Periodic Reporting for period 4 - AGNOSTIC (Actively Enhanced Cognition based Framework for Design of Complex Systems)
Période du rapport: 2022-04-01 au 2023-09-30
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.
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.
The work pursued during the reporting period (corresponding to the duration of the project) include:
1) Foundational WP (WP1): Investigations were undertaken 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 were also pursued. Mathematical approaches including iterative optimization for large data sets and emerging learning paradigms were considered.
2) Application Motivated WPs(WP2-4): Several avenues have been investigated into the use of data-driven and model-based system design for radar, communication and caching, beginning with the static environment and culminating in dynamic environments, have been initiated.
3) Implementation based WP (WP5): This WP started with an initial implementation and demonstration of the cognitive radar in the lab using Software defined Radios. As a use case, this cognitive radar was utilised in the emerging context of joint radar and communications to enable co-existence of radar and communications in real time.
The project has resulted in 110 publications till date. 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 (https://2018.ieeeicassp.org/SpecialSessions.asp). It provided visibility to the research and a platform for discussions. To foster interaction between the involved parties, a joint workshop was held in October 2018. A second workshop was held virtually on 30 October 2020 where the involved researchers summarized their work and ample time was provided for discussions. A full-day workshop was organized at the ICASSP 2023 in Rhodes on the Integrated Sensing and Communications which brought in elements of learning and model-based approaches to an emerging application. Participants in the project have organized Special Sessions in international conferences including EUSIPCO 22 and IEEE SPAWC 2021. Project demonstrations have also been undertaken in IEEE ICASSP 2020 & 21.
Further, a plenary keynote presentation at IEEE ICASSP 2020 was delivered by the PI, Björn Ottersten. Further, Dr Bhavani Shankar MYSORE delivered a series of tutorials at IEEE ICASSP 2020 & 2021/ IEEE RadarConf 2020 &21/ IEE VTC Spring 2021/ IEEE Globecom 2021/ IEEE SAM 2022/ IEEE SPCOM 2022/ EUSIPCO 2022/ IEEE ANTS 2020 & 23. An ERC PoC based on the research in WP2 has been successfully executed employing intelligent reconfiguration of sensors for self-localization and mapping.
1) Foundational WP (WP1): Investigations were undertaken 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 were also pursued. Mathematical approaches including iterative optimization for large data sets and emerging learning paradigms were considered.
2) Application Motivated WPs(WP2-4): Several avenues have been investigated into the use of data-driven and model-based system design for radar, communication and caching, beginning with the static environment and culminating in dynamic environments, have been initiated.
3) Implementation based WP (WP5): This WP started with an initial implementation and demonstration of the cognitive radar in the lab using Software defined Radios. As a use case, this cognitive radar was utilised in the emerging context of joint radar and communications to enable co-existence of radar and communications in real time.
The project has resulted in 110 publications till date. 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 (https://2018.ieeeicassp.org/SpecialSessions.asp). It provided visibility to the research and a platform for discussions. To foster interaction between the involved parties, a joint workshop was held in October 2018. A second workshop was held virtually on 30 October 2020 where the involved researchers summarized their work and ample time was provided for discussions. A full-day workshop was organized at the ICASSP 2023 in Rhodes on the Integrated Sensing and Communications which brought in elements of learning and model-based approaches to an emerging application. Participants in the project have organized Special Sessions in international conferences including EUSIPCO 22 and IEEE SPAWC 2021. Project demonstrations have also been undertaken in IEEE ICASSP 2020 & 21.
Further, a plenary keynote presentation at IEEE ICASSP 2020 was delivered by the PI, Björn Ottersten. Further, Dr Bhavani Shankar MYSORE delivered a series of tutorials at IEEE ICASSP 2020 & 2021/ IEEE RadarConf 2020 &21/ IEE VTC Spring 2021/ IEEE Globecom 2021/ IEEE SAM 2022/ IEEE SPCOM 2022/ EUSIPCO 2022/ IEEE ANTS 2020 & 23. An ERC PoC based on the research in WP2 has been successfully executed employing intelligent reconfiguration of sensors for self-localization and mapping.
Investigating cognition in such complex networks would have been a mere academic study, but for their continuously growing impact on our daily lives. The mobile communication system offers an ideal example where the number of users and their requirements are growing exponentially due to various applications. These must be served by limited resources, be it bandwidth, power or time. In addition, the nature of communication and the actors therein have changed from a homogenous (voice calls/ TV broadcast/ fixed phones) to heterogenous (bursty data, internet, satellites and flying objects in different orbits). There is a demand for seamless connection to these entities and to different types of traffic. Having identified all the available resources, the only way to meet the requirements and also to scale up is to make the systems intelligent through active cognition, realizing flexible designs allowing dynamic reconfiguration. In this setting, the significant and diverse research carried out in AGNOSTIC offers a framework to design and implement these heterogenous systems with varying traffic types to meet the growing demands with the available resources. This would offer critical communications to the needy and during natural disaster, as well as entertainment leading to an uptake in commercial activities in the communications arena.
Another application area that has truly moved from the niche to utilitarian is that of radar sensing. Automotive radars offer a cheap, yet powerful, sensing mechanism that enables driver safety and comfort. However, a road-side scenario with multiple cars, all equipped with radar, is one of the most dynamic and complex scenes; making radar operation efficient in such scenarios requires distributed cognition. This is considered essential in the concept of autonomous vehicles, a concept that would revolutionize travel making it less mundane and tiring for young and elderly alike. The waveform design, integrated sensing and communications, sensor placement and related effects investigated in AGNOSTIC, aids to the further proliferation of sensing in daily lives including vital sign monitoring, people counting and human action recognition.
There exist many other societal applications where all the system resources are being used but needs an additional degree of freedom for functional efficacy and efficiency. In this context, AGNOSTIC is relevant to the current developments in the society.
Another application area that has truly moved from the niche to utilitarian is that of radar sensing. Automotive radars offer a cheap, yet powerful, sensing mechanism that enables driver safety and comfort. However, a road-side scenario with multiple cars, all equipped with radar, is one of the most dynamic and complex scenes; making radar operation efficient in such scenarios requires distributed cognition. This is considered essential in the concept of autonomous vehicles, a concept that would revolutionize travel making it less mundane and tiring for young and elderly alike. The waveform design, integrated sensing and communications, sensor placement and related effects investigated in AGNOSTIC, aids to the further proliferation of sensing in daily lives including vital sign monitoring, people counting and human action recognition.
There exist many other societal applications where all the system resources are being used but needs an additional degree of freedom for functional efficacy and efficiency. In this context, AGNOSTIC is relevant to the current developments in the society.