications.
In the early 90's Paxson et al. [29,30] led the foundation of statistical feature based traffic classification. Later on, McGregor et al. [31] have proposed a method based on the Expectation-Maximization (EM) algorithm which utilizes the flow features for flows clustering the traces into single and multiple transactions, bulk transfer, and interactive traffic. Moore et al. [32] proposed one of the first studies considering machine learning for traffic classification. Naïve Bayes estimator and a Bayesian neural network were employed to categorize network traffic flows into different applications, respectively. Nguyen et al. [13] reviewed the emerging research from 2004 to early 2007 in the field of IP traffic classification approach based on machine learning techniques. Erman et al. [33] applied a semi-supervised technique (k-means algorithm with Euclidean distance) as an internet traffic classifier using flow statistics. The TCP flows contain few type-known examples and many type-unknown examples that were successfully classified into a variety of applications including EMAIL, FTP, HTTP, P2P, and CHAT. Este et al. [18] extensively studied SVM for network traffic classification. The proposed method has been validated with the three traces from different locations, namely: UNIBS dataset, LBNL dataset, and CAIDA dataset. A simple optimization procedure was used to derive the ideal parameters for SVM and further applied for classification of TCP bi-directional flows into multiple applications such as HTTP, HTTPS, POP3, eDonkey, SMTP, BitTorrent, MSN, Gnutella, and FTP. Li et al. [34] used a set of flow features derived from the initial few packet headers to classify network traffic flows using C4.5 and Naïve Bayes. C4.5 shown better results for the classification of network traffic flows compared with the other classifiers in all cases. Bacquet et al. [35] identified encrypted traffic via unsupervised learning techniques: DBSCAN, EM, basic k-means, Semi-supervised k-means, and MOFA. NetMate toolset [36] is used for flow generation and feature extraction. MOGA shows the best result with the detection rate (DR) of 93.5% and a false positive rate (FPR) of 0.7%. Alshammari et al. [22] tested AdaBoost, C4.5, and GP to identify traffic traces of encrypted applications: SSH and Skype. They employed public traffic traces to distinguish encrypted applications traffic and non-encrypted applications traffic. Furthermore, Jun Zhuang et al. [37] proposed a new bag-of-flow (BoF) based traffic classification scheme to agammaegate Naïve Bayes predictions of correlated flows generated by applications. Huang et al. [38] proposed an early identification method of different network application traffic based on L7 (application layer). Six machine learning algorithms were employed to classify 59 protocols from TCP/UDP flows. C4.5 tree algorithm attains better performance among the other schemes with the average overall accuracy of 92.88%.
Recently, the use of mobile messaging apps increased significantly because of many conveniences, such as sharing photos, texting each other, video chatting, booking tickets, paying bills, and shopping, etc. The widespread usage of mobile apps generates an enormous amount of mixed encrypted traffic every day. Therefore, network management and mobile companies need to classify the service usages of mobile applications (Apps). Fu et al. [39] developed a system, to classify traffic flow of famous Apps WhatsApp and WeChat into different services such as text, picture, voice note, stream video call, location sharing, and a short video. Features based on packet lengths and time delays were employed to ensemble classifiers for classification of service usages in mobile messaging Apps. Mauro et al. [40] studied Android encrypted network traffic generated by mobile apps to investigate the user's actions. They studied user's actions for different apps such as Facebook, Gmail, Twitter, Tumbler, Dropbox, Google+, and Evernote, and classified them through supervised and unsupervised machine learning algorithms. Zhen et al. [41] presented a method, named Extended Labeled Data (ELD), to identify unknown mobile traffic with the flow and byte accuracy more than 96%. Giuseppe et al. [42] used a multi-classification system to enhance the classification performance of encrypted traffic generated by mobile apps. Different classifier fusion techniques were used to classify a dataset composed of 607 real-traffic traces generated from 49 mobile apps.
Furthermore, Yu et al. [2] proposed a novel scheme to classify network video traffic using effective statistical features. Hierarchical KNN classifier employed to classify the captured internet video traffic consist of six different applications (ASD, AHD, QQ, HTTP-download, Xunlei, and Sopcast). Klenilmar et al. [43] presented an identification scheme for video streaming traffic based on Naïve Bayes algorithm, which was able to classify YouTube streaming video, Netflix streaming video, and background traffic with an average accuracy of 98.88%. Antônio et al. [44] used packet-based features to classify encrypted IoT traffic and characterize the behavior of IoT devices. The evaluation included five popular classifiers KNN, RF, DT, SVM, and MV applied to classify IoT devices from the non-IoT device. These mentioned works depicted that it was promising to use statistical features based on flows and packets for efficient classification of network traffic and applications. More studies are still required to identify encrypted traffic within VPN tunnels.
In this section, we highlighted some of the related work related to methods, techniques, and tools to analyze VoIP traffic. Traffic classification of VoIP services has been an enormous interest by several recent data analysts.
Some of the authors proposed signature-based identification schemes to identify the VoIP traffic [45,46]. Moreover, due to encryption and tunneling of VoIP services, it is impossible to do payload-based analysis. The machine learning algorithms, statistic clustering, and some other heuristic approaches have been presented to identify the VoIP traffic.
Toshiya et al. [47] presented one of the earlier studies of VoIP traffic identification. The authors considered the flow-level features were more suitable for real-time VoIP application identification. Five different applications were studied based on the packet size and inter-arrival time and classified them into voice, video and file-sharing applications.
Skype is one of the most popular peer-to-peer (P2P) VoIP service provider, developed in 2002. Skype comprises a robust encryption mechanism, making it difficult to be analyzed which attracted the attention of the research community. Some of the work related to Skype traffic analysis are listed here [48,49,50,51]. Bonfiglio et al. [48] tested two approaches for the identification of Skype voice calls. The first approach used Pearson's Chi-Square testing to explore the statistical characteristics of the traffic payload, which analyzed the randomness of the message content generated by cipher during encryption. The second approach applied the Naïve Bayesian classifier to characterize Skype traffic. The combination of these two approaches for detecting Skype voice traffic showed negligible FPR and FNR. Alshammari et al. [49] employed flow-based features to classify VoIP flows in the particular Skype and GTalk traffic. Three popular machine learning approaches, namely, C4.5, AdaBoost, and GP, were evaluated to accomplish the classification task. C4.5 provided an optimal result with the DR of 99% approximately and less than 1% FPR. Ibrahim et al. [50] studied ten machine learning algorithms to classify captured traffic successfully into Skype and OnlineTV traffic. Davide et al. [51] proposed a joint statistical and signature-based approach to classify Skype traffic and distinguish file transfer, voice, and video calls.
Moreover, Khan et al. [52] proposed a method to recognize a perpetrator from the encrypted VoIP conversation. The variable packet-length-based features are taken into account to identify the speaker from an encrypted conversation in case of the variable bit speech encoding mechanism. Taner et al. [53] proposed one of the earlier studies in the classification of VoIP traffic in the VPN tunnel. The authors proposed a simple technique to identify VoIP traffic in the IPSec tunnel for improving VoIP QoS. Li et al. [54] proposed a method based on the host behavior estimation and the flow statistical analysis to identify the VoIP traffic at the transport layer. The port numbers and IP addresses were tested as the host behavior while the inter-packet arrival time and entropy value (to model the packet size) were selected as the flow features to make identification. Express Talk, Tom-Skype, QQ voice and video, and windows live messenger (MSN) voice and video were selected as typical VoIP applications, and web-browser, multimedia applications, online gaming, and file-sharing applications were considered as Non-VoIP applications to accomplish the identification evaluation. Alshammari et al. [55] tested the C4.5, GP and Adaboost classifiers for encrypted VoIP traffic identification. They used the NetMate toolset [36] to generate flow from captured traffic traces and extract flow-based attributes. Two different experiments were studied. In the first experiment, the captured traffic traces were classified into Skype and non-Skype traffic. In the second experiment, the traffic traces were separately labeled as VoIP (GTalk, Skype, and Primus softphone) applications and Non-VoIP applications. Qin et al. [56] developed a traffic identification system based on the packet size distribution of the first few packets to divide the network traffic into VoIP and P2P applications. And further, give the specific application type, such as Skype. Recently, Mazhar et al. [57] proposed a method to detect the voice call flows from encrypted and tunneled traffic based on statistical features. The proposed scheme is with the real-time VoIP calls detection ability using 6 seconds captured traffic after the call initiation. The proposed method achieved a TPR of 97.54% and FPR of 0.00015%.
Most of the researchers focused on the classification of VoIP applications (Skype, YouTube, Facebook, GTalk, etc.) and further into specific application types (Voice calls, File Transfer, P2P, Streaming, Chat, etc.). The enormous growth of encrypted and tunneled VoIP traffic opens the new research area for the internet community. There are still quite a few studies on the VPN-tunneled VoIP traffic analysis. To improve the quality, block the forbidden traffic, to prevent the illegal use of network resources and characterize bandwidth used by VPN tunnels for different applications, we need more accurate solutions for efficient classification, identification, and characterization of VPN-tunneled VoIP traffic.
The proposed scheme aims to identify VoIP traffic from Non-VoIP traffic in the VPN tunnel accurately. The general framework of the proposed scheme is given in Figure 1. The key idea we leveraged is divided into five discrete phases. In the first phase, we extract the captured VPN traffic (.pcap files) from the mixed traffic traces containing both VPN and Non-VPN traffic traces. In the second phase, traffic flows are generated based on three different flow latency periods (FLPs). In the third phase, data cleaning, feature selection, and labeling of the flows are described. In the fourth phase, three exclusive datasets are generated based on 15, 60 and 120 sec FLPs with unique flow instances. In the last phase, different classifiers are employed to distinguish VPN-VoIP traffic from VPN-Non_VoIP traffic. The refined datasets obtained after several preprocessing steps are presented in Section 3.1. Section 3.2 describes the machine learning algorithms employed for classification in the proposed work. In Section 3.3, we explained the performance assessment metrics to evaluate the efficiency of the scheme. Furthermore, Section 3.4 summarized the experimental results.
The VPN-Non VPN dataset published by CIC, the University of New Brunswick, is employed to validate the aforementioned problem [58]. The extensive dataset consists of encrypted and VPN raw traffic traces generated by VoIP, Email, File Transfer, P2P, Streaming, Chat, and Web browsing (seven traffic traces for regular encrypted traffic and seven traffic traces for VPN traffic) in pcap format with particular application labels. However, the valid dataset has the actual number of 12 application classes because some files are related to two different classes, which can be considered in a single category at the same time. Wireshark and tcpdump are used to capture the traffic generated by most popular applications which yield diversity in service types. We choose only the VPN tunneled traffic traces files (about 2.3 GB) for our interest to accomplish the classification task. Furthermore, we divided all the raw traffic types into two categories, VPN-VoIP and VPN-Non_VoIP. The detailed content type of these two classes is listed in Table 1. The voice calls (for 1-hour duration) traffic flow generated by Skype, VoIPbuster, Hangouts, and Facebook are grouped into VPN-VoIP traffic. On the other hand, traffic traces collected from P2P, Chat, Email, Streaming, and File Transfer are considered to be VPN-Non_VoIP traffic.
Traffic Labels | Content Type | |
VPN-VoIP | Skype, VoIPbuster, Hangouts and Facebook voice calls for the duration of 1hour | |
VPN-Non_VoIP | P2P | μTorrent and Bittorent |
Chat | AIM, Skype, Facebook, Hangouts, ICQ | |
SMPTS, IMAPS, POP3S | ||
Streaming | Netflix, Vimeo, YouTube and Spotify | |
File Transfer | FTPS, SFTP and Skype |
For computing the flows and features, we use CICFlowmeter, an open-source java-based application [59] as the network flow generator. The input of CICFlowmeter is raw traffic traces captured in pcap format, and it will generate bidirectional flows based on forward (src to dst) and backward (dst to src) directions. The output of CICFlowmeter is the Comma Separated Values (CSV) file, where each flow is defined by Flow ID, consists of five parameters, i.e., Src IP, Dst IP, Src port, Dst port, and Protocol (TCP or UDP) with 76 statistical traffic features. The duration of the FLP can be controlled and adjusted by an individual in the source code. All the captured VPN raw traffic traces mentioned in Table 1 are agammaegated into traffic flows by using the java source code of CICFlowmeter with 15, 60 and 120 sec FLP to get the corresponding CSV files for both classes. In order to classify the traffic flows according to Table 1, we labeled the corresponding obtained instances according to the source applications: Streaming, P2P, Chat, Email, VoIP (Skype, Voipbuster, hangouts, and Facebook voice calls) and File Transfer. All the flows obtained from P2P, Chat, Email, Streaming, and File Transfer are labeled as VPN-Non_VoIP class, and the flows obtained from the Skype, Voipbuster, hangouts, and Facebook voice calls traces are labeled as VPN-VoIP class. In the end, to get a refined data set, we removed all the duplicate flows.
According to the 76 statistical features, we can make the Flow Spatio-Temopral Features selection. Firstly, we remove all the features with null values because features with zero values do not affect the classification task. Secondly, some distinct features are considered as ineffective because of zero values for more than 70% of flows. We only considered that statistical feature which has dispersed and majority non-zero values. A minimum number of features are selected to reduce computational time and complexity. In the past, most work [22,49,55] used NetMate toolset [36], which generates flows with the maximum number of 22 distinct flow features. CICFlowmeter produces a wide variety of features, and also the duration of the flow is changeable, which shows better performance in contrast to NetMate toolset. The selected effective feature set for classification is listed in Table 2.
Abbreviation | Description | |
1 | feduration | Duration of the flow in Microsecond |
2 | max_fpktl | Maximum size of packet in forward direction |
3 | std_fpktl | Standard deviation size of packet in forward direction |
4 | max_bpktl | Maximum size of packet in backward direction |
5 | std_bpktl | Standard deviation size of packet in backward direction |
6 | tot_fhlen | Total bytes used for headers in the forward direction |
7 | tot_bhlen | Total bytes used for headers in the backward direction |
8 | init_win_bytes | The total number of bytes sent in initial window in the backward |
_backward | direction | |
9 | fpkts_persec | Number of flow packets per second |
10 | min_flowiat | Minimum inter-arrival time of packet |
11 | tot_fiat | Total time between two packets sent in the forward direction |
12 | tot_biat | Total time between two packets sent in the backward direction |
13 | min_biat | Minimum time between two packets sent in the backward direction |
14 | max_active | Maximum time a flow was active before becoming idle |
15 | min_active | Minimum time a flow was active before becoming idle |
16 | max_idle | Maximum time a flow was idle before becoming active |
17 | min_idle | Minimum time a flow was idle before becoming active |
After the flow generation, labeling, and feature extraction, we obtain a reasonable amount of instances for each class listed in Table 3. We get three distinct data sets (D1, D2 and D3), as shown in Figure 1 based on the duration of the flow. All the datasets consist of the entry with 17-dimension FSTFs and the class label (i.e., VPN-VoIP and VPN-Non_VoIP). Furthermore, the experiments are based on two different scenarios. In Scenario 1, Cross-Validation with 10-Folds is implemented to test all the three datasets. In Scenario 2, the datasets are divided into training (Xtrain) and testing (Xtest) subsets. The Xtrain contains 70% of random instances, while the Xtest contains the remaining 30% instances. The Xtest comprises of unseen data which shows the generalization of the algorithm. The details of Xtrain and Xtest datasets are given in Table 3 according to different FLP.
Datasets | VPN-VoIP and VPN-Non_VoIP instances | ||
(based on FLP) | (Xtrain) | (Xtest) | (XTotal) |
D1 | 16124 | 7212 | 23336 |
D2 | 12102 | 5326 | 17428 |
D3 | 10799 | 4905 | 15704 |
According to the consideration of the practical implementation efficiency demand, instead of deep learning algorithms, only traditional machine-learning-based classifers are used and tested in this paper. Several typical machine learning algorithms are considered, they are decision trees (C4.5, RF), KNN, Bagging (via C4.5), Boosting (via C4.5), and MLP. This section provides a brief overview of them. The literature indicates the superior performance of these machine learning algorithms in the field of network traffic classification [22,34,38,49,55].
Decision trees are a famous supervised learning algorithm. It follows the divide and conquer strategy to classify input data. The decision tree structure is composed of decision nodes and terminal leaves, which represent the attributes and the final class or label, respectively. We used two popular decision trees based algorithms C4.5 and RF in the proposed work.
C4.5: C4.5 is a well-known classification tree algorithm that can perform both regression and classification tasks, developed by Ross Quinlan [60]. It is a successor to Iterative Dichotomiser (ID3) invented by Ross Quinlan. C4.5 uses training instances to construct a decision tree with the concept of information entropy evaluation function [61].
E(X)=−N∑m=1{[freq(Ym,X)/|X|]log2[freq(Ym,X)/|X|]} | (3.1) |
where X represent the training set. ID3 uses gains while the C4.5 relies on gain ratios to avoid overfitting as given below [61].
GainRatio(A)=Gain(A)/Splitinfo(A) | (3.2) |
where A represent the test attribute and the Splitinfo(A) which in turn is defined as[61]
Splitinfo(A)=−12L∑m=1[|Xm||X|log2|Xm||X|+(1−|Xm||X|)log2(1−|Xm||X|)] | (3.3) |
The attribute with large Gain Ratio is considered as the root of the classification tree.
Random Forest (RF): Random Forest (RF) is another supervised classification algorithm based on ensembles of decision trees proposed by Breiman [62]. RF makes the final prediction on the majority-voting concept visualized in Figure 2.
KNN is a supervised machine learning technique used for regression and classification problems. It classifies testing instance based on the k closest samples (where k = positive integer) from the training set. And a testing instance is categorized by calculating the distance to the nearest training case. The distance metric defines the similarity between two data points. Euclidean, Manhattan, and Minkowski distances are the used cases. Mostly used distance metrics are listed in Table 4. Mainly, two factors affect the performance of the KNN classifier, value of k, and the selection of distance metric.
Euclidean distance | D(y,y′)=√(y1−y′1)2+(y2−y′2)2+...+(yn−y′n)2 |
Manhattan distance | D(y,y′)=K∑m=1|ym−ym′| |
Minkowski distance | D(y,y′)=b√(K∑m=1|ym−ym′|b) |
In the proposed scheme, the k value is 1, and the Manhattan distance metric is selected for better performance.
Boosting is an ensemble machine learning technique proposed by Freund to train multiple weak models sequentially and estimate a set of weights to produce a strong estimator [63]. Adaptive Boosting (AdaBoost) in one of the most influential boosting algorithm. It uses a set of base classifiers during the training process and evaluates the weights for the instances for the final prediction. The weights of the correctly classified instances remain the same and the weights of the incorrectly classified instances increases. The normalized weights of all the instances are further employed for the next classifier. The final classification is agammaegated by weighted based classifiers.
Bagging (Bootstrap agammaegating) is proposed by Breiman [64]. In contrast to boosting, multiple base classifiers are trained in parallel. Each classifier is trained through tiny alteration in the training dataset (a.k.a bootstrap sampling). The majority voting over the class labels decides the final prediction. In this study, the ensemble classifier consists of C4.5 as a base classifier.
MLP is a kind of feedforward artificial neural network, consists of a minimum of three layers (input, hidden, and output layer), as shown in Figure 3. Mathematically, it can be express as
F(n)=Y(x(2)+W(2)(s(x(1)+W(1)n))) | (3.4) |
where x(1), x(2) are bias vectors; W(1), W(2) are weight matrices; Y and s are the activation functions. Each neuron work as a summation unit and an activation function.
The number of hidden layers increases the complexity of the model. We selected nine hidden layers for the used case.
The four basic parameters which are employed to compute the performance assessment metrics are [43]:
● True Positive (TP): When the predicted and actual instances are YES.
● False Positive (FP): When the predicted instance is YES and the actual instance is NO.
● True Negative (TN): When the predicted and actual instance are NO.
● False Negative (FN): When the predicted instance is NO and the actual instance is YES.
The number of performance assessment metrics are utilized to assess the predictive power of the machine learning algorithms. Four performance assessment metrics. i.e., Precision (Pr) or Positive Predictive Value (PPV), Sensitivity (Sen) or Recall (Rc), Accuracy (Acc), and F-measure (F-m) are utilized to assess the performance of the proposed algorithms, which are defined as follows [43].
Pr represents the classifier correctness and also known as Positive Predictive Value (PPV). It is the number of True Positive results divided by the total number of positive class results predicted by the classifier, defined as:
Pr = TP(TP+FP) | (3.5) |
Sen reflects the completeness rate of the classifier. It is the ratio of True Positive results to the True Positive and False Negatives results, as defined below:
Sen=TP(TP+FN) | (3.6) |
Acc is the ratio of the sum of all correctly classified flows to the sum of all the traffic flows tested.
Acc=∑Ti#Samples×100% | (3.7) |
where Ti denotes all correctly classified flows, and #Samples denotes the sum of all the traffic flows exists in the dataset.
F−m is the harmonic mean of Sen and Pr, which conveys the balance weight between Sen and the Pr. The value near to 1 is the best result, F−m can be calculated as:
F−m=2×Pr×RcPr+Rc0≤F−m≤1 | (3.8) |
This work runs on the computer with a 2.4 GHz Intel Core-i3 CPU, a 6.0 GB of Random Access Memory (RAM), and the operating system Windows 7. All the pcap traces are processed through CICFlowmeter, which is written in java language to get the CSV files. The FLP adjusted accordingly in the script. WEKA platform is used for further classification, and evaluation indexes, which are discussed in Section 3.3 are employed to examine the predictive ability of the proposed scheme.
In this section, a series of comparative experiments are performed to comprehensively highlight the effectiveness of the proposed FSTF identification method. The proposed identification process of encrypted VPN traffic comprises of two major stages. One is the flow generation followed by dataset generation based on FSTF set with different FLPs. And the other is to perform the experiments according to the developed models to characterize the VPN traffic into VoIP and Non-VoIP classes. We evaluated the performance of each of the six learning algorithms (decision trees (C4.5, RF), KNN, Bagging (via C4.5), Boosting (via C4.5), and MLP) discussed in Section 3.2 for identification of VoIP services through VPN tunnel to validate the effectiveness of the FSTF. In order to evaluate the performance of the developed system, two types of experiments have been implemented: a) K-Fold Cross-Validation, and b) training-testing scenario. The FSTF set listed in Table 2 outperformed to classify the three datasets based on FLP with regard to the four evaluation metrics Pr, Sen, F−m, and Acc described earlier.
K-Fold Cross-Validation is a resampling procedure that includes the random shuffling of the dataset, followed by splitting the dataset into k number of groups. In each run, one group is holded for testing and the remaining groups is used for training the model. The process continues until the last group used as the testing data, as described in Figure 4. During the first phase of experiments, the K-Fold Cross-Validation is employed to classify the VPN traffic into two categories, i.e., VoIP and Non-VoIP traffic flows. In our experiments, we selected k=10 for all the six learning algorithms.
Figure 5 lists the accuracy level achieved by all machine learning algorithms based on FSTF set for three different datasets generated with 15, 60 and 120 sec FLP. It is evident that we obtained the best results for 15 sec of FLP throughout the six algorithms. All the classification algorithms show an accuracy level of more than 95.72% except MLP. In contrast to these results, MLP shows average performance with the highest accuracy of 88.53% and a minimum accuracy of 74.99%. The results look very promising for better classification of traffic through a VPN tunnel. All the machine learning algorithms achieved better accuracy. During experiments, the decision trees, KNN, Boosting and Bagging have almost similar performance results. The VoIP traffic was detected with the highest accuracy of 98.27% obtained by applying boosting (Adaboost via C4.5) with 15 sec FLP. The experimental results depict that the features listed in Table 2, along with the shorter FLP (15 sec in this case) are the best options to identify and classify the VoIP traffic in a mixed VPN traffic flows.
In this section of the paper, we illustrate the identification method for VoIP traffic traces using trained models and later on tested their performance with unseen testing traffic flows using the same FSTF set. The experimental setup for this scenario is portrayed in Figure 6, where all the six models are initially trained with the Xtrain traffic flows and then tested with the unseen Xtest traffic flows with FSTF set given in Table 2. Details of the training and testing datasets (D1, D2 and D3) are given in Table 3. The generalization of the system is examined by four performance assessment metrics explained in Section 3.3.
Figure 7 shows classification accuracy obtained by testing a trained model with unseen VPN traffic flows. From Figure 7, it can be seen that all the six algorithms have consistent performance and show the accuracy level of more than 80.96%. Among these trained classifier models, the boosting model shows attractive results with the highest accuracy of 96.1% for FLP of 15 sec. Repeatedly, the shorter FLP enhanced the accuracy level and achieved better identification results. As a result, the boosting technique offered significantly better performance as an identification engine in both scenarios for all FLP values.
The Pr, Sen, and F−m have been calculated to assess the results further. Except for MLP, all the classifier models presents attractive identification results. Here we will only explain the performance metrics obtained from a trained model based on the boosting technique. Figure 8 illustrates the Pr of the VPN traffic traces, which are classified into VoIP and Non-VoIP traffic flows. The optimal Pr values for Non-VoIP and VoIP traffic flows are 0.954 and 0.97 respectively for 15 sec of FLP. On the other hand, when we increase the FLP to 120 sec, the Pr decreased to 0.936 and 0.939, respectively. Figure 9 exhibits similar behavior, as the Sen of boosting via C4.5 classifier for Non-VoIP and VoIP traffic flows have been decreased from 0.978, 0.937 to 0.95, and 0.923, respectively. Finally, the F−m count shown in Figure 10 for Non-VoIP and VoIP traffic flows are 0.966 and 0.954 for 15 sec FLP, while a slight decrease detected for 120 sec, which is 0.943 and 0.931 respectively.
Moreover, all the learning algorithms except MLP show prominent classification results for distinguishing the VPN traffic, the Pr is greater than 0.873, the Sen is greater than 0.891, and the F−m is greater than 0.884. The qualitative analysis proved that FSTF set along with shorter FLP are the best choice for the classification and identification of VPN traffic flows. Furthermore, VoIP traffic detection is easily and efficiently achievable.
During the experiments, MLP shows the lowest and inconsistent performance in all three experiments. The corresponding highest Pr, Sen, and F−m for Non-VoIP and VoIP traffic flows are 0.872, 0.779, 0.823 and 0.745, 0.85, and 0.794 with 15 sec FLP respectively. With the longer FLP i.e. 120 sec, the corresponding Pr, Sen, and F−m are 0.81, 0.88, and 0.841 for Non-VoIP, while 0.84, 0.746, and 0.79 in the case of VoIP. In this case, Non-VoIP Pr decreases while Sen and F−m shows increment with the increase of FLP from 15 to 120 sec. On the contrary, VoIP traffic flows show an increase in Pr, while decreasing results for Sen and F−m.
Additionally, the MLP shows clear improvement with the increase in the training time. Figure 11 clearly illustrates that increasing the training time of the model shows enhanced results in both scenarios. Here, we selected the number of epochs, t1 = 500 and t2 = 1000. Almost, all the FLP based datasets show improvement in the Acc. The number of epochs can be selected according to the requirement of the classification model to do a trade-off between computational time and accuracy.
The work of this paper is to distinguish the VPN tunneled traffic into VoIP and Non-VoIP traffic. Overall, our approach can identify VoIP traffic generated from different applications among diverse network traffic in the VPN tunnel. Two aspects of experiments represent the efficiency of the proposed identification method. The one is that shorter FLP show consistently better results. The other is the FSTF employed to machine learning algorithms for modeling and identification of network traffic instead of DPI or port numbers. The selected feature set exhibits accurate identification performance for a wide range of encrypted traffic flows. Six learning algorithms were employed during the identification process. During the series of experiments, among the tested six learning algorithms, the boosting technique consistently performed the best on all the given datasets using the FSTF. The boosting based classifier attains the acc of (98.27, 97.69 and 98.1%) by 10-Fold Cross-Validation for 15, 60 and 120 sec FLP respectively. Furthermore, the trained model achieved acc of (96.1, 93.82 and 93.76%) for 15, 60 and 120 sec FLP, respectively, with testing on unseen testing data. Training time in all the experiments took several seconds. The proposed method generates three different datasets consists of enough samples in the training data sets from which the mapping between traffic flows and traffic type may be learned. The shorter FLP enhanced the accuracy level and achieved better identification results. As a result, the boosting technique (via C4.5) exhibits significantly better performance as an identification model in both scenarios for all FLP values.
We conclude that our scheme can successfully identify VoIP traffic flows from the mixed network traffic flows generated by various applications based on the FSTF. Future work includes further analysis of application-based VPN-VoIP traffic with further less FLP. The more accurate approaches based on deep learning with less computation complexity are also future targets.
This work was supported in part by the National Natural Science Foundation of China under Grant no. U1836104, 61702235 and 61921004, and partly supported by the Fundamental Research Funds for the Central Universities under Grant no. 30918012204.
The authors declare that they have no conflicts of interests.
[1] | DOE (2016) Strategic Plan for a Thriving and Sustainable Bioeconomy. 56. |
[2] |
Lin L, Xu F, Ge X, et al. (2018) Improving the sustainability of organic waste management practices in the food-energy-water nexus: A comparative review of anaerobic digestion and composting. Renew Sustain Energy Rev 89: 151-167. doi: 10.1016/j.rser.2018.03.025
![]() |
[3] | Daher B, Mohtar R (2012) Water, energy, and food: The Ultimate Nexus, Encyclopedia of Agricultural, Food, and Biological Engineering, Second Edition. |
[4] |
Albrecht TR, Crootof A, Scott CA (2018) The Water-Energy-Food Nexus: A systematic review of methods for nexus assessment. Environ Res Lett 13: 043002. doi: 10.1088/1748-9326/aaa9c6
![]() |
[5] |
Popp J, Lakner Z, Harangi-Rákos M, et al. (2014) The effect of bioenergy expansion: Food, energy, and environment. Renew Sustain Energy Rev 32: 559-578. doi: 10.1016/j.rser.2014.01.056
![]() |
[6] | Gomo FF, Macleod C, Rowan J, et al. (2018) Supporting better decisions across the nexus of water, energy and food through earth observation data: \hack\break case of the Zambezi basin. Proc Int Assoc Hydrol Sci 376: 15-23. |
[7] | Belmonte BA, Benjamin MFD, Tan RR (2017) Biochar systems in the water-energy-food nexus: the emerging role of process systems engineering. Biotechnol Bioprocess Eng Process Syst Eng 18: 32-37. |
[8] | Hansen S, Mirkouei A (2019) Bio-Oil Upgrading Via Micro-Emulsification And Ultrasound Treatment: Examples For Analysis And Discussion, ASME 2019 International Design Engineering Technical Conferences & Computers and Information in Engineering Conference. |
[9] | Hansen S, Mirkouei A, Diaz LA (2020) A Comprehensive State-of-Technology Review for Upgrading Bio-oil to Renewable or Blended Hydrocarbon Fuels. Renew Sustain Energy Rev in press. |
[10] |
Roy P, Dias G (2017) Prospects for pyrolysis technologies in the bioenergy sector: A review. Renew Sustain Energy Rev 77: 59-69. doi: 10.1016/j.rser.2017.03.136
![]() |
[11] | Mirkouei A, Haapala KR (2015) A network model to optimize upstream and midstream biomass-to-bioenergy supply chain costs, ASME 2015 International Manufacturing Science and Engineering Conference (MSEC), MSEC2015-9355. |
[12] | Mirkouei A (2016) Techno-Economic Optimization and Environmental Impact Analysis for a Mixed-Mode Upstream and Midstream Forest Biomass to Bio-Products Supply Chain. |
[13] |
Cha JS, Park SH, Jung S-C, et al. (2016) Production and utilization of biochar: A review. J Ind Eng Chem 40: 1-15. doi: 10.1016/j.jiec.2016.06.002
![]() |
[14] |
Krajačić G, Vujanović M, Duić N, et al. (2018) Integrated approach for sustainable development of energy, water and environment systems. Energy Convers Manag 159: 398-412. doi: 10.1016/j.enconman.2017.12.016
![]() |
[15] | Hugh McLaughlin (2016) An Overview of the current Biochar and Activated Carbon Markets : Biofuels Digest, 2016. Available from: https://www.biofuelsdigest.com/bdigest/2016/10/11/an-overview-of-the-current-biochar-and-activated-carbon-markets/. |
[16] | Mirkouei A (2019) Cyber-Physical Real-time Monitoring and Control for Biomass-based Energy Production, Emerging Frontiers in Industrial and Systems Engineering: Growing Research and Practice, Taylor & Francis. |
[17] | Hansen S, Mirkouei A (2018) Past Infrastructures and Future Machine Intelligence (MI) for Biofuel Production: A Review and MI-Based Framework. ASME 2018 Int Des Eng Tech Conf Comput Inf Eng Conf V004T05A022. |
[18] | Mirkouei A, Kardel K (2017) Enhance Sustainability Benefits Through Scaling-up Bioenergy Production from Terrestrial and Algae Feedstocks, Proceedings of the 2017 ASME IDETC/CIE: 22nd Design for Manufacturing and the Life Cycle Conference. |
[19] | Mirkouei A, Haapala KR, Murthy GS, et al. (2017) Evolutionary Optimization of Bioenergy Supply Chain Cost with Uncertain Forest Biomass Quality and Availability, Proceedings of the 2016 Industrial and Systems Engineering Research Conference H. Yang, Z. Kong, and MD Sarder, eds., May 21-24, Anaheim, California, USA. |
[20] | Mirkouei A, Haapala K (2014) Integration of machine learning and mathematical programming methods into the biomass feedstock supplier selection process, Proc. 24th Int. Conf. Flex. Autom. Intell. Manuf. FAIM May, 20-23. |
[21] |
Biggs EM, Bruce E, Boruff B, et al. (2015) Sustainable development and the water-energy-food nexus: A perspective on livelihoods. Environ Sci Policy 54: 389-397. doi: 10.1016/j.envsci.2015.08.002
![]() |
[22] | Wallington K, Cai X (2017) The Food-Energy-Water Nexus: A Framework to Address Sustainable Development in the Tropics. Trop Conserv Sci 10: 1940082917720665. |
[23] | GVR (2017) Biochar Market Size, 2017. Available from: https://www.marketwatch.com/press-release/biochar-market-size-worth-31461-million-by-2025-cagr-132-grand-view-research-inc-2018-05-30. |
[24] |
Jeffery S, Verheijen FGA, van der Velde M, et al. (2011) A quantitative review of the effects of biochar application to soils on crop productivity using meta-analysis. Agric Ecosyst Environ 144: 175-187. doi: 10.1016/j.agee.2011.08.015
![]() |
[25] |
Atkinson CJ, Fitzgerald JD, Hipps NA (2010) Potential mechanisms for achieving agricultural benefits from biochar application to temperate soils: a review. Plant Soil 337: 1-18. doi: 10.1007/s11104-010-0464-5
![]() |
[26] |
Ahmad M, Rajapaksha AU, Lim JE, et al. (2014) Biochar as a sorbent for contaminant management in soil and water: A review. Chemosphere 99: 19-33. doi: 10.1016/j.chemosphere.2013.10.071
![]() |
[27] |
Liu W-J, Jiang H, Yu H-Q (2015) Development of Biochar-Based Functional Materials: Toward a Sustainable Platform Carbon Material. Chem Rev 115: 12251-12285. doi: 10.1021/acs.chemrev.5b00195
![]() |
[28] |
Das O, Sarmah AK, Bhattacharyya D (2015) A sustainable and resilient approach through biochar addition in wood polymer composites. Sci Total Environ 512-513: 326-336. doi: 10.1016/j.scitotenv.2015.01.063
![]() |
[29] |
Carpenter AW, de Lannoy C-F, Wiesner MR (2015) Cellulose Nanomaterials in Water Treatment Technologies. Environ Sci Technol 49: 5277-5287. doi: 10.1021/es506351r
![]() |
[30] |
Mousa E, Wang C, Riesbeck J, et al. (2016) Biomass applications in iron and steel industry: An overview of challenges and opportunities. Renew Sustain Energy Rev 65: 1247-1266. doi: 10.1016/j.rser.2016.07.061
![]() |
[31] | Nemati M, Simard F, Fortin J-P, et al. (2015) Potential Use of Biochar in Growing Media. Soil Sci Soc Am J 14. |
[32] | Berry M, Seppalaomem O (2014) OSUAA_Biochar_MarketStudy_6_6_14.pdf. |
[33] | Delaney M (2015) Northwest Biochar commercialization strategy paper, U.S. Forest Service and the Oregon Department of Forestry. |
[34] | Ferroukhi R, Nagpal D, Alvaro Lopez-Peña, et al. (2015) Renewable Energy in the Water, Energy and Food Nexus, IREA. |
[35] | Jha P, Biswas AK, Lakaria BL, et al. (2010) Biochar in agriculture - prospects and related implications. Curr Sci 99: 1218-1225. |
[36] | Jeffery S, Bezemer TM, Cornelissen G, et al. (2013) The way forward in biochar research: targeting trade-offs between the potential wins. GCB Bioenergy 7: 1-13. |
[37] |
Gwenzi W, Musarurwa T, Nyamugafata P, et al. (2014) Adsorption of Zn2+ and Ni2+ in a binary aqueous solution by biosorbents derived from sawdust and water hyacinth (Eichhornia crassipes). Water Sci Technol 70: 1419. doi: 10.2166/wst.2014.391
![]() |
[38] |
Yoo G, Kim H, Chen J, et al. (2014) Effects of Biochar Addition on Nitrogen Leaching and Soil Structure following Fertilizer Application to Rice Paddy Soil. Soil Sci Soc Am J 78: 852-860. doi: 10.2136/sssaj2013.05.0160
![]() |
[39] |
Bucheli TD, Bachmann HJ, Blum F, et al. (2014) On the heterogeneity of biochar and consequences for its representative sampling. J Anal Appl Pyrolysis 107: 25-30. doi: 10.1016/j.jaap.2014.01.020
![]() |
[40] | Gerlach A, Schmidt HP (2014) The use of biochar in cattle farming. Biochar J 2014. |
[41] |
Xu R, Ferrante L, Hall K, et al. (2011) Thermal self-sustainability of biochar production by pyrolysis. J Anal Appl Pyrolysis 91: 55-66. doi: 10.1016/j.jaap.2011.01.001
![]() |
[42] |
Han J, Elgowainy A, Dunn JB, et al. (2013) Life cycle analysis of fuel production from fast pyrolysis of biomass. Bioresour Technol 133: 421-428. doi: 10.1016/j.biortech.2013.01.141
![]() |
[43] |
Agegnehu G, Srivastava AK, Bird MI (2017) The role of biochar and biochar-compost in improving soil quality and crop performance: A review. Appl Soil Ecol 119: 156-170. doi: 10.1016/j.apsoil.2017.06.008
![]() |
[44] | Biederman LA, Harpole WS (2012) Biochar and its effects on plant productivity and nutrient cycling: a meta-analysis. GCB Bioenergy 5: 202-214. |
[45] |
Agegnehu G, Nelson PN, Bird MI (2016) Crop yield, plant nutrient uptake and soil physicochemical properties under organic soil amendments and nitrogen fertilization on Nitisols. Soil Tillage Res 160: 1-13. doi: 10.1016/j.still.2016.02.003
![]() |
[46] |
Aldridge CA, Baker BH, Omer AR (2019) Investigation of short-term effects of winter cover crops on compaction and total soil carbon in a long-term no-till agricultural system. J Soil Water Conserv 74: 77-84. doi: 10.2489/jswc.74.1.77
![]() |
[47] |
Boateng AA, Mullen CA, Goldberg NM, et al. (2010) Sustainable production of bioenergy and biochar from the straw of high-biomass soybean lines via fast pyrolysis. Environ Prog Sustain Energy 29: 175-183. doi: 10.1002/ep.10446
![]() |
[48] |
Ali I (2014) Water Treatment by Adsorption Columns: Evaluation at Ground Level. Sep Purif Rev 43: 175-205. doi: 10.1080/15422119.2012.748671
![]() |
[49] |
Barber ST, Yin J, Draper K, et al. (2018) Closing nutrient cycles with biochar- from filtration to fertilizer. J Clean Prod 197: 1597-1606. doi: 10.1016/j.jclepro.2018.06.136
![]() |
[50] |
Shackley S, Hammond J, Gaunt J, et al. (2011) The feasibility and costs of biochar deployment in the UK. Carbon Manag 2: 335-356. doi: 10.4155/cmt.11.22
![]() |
[51] |
Sessions J, Smith D, Trippe KM, et al. (2019) Can biochar link forest restoration with commercial agriculture? Biomass Bioenergy 123: 175-185. doi: 10.1016/j.biombioe.2019.02.015
![]() |
[52] |
Rosas JG, Gómez N, Cara J, et al. (2015) Assessment of sustainable biochar production for carbon abatement from vineyard residues. J Anal Appl Pyrolysis 113: 239-247. doi: 10.1016/j.jaap.2015.01.011
![]() |
[53] | Shackley S, Carter S, Knowles T, et al. (2012) Sustainable gasification-biochar systems? A case-study of rice-husk gasification in Cambodia, Part I: Context, chemical properties, environmental and health and safety issues. Energy Policy 42: 49-58. |
[54] |
Qin K, Jensen PA, Lin W, et al. (2012) Biomass gasification behavior in an entrained flow reactor: gas product distribution and soot formation. Energy Fuels 26: 5992-6002. doi: 10.1021/ef300960x
![]() |
[55] |
Gwenzi W, Chaukura N, Noubactep C, et al. (2017) Biochar-based water treatment systems as a potential low-cost and sustainable technology for clean water provision. J Environ Manage 197: 732-749. doi: 10.1016/j.jenvman.2017.03.087
![]() |
[56] |
Zygourakis K (2017) Biochar soil amendments for increased crop yields: How to design a "designer" biochar. AIChE J 63: 5425-5437. doi: 10.1002/aic.15870
![]() |
[57] |
Suter GW (2013) Review papers are important and worth writing. Environ Toxicol Chem 32: 1929-1930. doi: 10.1002/etc.2316
![]() |
[58] |
Mirkouei A, Haapala KR, Sessions J, et al. (2017) A review and future directions in techno-economic modeling and optimization of upstream forest biomass to bio-oil supply chains. Renew Sustain Energy Rev 67: 15-35. doi: 10.1016/j.rser.2016.08.053
![]() |
[59] | Taticchi P, Garengo P, Nudurupati SS, et al. (2014) A review of decision-support tools and performance measurement and sustainable supply chain management. Int J Prod Res 1-22. |
[60] | WEF (2011) Water Security: the water-food-energy-climate nexus, Island Press. |
[61] | Franz M, Schlitz N, Schumacher KP (2017) Globalization and the water-energy-food nexus - Using the global production networks approach to analyze society-environment relations. Environ Sci Policy. |
[62] | Endo A, Tsurita I, Burnett K, et al. (2017) A review of the current state of research on the water, energy, and food nexus. Water Energy Food Nexus Asia-Pac Reg 11: 20-30. |
[63] | Venghaus S, Hake J-F (2018) Nexus thinking in current EU policies - The interdependencies among food, energy and water resources. Environ Sci Policy. |
[64] |
Bazilian M, Rogner H, Howells M, et al. (2011) Considering the energy, water and food nexus: Towards an integrated modelling approach. Energy Policy 39: 7896-7906. doi: 10.1016/j.enpol.2011.09.039
![]() |
[65] |
Leck H, Conway D, Bradshaw M, et al. (2015) Tracing the Water-Energy-Food Nexus: Description, Theory and Practice. Geogr Compass 9: 445-460. doi: 10.1111/gec3.12222
![]() |
[66] |
Al-Saidi M, Elagib NA (2017) Towards understanding the integrative approach of the water, energy and food nexus. Sci Total Environ 574: 1131-1139. doi: 10.1016/j.scitotenv.2016.09.046
![]() |
[67] | Bates B, Kundzewicz ZW, Wu S, et al. (Eds.) (2008) Climate Change and Water. Technical Paper of the Intergovernmental Panel on Climate Change, Geneva, IPCC Secretariat. |
[68] | Alexandratos N, Bruinsma J (2012) World agriculture towards 2030/2050: the 2012 revision. 160. |
[69] | Mateo-Sagasta J, Burke J (2008) Agriculture and water quality interactions: a global overview. |
[70] | Ongley ED (1996) Control of water pollution from agriculture; Chapter 1: Introduction to agricultural water pollution. |
[71] |
Wiedmann T (2018) Eutrophication's neglected drivers. Nat Sustain 1: 273-274. doi: 10.1038/s41893-018-0084-2
![]() |
[72] | Yue Y, Guo WN, Lin QM, et al. (2016) Improving salt leaching in a simulated saline soil column by three biochars derived from rice straw (Oryza sativa L.), sunflower straw (Helianthus annuus), and cow manure. J Soil Water Conserv 71: 467-475. |
[73] |
Bell JM, Schwartz R, McInnes KJ, et al. (2018) Deficit irrigation effects on yield and yield components of grain sorghum. Agric Water Manag 203: 289-296. doi: 10.1016/j.agwat.2018.03.002
![]() |
[74] | Wagner K (2012) Status and trends of irrigated agriculture in Texas. Spec Rep Tex Water Resour Inst Tex AM Univ Coll Stn Tex Tex Water Resour Inst. |
[75] |
Bordovsky JP, Mustian JT, Cranmer AM, et al. (2011) Cotton-grain sorghum rotation under extreme deficit irrigation conditions. Appl Eng Agric 27: 359-371. doi: 10.13031/2013.37079
![]() |
[76] |
Grafton RQ, Williams J, Perry CJ, et al. (2018) The paradox of irrigation efficiency. Science 361: 748-750. doi: 10.1126/science.aat9314
![]() |
[77] | Vadez V, Kholová J, Yadav RS, et al. (2013) Small temporal differences in water uptake among varieties of pearl millet (Pennisetum glaucum (L.) R. Br.) are critical for grain yield under terminal drought. Plant Soil 371: 447-462. |
[78] |
Compton M, Willis S, Rezaie B, et al. (2018) Food processing industry energy and water consumption in the Pacific northwest. Innov Food Sci Emerg Technol 47: 371-383. doi: 10.1016/j.ifset.2018.04.001
![]() |
[79] | Ralph E. H. Sims (2011) Energy-Smart Food for People and Climate. UN Food and Agriculture Organisation. |
[80] | FAO (2018) Food and Agriculture Organization of the United Nations, 2018.Available from: http://www.fao.org/home/en/. |
[81] |
Reganold JP, Wachter JM (2016) Organic agriculture in the twenty-first century. Nat Plants 2: 15221. doi: 10.1038/nplants.2015.221
![]() |
[82] |
Dong X, Vuran MC, Irmak S (2013) Autonomous precision agriculture through integration of wireless underground sensor networks with center pivot irrigation systems. Ad Hoc Netw 11: 1975-1987. doi: 10.1016/j.adhoc.2012.06.012
![]() |
[83] | Helu M, Hedberg T, Barnard Feeney A (2017) Reference architecture to integrate heterogeneous manufacturing systems for the digital thread. CIRP J Manuf Sci Technol. |
[84] | Vogl GW, Weiss BA, Helu M (2016) A review of diagnostic and prognostic capabilities and best practices for manufacturing. J Intell Manuf 1-17. |
[85] |
Wolfe ML, Ting KC, Scott N, et al. (2016) Engineering solutions for food-energy-water systems: it is more than engineering. J Environ Stud Sci 6: 172-182. doi: 10.1007/s13412-016-0363-z
![]() |
[86] |
Lehmann J, Gaunt J, Rondon M (2006) Bio-char Sequestration in Terrestrial Ecosystems - A Review. Mitig Adapt Strateg Glob Change 11: 403-427. doi: 10.1007/s11027-005-9006-5
![]() |
[87] | Napolitano G, Isaac J, Bizzarri G, et al. (2010) Bioenergy and Food Security. Food and Agriculture Organization of the United Nations. |
[88] | Khan Z, Linares P, García-González J (2017) Integrating water and energy models for policy driven applications. A review of contemporary work and recommendations for future developments. Renew Sustain Energy Rev 67: 1123-1138. |
[89] | Pate R, Hightower M, Cameron C, et al. (2007) Overview of Energy-Water Interdependencies and the emerging energy demands on Water Resources. |
[90] | Scott CA, Pierce SA, Pasqualetti MJ, et al. (2011) Policy and institutional dimensions of the water-energy nexus. Sustain Biofuels 39: 6622-6630. |
[91] |
Hamiche AM, Stambouli AB, Flazi S (2016) A review of the water-energy nexus. Renew Sustain Energy Rev 65: 319-331. doi: 10.1016/j.rser.2016.07.020
![]() |
[92] | Erik Mielke, Laura Diaz Anadon, Vankatesh Narayanamurti (2010) Water Consumption of Energy Resource Extraction, Processing, and Conversion, A review of the literature for estimates of water intensity of energy-resource extraction, processing to fuels,and conversion to electricity, Belfer Center for Science and International Affairs, Harvard University. |
[93] |
Bartos MD, Chester MV (2014) The Conservation Nexus: Valuing Interdependent Water and Energy Savings in Arizona. Environ Sci Technol 48: 2139-2149. doi: 10.1021/es4033343
![]() |
[94] |
Wu M, Mintz M, Wang M, et al. (2009) Water Consumption in the Production of Ethanol and Petroleum Gasoline. Environ Manage 44: 981. doi: 10.1007/s00267-009-9370-0
![]() |
[95] | Ernst KM, Preston BL (2017) Adaptation opportunities and constraints in coupled systems: Evidence from the U.S. energy-water nexus. Environ Sci Policy 70: 38-45. |
[96] |
Liu J, Hull V, Godfray HCJ, et al. (2018) Nexus approaches to global sustainable development. Nat Sustain 1: 466. doi: 10.1038/s41893-018-0135-8
![]() |
[97] |
Jägerskog A, Lindström A, Björklund G, et al. (2012) Regional Options for Addressing the Water, Energy and Food Nexus in Central Asia and the Aral Sea Basin AU - Granit, Jakob. Int J Water Resour Dev 28: 419-432. doi: 10.1080/07900627.2012.684307
![]() |
[98] |
Villarroel Walker R, Beck MB, Hall JW (2012) Water - and nutrient and energy - systems in urbanizing watersheds. Front Environ Sci Eng 6: 596-611. doi: 10.1007/s11783-012-0445-4
![]() |
[99] | Siddiqi A, Anadon LD (2011) The water-energy nexus in Middle East and North Africa. Crossroads Pathw Renew Nucl Energy Policy North Afr 39: 4529-4540. |
[100] | Ringler C, Bhaduri A, Lawford R (2013) The nexus across water, energy, land and food (WELF): potential for improved resource use efficiency? Aquat Mar Syst 5: 617-624. |
[101] | Lawford R, Bogardi J, Marx S, et al. (2013) Basin perspectives on the Water-Energy-Food Security Nexus. Aquat Mar Syst 5: 607-616. |
[102] |
Rasul G (2014) Food, water, and energy security in South Asia: A nexus perspective from the Hindu Kush Himalayan region☆. Environ Sci Policy 39: 35-48. doi: 10.1016/j.envsci.2014.01.010
![]() |
[103] | Stein C, Barron J, Nigussie L, et al. (2014) Advancing the water-energy-food nexus: social networks and institutional interplay in the Blue Nile, International Water Management Institute (IWMI). CGIAR Research Program. |
[104] |
Villarroel R, Beck MB, Hall JW, et al. (2014) The energy-water-food nexus: Strategic analysis of technologies for transforming the urban metabolism. J Environ Manage 141: 104-115. doi: 10.1016/j.jenvman.2014.01.054
![]() |
[105] |
Biggs EM, Bruce E, Boruff B, et al. (2015) Sustainable development and the water-energy-food nexus: A perspective on livelihoods. Environ Sci Policy 54: 389-397. doi: 10.1016/j.envsci.2015.08.002
![]() |
[106] |
Conway D, van Garderen EA, Deryng D, et al. (2015) Climate and southern Africa's water-energy-food nexus. Nat Clim Change 5: 837. doi: 10.1038/nclimate2735
![]() |
[107] | Daher BT, Mohtar RH (2015) Water-energy-food (WEF) Nexus Tool 2.0: guiding integrative resource planning and decision-making. Water Int 40: 748-771. |
[108] |
Jeswani HK, Burkinshaw R, Azapagic A (2015) Environmental sustainability issues in the food-energy-water nexus: Breakfast cereals and snacks. Sustain Prod Consum 2: 17-28. doi: 10.1016/j.spc.2015.08.001
![]() |
[109] |
Kraucunas I, Clarke L, Dirks J, et al. (2015) Investigating the nexus of climate, energy, water, and land at decision-relevant scales: the Platform for Regional Integrated Modeling and Analysis (PRIMA). Clim Change 129: 573-588. doi: 10.1007/s10584-014-1064-9
![]() |
[110] |
Mukuve FM, Fenner RA (2015) Scale variability of water, land, and energy resource interactions and their influence on the food system in Uganda. Sustain Prod Consum 2: 79-95. doi: 10.1016/j.spc.2015.07.009
![]() |
[111] |
Ozturk I (2015) Sustainability in the food-energy-water nexus: Evidence from BRICS (Brazil, the Russian Federation, India, China, and South Africa) countries. Energy 93: 999-1010. doi: 10.1016/j.energy.2015.09.104
![]() |
[112] |
Endo A, Burnett K, Orencio P, et al. (2015) Methods of the water-energy-food nexus. Water 7: 5806-5830. doi: 10.3390/w7105806
![]() |
[113] | Middleton C, Allouche J, Gyawali D, et al. (2015) The Rise and Implications of the Water-Energy-Food Nexus in Southeast Asia through an Environmental Justice Lens. |
[114] | Villamayor-Tomas S, Grundmann P, Epstein G, et al. (2015) The Water-Energy-Food Security Nexus through the Lenses of the Value Chain and the Institutional Analysis and Development Frameworks. 8: 21. |
[115] | Keskinen M, Someth P, Salmivaara A, et al. (2015) Water-Energy-Food Nexus in a Transboundary River Basin: The Case of Tonle Sap Lake, Mekong River Basin. |
[116] | Garcia DJ, You F (2016) The water-energy-food nexus and process systems engineering: A new focus. 12th Int Symp Process Syst Eng 25th Eur Symp Comput Aided Process Eng PSE-2015ESCAPE-25 31 May - 4 June 2015 Cph Den 91: 49-67. |
[117] |
Rasul G (2016) Managing the food, water, and energy nexus for achieving the Sustainable Development Goals in South Asia. Environ Dev 18: 14-25. doi: 10.1016/j.envdev.2015.12.001
![]() |
[118] |
De Laurentiis V, Hunt D, Rogers C (2016) Overcoming food security challenges within an energy/water/food nexus (EWFN) approach. Sustainability 8: 95. doi: 10.3390/su8010095
![]() |
[119] |
Cairns R, Krzywoszynska A (2016) Anatomy of a buzzword: The emergence of 'the water-energy-food nexus' in UK natural resource debates. Environ Sci Policy 64: 164-170. doi: 10.1016/j.envsci.2016.07.007
![]() |
[120] |
Yang YE, Wi S, Ray PA, et al. (2016) The future nexus of the Brahmaputra River Basin: climate, water, energy and food trajectories. Glob Environ Change 37: 16-30. doi: 10.1016/j.gloenvcha.2016.01.002
![]() |
[121] |
de Strasser L, Lipponen A, Howells M, et al. (2016) A methodology to assess the water energy food ecosystems nexus in transboundary river basins. Water 8: 59. doi: 10.3390/w8020059
![]() |
[122] |
Fasel M, Bréthaut C, Rouholahnejad E, et al. (2016) Blue water scarcity in the Black Sea catchment: Identifying key actors in the water-ecosystem-energy-food nexus. Environ Sci Policy 66: 140-150. doi: 10.1016/j.envsci.2016.09.004
![]() |
[123] |
Perrone D, Hornberger G (2016) Frontiers of the food-energy-water trilemma: Sri Lanka as a microcosm of tradeoffs. Environ Res Lett 11: 014005. doi: 10.1088/1748-9326/11/1/014005
![]() |
[124] |
Mortensen JG, González-Pinzón R, Dahm CN, et al. (2016) Advancing the Food-Energy-Water Nexus: Closing Nutrient Loops in Arid River Corridors. Environ Sci Technol 50: 8485-8496. doi: 10.1021/acs.est.6b01351
![]() |
[125] |
Wichelns D (2017) The water-energy-food nexus: Is the increasing attention warranted, from either a research or policy perspective? Environ Sci Policy 69: 113-123. doi: 10.1016/j.envsci.2016.12.018
![]() |
[126] | Endo A, Tsurita I, Burnett K, et al. (2017) A review of the current state of research on the water, energy, and food nexus. Water Energy Food Nexus Asia-Pac Reg 11: 20-30. |
[127] |
Howarth C, Monasterolo I (2017) Opportunities for knowledge co-production across the energy-food-water nexus: Making interdisciplinary approaches work for better climate decision making. Environ Sci Policy 75: 103-110. doi: 10.1016/j.envsci.2017.05.019
![]() |
[128] | Flammini A, Puri M, Pluschke L, et al. (2014) Walking the nexus talk: assessing the water-energy-food nexus in the context of the sustainable energy for all initiative, Rome, Climate, Energy and Tenure Division (NRC), Food and Agriculture Organization of the United Nations. |
[129] |
Pahl-Wostl C (2019) Governance of the water-energy-food security nexus: A multi-level coordination challenge. Environ Sci Policy 92: 356-367. doi: 10.1016/j.envsci.2017.07.017
![]() |
[130] |
El Gafy I, Grigg N, Reagan W (2017) Dynamic Behaviour of the Water-Food-Energy Nexus: Focus on Crop Production and Consumption. Irrig Drain 66: 19-33. doi: 10.1002/ird.2060
![]() |
[131] | Wicaksono A, Jeong G, Kang D (2017) Water, energy, and food nexus: Review of global implementation and simulation model development. |
[132] | Dhaubanjar S, Davidsen C, Bauer-Gottwein P (2017) Multi-Objective Optimization for Analysis of Changing Trade-Offs in the Nepalese Water-Energy-Food Nexus with Hydropower Development. Water 9. |
[133] |
Hussien WA, Memon FA, Savic DA (2017) An integrated model to evaluate water-energy-food nexus at a household scale. Environ Model Softw 93: 366-380. doi: 10.1016/j.envsoft.2017.03.034
![]() |
[134] |
Johnson OW, Karlberg L (2017) Co-exploring the Water-Energy-Food Nexus: Facilitating Dialogue through Participatory Scenario Building. Front Environ Sci 5: 24. doi: 10.3389/fenvs.2017.00024
![]() |
[135] |
White DJ, Hubacek K, Feng K, et al. (2018) The Water-Energy-Food Nexus in East Asia: A tele-connected value chain analysis using inter-regional input-output analysis. Appl Energy 210: 550-567. doi: 10.1016/j.apenergy.2017.05.159
![]() |
[136] | Albrecht TR, Crootof A, Scott CA (2018) The water-energy-food nexus: A comprehensive review of nexus-specific methods. Environ Res Lett. |
[137] | Siddiqi A, Anadon LD (2011) The water-energy nexus in Middle East and North Africa. Crossroads Pathw Renew Nucl Energy Policy North Afr 39: 4529-4540. |
[138] |
Villarroel Walker R, Beck MB, Hall JW (2012) Water - and nutrient and energy - systems in urbanizing watersheds. Front Environ Sci Eng 6: 596-611. doi: 10.1007/s11783-012-0445-4
![]() |
[139] | Ringler C, Bhaduri A, Lawford R (2013) The nexus across water, energy, land and food (WELF): potential for improved resource use efficiency? Aquat Mar Syst 5: 617-624. |
[140] |
Rasul G (2016) Managing the food, water, and energy nexus for achieving the Sustainable Development Goals in South Asia. Environ Dev 18: 14-25. doi: 10.1016/j.envdev.2015.12.001
![]() |
[141] |
Wichelns D (2017) The water-energy-food nexus: Is the increasing attention warranted, from either a research or policy perspective? Environ Sci Policy 69: 113-123. doi: 10.1016/j.envsci.2016.12.018
![]() |
[142] |
White DJ, Hubacek K, Feng K, et al. (2018) The Water-Energy-Food Nexus in East Asia: A tele-connected value chain analysis using inter-regional input-output analysis. Appl Energy 210: 550-567. doi: 10.1016/j.apenergy.2017.05.159
![]() |
[143] |
Qambrani NA, Rahman MdM, Won S, et al. (2017) Biochar properties and eco-friendly applications for climate change mitigation, waste management, and wastewater treatment: A review. Renew Sustain Energy Rev 79: 255-273. doi: 10.1016/j.rser.2017.05.057
![]() |
[144] |
Bruun EW, Müller-Stöver D, Ambus P, et al. (2011) Application of biochar to soil and N2O emissions: potential effects of blending fast-pyrolysis biochar with anaerobically digested slurry. Eur J Soil Sci 62: 581-589. doi: 10.1111/j.1365-2389.2011.01377.x
![]() |
[145] |
Laird DA (2008) The Charcoal Vision: A Win-Win-Win Scenario for Simultaneously Producing Bioenergy, Permanently Sequestering Carbon, while Improving Soil and Water Quality. Agron J 100: 178-181. doi: 10.2134/agronj2007.0161
![]() |
[146] |
Kuppusamy S, Thavamani P, Megharaj M, et al. (2016) Agronomic and remedial benefits and risks of applying biochar to soil: Current knowledge and future research directions. Environ Int 87: 1-12. doi: 10.1016/j.envint.2015.10.018
![]() |
[147] | Vanholme B, Desmet T, Ronsse F, et al. (2013) Towards a carbon-negative sustainable bio-based economy. Front Plant Sci 4: 174. |
[148] |
Tan Z, Lin CSK, Ji X, et al. (2017) Returning biochar to fields: A review. Appl Soil Ecol 116: 1-11. doi: 10.1016/j.apsoil.2017.03.017
![]() |
[149] |
Beesley L, Moreno-Jiménez E, Gomez-Eyles JL, et al. (2011) A review of biochars' potential role in the remediation, revegetation and restoration of contaminated soils. Environ Pollut 159: 3269-3282. doi: 10.1016/j.envpol.2011.07.023
![]() |
[150] | Lehmann J, Joseph S (2009) Biochar for Environmental Management: An Introduction. 12. |
[151] |
Mohan D, Pittman Charles U, Steele PH (2006) Pyrolysis of Wood/Biomass for Bio-oil: A Critical Review. Energy Fuels 20: 848-889. doi: 10.1021/ef0502397
![]() |
[152] |
Kong S-H, Loh S-K, Bachmann RT, et al. (2014) Biochar from oil palm biomass: A review of its potential and challenges. Renew Sustain Energy Rev 39: 729-739. doi: 10.1016/j.rser.2014.07.107
![]() |
[153] |
Qian K, Kumar A, Zhang H, et al. (2015) Recent advances in utilization of biochar. Renew Sustain Energy Rev 42: 1055-1064. doi: 10.1016/j.rser.2014.10.074
![]() |
[154] |
Shea EC (2014) Adaptive management: The cornerstone of climate-smart agriculture. J Soil Water Conserv 69: 198A-199A. doi: 10.2489/jswc.69.6.198A
![]() |
[155] | Tenenbaum DJ (2009) Biochar: Carbon Mitigation from the Ground Up. Environ Health Perspect 117: A70-A73. |
[156] | Liang B, Lehmann J, Solomon D, et al. (2006) Black Carbon Increases Cation Exchange Capacity in Soils. Soil Sci Soc Am J 70. |
[157] | Jeffery S, Bezemer TM, Cornelissen G, et al. (2013) The way forward in biochar research: targeting trade-offs between the potential wins. GCB Bioenergy 7: 1-13. |
[158] |
Taha SM, Amer ME, Elmarsafy AE, et al. (2014) Adsorption of 15 different pesticides on untreated and phosphoric acid treated biochar and charcoal from water. J Environ Chem Eng 2: 2013-2025. doi: 10.1016/j.jece.2014.09.001
![]() |
[159] |
Taha SM, Amer ME, Elmarsafy AE, et al. (2014) Adsorption of 15 different pesticides on untreated and phosphoric acid treated biochar and charcoal from water. J Environ Chem Eng 2: 2013-2025. doi: 10.1016/j.jece.2014.09.001
![]() |
[160] |
Liu J, Schulz H, Brandl S, et al. (2012) Short-term effect of biochar and compost on soil fertility and water status of a Dystric Cambisol in NE Germany under field conditions. J Plant Nutr Soil Sci 175: 698-707. doi: 10.1002/jpln.201100172
![]() |
[161] |
Van Zwieten L, Kimber S, Morris S, et al. (2010) Effects of biochar from slow pyrolysis of papermill waste on agronomic performance and soil fertility. Plant Soil 327: 235-246. doi: 10.1007/s11104-009-0050-x
![]() |
[162] | Kookana RS, Sarmah AK, Van Zwieten L, et al. (2011) Chapter three - Biochar Application to Soil: Agronomic and Environmental Benefits and Unintended Consequences, In: Sparks DL (Ed.), Advances in Agronomy, Academic Press, 103-143. |
[163] |
Major J, Rondon M, Molina D, et al. (2010) Maize yield and nutrition during 4 years after biochar application to a Colombian savanna oxisol. Plant Soil 333: 117-128. doi: 10.1007/s11104-010-0327-0
![]() |
[164] |
Mohammadi A, Cowie AL, Mai TLA, et al. (2017) Climate-change and health effects of using rice husk for biochar-compost: Comparing three pyrolysis systems. J Clean Prod 162: 260-272. doi: 10.1016/j.jclepro.2017.06.026
![]() |
[165] |
Mohammadi A, Cowie A, Mai TLA, et al. (2016) Biochar use for climate-change mitigation in rice cropping systems. J Clean Prod 116: 61-70. doi: 10.1016/j.jclepro.2015.12.083
![]() |
[166] |
Mohammadi A, Cowie A, Mai TLA, et al. (2016) Quantifying the greenhouse gas reduction benefits of utilising straw biochar and enriched biochar. Energy Procedia 97: 254-261. doi: 10.1016/j.egypro.2016.10.069
![]() |
[167] |
Agegnehu G, Bass AM, Nelson PN, et al. (2016) Benefits of biochar, compost and biochar-compost for soil quality, maize yield and greenhouse gas emissions in a tropical agricultural soil. Sci Total Environ 543: 295-306. doi: 10.1016/j.scitotenv.2015.11.054
![]() |
[168] |
Schimmelpfennig S, Glaser B (2012) One Step Forward toward Characterization: Some Important Material Properties to Distinguish Biochars. J Environ Qual 41: 1001-1013. doi: 10.2134/jeq2011.0146
![]() |
[169] | Kloss S, Zehetner F, Wimmer B, et al. (2013) Biochar application to temperate soils: Effects on soil fertility and crop growth under greenhouse conditions. J Plant Nutr Soil Sci 177: 3-15. |
[170] |
Zhou H, Zhang D, Wang P, et al. (2017) Changes in microbial biomass and the metabolic quotient with biochar addition to agricultural soils: A Meta-analysis. Agric Ecosyst Environ 239: 80-89. doi: 10.1016/j.agee.2017.01.006
![]() |
[171] |
Burns RG, DeForest JL, Marxsen J, et al. (2013) Soil enzymes in a changing environment: current knowledge and future directions. Soil Biol Biochem 58: 216-234. doi: 10.1016/j.soilbio.2012.11.009
![]() |
[172] |
Lehmann J, Rillig MC, Thies J, et al. (2011) Biochar effects on soil biota - A review. Soil Biol Biochem 43: 1812-1836. doi: 10.1016/j.soilbio.2011.04.022
![]() |
[173] |
Caroline A, Debode J, Vandecasteele B, et al. (2016) Biological, physicochemical and plant health responses in lettuce and strawberry in soil or peat amended with biochar. Appl Soil Ecol 107: 1-12. doi: 10.1016/j.apsoil.2016.05.001
![]() |
[174] |
Kolton M, Graber ER, Tsehansky L, et al. (2017) Biochar-stimulated plant performance is strongly linked to microbial diversity and metabolic potential in the rhizosphere. New Phytol 213: 1393-1404. doi: 10.1111/nph.14253
![]() |
[175] |
Graber ER, Frenkel O, Jaiswal AK, et al. (2014) How may biochar influence severity of diseases caused by soilborne pathogens? Carbon Manag 5: 169-183. doi: 10.1080/17583004.2014.913360
![]() |
[176] |
Jaiswal AK, Frenkel O, Elad Y, et al. (2015) Non-monotonic influence of biochar dose on bean seedling growth and susceptibility to Rhizoctonia solani: the "Shifted R max-Effect". Plant Soil 395: 125-140. doi: 10.1007/s11104-014-2331-2
![]() |
[177] |
Zhu X, Chen B, Zhu L, et al. (2017) Effects and mechanisms of biochar-microbe interactions in soil improvement and pollution remediation: a review. Environ Pollut 227: 98-115. doi: 10.1016/j.envpol.2017.04.032
![]() |
[178] |
Lal R, Smith P, Jungkunst HF, et al. (2018) The carbon sequestration potential of terrestrial ecosystems. J Soil Water Conserv 73: 145A-152A. doi: 10.2489/jswc.73.6.145A
![]() |
[179] |
Karhu K, Mattila T, Bergström I, et al. (2011) Biochar addition to agricultural soil increased CH4 uptake and water holding capacity - Results from a short-term pilot field study. Agric Ecosyst Environ 140: 309-313. doi: 10.1016/j.agee.2010.12.005
![]() |
[180] |
Bu XL, Su J, Xue JH, et al. (2019) Effect of rice husk biochar addition on nutrient leaching and microbial properties of Calcaric Cambisols. J Soil Water Conserv 74: 172-179. doi: 10.2489/jswc.74.2.172
![]() |
[181] |
Uchimiya M, Lima IM, Klasson KT, et al. (2010) Contaminant immobilization and nutrient release by biochar soil amendment: Roles of natural organic matter. Chemosphere 80: 935-940. doi: 10.1016/j.chemosphere.2010.05.020
![]() |
[182] |
Xu S, Adhikari D, Huang R, et al. (2016) Biochar-facilitated microbial reduction of hematite. Environ Sci Technol 50: 2389-2395. doi: 10.1021/acs.est.5b05517
![]() |
[183] |
Downie A, Munroe P, Cowie A, et al. (2012) Biochar as a Geoengineering Climate Solution: Hazard Identification and Risk Management. Crit Rev Environ Sci Technol 42: 225-250. doi: 10.1080/10643389.2010.507980
![]() |
[184] |
Campbell JL, Sessions J, Smith D, et al. (2018) Potential carbon storage in biochar made from logging residue: Basic principles and Southern Oregon case studies. PLOS ONE 13: e0203475. doi: 10.1371/journal.pone.0203475
![]() |
[185] |
Creamer AE, Gao B, Zhang M (2014) Carbon dioxide capture using biochar produced from sugarcane bagasse and hickory wood. Chem Eng J 249: 174-179. doi: 10.1016/j.cej.2014.03.105
![]() |
[186] |
Hasler K, Bröring S, Omta SWF, et al. (2015) Life cycle assessment (LCA) of different fertilizer product types. Eur J Agron 69: 41-51. doi: 10.1016/j.eja.2015.06.001
![]() |
[187] |
Duku MH, Gu S, Hagan EB (2011) Biochar production potential in Ghana-A review. Renew Sustain Energy Rev 15: 3539-3551. doi: 10.1016/j.rser.2011.05.010
![]() |
[188] |
Lee JW, Hawkins B, Day DM, et al. (2010) Sustainability: the capacity of smokeless biomass pyrolysis for energy production, global carbon capture and sequestration. Energy Environ Sci 3: 1695-1705. doi: 10.1039/c004561f
![]() |
[189] |
Yang Q, Han F, Chen Y, et al. (2016) Greenhouse gas emissions of a biomass-based pyrolysis plant in China. Renew Sustain Energy Rev 53: 1580-1590. doi: 10.1016/j.rser.2015.09.049
![]() |
[190] | Mirkouei A, Haapala KR, Sessions J, et al. (2016) Reducing greenhouse gas emissions for sustainable bio-oil production using a mixed supply chain, ASME 2016 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, American Society of Mechanical Engineers, V004T05A031. |
[191] | Dougherty B (2016) Biochar as a cover for dairy manure lagoons: reducing odor and gas emissions while capturing nutrients, Oregon State University. |
[192] | Stetina K (2017) Control of Fecal Malodor by Adsorption onto Biochar. |
[193] | Tsutomu I, Asada T, Kuniaki K, et al. (2004) Comparison of Removal Efficiencies for Ammonia and Amine Gases between Woody Charcoal and Activated Carbon. J Health Sci - J Health SCI 50: 148-153. |
[194] |
Parham J, Deng S, Raun W, et al. (2002) Long-term cattle manure application in soil. Biol Fertil Soils 35: 328-337. doi: 10.1007/s00374-002-0476-2
![]() |
[195] | Gasser JKR (1987) The future of animal manures as fertilizer or waste, In: Van Der Meer HG, Unwin RJ, Van Dijk TA, et al. (Eds.), Animal Manure on Grassland and Fodder Crops. Fertilizer or Waste? Proceedings of an International Symposium of the European Grassland Federation, Wageningen, The Netherlands, 31 August-3 September 1987, Dordrecht, Springer Netherlands, 259-278. |
[196] |
Kleinman PJ, Sharpley AN, Wolf AM, et al. (2002) Measuring water-extractable phosphorus in manure as an indicator of phosphorus in runoff. Soil Sci Soc Am J 66: 2009-2015. doi: 10.2136/sssaj2002.2009
![]() |
[197] | Sharpley A, Moyer B (2000) Phosphorus forms in manure and compost and their release during simulated rainfall. J Environ Qual 29: 1462-1469. |
[198] |
Heuer H, Schmitt H, Smalla K (2011) Antibiotic resistance gene spread due to manure application on agricultural fields. Curr Opin Microbiol 14: 236-243. doi: 10.1016/j.mib.2011.04.009
![]() |
[199] | Spencer JL, Guan J (2004) Public Health Implications Related to Spread of Pathogens in Manure From Livestock and Poultry Operations, In: Spencer JFT, Ragout de Spencer AL (Eds.), Public Health Microbiology: Methods and Protocols, Totowa, NJ, Humana Press, 503-515. |
[200] | You Y, Silbergeld EK (2014) Learning from agriculture: understanding low-dose antimicrobials as drivers of resistome expansion. Front Microbiol 5: 284. |
[201] |
Casey JA, Curriero FC, Cosgrove SE, et al. (2013) High-density livestock operations, crop field application of manure, and risk of community-associated methicillin-resistant Staphylococcus aureus infection in Pennsylvania. JAMA Intern Med 173: 1980-1990. doi: 10.1001/jamainternmed.2013.10408
![]() |
[202] |
Manyi-Loh CE, Mamphweli SN, Meyer EL, et al. (2016) An Overview of the Control of Bacterial Pathogens in Cattle Manure. Int J Environ Res Public Health 13: 843. doi: 10.3390/ijerph13090843
![]() |
[203] |
Nicholson FA, Groves SJ, Chambers BJ (2005) Pathogen survival during livestock manure storage and following land application. Bioresour Technol 96: 135-143. doi: 10.1016/j.biortech.2004.02.030
![]() |
[204] | Qiu M, Sun K, Jin J, et al. (2014) Properties of the plant- and manure-derived biochars and their sorption of dibutyl phthalate and phenanthrene. Sci Rep 4: 5295. |
[205] |
Batista EMCC, Shultz J, Matos TTS, et al. (2018) Effect of surface and porosity of biochar on water holding capacity aiming indirectly at preservation of the Amazon biome. Sci Rep 8: 10677. doi: 10.1038/s41598-018-28794-z
![]() |
[206] |
Brown TR, Thilakaratne R, Brown RC, et al. (2013) Techno-economic analysis of biomass to transportation fuels and electricity via fast pyrolysis and hydroprocessing. Fuel 106: 463-469. doi: 10.1016/j.fuel.2012.11.029
![]() |
[207] |
Huang Y-F, Syu F-S, Chiueh P-T, et al. (2013) Life cycle assessment of biochar cofiring with coal. Bioresour Technol 131: 166-171. doi: 10.1016/j.biortech.2012.12.123
![]() |
[208] | Bergman PCA, Boersma AR, Zwart RWR, et al. (2005) Torrefaction for biomass co-firing in existing coal-fired power stations, Energy research Centre of the Netherlands. |
[209] | Abdullah H, Wu H (2009) Biochar as a Fuel: 1. Properties and Grindability of Biochars Produced from the Pyrolysis of Mallee Wood under Slow-Heating Conditions. Energy Fuels 23: 4174-4181. |
[210] |
Yang X, Wang H, Strong P, et al. (2017) Thermal properties of biochars derived from waste biomass generated by agricultural and forestry sectors. Energies 10: 469. doi: 10.3390/en10040469
![]() |
[211] | Engineering ToolBox (2003) Classification of Coal, 2003.Available from: https://www.engineeringtoolbox.com/classification-coal-d_164.html. |
[212] |
Vikrant K, Kim K-H, Ok YS, et al. (2018) Engineered/designer biochar for the removal of phosphate in water and wastewater. Sci Total Environ 616-617: 1242-1260. doi: 10.1016/j.scitotenv.2017.10.193
![]() |
[213] |
Tan X, Liu Y, Gu Y, et al. (2016) Biochar-based nano-composites for the decontamination of wastewater: A review. Bioresour Technol 212: 318-333. doi: 10.1016/j.biortech.2016.04.093
![]() |
[214] | Sizmur T, Fresno T, Akgül G, et al. (2017) Biochar modification to enhance sorption of inorganics from water. Spec Issue Biochar Prod Charact Appl - Soil Appl 246: 34-47. |
[215] |
Husk BR, Sanchez JS, Anderson BC, et al. (2018) Removal of phosphorus from agricultural subsurface drainage water with woodchip and mixed-media bioreactors. J Soil Water Conserv 73: 265-275. doi: 10.2489/jswc.73.3.265
![]() |
[216] |
Inyang MI, Gao B, Yao Y, et al. (2016) A review of biochar as a low-cost adsorbent for aqueous heavy metal removal. Crit Rev Environ Sci Technol 46: 406-433. doi: 10.1080/10643389.2015.1096880
![]() |
[217] |
Jing X-R, Wang Y-Y, Liu W-J, et al. (2014) Enhanced adsorption performance of tetracycline in aqueous solutions by methanol-modified biochar. Chem Eng J 248: 168-174. doi: 10.1016/j.cej.2014.03.006
![]() |
[218] |
Regmi P, Garcia Moscoso JL, Kumar S, et al. (2012) Removal of copper and cadmium from aqueous solution using switchgrass biochar produced via hydrothermal carbonization process. J Environ Manage 109: 61-69. doi: 10.1016/j.jenvman.2012.04.047
![]() |
[219] |
Ahmed MB, Zhou JL, Ngo HH, et al. (2016) Progress in the preparation and application of modified biochar for improved contaminant removal from water and wastewater. Bioresour Technol 214: 836-851. doi: 10.1016/j.biortech.2016.05.057
![]() |
[220] |
DeBoe G, Bock E, Stephenson K, et al. (2017) Nutrient biofilters in the Virginia Coastal Plain: Nitrogen removal, cost, and potential adoption pathways. J Soil Water Conserv 72: 139-149. doi: 10.2489/jswc.72.2.139
![]() |
[221] |
Cao X, Ma L, Gao B, et al. (2009) Dairy-Manure Derived Biochar Effectively Sorbs Lead and Atrazine. Environ Sci Technol 43: 3285-3291. doi: 10.1021/es803092k
![]() |
[222] |
Xu X, Cheng K, Wu H, et al. (2019) Greenhouse gas mitigation potential in crop production with biochar soil amendment-a carbon footprint assessment for cross-site field experiments from China. GCB Bioenergy 11: 592-605. doi: 10.1111/gcbb.12561
![]() |
[223] | Wijitkosum S, Jiwnok P, UNISEARCH C (2019) Effect of biochar on Chinese kale and carbon storage in an agricultural area on a high rise building. |
[224] |
Shaheen A, Turaib Ali Bukhari S (2018) Potential of sawdust and corn cobs derived biochar to improve soil aggregate stability, water retention, and crop yield of degraded sandy loam soil. J Plant Nutr 41: 2673-2682. doi: 10.1080/01904167.2018.1509092
![]() |
[225] |
Sarma B, Farooq M, Gogoi N, et al. (2018) Soil organic carbon dynamics in wheat-Green gram crop rotation amended with vermicompost and biochar in combination with inorganic fertilizers: A comparative study. J Clean Prod 201: 471-480. doi: 10.1016/j.jclepro.2018.08.004
![]() |
[226] | Hersh B, Mirkouei A (2019) Life Cycle Assessment of Pyrolysis-Derived Biochar from Organic Wastes and Advanced Feedstocks, Proceedings of the ASME 2019 International Design Engineering Technical Conferencesand Computers and Information in Engineering Conference, IDETC2019-97896. |
[227] |
Uusitalo V, Leino M (2019) Neutralizing global warming impacts of crop production using biochar from side flows and buffer zones: A case study of oat production in the boreal climate zone. J Clean Prod 227: 48-57. doi: 10.1016/j.jclepro.2019.04.175
![]() |
[228] |
Thers H, Djomo SN, Elsgaard L, et al. (2019) Biochar potentially mitigates greenhouse gas emissions from cultivation of oilseed rape for biodiesel. Sci Total Environ 671: 180-188. doi: 10.1016/j.scitotenv.2019.03.257
![]() |
[229] |
Alotaibi KD, Schoenau JJ (2019) Addition of Biochar to a Sandy Desert Soil: Effect on Crop Growth, Water Retention and Selected Properties. Agronomy 9: 327. doi: 10.3390/agronomy9060327
![]() |
[230] |
Wang D, Li C, Parikh SJ, et al. (2019) Impact of biochar on water retention of two agricultural soils-A multi-scale analysis. Geoderma 340: 185-191. doi: 10.1016/j.geoderma.2019.01.012
![]() |
[231] | Ding Y, Liu Y, Liu S, et al. (2017) Potential Benefits from Biochar Application for Agricultural Use: A Review. Pedosphere. |
[232] |
Wang L, Li L, Cheng K, et al. (2018) An assessment of emergy, energy, and cost-benefits of grain production over 6 years following a biochar amendment in a rice paddy from China. Environ Sci Pollut Res 25: 9683-9696. doi: 10.1007/s11356-018-1245-6
![]() |
[233] |
Shahzad K, Abid M, Sintim HY (2018) Wheat productivity and economic implications of biochar and inorganic nitrogen application. Agron J 110: 2259-2267. doi: 10.2134/agronj2018.01.0055
![]() |
[234] |
Jalal F, Arif M, Ahmad I, et al. (2018) Increasing Farm Productivity and Soil Fertility on Sustainable Basis Through "Summer Gap" Utilization with Biochar and Legumes. Gesunde Pflanz 70: 45-53. doi: 10.1007/s10343-017-0412-x
![]() |
[235] |
Mohammadi A, Cowie AL, Cacho O, et al. (2017) Biochar addition in rice farming systems: Economic and energy benefits. Energy 140: 415-425. doi: 10.1016/j.energy.2017.08.116
![]() |
[236] |
El-Naggar A, Lee SS, Rinklebe J, et al. (2019) Biochar application to low fertility soils: A review of current status, and future prospects. Geoderma 337: 536-554. doi: 10.1016/j.geoderma.2018.09.034
![]() |
[237] |
Al-Wabel MI, Hussain Q, Usman AR, et al. (2018) Impact of biochar properties on soil conditions and agricultural sustainability: a review. Land Degrad Dev 29: 2124-2161. doi: 10.1002/ldr.2829
![]() |
[238] |
Robb S, Dargusch P (2018) A financial analysis and life-cycle carbon emissions assessment of oil palm waste biochar exports from Indonesia for use in Australian broad-acre agriculture. Carbon Manag 9: 105-114. doi: 10.1080/17583004.2018.1435958
![]() |
[239] |
Dai L, Li H, Tan F, et al. (2016) Biochar: a potential route for recycling of phosphorus in agricultural residues. Gcb Bioenergy 8: 852-858. doi: 10.1111/gcbb.12365
![]() |
[240] | Munoz C, Gongora S, Zagal E (2016) Use of biochar as a soil amendment: a brief review. Chil J Agric Anim Sci Ex Agro-Cienc 32: 37-47. |
[241] | Li W (2018) Investigating the development of the bioproducts production platform from a techno-economic and environmental perspective. |
[242] |
Ji C, Cheng K, Nayak D, et al. (2018) Environmental and economic assessment of crop residue competitive utilization for biochar, briquette fuel and combined heat and power generation. J Clean Prod 192: 916-923. doi: 10.1016/j.jclepro.2018.05.026
![]() |
[243] |
Aller DM, Archontoulis SV, Zhang W, et al. (2018) Long term biochar effects on corn yield, soil quality and profitability in the US Midwest. Field Crops Res 227: 30-40. doi: 10.1016/j.fcr.2018.07.012
![]() |
[244] |
Woolf D, Amonette JE, Street-Perrott FA, et al. (2010) Sustainable biochar to mitigate global climate change. Nat Commun 1: 56. doi: 10.1038/ncomms1053
![]() |
[245] | Mohan D, Sarswat A, Ok YS, et al. (2014) Organic and inorganic contaminants removal from water with biochar, a renewable, low cost and sustainable adsorbent - A critical review. Spec Issue Biosorption 160: 191-202. |
[246] |
Woolf D, Amonette JE, Street-Perrott FA, et al. (2010) Sustainable biochar to mitigate global climate change. Nat Commun 1: 56. doi: 10.1038/ncomms1053
![]() |
[247] | Mohan D, Sarswat A, Ok YS, et al. (2014) Organic and inorganic contaminants removal from water with biochar, a renewable, low cost and sustainable adsorbent - A critical review. Spec Issue Biosorption 160: 191-202. |
[248] |
Beesley L, Moreno-Jiménez E, Gomez-Eyles JL, et al. (2011) A review of biochars' potential role in the remediation, revegetation and restoration of contaminated soils. Environ Pollut 159: 3269-3282. doi: 10.1016/j.envpol.2011.07.023
![]() |
[249] |
Van Zwieten L, Kimber S, Morris S, et al. (2010) Effects of biochar from slow pyrolysis of papermill waste on agronomic performance and soil fertility. Plant Soil 327: 235-246. doi: 10.1007/s11104-009-0050-x
![]() |
[250] |
Laird DA, Fleming P, Davis DD, et al. (2010) Impact of biochar amendments on the quality of a typical Midwestern agricultural soil. Geoderma 158: 443-449. doi: 10.1016/j.geoderma.2010.05.013
![]() |
[251] |
Steinbeiss S, Gleixner G, Antonietti M (2009) Effect of biochar amendment on soil carbon balance and soil microbial activity. Soil Biol Biochem 41: 1301-1310. doi: 10.1016/j.soilbio.2009.03.016
![]() |
[252] |
Laird DA, Brown RC, Amonette JE, et al. (2009) Review of the pyrolysis platform for coproducing bio-oil and biochar. Biofuels Bioprod Biorefining 3: 547-562. doi: 10.1002/bbb.169
![]() |
[253] | Kookana RS, Sarmah AK, Van Zwieten L, et al. (2011) Chapter three - Biochar Application to Soil: Agronomic and Environmental Benefits and Unintended Consequences, In: Sparks DL (Ed.), Advances in Agronomy, Academic Press, 103-143. |
[254] |
Regmi P, Garcia Moscoso JL, Kumar S, et al. (2012) Removal of copper and cadmium from aqueous solution using switchgrass biochar produced via hydrothermal carbonization process. J Environ Manage 109: 61-69. doi: 10.1016/j.jenvman.2012.04.047
![]() |
[255] |
Méndez A, Gómez A, Paz-Ferreiro J, et al. (2012) Effects of sewage sludge biochar on plant metal availability after application to a Mediterranean soil. Chemosphere 89: 1354-1359. doi: 10.1016/j.chemosphere.2012.05.092
![]() |
[256] |
Laird DA, Fleming P, Davis DD, et al. (2010) Impact of biochar amendments on the quality of a typical Midwestern agricultural soil. Geoderma 158: 443-449. doi: 10.1016/j.geoderma.2010.05.013
![]() |
[257] |
Steinbeiss S, Gleixner G, Antonietti M (2009) Effect of biochar amendment on soil carbon balance and soil microbial activity. Soil Biol Biochem 41: 1301-1310. doi: 10.1016/j.soilbio.2009.03.016
![]() |
[258] |
Laird DA, Brown RC, Amonette JE, et al. (2009) Review of the pyrolysis platform for coproducing bio-oil and biochar. Biofuels Bioprod Biorefining 3: 547-562. doi: 10.1002/bbb.169
![]() |
[259] |
Méndez A, Gómez A, Paz-Ferreiro J, et al. (2012) Effects of sewage sludge biochar on plant metal availability after application to a Mediterranean soil. Chemosphere 89: 1354-1359. doi: 10.1016/j.chemosphere.2012.05.092
![]() |
[260] | Hansen S, Mirkouei A (2019) Prototyping of a Laboratory-scale Cyclone Separator for Biofuel Production from Biomass Feedstocks Using a Fused Deposition Modeling Printer, The Minerals, Metals & Materials Society (TMS) Conference Proceedings. |
[261] | Hansen S, Mirkouei A, Xian M (2019) Cyber-Physical Control and Optimization for Biofuel 4.0.Proc 2019 IISE Annu Conf H.E. Romeijn, A Schaefer, R. Thomas, eds. |
[262] |
Tripathi M, Sahu JN, Ganesan P (2016) Effect of process parameters on production of biochar from biomass waste through pyrolysis: A review. Renew Sustain Energy Rev 55: 467-481. doi: 10.1016/j.rser.2015.10.122
![]() |
[263] |
Bis Z, Koby\lecki R, Ścis\lowska M, et al. (2018) Biochar-Potential tool to combat climate change and drought. Ecohydrol Hydrobiol 18: 441-453. doi: 10.1016/j.ecohyd.2018.11.005
![]() |
1. | Faiz Ul Islam, Guangjie Liu, Jiangtao Zhai, Weiwei Liu, VoIP Traffic Detection in Tunneled and Anonymous Networks Using Deep Learning, 2021, 9, 2169-3536, 59783, 10.1109/ACCESS.2021.3073967 | |
2. | Faiz Ul Islam, Guangjie Liu, Weiwei Liu, Qazi Mazhar ul Haq, A deep learning‐based framework to identify and characterise heterogeneous secure network traffic, 2023, 17, 1751-8709, 294, 10.1049/ise2.12095 | |
3. | Xin Tong, Changlin Zhang, Jingya Wang, Zhiyan Zhao, Zhuoxian Liu, Dark-Forest: Analysis on the Behavior of Dark Web Traffic via DeepForest and PSO Algorithm, 2023, 135, 1526-1506, 561, 10.32604/cmes.2022.022495 | |
4. | Panpan Zhao, Gaopeng Gou, Chang Liu, Yangyang Guan, Mingxin Cui, Gang Xiong, 2021, Chapter 21, 978-3-030-90018-2, 418, 10.1007/978-3-030-90019-9_21 | |
5. | Liu Chaoqun, Luo Jie, Dong Bin, 2022, Research on Voip-Over- Vpn s Classification And Mitigation, 978-1-6654-5174-1, 63, 10.1109/ICIM56520.2022.00019 | |
6. | S. Ishkuvatov, Method and Algorithm for Determining the Type of Traffic in an Encrypted Communication Channel, 2023, 8, 2712-8830, 109, 10.31854/1813-324X-2022-8-4-109-118 | |
7. | Baoqiang Zhong, Yikai Tan, Wei Wang, Weihong Cui, Zehan Cai, Lai Wei, Qian He, 2022, Research on the identification of network traffic anomalies in the access layer of power IoT based on extreme learning machine, 978-1-6654-6287-7, 374, 10.1109/AIIPCC57291.2022.00085 | |
8. | Setiyo Budiyanto, Dadang Gunawan, Comparative Analysis of VPN Protocols at Layer 2 Focusing on Voice Over Internet Protocol, 2023, 11, 2169-3536, 60853, 10.1109/ACCESS.2023.3286032 |
Traffic Labels | Content Type | |
VPN-VoIP | Skype, VoIPbuster, Hangouts and Facebook voice calls for the duration of 1hour | |
VPN-Non_VoIP | P2P | μTorrent and Bittorent |
Chat | AIM, Skype, Facebook, Hangouts, ICQ | |
SMPTS, IMAPS, POP3S | ||
Streaming | Netflix, Vimeo, YouTube and Spotify | |
File Transfer | FTPS, SFTP and Skype |
Abbreviation | Description | |
1 | feduration | Duration of the flow in Microsecond |
2 | max_fpktl | Maximum size of packet in forward direction |
3 | std_fpktl | Standard deviation size of packet in forward direction |
4 | max_bpktl | Maximum size of packet in backward direction |
5 | std_bpktl | Standard deviation size of packet in backward direction |
6 | tot_fhlen | Total bytes used for headers in the forward direction |
7 | tot_bhlen | Total bytes used for headers in the backward direction |
8 | init_win_bytes | The total number of bytes sent in initial window in the backward |
_backward | direction | |
9 | fpkts_persec | Number of flow packets per second |
10 | min_flowiat | Minimum inter-arrival time of packet |
11 | tot_fiat | Total time between two packets sent in the forward direction |
12 | tot_biat | Total time between two packets sent in the backward direction |
13 | min_biat | Minimum time between two packets sent in the backward direction |
14 | max_active | Maximum time a flow was active before becoming idle |
15 | min_active | Minimum time a flow was active before becoming idle |
16 | max_idle | Maximum time a flow was idle before becoming active |
17 | min_idle | Minimum time a flow was idle before becoming active |
Datasets | VPN-VoIP and VPN-Non_VoIP instances | ||
(based on FLP) | (Xtrain) | (Xtest) | (XTotal) |
D1 | 16124 | 7212 | 23336 |
D2 | 12102 | 5326 | 17428 |
D3 | 10799 | 4905 | 15704 |
Euclidean distance | D(y,y′)=√(y1−y′1)2+(y2−y′2)2+...+(yn−y′n)2 |
Manhattan distance | D(y,y′)=K∑m=1|ym−ym′| |
Minkowski distance | D(y,y′)=b√(K∑m=1|ym−ym′|b) |
Traffic Labels | Content Type | |
VPN-VoIP | Skype, VoIPbuster, Hangouts and Facebook voice calls for the duration of 1hour | |
VPN-Non_VoIP | P2P | μTorrent and Bittorent |
Chat | AIM, Skype, Facebook, Hangouts, ICQ | |
SMPTS, IMAPS, POP3S | ||
Streaming | Netflix, Vimeo, YouTube and Spotify | |
File Transfer | FTPS, SFTP and Skype |
Abbreviation | Description | |
1 | feduration | Duration of the flow in Microsecond |
2 | max_fpktl | Maximum size of packet in forward direction |
3 | std_fpktl | Standard deviation size of packet in forward direction |
4 | max_bpktl | Maximum size of packet in backward direction |
5 | std_bpktl | Standard deviation size of packet in backward direction |
6 | tot_fhlen | Total bytes used for headers in the forward direction |
7 | tot_bhlen | Total bytes used for headers in the backward direction |
8 | init_win_bytes | The total number of bytes sent in initial window in the backward |
_backward | direction | |
9 | fpkts_persec | Number of flow packets per second |
10 | min_flowiat | Minimum inter-arrival time of packet |
11 | tot_fiat | Total time between two packets sent in the forward direction |
12 | tot_biat | Total time between two packets sent in the backward direction |
13 | min_biat | Minimum time between two packets sent in the backward direction |
14 | max_active | Maximum time a flow was active before becoming idle |
15 | min_active | Minimum time a flow was active before becoming idle |
16 | max_idle | Maximum time a flow was idle before becoming active |
17 | min_idle | Minimum time a flow was idle before becoming active |
Datasets | VPN-VoIP and VPN-Non_VoIP instances | ||
(based on FLP) | (Xtrain) | (Xtest) | (XTotal) |
D1 | 16124 | 7212 | 23336 |
D2 | 12102 | 5326 | 17428 |
D3 | 10799 | 4905 | 15704 |
Euclidean distance | D(y,y′)=√(y1−y′1)2+(y2−y′2)2+...+(yn−y′n)2 |
Manhattan distance | D(y,y′)=K∑m=1|ym−ym′| |
Minkowski distance | D(y,y′)=b√(K∑m=1|ym−ym′|b) |