09 August 2018

The bigger problem you solve, the bigger business you have


Nothing more true than this quotation.

The tremendous opportunities created by the technology advances in IT, ICT, A.I. ... must be directed to solve real socio-economic and cultural bottlenecks and problems.  

Quality of life will improve, overall develpment will further diffuse while making business.Question is agreeing on and taking care seriously of the real problems of the world...


At this link the "The 10 most critical problems in the world, according to millennials". 

03 May 2018

Multi Access Edge Computing: Telcos cooperation in IaaS vs PaaS scenarios


It was 2013 when IEEE Com. Magazine (2013) published my paper “Clouds of Virtual Machines in Edge Networks”, a first seminal work on Edge Computing. Paper claimed (for the first time to my knowledge) the advantages of bringing the Cloud model towards the edge of a SDN-NFV network.

In 2014 ETSI released an introductory technical white paper on MEC, at that time meaning Mobile Edge Computing, then renamed  Multi Access Edge Computing (MEC). 

Technical speaking, MEC can be seen as an extension of the Cloud Computing paradigm towards the edge (i.e., aggregation and access segments) of the Telecommunications Networks. 

As a matter of fact, the use of IT resources (computing, storage/memory and networking), allocated at the edge of the infrastructure, can bring a number of advantages, for example: to improve QoS/QoE by reducing the network latencies, to reduce costs in the Cloudification of the infrastructure towards 5G, to enable new service/biz models, etc:

In fact, in the medium-long term, it is likely that the network infrastructures will be composed by a physical layer (i.e., IT and network hardware and physical links) hosting dynamic software platforms executing millions of software processes, implementing both network and services components/functionalities (e.g., VNFs, Virtualized Network Functions).

In these scenario, MEC aims at complementing Cloud Computing (it is not replacing it, obviously…): for example, the so-called “slices” of the network infrastructure (e.g. future networks, 5G) may integrate both Cloud Computing and MEC resource, requiring this orchestration capabilities spanning across the overall infrastructure.

Today, Operators are exploring different strategies for MEC adoption, motivated by:

  • costs savings in the Cloudification (SDN-NFV) of the infrastructure: for example  using MEC for deploying smaller Central Offices at the edge (for example., Cloud CO inititative of Broadband Forum);
  • revenues generation.


Regarding the latter, among the various approaches and biz models for revenue generations, decoupling MEC IaaS vs PaaS appear to enable cooperation between Telcos, who can join forces to boost the development of multi-domains open/ecosystems (e.g., for V2X, Industry 4.0, etc.).

In particular, decoupling MEC IaaS vs PaaS means:

  • Telcos deploy MEC servers (e.g. Cloudlets) for providing infrastructure services (i.e., MEC IaaS);
  • Third Parties (or also Telcos, themselves) deploy a MEC software platform/framework for providing platform services (i.e., MEC PaaS);




To achieve this decoupling, and to allow different companies to develop and to interwork,  it is crucial well-defining what's MEC IaaS, MEC PaaS and standardizing interfaces between MEC IaaS and PaaS.  

In the menawhile, a numer of initiatives on MEC, in general Edge Computing, are emerging and flourishing, such as this one: https://www.akraino.org/ 

15 February 2018

When Will Networks Have Common Sense? Generative Adversarial Networks are on the way…


Generative Adversarial Networks (GANs) is a relatively new Machine Learning architecture for neural networks: it was first introduced in 2014 by University of Montreal (see this paper).

In order to better capture the value of GANs, one has to consider the difference between Supervised and Unsupervised learning. Supervised neural machineries are trained and tested based on large quantities of “labeled” samples. For example, a supervised image classifier engine would require a set of images with correct labels (e.g. cats, dogs, birds, . . .). Unsupervised neural machineries learn on the job from mistakes and try avoiding errors in the future. One can view a GAN as a new architecture for an unsupervised neural network able to achieve far better performance compared to traditional ones.

Main idea of GAN is to let two neural networks competing in a zero-sum game framework. A first network takes noise as input and generates samples (generator). The second one (discriminator) receives samples from both the generator and the training data, and has to be able to distinguish between the two sources.
The two networks play a game, where the generator is learning to produce more and more realistic samples, and the discriminator is learning to get better and better at distinguishing generated data from real data. These two networks are trained simultaneously, in order to drive the generated samples to be indistinguishable from real data.

GANs will allow training a discriminator as an unsupervised “density estimator”, i.e. a contrast function that gives us a low value for data and higher output for everything else: discriminator has to develop a good internal representation of the data to solve this problem properly. More details here.

GANs were previously thought to be unstable. Facebook AI Research (FAIR) published a set of papers on stabilizing adversarial networks, starting with image generators using Laplacian Adversarial Networks (LAPGAN) and Deep Convolutional Generative Adversarial Networks (DCGAN), and continuing into the more complex endeavor of video generation using Adversarial Gradient Difference Loss Predictors (AGDL).

As claimed here, it seems that GANs can provide a strong algorithmic framework for building unsupervised learning models that incorporate properties such as common sense.

There is a nice metaphor here about GANs: “In a way of an analogy, GANs act like the political environment of a country with two rival political parties. Each party continuously attempts to improve on its weaknesses while trying to find and leverage vulnerabilities in their adversary to push their agenda. Over time both parties become better operators”.

30 January 2018

A.I. for mitigating the "complexity" of the Digital Transformation

Several techno-economic drivers which are paving the way to a profound digital transformation of the Telecommunications infrastructures. Among these drivers, there are: the diffusion of ultra-broadband, the increasing of performance of IT systems vs the down-spiralling costs, emerging of innovative networks and services paradigms such as SDN and NFV, the growing availability of open source software but also the impressive advances of Machine Learning and Artificial Intelligence.

This digital transformation will lead the current legacy Telecommunications infrastructures to evolve towards the 5G (the 5th generation of network infrastructures) as an end-to-end network and service platform: in the long term, 5G is set to integrate processing, memory/storage and networking resources, functions and services through a “transparent”, hyper-connected ultra-broadband programmable “fabric”.

In this direction, in several Standardization Bodies and Fora, an high-level reference model is emerging, based on two main pillars: 1) an infrastructure physical layer, which will include computing, memory/storage and network resources (up to the edge/fog resources and even the Users’ terminals, devices, smart things); 2) a software virtualization layer which will allow providing high-level abstractions of all the infrastructure resources, functions and services.

It is well known that Software-Defined Networks (SDN) and Network Function Virtualization (NFV) are two of the key enabling technologies. Their exploitation in 5G will allow Virtualized Network Function (VNF) and services will be dynamically combined and orchestrated to create specific end-to-end “service chains” for the vertical applications; moreover the infrastructure will provide “slices” of logical resources where to execute multiple chains to serve applications (specific QoS requirements).

It is also reasonable to expect that this network transformation will reduce the costs (e.g., CAPEX and OPEX) and increase the flexibility of the infrastructure, ensuring high levels of programmability (through APIs) and the performance and security levels required by future 5G scenarios and applications (e.g., Internet of Things, Tactile Internet, Immersive Communications, Automotive, Indutry4.0, Smart Agriculture, Genomics/Omics and E-Health, etc).

So 5G will be much more than one step beyond today’s 4G-LTE networks: it is expected to become a an end-to-end network and service platform where multi-level APIs will allow Operators/Providers, Third Parties or even end-Users to create/operate “service chains”, made of elementary services/functions component capable of meeting on-demand the applications’ requirements. As a matter of fact, 5G architectural and functional disaggregation is one of the most debated avenues in innovation and standardization activities. 

We are witnessing a rapidly increasing in the “complexity” of the infrastructures subjected to this process of digital transformation, a complexity which will be too high just for human-made operators.

In fact, configuration, control and management of current physical pieces of equipment (in most cases closed boxes) will have to be replaced by automated processes acting over millions of virtual/logical entities (e.g., Virtual Machines, Containers, appliances etc). Management (e.g., Fault, Configuration, Accounting, Performance and Security) control and orchestration functions of such future infrastructures will require innovative methods and systems (e.g., self-organizing, adaptive control, machine learning, neural networks, etc.) capable of using the big data to mitigate this “complexity”.


It is not only a technical “complexity” but also an economic one, about biz sustainability. The increasing competition pressure in the Telecommunications market is pushing Network Operators and Service Providers to look for new services scenarios and solutions for reducing/optimising the overall operations costs to compensate the cases where revenues are declining.    

It is expected that A.I. (e.g., ML over actionable Big Data, etc...) will help for mitigating the "complexity" of this Digital Transformation, but what will be its impact on the networks and services platforms architectures ?

It's not just a matter of mathematical methods or algorithms, heuristics, etc.. What A.I. functions, what interfaces to what have to be standardized ?   

17 December 2017

The rise of a 5G Networked AI with humans-in-the-loop


The programmability, flexibility and high levels of automation of 5G operations will reduce costs (e.g., OPEX) and create new service paradigms which might be even beyond our imagination. Some examples concern the applications of the Internet of Things, Tactile Internet, advanced Robotics, Immersive Communications and, in general, the X-as-a-Service paradigm.

Let us consider some examples. Cloud Robotics and 5G-controlled robotics will have huge impacts in several sectors, such as industrial and agricultural automation, in smart cities and in many domestic applications. In agriculture, autonomous machines will be used for tasks like crop inspection, the targeted use of water and pesticides, and for other actions and monitoring activities that will assist farmers, as well as in data gathering, exchange and processing for process optimization. Interestingly, Cloud Robotics and 5G APIs can be opened to end-users and third-parties to develop, program and provide any type of related service or application for pursuing specific tasks. In industry, this will pave the way to process automation, data exchange and robotics manufacturing technologies (e.g., Industry 4.0). It is likely that we will soon see robotic applications in the domestic environment: it is estimated that by 2050-2060 one third of European people will be over 65. The cost of the combined pension and health care system could be close to 29% of the European GDP. Remotely controlled and operated robots will enable remote medical/supportive care and open up a new world of domestic applications which may also be incorporated by the entire population (e.g. cleaning, cooking, playing, communicating, etc.).

5G will have a big impact also on the automotive and transportation markets. Nevertheless there are still open issues. In fact, even if significant progresses have been made in developing self-driving/autonomous machines, equipped with sensors, actuators and ICT capabilities, the achievement of very low reaction times still represent an open challenge. As a matter of fact, the autonomous driving in real traffic is a very challenging problem: reaction time in units of milliseconds, or even less, are needed for safety reasons to avoid sudden and unpredictable obstacles. This means that a considerable amount of computing and storage power must be always available through ultra-low latency links. Today, the amount of computing and storage power that can be equipped locally in a machine/vehicle is not enough (for several reasons, e.g., space, dissipation limits, costs restraints, etc.) to cope with these requirements. Huge amounts of data needs to be stored and accessed and the AI methods have to be executed very quickly to exploit such levels of reactive autonomy. An ultra-low latency 5G network will allow exploiting the best balance of resources in the Cloud and Edge Computing systems, thus offering trade-offs between a local vs global cognition execution, essential to minimize reaction times.

In a similar direction, images/video real-time processing, for example for recognizing forms, faces or even emotions in photos or live-streamed video, represents another challenging case study or AI in 5G infrastructures. In fact, this could be radically improved from the distributed execution of deep learning solutions in a 5G infrastructure capable of providing ultra-low latency connectivity links.  Also in this case, performances will be improved by the flexibility of 5G in dynamically allocating/moving either huge data sets and software tasks/service where/when it is more effective to have them.

Another example is Immersive Communications, which refers to a paradigm going beyond the “commoditization” of current communication means (e.g., voice, messaging, social media, etc.). Immersive Communications will be enabled by new advanced technologies of social communication interactions, for example through artificially intelligent avatars, cognitive robot-human interfaces, etc. Eventually, the term X-as-a-Service will refer to the possibility of providing (anytime and anywhere) wider and wider sets of 5G services by means of anything from machines to smart things, from robots to toys, etc. If today we are already linking our minds with laptops, tablets, smartphones, wearable devices, and avatars, in the future we will see enhanced forms of interactions between humans, intelligent machines and software processes.

Current socio-economic drivers and ICT trends are already bringing to a convergence Computer Science, Telecommunications and AI.

In this profound transformation, mathematics will be the language, computation will be about running that language (coded in software), storage will be about saving this encoded information, and, eventually, the network will be creating relationships – at almost zero latency -- between these sets of functions. This trend will also see the rise of the so-called Networked AI with humans-in-the-loop. Today there are already some examples, such as analyst-in-the-loop security systems, which combine human experts’ intuition with machine learning capable of predicting infrastructure cyber-attacks.

Although security and privacy are out of the scope of this work (focusing on 5G enabling capabilities), these two strategic areas deserve some further considerations. On one side 5G could provide the means for improving security, for example as information will be available everywhere and the context needed to detect anomalous behavior will be more easily provided; nevertheless on the other side, enabling technologies such as SDN and NFV have the potential to create situations where all primary personal data and information is held and controlled at a global level, even outside the national jurisdiction of individual citizens. It has been mentioned, as an example, the real-time processing of several thousands of images per second and live-streamed video: this will have wide-ranging, but also controversial applications: from predicting crimes, terrorist acts and social upheaval to law enforcement and psychological analysis. Eventually, in the long term, this might transform everything from policing to the way people interact every day with banks, stores, and transportation services: this will have huge security and privacy implications.

Reasonably privacy and security concerns should be considered by-design, with  systemic solutions capable of operating at different levels in future 5G infrastructures: for example, such design will need to consider issues such as automated mutual authentication, isolation, data access and management of multiple virtual network slices coexisting onto the same 5G infrastructure.

04 December 2017

How cloudy and green will mobile network and services be?

The CLEEN international workshop series is about “Cloud Technologies and Energy Efficiency in Mobile Communication Networks” and during all these years obtained a great interest from both research and industry. Every year the CLEEN workshops collaborated with EU projects and provided a great opportunity for researchers and industry practitioners to share their state-of-the-art research and development results in areas of particular interest.
Next edition, the CLEEN2018 workshop will be co-located with IEEE VTC2018-spring (Porto, 3 June 2018, http://www.ieeevtc.org/vtc2018spring/index.php), where a particular emphasis to edge cloud, MEC and vertical segments will be given, due to the growing interest of these topics toward 5G networks.
CLEEN2018 will have the objective to explore novel concepts to allow for flexibly centralised radio access networks using cloud-processing based on open IT platforms, in coordination with network function virtualization technologies and MEC (Multi-Access Edge Computing), which are recognized as key enablers for the definition of future 5G systems. The aim is to allow for a guaranteed high quality of experience for mobile access to cloud-processing resources and services, and to allow a future network evolution focused on energy efficiency and cost-effectiveness. In fact, all future innovative network solutions will be conceived and deployed with a long term perspective of sustainability, both in terms of energy consumption of mobile network (and related interoperability with terminals) and cost efficiency of the different deployment and management options. This requires new concepts for the design, operation, and optimization of radio access networks, backhaul networks, operation and management algorithms, and architectural elements, tightly integrating mobile networks and cloud-processing. This workshop will cover technologies across PHY, MAC, and network layers, technologies which translate the cloud-paradigm to the radio access and backhaul network, and will analyse the network evolution from the energy efficiency perspective. It will study the requirements, constraints, and implications for mobile communication networks, and also potential relationship with the offered service, both from the academic and the industrial point of view.
Here below the link to the call-for-papers, that we would kindly ask you to promptly forward to your projects/colleagues and interested people.
The CLEEN2018 workshop program is under definition, and we are working hard to organize a great panel discussion with key note speakers selected from highly qualified representatives in the international field.
Stay tuned!
Dario Sabella

INTEL, General Chair of CLEEN2018 workshop

30 November 2017

"Nervous Systems" for Smart Cities... but what about jellyfishes ?


The metaphor of future networks (e.g., SDN/NFV, 5G) becoming the "Nervous System" of Digital Society and Economy has been mentioned several time in this blog.

I remember I made a welcome presentation at EuCNC-2014 showing this picture, elaborating this vision for the first time (at least to my knowledge). In the talk, my take was that technology advances (SDN, NFV, Cloud Computing, AI) are creating the conditions to deploy - in the Digital Society and Economy - a sheer number of pervasive "control-loops" (or if you prefer autonomic control-loops a la MAPE-K) mimicking the role of a "nervous system" in a living being. 


...and in this paper on December, 2014:


...and more recently in this piece:


Today I've stumbled upon this press:



...so we are witnessing progresses in exploiting this vision !

We could even extend this biogical metaphor considering that the traditional view of central nervous system is not valid for some living being, e.g., jellyfish: in fact, they have a ring nervous system, located along the margin of the bell !


It this a lesson learnt from Nature about the value of decentralization in case of asymmetry ?