#	Pagina
attuale pagina	/open-h2020/projects/211402/results.html
-1	/open-h2020/projects/195905/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - AGNOSTIC (Actively Enhanced Cognition based Framework for Design of Complex Systems)

Teaser

Summary

Summary of the project:

Parameterized mathematical models have been central to the understanding and design of communication, networking, and radar systems. However, they often lack the ability to model intricate interactions innate in complex systems. On the other hand, data-driven approaches do not need explicit mathematical models for data generation and have a wider applicability at the cost of flexibility. These approaches need labelled data, representing all the facets of the system interaction with the environment. With the aforementioned systems becoming increasingly complex with intricate interactions and operating in dynamic environments, the number of system configurations can be rather large leading to paucity of labelled data. Thus, there are emerging networks of systems of critical importance whose cognition is not effectively covered by traditional approaches. AGNOSTIC uses the process of exploration through system probing and exploitation of observed data in an iterative manner drawing upon traditional model-based approaches and data-driven discriminative learning to enhance functionality, performance, and robustness through the notion of active cognition. AGNOSTIC clearly departs from a passive assimilation of data and aims to formalize the exploitation/exploration framework in dynamic environments. The development of this framework in three applications areas is central to AGNOSTIC. The project aims to provide active cognition in radar to learn the environment and other active systems to ensure situational awareness and coexistence; to apply active probing in radio access networks to infer network behaviour towards spectrum sharing and self-configuration; and to learn and adapt to user demand for content distribution in caching networks, drastically improving network efficiency. Although these cognitive systems interact with the environment in very different ways, sufficient abstraction allows cross-fertilization of insights and approaches motivating the joint treatment in this proposal.

Objectives:

1) To devise a framework and methodology, specify the component blocks and formalize the joint design and interaction thereof. In this context, development of a toolset of learning-based optimization approaches for complex and dynamic wireless networks is envisaged.
2) To enhance situational awareness and coexistence of cognitive radar by integrated and iterative dynamic radar waveform adaptation enabled by unsupervised target discrimination within cognitive radar
3) To optimize cognitive RAN by facilitating autonomous deployment and optimization, along with self-organization and coexistence of future radio access networks
4) To efficiently deliver content through active caching networks by providing seamless delivery of content for non-linear consumption over heterogeneous networks through efficient broad/multicasting and caching of popular content at the network edges taking the dynamic spatio-temporal user demand into account.
5) To implement and experimentally validate key aspects of the active cognitive engine

Towards addressing these objectives, the following aspects have been identified,

1) For methodology development in WP1, initial focus is on developing a generic learning-optimization framework for addressing the resource optimization problems in complex wireless networks.
We investigate the interplay between optimization and deep learning. Basically, we try to provide answers for two key research questions: what can learning approaches do for conventional optimization techniques, and what can advanced optimization approaches do for machine/deep learning?

2) In the context of cognitive RAN, ongoing research addresses Active Learning (AL) techniques for Cognitive RadioNetwork (CRN) scenarios. Intelligent radio device networks, also called Secondary Users (SUs), havean important deployment weakness: the absence of communication with legacy telecommunicationsystems whose resources they try to e

Work performed

AGNOSTIC envisions a computationally efficient framework for enhancing the performance of complex dynamical cognitive systems by actively probing the environment/system and allowing model-based and data-driven approaches to work in tandem to overcome their respective shortcomings. Five work packages have been envisioned to achieve this objective. The work pursued during the reporting period (corresponding to about 18 months of the project) include:

1) (WP1) Investigations to devise a framework for formalizing the active learning using data-driven and model-based approaches which would result in the AGNOSTIC framework. In addition to tackling the problem from this objective, a top-down approach of investigation how this aspect has been dealt in the various AGNOSTIC application areas has been pursued. Mathematical approaches including iterative optimization for large data sets are being considered.

2) (WP2-4) Initial investigations into the use of data-driven and model-based system design for radar, communication and caching in static environments have been initiated. This study is a precursor to the active learning for dynamic systems. Results from WP2-4 will be analyzed and considered for in-lab demonstration in WP5 (to start in year 4).

The project has resulted in 11 conference articles, 7 journal papers and one book chapter during the reporting period. The work pursued in these publications are summarized in the other sections of this form appropriately.

The topic of AGNOSTIC was selected for a special session at the prestigious IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018, Calgary, Canada. It provided visibility to the research and also a platform for discussions. To foster interaction between the involved parties, a joint workshop was held in October 2018, followed by planned research visits.

Final results

In the literature, a large amount of works of applying deep learning to wireless networks focus on end-to-end learning, i.e., using DL model to directly predict decisions in simple optimization tasks, e.g., applying ML/DL to predict channel estimation, traffic classification, modulation scheme. However, for addressing the difficult optimization problems with various constraints, e.g., combinatorial, non-convex, non-linear problems, there is no clear roadmap available so far. Beyond the current state-of-the-art, our contributions/progresses can be from investigating the following issues:
-- In what way the optimization and learning techniques can be integrated.
-- For solving a hard optimization problem, which features are learnable/unlearnable, and which features can lead to good performance.
-- How to guarantee the solutionâ€™s performance and feasibility.
These are the important open issues and studied to a limited extent in the literature.

With regards to Cognitive Radio Networks, the aforementioned â€œproperâ€ behaviour of the CRN is considered as the SUs transmitting underan interference PU threshold where the interference channel gains are unknown and therefore need tobe learned. The learning objective though is not only to learn these channel gains, but also to trigger PUreactions in a way which is the most informative about these channels. This happens throughintelligently probing the legacy system so that the number of these probings, which are necessary tolearn the channels, is minimum.State-of-the-art AI methods applied in CRN scenarios have been presented in numerous worksin the Telecommunication field. Nevertheless, the planned approaches managed to outrun them by employing faster ALtechniques which we further enhanced. Their inference part was implemented using well knownBayesian ML tools where we utilized recent advances from Econometrics to make them more efficient.