The Emergence of Experience Packages in the 5G Era

By Eirini Liotou, Nikos Passas, and Lazaros Merakos, National and Kapodistrian University of Athens

IEEE 5G Tech Focus: Volume 1, Number 3, September 2017


Handling Quality of Experience as a mean opinion score is not sufficient to meet the requirements of the 5th generation of communication networks. In this direction, the concept of “experience package” emerges. Experience packages may be configured and delivered in a way that fine-grained differentiation is achieved, respecting the user, application and communication context.

1. QoE’s role in 5G

Quality of Experience (QoE) is defined as “the degree of delight or annoyance of the user of an application or service” and is influenced by three main pillars: the human factor, the system technical characteristics, and the context of use [1]. The inherent subjectivity in this definition explains the exploding interest in QoE research from multiple disciplines. In this article, we address QoE from a telecom operator’s perspective, providing insights towards improving QoE in the next, i.e. 5th generation of communications. It is expected that 5G will be more user-centric, designed to deliver consistent experience to the users, giving them the perception of “infinite capacity” [2].

In this direction, QoE intelligence is invaluable to telecom operators, since it can: a) Enable customer experience management through QoE-oriented data analytics (e.g. automate service configuration, facilitate self-care and self-diagnosis), b) Drive business operations and build more meaningful SLAs, c) Decrease churn, by comprehending end-users’ and applications’ requirements and controlling the network accordingly (namely, avoid under-engineering), and finally, d) Increase network efficiency through identifying and exploiting the non-linear relationships between QoS and QoE (namely, avoid over-engineering) (Figure 1).

2. The concept of experience packages

QoE should be provisioned in a way that end-users create the impression that the best experience “always follows them”, regardless a) the application they use, b) the communication context in which they are currently involved, and c) the current network conditions. What is more, users are not uniform in their QoE expectations or requirements; Their satisfaction with a service is the result of their psychology, cognitive and psychophysical characteristics, and current state [3]. Nevertheless, so far, mobile cellular networks do not allow a fine-grained differentiation in the offered experience, since:

  • All users are represented by an “average user”;
  • Most applications are treated as best effort;
  • The context of communication (e.g. task urgency, environment, billing, etc.) are not actually taken into consideration in service provisioning.

Nevertheless, a rational development in the current communication paradigm as well as a key for reaching 5G requirements is the support of user personalization, application differentiation and quality adjustment based on the current communication scenario. User personalization is currently performed by some service providers, who distinguish gold/silver/bronze users based on subscription profiles, and configure the offered quality accordingly. However, telecom operators do not really engage in such a per-user differentiation on a monetary basis, at least for the time being.

Furthermore, if we move one step further, we can consider QoE not only as a single Mean Opinion Score (MOS), but more generally as an “experience package”. For instance, we could easily claim that a user is not only interested in receiving the best quality, but may be equally (or more) interested in communicating in an energy-efficient way through his/her device, in minimizing the charges imposed when using a service, in being prioritized for a specific task with respect to another, in being served in the securest way possible, or in combinations of the previous. Similarly, other dimensions can be integrated into an experience package.

3. Experience-package provisioning scheme

Capitalizing on this observation, we envision future architectures, where the network builds “experience packages” based on actual communication scenarios and user profiles. We assume that such profiles are built based on the users’ communication habits, preferences, mobility patterns, physical environment, used equipment, etc., while a group of users will fit a specific profile. These profiles can be built offline, based on accumulated information about the users, while they will be updated in case of different or unusual user behavior. Once such a “pool of profiles” is deduced and becomes available at the operator’s side, then configuring (i.e. tuning) the offered “experience package” during any communication session is a 4-step process:

  • Step 1: Match a user to a suitable profile based on demographics, preferences, subscription types or other factors that are meaningful and measurable;
  • Step 2: Derive the application unique characteristics and requirements, such as tolerability and adaptability to various network conditions;
  • Step 3: Deduce the context of the current communication session (based on insights from the past, sensed environment, device info, etc.);
  • Step 4: Finally, build an “experience package” in real-time upon service delivery, which will remain valid throughout the user’s session. The package building decision is inevitably a compromise between the requirements derived from steps 1-3 (i.e. demand) and the actual capabilities of the network in terms of momentary available resources (i.e. supply).


Therefore, as depicted in Figure 2, a network’s decision about how to build an experience package is a function of i) the user profile, ii) the specific application requirements, iii) the current context, and on the other side, iv) the current network state and capabilities. The latter includes any restrictions or limitations in the network, such as resource availability, current load, energy constraints, operator policies, etc. The experience packages may be built at a central node within the operator’s core network, while the 3GPP Policy and Charging Rules Function (PCRF) can be considered as a potential host of this procedure, as it is already responsible for the creation of rules and policies for each subscriber [4]. Similarly, the Policy and Charging Enforcement Function (PCEF) and the Access Network Discovery and Selection Function (ANDSF) can be considered as candidates for the enforcement of these packages to the end-users.

An experience package does not refer to entities sent to the users; it is a reference of how service provisioning should be adjusted per user’s session in order to provide special treatment. In that sense, it reminds of the QoS-centric “bearer” concept of LTE [5], while it embraces much broader QoE-centric intelligence with respect to the user, application and context. As a result, a user will perceive a much more personalized and friendly experience, tailored to his/her specific session needs.

As an example, we consider experience packages as a weighted sum of the following three dimensions: {QoE, price, energy cost}, while this sum is subject to the actual network capabilities, i.e. 

In Figure 3, we abstractly present five experience packages that derive from matching these three weighted dimensions to the actual network capabilities (note that, for simplicity, context and application awareness are ignored). We can observe various possible configurations: Delivering experience package “1” puts more emphasis (weight) on QoE, so it represents a situation where a user is mostly interested in quality, regardless of the price- or energy-to-pay. Similarly, packages “2” and “3” represent users who care more about charges and energy costs, respectively, at the expense of quality. Finally, packages “4” and “5” imply some trade-offs between all dimensions, subject to the current network conditions. In real-life, package “2” could lead to a reduction of the video quality layers in an HTTP Adaptive Streaming session in order to reduce data consumption and subsequent charges, while package “3” could handover a user’s uplink to the closest access-point (e.g. femto- instead of macro-base station) to reduce the device’s transmitted power.

4. Technological enablers

To enable the creation and empowerment of experience packages in a network, various enablers can be called forth. First, a QoE management entity is required that implements the functions of QoE collection, monitor and management, that help evaluate and control the users’ QoE [3]. In terms of QoE evaluation, traditionally used models such as the E-model [6] need to be confirmed or revisited (e.g. regarding the model’s default values and permitted ranges), while new QoE estimation models for the emerging services towards the 5G era, such as immersive and 360° video, need to be created.

Apart from this, network slicing appears to play an important role [7]. The concept of slicing has been proposed towards 5G as a way to handle the plethora of verticals that are integrated into this new ecosystem. Slicing per vertical can be seen, however, as a coarse-grained solution. We can however envision a more fine-grained slicing concept, that is slicing per user, application and context. This would imply that two neighbor users might be served by different softwarized-eNBs or softwarized-EPCs, on the basis of successfully delivering each one’s experience packages. These decisions will be driven by the creation of a fine-grained slice per user flow.

Resource allocation mechanisms also need to evolve in order to serve the experience packages’ provisioning. Thinking out of the box, we may envision “elastic resources” as a superset of spectrum, processing power, memory, energy, etc., namely of any dimensions that build an experience package. Then, through the virtualization of elastic resources in combination with an abstraction of the wireless medium, flexible experience package provisioning could be enabled. However, technical feasibility and challenges need to be addressed first.


  1. Qualinet COST Action IC 1003, “White Paper on Definitions of Quality of Experience”, 2012.
  2. NGMN Alliance, “5G White Paper”, 2015.
  3. Liotou, D. Tsolkas, N. Passas and L. Merakos, “Quality of Experience Management in Mobile Cellular Networks: Key Issues and Design Challenges”, IEEE Communications Magazine, vol. 53, no. 7, pp. 145-153, July 2015.
  4. 3GPP TS 23.203: “Policy and Charging Control Architecture”.
  5. Ekstrom, “QoS Control in the 3GPP Evolved Packet System”, IEEE Communications Magazine, vol. 47, no. 2, pp. 76-83, Feb. 2009.
  6. ITU-T Recommendation G.107: “The E-Model, a computational model for use in transmission planning”.
  7. Richart, J. Baliosian, J. Serrat, and J.-Luis Gorricho, “Resource Slicing in Virtual Wireless Networks: A Survey”, IEEE Transactions on Network and Service Management, vol. 13, no. 3, pp. 462-476, Sept. 2016.

Eirini Liotou received the Diploma in Electrical & Computer Engineering from the National Technical University of Athens in 2006. She obtained the MSc in New Technologies of Informatics & Telecommunications from the University of Athens and the MSc in Communications & Signal Processing from the Imperial College of London. She has worked as a Software Engineer in Siemens AG and as a Senior Software Engineer in Siemens Enterprise Communications within the R&D department. Currently she is a Researcher in the Department of Informatics & Telecommunications in University of Athens. Her research is focused on SDN-enabled QoE provisioning in 4G networks and beyond. 


Nikos Passas received his Diploma (honors) from the Department of Computer Engineering, University of Patras, and his Ph.D. degree from the Department of Informatics and Telecommunications, University of Athens, in 1992 and 1997, respectively. From 1992 to 1995 he was a research engineer at the Greek National Research Center “Demokritos”. Since 1995, he has been with the Communication Networks Laboratory of the University of Athens, working as a sessional lecturer and senior researcher in a number of national and European research projects. He has also served as a guest editor and technical program committee member in prestigious magazines and conferences, such as IEEE Wireless Communications Magazine, IEEE Vehicular Technology Conference, IEEE Globecom, etc. He has published more than 100 papers in peer-reviewed journals and international conferences. 

 Lazaros Merakos received the Diploma in Electrical and Mechanical Engineering from the National Technical University of Athens, Athens, Greece, in 1978, and the M.S. and Ph.D. degrees in Electrical Engineering from the State University of New York, Buffalo, in 1981 and 1984, respectively. From 1983 to 1986, he was on the faculty of the Electrical Engineering and Computer Science Department, University of Connecticut, Storrs. From 1986 to 1994, he was on the faculty of the Electrical and Computer Engineering Department, Northeastern University, Boston, MA. During the period 1993–1994, he served as Director of the Communications and Digital Processing Research Center, Northeastern University. During the summers of 1990 and 1991, he was a Visiting Scientist at the IBM T. J. Watson Research Center, Yorktown Heights, NY. In 1994, he joined the faculty of the University of Athens, Athens, Greece, where he is presently a Professor in the Department of Informatics and Telecommunications, and Scientific Director of the Networks Operations and Management Center.

Editor: Anwer Al-Dulaimi 



Pageviews: 8033

Subscribe to Tech Focus

Join our IEEE Future Networks Technical Community and receive IEEE Future NetworksTech Focus delivered to your email.

Subscribe Now

Article Contributions Welcome


Submit Manuscript

Submit Manuscript via Track Chair

Author guidelines can be found here

Other Future Networks Publications 

Archived Issues 

Contact Us

IEEE Future Networks Tech Focus Editorial Board

Rod Waterhouse, Editor-in-Chief
Mithun Mukherjee, Managing Editor
Imran Shafique Ansari
Anwer Al-Dulaimi
Stefano Buzzi
Yunlong Cai
Zhi Ning Chen
Panagiotis Demestichas
Ashutosh Dutta
Yang Hao
Gerry Hayes
Chih-Lin I
James Irvine
Meng Lu
Amine Maaref
Thas Nirmalathas
Sen Wang
Shugong Xu
Haijun Zhang