Internet Engineering Task Force X. W. Zhou Internet-Draft Z. M. Cheng X. N. Li Intended status: Informational U. S. T. B Expires: November 9 , 2014 May 9 , 2014 Network Technology- Wisdom Network draft-xwzhou-ntwn-00.txt Abstract In this paper, a new form of network technology with wisdom, the wisdom network, is presented, defined and described. It is a collaborative network; it can distinguish, judgment;it can process information resources into knowledge, achieve mastery of knowledge; it can self- manage, self-repair and self-adapt; it can predict the future changes of network environment and people's emotional state. Network can self- learn,self-grow and self-innovate. The network has humanChengke abiChengty of observation,understanding people's emotions and intentions. On the base of the definition of wisdom network, we describe the basic characteristics and architecture of it, and detailedly depict wisdom framework of Wisdom network, finally, we propose a method to implement wisdom network, namely, multi-agent technology. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119]. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http://datatracker.ietf.org/drafts/current/. zhou, et al. Expires November 9, 2014 [Page 1] Internet-Draft Wisdom Network May 2014 Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on November 9, 2014. Copyright Notice Copyright (c) 2013 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD license text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the SimpChengfied BSD license. Table of Contents 1. Terminology . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 3. Basic Concepts of Wisdom Network . . . . . . . . . . . . . . 3 3.1. Architecture of Wisdom Network . . . . . . . . . . . . . 5 3.2. Wisdom Framework of Wisdom Network . . . . . . . . . . . 6 4. Implementation Scheme of Wisdom Network. . . . . . . . . . . 7 5. Major Technical Challenges of Wisdom Network . . . . . . . . 7 6. Security Considerations. . . . . . . . . . . . . . . . . . . 9 7. IANA Considerations. . . . . . . . . . . . . . . . . . . . . 9 8. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . 9 9. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 9 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 9 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 10 zhou, et al. Expires November 9, 2014 [Page 2] Internet-Draft Wisdom Network May 2014 1. Terminology Generic Routing Encapsulation (GRE) [RFC2784] can be used to carry any network layer protocol over any network layer protocol. GRE has been implemented by many vendors and is widely deployed on the Internet. [RFC2784], by design, does not describe procedures that affect fragmentation. Lacking guidance from the specification, vendors have developed implementation-specific fragmentation strategies. For the most part, devices implementing one fragmentation strategy can interoperate with devices that implement another fragmentation strategy. Operational experience has demonstrated the relative merits of each strategy. Section 3 of [RFC4459] describes four fragmentation strategies and evaluates the relative merits of each. 2. Introduction From the post house and beacon tower of ancient china to the modern information network, we have felt the rapid development of information technology. On the other hand, in order to meet the different needs and effective transmission of information in different environments, we hope information network have wisdom like people. Now there have been various intelligent networks. For example, in 2005, based on cognitive radio, Thomas proposed Cognitive Network. A cognitive network has a cognitive process that can perceive current network conditions,and then plan, decide and act on those conditions. It has the adaptive capacity of the network environment and learning abiChengty form the evaluation of previous and future decision-making, all while takes into account end-to-end goals-network targets [1]. Although these information networks are more and more intelligent, the information network has distance for our ideal network which is like human beings with wisdom. Then we can not help but ask what is "wisdom" of the information network? Information network with "wisdom" should be a what kind of network? How to realize the information network with "wisdom"? In this article, we give our ideas of these issues. zhou, et al. Expires November 9, 2014 [Page 3] Internet-Draft Wisdom Network May 2014 3. Basic Concepts of Wisdom Network Wisdom network, which is a collaborative network, can fast,flexibly and appropriately distinguish, judgment and choose the information of person's emotional state and network status form a lot of sense information, then processes these information resources into knowledge combined with the information of the user's application requirement, achieves mastery of knowledge, and ultimately finds the right decision to adjust the current network configuration and predicts the future changes of network environment and user emotional state. These actions make network flexibly adapt to changes in network environment, properly response to the changes of environment that the network will face with and future changes in users emotional state, and in the meanwhile provide innovative services for users. Network in the process can self-learn, self-grow and self-innovate. "Wisdom" of wisdom network is that it knows all the "network" capabilities and "network" resources usage condition, performances and security capabilities of "network", and analyzes, judges, predicts changes in the network environment; It uses the most reasonable and best way to co-configure and use the "network",and makes the network adapt the changes in environment; it makes operators obtain highest return from the most rationally utilization of resources, and lets users get best service from the most reasonable price; in addition, it endows the network with the humanlike ability of observation, understanding people's emotions and intentions, reflects ideas of the integration of people and network; meanwhile, it makes the network with the ability of self-learn, growth and innovation. Wisdom network has the following characteristics: It has more complete behavioral consciousness, control capability and collaboration capability; It has intelligent perception, context-aware and wisdom ability; It has mature information-knowledge-wisdom conversion mechanism, and capability of judge, analyze and decision make; It has the capability of self-learning,self-growth and self-innovation; Uncertainty of the network's living environment comes up with new requests for the zhou, et al. Expires November 9, 2014 [Page 4] Internet-Draft Wisdom Network May 2014 development of network architecture. The architecture which possesses predictable and adaptive trait is urgently needed, it make the network better sustain architecture ability of the self-management, self-repair and self-adapt in the changing living environment, that is to say, the network has "future eye", automatically adjusts its parameters to cope with the future changes; It can perceive,recognize and understand human emotions, and make smart, sensitive and friendly response for human emotions, that is, we endow the network with humanlike ability of observe,understand user emotional characteristics, in the meantime, it can predict the emotional intent behind the change based on trend of changes in the user emotional state. Wisdom loop has six parts. There are sense, analyze judge and infer predict,decide, act, learn, grow and innovate, policy. Wisdom network obtains sensing datum of ambient and users emotional state through the sensors, it uses these sensing datum to judge, analyze, infer and forecast, and make future decisions for decision-making module. Based on information, the analysis, judgment and inference and forecasting module obtains the signals of physical and behavioral characteristics caused by human emotions, analyzes the relationships among human emotions and a variety of sensing signals, establishes emotional model; and this module also deals with status information which affect the performance of end to end transmission, namely, the network type, network topology, available resources,interface protocols, network traffic, and network error rate, node residual energy,data rate, end to end delay and network throughput etc; combining with actions that strategic module may take, this module determines whether the current network meets user requirements, if not, then it would take appropriate re-configured measures to ensure that the network meets user requirements. Decision-making module decides to take corresponding actions based on the previous study and the result of analysis judgments inference prediction. Action module is responsible for taking actions(reconfiguration) made by the decision-making module. Learning, growth and innovation module is the core of wisdom loop, the network can learn, grow, innovate in the dynamic adaptive process of sense-reasoning-predictable-decision-action", gains experiences and knowledge in process of learning, growth and innovation, and uses its experiences and knowledge to master knowledge and analyze judge, infer and forecast and decision-make in the future. zhou, et al. Expires November 9, 2014 [Page 5] Internet-Draft Wisdom Network May 2014 Basic Characteristics and Architecture of the Wisdom Network Besides self-perception, self-management, self-learn, self-optimization, self-heal and selfconfiguration[6], wisdom network has following characteristics: Perceive of people's emotional state: in addition to actively perceiving its own behavior, condition and environment, wisdom network also acquires signals of physiological and behavioral characteristics caused by the user emotions through a variety of sensors, then establishes the "emotional model". We hope the wisdom network that has the ability of perception, identify and understand human emotions,and can make intelligent, sensitive, friendly response to the users; our goal is to achieve shorten the distance between user and network and create a truly harmonious environment for users. Self-grow and self-innovate: self-growth is that the network can arm itself with knowledge which is learned from the learning and innovative process and stored in the Knowledge Base; self-innovation is that the network provides innovative services for users on the basis of comprehensive analysis of network status, environment,users emotional state and its knowledge. Self-predict: According to the trend of changes in environment and people emotional state, the network makes prediction and judgment, so that the network can control impending events. It mainly reflects the network can cognize and grasp the future changes of environment and user emotional, achieve proactively self-management and reduce manual intervention. Architecture and Wisdom Framework of Wisdom Network 3.1. Architecture of Wisdom Network In this section, we give the architecture of wisdom network. User (Data) plane: it has data information transfer logical function, the date information are network environment, network status, application requirements and emotional state. zhou, et al. Expires November 9, 2014 [Page 6] Internet-Draft Wisdom Network May 2014 Control plane: it has signaling control information logic function, the signal of data transmission relate to data information. Wisdom plane: it provides a complete view of the information of entire network, and processes data obtained from the user (data) plane into network knowledge in order to guide adaptive control of the control plane; it delivers, stores, processes, transmits, analyzes and judges information of perception for network environment, application requirements and user emotional state; it provides adequate reference information for infering and forecasting the decisions and actions taken by network; it analyzes and judges emotional changes according. to perceptive information, forms predictions, takes network adjustments, makes real-time feedback to the current operation, at the same time, also forms new prediction for the intent behind emotional changes, activates its knowledge base, timely and initiatively provides new information for users; it supplies knowledge for self-learning,self-growth and self-innovation; in addition, it possesses information-knowledge-wisdom sophisticated conversion mechanism, establishes unified protocol description language knowledge representation and efficient integration of business; wisdom is higher than the cognitive, so the wisdom plane possesses more advanced functions, such as analyze, judge, predict, grow and innovate ect, therefore, it can effectively resolve integration of heterogeneity and cooperativity of networks, achieve resource sharing among nodes and make their respective advantages complementary to each other, more rationally and efficiently use network resources. 3.2. Wisdom Framework of Wisdom Network In this section, we give the wisdom framework of wisdom network. Application layer: it is responsible for providing network services to application, and providing perceptive information of user's application requirements and emotional state for wisdom processing layer. Wisdom process layer: it has a flexible information-knowledge-wisdom mechanism. After handled by affective computing platform, the perceptive information of user's application requirements and emotional state are delivered to the wisdom processing layer. The wisdom process layer accurately obtains changes of emotional state and application requirement by analyzing, judging and inferring; in the meantime, it accurately knows zhou, et al. Expires November 9, 2014 [Page 7] Internet-Draft Wisdom Network May 2014 information of network status via network state sensor, including all the "network" capabilities and "network" resource usage, "network" performance and security capabilities. After processing the above information together, it makes decision-making of network configuration, responds to application requirements and changes in emotional state. It processes these successful decision-making information into knowledge, then the knowledge are stored in its knowledge base. In addition, on base of gaining experience and current information of user emotions status and network status, it predicts changes that the network will face, in the same time, processes these information into knowledge, then the knowledge are also stored in its knowledge base for providing reference for the future decision-making and strategy. It can learn during processing, the learned knowledge is also stored in the knowledge base for growth and mastery and innovation. This layer makes network accurately analyze, judge, infer, forecast, grow and innovate. Learning, growth and innovation are in order to provide quality services. It also has automatically filter and abstract useful information from a lot of perceptive information, then processes these useful information into knowledge, then combining with its knowledge, it masters these knowledge for obtaining optimal decision-making and strategy, at same time stores these knowledge in the knowledge base. Software adaptive networks: the functions of this layer are similar to cognitive network, it is no longer introduced . 4. Implementation Scheme of Wisdom Network Multi-agent technology is a very important application field in Artificial intelligence, it has the distributed characteristics. Multi-agent systems can accomplish one or more tasks together, and are commonly used to deal with complex environment and non-deterministic problems [7]. As the wisdom network has complexity and uncertainty of network environment and user emotional state, so we can implement wisdom network by borrowing ideas from multi-agent technology. Learning from multi-agent technology. The above agent model is constituted by observe, perceive, act, analyze, judge, infer, predict and knowledge base. Analysis, judgment, inference, prediction combine with the knowledge base achieve reason, grow and innovate. We can see that it contains the main part of the wisdom network. zhou, et al. Expires November 9, 2014 [Page 8] Internet-Draft Wisdom Network May 2014 Wisdom processing layer can be considered to be formed by multiple agents. These agents mutually cooperate in the process of analysis, judgment, inference, prediction, learn, growth and innovation, jointly complete the "emotional model" of user's emotional characteristics,guide network automatically configure its various parameters and adapt to the changing environment or application. 5. Major Technical Challenges of Wisdom Network Wisdom network, which compares to the cognitive network [1] and ubiquitous network [6], faces the following major technical challenges: Affective Computing is the computing that relates to, arises from, or influences emotions. The goal of it is an attempt to create a computing system which can perceive, recognize and understand human emotions, can make smart, sensitive,friend response to human emotions, that is, endow computer with humanlike ability of observing, understanding and generating a variety of emotions characteristics[8]. We hope the wisdom network that can have humanlike "brain" thinking and understanding of human emotion, it should be able to identify the user's emotional state and aware of people's feelings changes in the interactive process with human. But people's feelings change is fluctuating[9], then how to use perceptive data of people's emotional state to build an appropriate "emotional model", and forecast for user's intention of emotional changes, these are the challenges of the wisdom network. Predictive self-adaptation technology: it can predict user future behaviors and changes of network environment, adjust system properties to adapt to the new environment. Combining its ability of observations and cognitive, it uses rational policy to achieve adaption. There are two challenges: one is how to build people's emotional model and network environment model which can learn from environment; the other is how to solve potential conflicts, when considering problems of multiple users. Knowledge representation: the wisdom network should have its own knowledge, and knowledge must to be expressed in the form of information which are understood by Wisdom process layer,so that the wisdom network is able to analyze, judge, infer, forecast, learn, grow and innovate. Clark, who first proposed the concept of knowledge plane, believed "Knowledge Plane which dealt with knowledge sharing, a pervasive system within the network that used cognitive information processing to build a self-managing network"[10]. zhou, et al. Expires November 9, 2014 [Page 9] Internet-Draft Wisdom Network May 2014 Wisdom plane possesses local and overall knowledge of network and network elements. Knowledge in different fields are different, and their representations are also different. How to find a unified knowledge representation method is our facing problem. 6. Security Considerations Security can not be considered in the emergency command and dispatch communication software. 7. IANA Considerations This document does not have any implications for IANA. 8. Conclusion This paper has presented, defined and described a new form of network technology with wisdom, namely, wisdom network. It is a collaborative network; it can distinguish, udgment;it can process information resources into knowledge, achieve mastery of knowledge; it can self-manage, self-repair and self-adapt in the changing living environment; it can predict the future changes of network environment and people's emotional state. Network in the process can self-learn, self-grow and self-innovate. The network has humanlike ability of observation, understanding people's emotions and intentions. On the base of the definition of wisdom network, we describe the basic characteristics and architecture of it, and detailedly depict wisdom framework of it, finally, we propose a method to implement it, namely, multi-agent technology. 9. Acknowledgements This work is supported by the Project supported by the Foundation for Key Program of Ministry of Education, P. R. China(No.311007), National Science Foundation Project of P. R. China (No. 60903004, 60902042, 61170014), National Science and Technology Key Projects (No. 2011 ZX03003-002-03) and the National Research Foundation for the Doctoral Program of Higher Education of P. R.China under Grant (No. 20090006110014). zhou, et al. Expires November 9, 2014 [Page 10] Internet-Draft Wisdom Network May 2014 10.References [1]Thomas. R.W, DaSilva. L.A, MacKenzie. A. B. Cognitive Networks [J]. Proc. IEEE Dyspan2005 [C] , 2005, pp: 352-360. [2]Qicui Gan, J. Chris Harreld, Yiwei Jiang, Yuheng Cheng, Jianjun Zhao. The Smart Planet will win in China, 2008. [3]International Telecommunication Union UIT. ITU Internet Reports 2005: The Internet of Things [R], 2005. [4]Gustavorg Mariom O, Carlos D K, Early infrastructure of all Internet of Things in Spaces for Learning [C]. Eighth IEEE International Conference on Advanced Learning Technologies, 2008, pp:381-383. [5]Amardeo C, Sarma J G, Identities in the Future Internet of Things [J]. Wireless Pers Commun, Vol. 49, pp: 353-363, 2009. Author's Addresses Xianwei Zhou Department of Communication Engineering School of Computer & Communication Engineering University of Science & Technology Beijing, Beijing, P.R. China E-mail address: zilengqier@sohu.com zhou, et al. Expires November 9, 2014 [Page 11]