 Welcome to the second webinar series of the IT Journal on Future and Evolving Technologies. My name is Alessia Magniarditi from ITU, the International Telecommunication Union. ITU is the United Nations Specialized Agency for Information and Communication Technologies. ITU allocates frequencies to the services that make use of the radio communication spectrum, it develops standards and assists developing countries in setting up their information and communication infrastructure. ITU and academia share a commitment to the public interest, and this commitment is embodied by the ITU Journal, which offers complete coverage of communications and networking paradigms free of charge for readers and authors. Our journal welcomes submissions at any time on any topic within its scope. And we believe that this webinar series launched in March this year will inspire more contributions from researchers around the world. It is my pleasure to open this new series today with Dr. Qin Li from China Mobile Research Institute. We count on your support to make this webinar an interesting experience. So please submit your questions via the Q&A channel. We will address them to our speaker during the Q&A session. And after the talk and the Q&A, please stay online. We have something very special for you, the wisdom corner, live life lessons. Dr. Qin Li agreed to a very personal chat. She will share with us some lessons learned over the years that might perhaps be useful for some of you. I'm very pleased now to introduce Professor Yanakildi, editor-in-chief of the ITU Journal and president and founder of Truva from the United States. So Professor Akildi is Chen Bayer's chair professor in telecommunications emeritus at Georgia Institute of Technology. In the last two decades, he established many research centers worldwide and is editor-in-chief emeritus of impact factor journals. Highly cited and at the top of the most prestigious international rankings, Yan is a visiting distinguished professors in several universities around the world. According to Google Scholar, his age index is 133 and the total number of his citations is over 136k. His current research interests include 6G, 7G wireless communication systems, hologram communication, terahertz, Internet of Bionano Things, molecular communication, Internet of Things, in challenged environments and many other topics. So Professor Akildi, the floor is yours for your opening remarks. Thank you so much and enjoy this webinar. Thank you Alessia. Good morning, good afternoon and good evening worldwide. I welcome you all to the second season and first episode of our ITU Journal Future and Evolving Technologies webinar series. I have the great pleasure to introduce you, one of the leading researchers in our era, Dr. Qin Lin Yi. Qin Lin has over 40 years, a super career. She's currently a chief scientist of wireless technologies at China Mobile Research Institute. She is very active or it has been very active in many organizations. She's the chair of the Oran Technical Steering Committee and a member of its executive committee chair of the future 5G 6G special interest group chair of the wireless AI Alliance Executive Committee executive board member of GreenTouch, steering board member and vice chair of World Wide Radio Forum, steering committee member of IEEE 5G and future networks initiatives, scientific advisory board of Singapore NRF, I assume that's like NSF, and IEEE comes up in many, many capacities since over three decades. She published over 200 papers and holds remarkably 100 patents. She wrote many books. Qin Lin received PhD from double E from Stanford University. I can say in the 80s and not to reveal her age. We met in the late 80s when she was working at Bell Labs in New Jersey. We also worked at the IEEE conference organizations together. After a couple of decades at Bell Labs, Qin Lin moved to Asian region. And I remember we met during my visit in Taiwan in 2002. She was leading this Taiwan big government research organization as I can recall. In 2006, with a coincidence, in millions of people in Beijing, during my visit to Tsinghua University in Beijing, we accidentally bumped each other in the lobby of the hotel while she was visiting China's NSF as I can recall again. Qin Lin received many prestigious awards. For example, 2005 IEEE Stephen Rice Prize, Best Paper Award, and also 2018 IEEE Fred Ellersick Prize, again Best Paper Award, and also 2015 Industrial Innovation Award for the next generation seller wireless networks. I express my sincere thanks for accepting our invitation and giving the speech, which is entitled a peek into the deep convergence towards 6G. Thanks Qin Lin. It's actually indeed a migrate pleasure to be on this forum. And let me share the screen first and get the show rolling. So, 6G, I think it's on most everybody's mind that are working in communications, especially mobile and wireless communications. And so today I would like to first peek into a little bit of the 6G and kind of focus on the key feature of 6G that is convergence, actually convergence in more way than one. And then finally, zooming on a specific aspect of convergence, which is the IT city and DT convergence, which I think is bring forth fundamental transformation of not just our industry by the whole cross domain ecosystem, and that will impact us all. So there will be kind of three parts in this sharing 6G and the general convergence trends and also the journey of the ICDT convergence. Okay. It's clear. I think you all agree that 5G deployment actually is still at the kind of early stage, even though there are more than 200 operators in close to 80 countries that have deployed the 5G network. And there are some, however, majority of the deployment had to be in non standalone mode, meaning that 5G is only kind of a add on pipe to a generically 4G system. And only 10% of the operators that have deployed 5G have deployed the standalone, meaning the real pure 5G system. And primarily is the first version of the 5G standards release 15 that's in the field these days. And in China, it's interesting to see that actually it's all the operators in China have deployed standalone 5G for their commercial services since a couple years ago. And the company I work for China Mobile specifically currently has deployed close to 1 million 5G based stations and that alone, it's about 30% of the global 5G deploying today. And in this network, we currently have a little less than a billion subscribers where half a billion, more than half a billion have already upgraded to the 5G services, enjoying new way for the video audio gaming and even video shooting capabilities. However, 5G, it's the most distinctive value and the value proposition to us is really to transform society by serving the various verticals. We have already developed more than 100 headquarter level SOP projects for this purpose and plus more than 2,000 provincial level projects serving verticals with regional characteristics. And many of it are replicated and promoted across different segments. So currently, there are more than 11,000 commercially deployed cases serving various verticals. However, this number considering the potential market is still very small number. And currently, there is still challenges in fast scaled up deployment SOPs for different verticals because the various verticals really has very different needs. And the effort that needs to customize and provide agile deployment and very flexible customization ability with the cost effective solution is still an ongoing effort across I think the globe, China the same. While this is still a major effort in our industry, the global 6G research initiative has already started actually since 2018, various regions have launched different initiatives looking into the early phase of 6G research. And on this page, I'm just showing you some examples. You can see that all over the world, somehow the 6G research initiative has become a really heavy focus and it's being accelerated by various initiative and fundings in many countries and regions. Okay, but a pragmatic 6G development timeline shows that more or less the actual 6G standard effort will officially start somewhere around 2015. Probably not before that between 2015 and 2016. That's when the actual standard effort will start. And which means that actually there are a few years yet for real research and real, I would say, hopefully a global soul searching to identify what's really needed and what problem we really want to solve and what capability we really want to bring. Okay. And of course, the world works in a continuous mode so it's not like that we will see 6G popping up that has nothing to do with what we have learned and what we have evolved. And the interesting thing to see is that the 5G, the 5G and beyond, which I think I should have put the title there, it's called officially the 5G Advanced, starting research in release 18 in CGPP is characterized by three aspect, extreme performance, cloud intelligence, and green and efficiency. And so, and the sum of this topic you will see that really is going to be a continual theme evolving forwards even into the 6G era. Actually, my company, China Mobile's vision of 6G era is an era of digital twin with ubiquitous intelligence. And central of that is that the network will be made of technology symbolizing the ICDT deep convergence. And some interesting scenario that's being contemplated globally includes a metaverse where the real world are with the meta world are kind of parallel and highly interconnected and that you can in the metaverse world, we would enjoy immersive experience including shopping, social interaction, and the collective work. And this is not only for personal convenience and entertainment, but we can envision that this really provides many opportunities for innovative services for the vertical industry as well. And another often talked about capability in 6G era is holographic interaction, okay, in communication, in industrial design, even in medicine and also of course for a lot of us for meetings, you will enjoy holographic partners presence. This is something that I think we can look forward to probably even before the 6G network become widely spread available existence. And another very interesting scenario that people often talk about for the 6G era is the interconnection of the five senses. And historically our 1G through 4G, 5G till now, we know that mostly it's our audio sense and the video sense are being connected. But there are various attempts, even since the 5G research started for the touch, the additional sense, and then maybe even involving smell and the taste in the future. There's very aggressive conjecture that in the 6G era sometime in there, we could extend to even synesthesia perception that is really go across our multiple model of senses. How this could be accomplished, I think it's something very exciting for us to look forward to. And definitely, we already see certain level of internet of skills that's been provided in the 5G era. And in the future, the internet of experiences, the internet of actually multi-modal senses interaction may become a reality as well. And to accomplish all of that, there is lots of early stage research and effort trying to figure out what exactly the 6G APIs are. And this I'm just showing one example, which was released a few months ago by NGMN as the 6G requirements that their first project delivered. And I think this number, probably this set of KPIs are probably in similar range compared to the similar white papers that's been published by other initiatives and organizations. However, one of the key characteristics for 6G design, I think, is from the onset. It's not just the individual KPIs. It's an important consideration, but also combinations of multiple KPIs, a subset of KPIs combined together would also be very important design target in the 6G system. Okay. Now, before we zoom in, let's quickly go over some of the interesting trends of convergence in different kind of cross-field, cross-domain, and cross-intersegment, cross-multiple-expect. The first one that often being talked about is multi-bands and also space, air, ground, all converged networking. At this point, people are very excited about this future that where we would use maybe the sub-6G or sub-7Hz band to provide primarily the seamless and basic network coverage, and then using minimal wave, terahertz, even VLC, visible light to complement wherever it's appropriate for localized high-capacity needs. Shorter range and the lower mobility scenario, of course, and then complemented with satellite where right now, I think, people's imagination covers Geo, Mio, and Leo, all inclusive for kind of global coverage over the water, over the mountainous area, over basic provides truly 100% seamless for global coverage. One thing I just, I want to caution is I think the complementary services between the space or air-based system satellites, that is, with the terrestrial mobile cellular system has always existed. I mean, in reality, it's just that it was very, very loosely coupled or decoupled. For 6G, it seems that the intent is to make the coverage much more tight, tightly coupled together, integrated together. I think it may be an interesting research topic and engineering effort, however, I would caution to consider how much disconvergence, how tight disconvergence really needs to be, to be cost-effective. So that's one thing I do want to bring up. And the other one is in terms of the wave, Terahertz, VLC. Personally, I think we saw, we all remember freshly our experience of a minute wave in the 5G efforts. Initially, there was great excitement and the expectation about the meter wave, and maybe we've made eventually play some role, I think, in the 5G network, and going forward for 6G as well. But I think we need to be more conservative in our expectation that includes meter wave and the Terahertz, given the fact that it's really coverage-limited and mobility-limited solutions. And on the other hand, VLC may be a surprising factor to combine VLC with a microwave system or cellular system as a downlink enhancement could become a cost-effective element in the 6G if we make good use of it. Okay, and another convergence is actually introducing sensing into the traditional communication-only network, that is our network, the integrated sensing and the communication. This has generated a lot of excitement. Our view is that probably the cooperative sensing in a cellular system would be an interesting addition. But we are still at the early stage that we have not been able to have a systematic way to quantify the cost in terms of both not just the hardware, but how to minimize the hardware addition. But also, actually, how much spectral efficiency for communications that needs to be sacrificed when we incorporate the sensing capability. Hopefully not too much, but there needs to be some effort to clearly quantify it. And moving into a further direction, which is now it's becoming also popular, micro-area communication technology and some calls it PRN. However, this kind of reminds me of Ben, which is, I think in the 90s, body area network and view it as extension of end-to-end network. It's interesting, but I would still caution not to over-engineer a tight integration of this piece into our 6G design. I think it would be a helpful element, helpful addition, but probably doesn't need to be a tightly integrated piece in the 6G network. And there's a lot of effort in material science these days, since we feel that it could bring us the next, I guess, quantum leap of the capabilities and the performance of our system. One hot topic that made a comeback is metamaterial. The reason why I say made a comeback is because I think at least 20 years ago, we were looking into metamaterial capability in miniaturizing our devices for wireless communications, and there was some interesting finding in that effort. We found out that we could make use of metamaterial to make one antenna behave like if it's two antenna. However, we needed to make a metamaterial mask, which still requires the volume of two antenna. So I hope you understand what I'm saying. So how to make use of metamaterial in our effort in 6G system is an interesting topic. And currently the most successful one is RIS, reconfigurable RIS. But it's very interesting. It divides technology combined with system technology and the RIS based on wireless relay has been prototyped and demoed in many places. It could provide coverage, hope, coverage, and the regional traffic increase may also be aided by this device. However, I can't help but think about the tremendous effort that was put in for various wireless relay during the 3G and 4G era. And it didn't seem to become a very important part of our large scale system. So that's just a caution about looking for opportunity for RIS based systems. And another aspect of the material science that may bring major transformational impact is the flexible material from creating flexible device and flexible systems. This would enable genuinely wearable computer and smart clothes and provide many opportunities to fit the conformal services of some of the important vehicles. But it seems that there are still great challenges to deal with the necessary flexibility and the necessary performance stability because the dielectric properties of this material at the different deformation states is still hard to control. And we have some interesting research effort going on and we look forward to good progress in this. And then another, I guess you could consider conversions also even though semantic communication was not new. Shannon had mentioned it, talked about it even as early as his classic paper was published. However, Shannon went on focusing on the bits but not the meaning of the communication. So now everyone is revisiting the possibility of a semantic communications. I think this particularly can become handy when we look at services in 6G area for vertical applications because many of the application would be task oriented. So there if we can find a way to compress redundant and irrelevant info out and so that we would be communicating at the semantic meaning level. This would drastically reduce I guess the bits that we would need to transmit reliably. So if whether we could have breakthrough in semantic communications I think would be a very interesting, exciting topic to look at. Even though some of the I think the current so-called semantic communications research work is not that different from a traditional source coding. And so it's a very promising direction and hopefully we will find the proper way to do it. And then remember that I mentioned the 6G digital twin pervasive intelligence. So digital twin could be considered for many objects or systems. However, with I think the most interesting part is converging this concept with four 6G network. We mean building the digital twin network, the DTN that can facilitate autonomous optimization autonomous operation. And then also the digital twin enabled by 6G network. One most interesting example would be the personal digital twin. And that could be at a different at appearance level at physical level or even eventually at emotional level. But all of this will require a lot of across domain technology breakthrough or so-called convergence. And then I want to talk a little bit about energy saving green systems. We had a very early on as early as 2015 worked on convergence of multiple network with that point was 2G, 3G, 4G, and now of course is 5G convergence of the energy saving system across multiple network. And we have deployed that in most of the areas that our network cover and the just activated system, 6 hours per day, we were able to achieve 20% of energy saving. So this is a very important I guess direction of convergence as well. And I think this is the last general sense of convergence case that I want to share. This is actually something that we did in 2012, which is convergence of the energy sources, where we combine the solar panel and wind turbine with fuel cell for a base station and energy supply. And actually, theoretically, it's possible to build a completely self-sustainable standalone base station using those resources. And the only thing that we often need to resupply is pure water. This was a very exciting project, but cost was really high. Part of this system components is very expensive. So we only deployed two of this. And this was exactly 10 years ago. And if the related material could become a lower cost one in the lower cost module, hopefully this would be something we can see more widely applied in the 6G network. Okay, now let me zoom into specifically the IT city and DT convergence. Actually, if you look at our core network of our network, very early on 2G, 3G, 4G era, the core network already adopted IP, pure IP, which is basically the IT technology. And then 4G to 5G, we see that the service-based architecture and the codification are being adopted. And then currently, also, actually, there was this NWDAF very early on in the 5G era being designed that is bringing AI machine learning technologies into the core network. So from the core network, the IT, CT, DT convergence actually started much earlier than in the access network. In the access network, for most people, I guess you could see that our traditional base station actually has the individual base station as distributed standalone by itself. Whereas, then, about 10 years ago, we started working on CRAN, where we bring part of the base station into a pool, which is the BBO pool. This CRAN is centralized and provide also opportunity for collaborative processing across multiple base station, and also build up foundation for going into a cloud platform, virtualized realization, which eventually actually move into open and smart O-RAN, where actually you see that you have edge cloud where you can bring in co-located BBO function with UPF, which comes down from the core network and also adding the intelligent controller and applications O in 1. Actually, let me go into a little more clear detail. The ICDT journey really, we see that into two parts. One is the journey of the cloud RAN, which started since 2011, which is characterized by being efficient, green, and being agile, soft, green and soft. And this was first started in 2011, and then eventually in 2013, Etsy has launched its NFV ISD, where initially it was focusing on core network, as I said earlier, but we expanded into RAN. And finally, we see that it progressed into a several cross ecosystem partnership to O-RAN Alliance establishment. And also, even though it started in the 4G era, but in 5G, the commercial deployment, 70% are based on this approach. And the second phase of this journey is the journey of intelligent RAN that started a little bit later, 2014, 2015 timeline, and features open and smart, started from the wireless big data research. I think this was in 2014. Yeah. And then there was an opportunity actually for 5G to start with a RAN DA, RDA, to couple with the NWDA in the core network, so that the AI machine learning capability could be built in in the 5G RAN. But that opportunity was not materialized, so eventually there was the O-RAN Alliance being established, and the currently O-RAN Alliance is building on the RIC intelligent controller in a hierarchical manner to fulfill the, to bring intelligence into the RAN, and it's built upon the existing 5G architecture. So, if you look at the big picture, you could see clearly that from 4G to 5G, the ICT convergence, which is green and soft, is the main trend, the main convergence, but from 5G to 5G at once, and to 6G, we will see that the main theme building on the green and soft is open and smart. And this is the second phase of this journey. And the most important cornerstone event, I think for the ecosystem, is the launch of O-RAN Alliance in 2018. And from day one, it is targeting to transform the radio access network so that it's open, intelligent, virtualized and fully interoperable. And it has moved very fast. Actually, it's already close to 350 members today. And with four and a half thousand global technical experts working together. And all the technical work under the technical steering committee of O-RAN is divided up by 11 working groups, and three focus groups, three subcommittees of TSC with one open source community and another newly established research group. And currently already 85 current version of technical specs building on the baseline 3G, BPP 5G standards was delivered. And there are open testing integration centers, OTICS established for a fair and transparent and stable testing environment for various software and hardware solution providers to test it out. Globally, there are seven established, including four in Europe and two in APEC area and one in North America. And this alliance also has provided so far six releases of open source software and have set up three open labs. The open labs are I think one West Coast, one East Coast, both in North America, and the third one also in APEC area. And also held for global plug fest. Every time there's hundreds of global vendors in the half a dozen to a dozen different sites across the globe to participate. More importantly, that this alliance work very actively with the industry partners, including Etsy, GSMA, ONF, and CTPP and ITOT, etc. A lot of the software hardware solution has been put on interesting virtual booths for virtual exhibition, 278 demos specifically. And there is the address for you to visit if you're interested. And on openness, Orion Alliance looked for open interface, open infrastructure, I mean open cloud and open source software and the open hardware reference design, etc. And this is a, and on cloud cloudification, this alliance looked into the O2 interface where it's an open interface to provide infrastructure management services and deployment services, as well as the lifecycle management for all of the virtualized applications. And more importantly, it is a container, many containerized infrastructure with accelerator abstraction layer being defined as for different profiles for the accelerators that the relatively real time load latency processing that ran processing needs. And this is one example of the latest open Orion cloud platform, version 1.0, which is specifically targeting for verticals and small sales usage. And as you can see that it has optimized its throughput and average delay and the jitter. Actually, the interrupt response is kept below 10 microseconds, the interrupt, and also the nanosecond level synchronization. And in terms of intelligence, it includes the SMO and non real time rig in the management plan, where you'll have O1 for the intelligent F caps, you'll have O2 for the open cloud IMS and DMS. And then you'll have A1 for to support a service based non real time rig architecture with R1 services, R1 as the interface with all the intelligent applications, R apps, and providing 10 driven automation also. And what this non real time rig is set up to complement near real time rig for different, primarily different role. Most of the offline training would be done in non real time rig, whereas the near real time inference, the online inference would be done in the near real time rig. So the non real time rig is characterized in the management domain, whereas the near real time rig is in the control control domain. And currently, there are some good examples, traffic theory, QoS QoE optimization, and the rain slicing assurance are already realized by this near real time rig. And the next thing that will be made available soon is ran literally information exposure, which would expose run performance analytics and some predictions so that external third party applications could make yourself. And this is a collection of some of the cases in application in management in the base station control use cases of the rain intelligent and controller and the SMO platform. And some of the, not field trial but some of the kind of SMO self deployment that was done in Nanjing last year using the hierarchy of rigs provide rain slice SLA assurance where actually the average satisfaction ratio of 94.6% was achieved. And also as you can see that it's compared to the other two, one is wrong robbing, and the other is the basic algorithm, it's outperforms both of them. And then for the, the rig was also used in in managing for the intercell down in the fields optimization where you could accurately predict the user interference in the darling and the accuracy is close to 93% and the provided with that prediction. It's managed you to provide 10% reduction in bit error rate and also 20% increase in the throughput. I don't have time to talk about how it is done but. Okay, and then for this one is also very interesting. We used the orange hierarchy Rick to provide the illness video quality assurance where you can see that we can predict the uplink. Well, the various surface camera moving around in different environment, we could not predict accurately the uplink quality with 98% accuracy, and with that, we can provide timely adjustment of the appropriate the colder the video code rate. This is without the hierarchical Rick, and this is with hierarchical Rick and you can see that is really matches the shape of the actual uplink. And with that, we managed you to maintain relatively stable the video frame arrival intervals meaning avoided major delays like what's shown here without the hierarchical Rick when then you will suffer from frame loss and the stalling of the surveillance camera. Okay, several releases has been has been provided to the ecosystem. The first one was focusing on open second one intelligent. And the third one will come up as a package of several important features that our ecosystem are waiting for, including massive MIME optimization and saving share or you relative exposure and the assessment on this slicing in the SMO. And all ticks I think I have described a little bit earlier, but this is the map of where they are and there are two more coming up one more in from North America and one more from a pack. And this is a typical setup of our opens, open labs of open source community, where you can bring in your latest software realization to try out on this testing for CI CD pipeline in the open labs. And some simple deployment and trials of orange technology by various operators around the globe for references. There's a lot of activities going on. Now, still, all of that work from the working group focus group, solving and providing capabilities, building on 5G. And in the meantime, also getting ready for 60, the NGRG was set up officially in June this year. And the, it's a technical oversight committee are composed of six operators to from each region and six vendors to to from each of the city IT and DT in the segment and three academic thought leaders went from each region and then a cyber scholar to together they shepherd now currently five research streams for NGRG. The first one is on gap analysis for interested 60 use cases. And second one is on the overall architecture. And third one on native AI source one native security and the fifth one is on the platform and the evaluation methodology for 60 concept. And the primary goal is to make sure that first of all that their effort would take into consideration of regional research efforts, including ITU and the switch PPP regarding 60 and also avoid incompatibility and make sure that some of the experience and the capability that has been built up in this could be carried forward into the 60 design. Okay, very quickly, I just want to talk a little bit about a very important aspect of a convergence, which is cross standards organizations. As an example, our intelligence, you can see that ITU T, Etsy, switch PPP, and all run. They all have various efforts. I think over I have talked a lot about it sufficiently. So let me just quickly show you the switch PPP work intelligence, not only as a tool for the core network and as a five for the management plan, but also now rain three on data collection and enhance the song and rain one on the CSI feedback and be management position accuracy improvement are all making use are all bringing AI and machine learning capability into the standards. And Etsy has many committees that I have it all listed here that has AI related work, particularly the experience, experiential network intelligence and the zero touch network and service management. And then in ITU SD 13, 16, 17, 20, all have activities related to AI and machine learning. Now, for clarification, you'll see that the primary actor are not really SDOs. Yeah, you have Etsy here and you have, I guess, or could be kind of pseudo SDO, but mostly primarily it is among various open source communities open source initiatives. And you'll see that they are also kind of complimentary and interconnected with different focuses. The reason why I'm showing this is because I think going forward. For 6G, unlike previous generations, I think in addition to 3GPP, there will, there will need to be explicit and expanded collaboration among multiple SDOs, which would include not just ITU, but Etsy and IETF to cover fully the ITCT and the DT aspects, because this convergence is really major and not all of the aspect could be or should be covered by one SDOs. Even though 3GPP will still be the main stage where the fundamental 6G spec will come from there, but I really am urging everyone to seriously consider how we can form a efficient explicit collaboration across different SDOs so that we can divide and conquer. And the rest of the reflection here, I think it's kind of a continuation of what I have covered, so I'm not going to spend time because I think time is up, but I do want to focus on this particular aspect on this page. Okay. As a reference, there's a book on 6G that was published early this year. If you're interested, please take a look. And then also about the ORAN Alliance's 6G effort to the research group will be holding its first workshop October 20th, Agenda is here. And also there is a NGRG session in an industry summit on open run October 26th. Hopefully you will be interested enough to take a look to see if you will be able to join. Both of these are going to take place in Madrid, Spain. Okay, this is my last page. I think the future belongs to all of us, and it has to be created by all of us. Again, all the ecosystem players and the collaboration from among the multiple SDOs will be essential for such a future of multifaceted convergence. Okay. I hope I'm confirming time well. Thank you for your attention. Thank you, very nice talk. You mentioned about these holograms and also the metaverse as well as the, you did not use the word mouse media but you know, like this, these five senses. Can you tell us a little bit more about your activities within China mobile. For example, in most media, are there some hardware design issues for like, sensing, I mean, audio or taste or, you know, these devices or, or platforms for metaverse. What are the activities within the China mobile if they are not, of course, confidential right. Well, yeah, sure. We actually. First, I want to tell you a story of the internet of a taste that was prototyped in the 90s using 2G network. That I had the pleasure of visiting by a different company. I was, I was very curious about it when they mentioned that okay, and then it turned out that it was you are just transmitting some simple bits, because at the receiver side, there is a kind of rotating plate with eight different fragrance of perfumes. So now you know the solution right just you just send the bits that will tell you which of the eight fragrance that is being transmitted. And that would be that would be activated. Yeah, but that's the in ancient times with 2G. This mapping problem right. Okay, but now we have much more sophisticated technology right with like the sensing sensor devices. I think there are some hardware design going on in the world. I'm not sure about in China mobile that you know they have these sensors they can get the taste or the smell, and then they can pack that and transmit as data. There are some work going on so I want to know, do you have those activities within the China mobile or like these metaverse platforms. I assume, I mean China mobile China is a big market right so you must have some platforms that we do not know here so most of the platforms we know they're all Americanize you know, but in China. You know, is there. I'm sure there are many of them. So, how far are you are you leading that better than Americans or what's going on. First, I don't think we are leading and then second that if if we are leading probably not to not done by China mobile that so that I'm not aware of. But I think certainly the touch the senses are being, I think, closer to to being materialized. And, but in terms of the taste and the smell. I haven't I haven't personally seen any but the closest one was the one I just told you this was in the 90s. But lots of people are very excited about this possibility and also something that always gets me very curious and it's about this so called synesthesia perception. And I'm not sure how much experience and data we have for that, because supposedly, maybe, when you see something, you'll feel like you smell something. So you use maybe your visual stimulant to trigger a sense of a smell. And whether that that is something that can can provide kind of the transmission of the senses, maybe that that could be one possibilities. But no, I don't I don't I don't personally, I'm not privileged to know anything that's that is close to reality even in the labs. No. I, I changed the focus now I want to go to this open network foundation type you know the STN NFVs etc. And I used to work with Huawei in Shenzhen 10 years ago we got like many patterns during that time. Like all these you know controllers multiple controllers load balancing etc. But those years we did not use any AI or machine learning techniques. So since you are leading many efforts in that front. I don't follow the literature anymore the last four years. Are there a lot of. Is there a lot of research going on that these people are looking into how AI and ML can be used to have much more sophisticated, like, for example, slicing, or software defined networking. Like, I'm talking about. Well, you know, you know that there is this hot topic called the autonomous driving network. Yeah, yes. I don't personally like that term, but I think certain level of autonomous operation and optimization will be hotly pursued. Not just for 6G but I mean actually 5G advance will be doing that as well. And the only thing is maybe it's currently the step that's been taken by 5G advance is still like very small steps, very. But like, what we do in Orion Alliance is not something that overnight happens, because we started looking at in the initially we call it the wireless data. What we can do, and what kind of data we need, right where can we get it. And so, we actually started this in 2014 2015 time frame and so we actually used this AI machine learning technology to try to solve lots of optimization problem in management domain in control domain as well, including like, like you said load balancing, we actually do it, and in energy saving we actually do it. And the immobility management interference optimization and in QoS QoE actually we like a cross domain optimization I showed some examples that we did in Nanjing and Hangzhou, which the cities you know right you're being injured. Yeah, so we actually can use the AI machine learning to do certain predict level prediction and the provide in turn with that prediction optimization. And I think more interestingly is we are trying to work out a way to be able to share and expose the prediction and the state for third party application developers so that they can use that information to also optimize the performance of their predictions. So, it is as we are talking it is going on right now, but you will see that it would be easier to be put into the actual network for the management plan optimization because you know the time is not usually is much coarser time resolution that is more realistic, but to build it into the control plan is probably something that the 6G era 6G network design should start with. I used to say, CRAN started after 4G was already spiked out right so CRAN capability, the green and the soft capability, and also the centralized deployment, etc, would be a starting point for 5G. And now I think the analogy that I would apply is ORAN, which would have embedded intelligence machine learning. This really as far as CRAN goes only started after the 5G spec is already spiked out without it, but this would be the starting point for 6G. And as you can see that the 5G advance is already slowly bringing some of the element except with very small steps. But it should be, I mean all of this accomplishment and capability should be definitely the starting point for 6G design. The 6G should be designed with embedded intelligence, which means that a lot more data in the network needs to be available as a built-in capability. I mean today there's a lot of data that's not available to operator or to third party. It's kind of closed in a black box. But going forward, a data availability being a built-in characteristic needs to be there, so that 6G would be designed from day one with native AI. You know, you mentioned about data. You know, we have all this quality of service notions and quality of expectations. Also the quality of data is becoming more and more important because we have terabyte or zeta byte of data floating around. There are these data brokers and it's really time to also distinguish between the quality of data. That's maybe the semantic communication maybe going in that direction too. Like, you know, how meaningful is the data that you are getting or collecting. And so, especially for 6G, we'll have all these platforms you mentioned about holograms and avatars. I call them avatars by the way, digital twins. And also the like metaverse, there will be gigantic data floating around right. So it's really good to look at the quality of data also like you know what is the meaning of quality of data right. It's because it changes based on person's interests. I mean, some data that it's interesting important quality for you. Maybe it's not for me. So it's really a difficult question actually, you know. That's indeed true. I think data as the core as the base for everything going forward will be I think one of the major challenge because our experience shows that even when we were doing our early stage data, which was not really the control or scheduling domain, it's the management domain. And the data we can get from the actual network. You will see that there's a lot of polluted or incomplete or missing or redundant. Any problem you can think of for the data, they exist and it turned out that the effort that you need to sort out the data and get a set of useful clean data in shape. So that effort actually it may be much greater than effort of coming up with the best algorithm or best AI machine learning models from this data. So this is a major challenge for sure. And another thing is what I remember when we first started the launch of the Oran Alliance, one of the focus group that we want to set up. Because we have standard development focus group open source focus group and also testing integration focus group in addition to all the working groups. One of the very first focus group we wanted to set up was called open data focus group. Yes, first recognizing how important data is second also there was a crying out for use for open data set from our ecosystem from academic community to industry, like, but that focus group has not been set up. Even today, it's very difficult. I have one more question by the mid 2025 China was planning to have these autonomous vehicular traffic. What's going on there you know is it are there still developing it because communication is very important subject there as you know like the driverless cars in other words. Is it still the target or what's going on. Well, you know that actually the driverless vehicles already are in use in some some area, but in general, in general, it's a closed campus. And, and there are also very large scale trial field where is a hybrid with some of the kind of a private isolated territory combined sometimes with the open road. Okay. Yeah, but whether by 2025, there will be large scale nationwide deployment. I don't know because there's a lot of interesting this some various. There's at least I think the three forces right. When are the automobile manufacturers. They feel that they don't need a network. They can just feel the intelligence with enough sensors. I remember I was very, I don't want to say traumatized very whatever impressed by NVIDIA's God. They told me that they don't need any network because as long as you have a media in the vehicle coupled with enough sensors and later as well then any cars can be made independent the driverless. I don't know. I don't know about that. Okay. The second, the second force is, of course, the operators like us that we would like to be the one facility enable building that ability with our own encompassing coverage of our network. And but there is yet a third one, which is the transportation department, the transportation industry, right, because the road. The roadside unit is generically, they don't belong to operators. The roadside units infrastructure is easily deployed by the transportation industry or the road. The public road bureau. So I think that there are all these forces trying to make this happen. And maybe I don't know about 25, but I think there must be a backbone network because some the some data must be collected what's going on in the on the roads right so maybe they say we don't need it these drivers or whoever the owners, but the transportation department may need the statistical data what's going on in terms of accidents or any problems. So there must be some network in my opinion. But the question is, we'll pay for the services right so the users may not pay. Yeah. You know, let me say this also because I've shared this in some other occasions a while back. When autonomous driving became a topic. I always thought it would be easier if we identify one lane on the road as a Thomas driving lane, so that anybody who is tired of driving can just enter that lane, and then all the vehicle on that lane can be controlled by you know, some special red dates roadside unit or it's the intervehicle communication, so that they are kind of all behaving properly. And then so the challenge it would just be how you enter such a lane safely and how you want to drive it over if you want to turn over exit how you come out of that lane safely. So I thought this would be a scenario that can be deployed globally the easiest everywhere, but somehow I haven't seen that that that happens. I agree with you. That's the easiest and the fastest solution like we have this H O V lanes here in the US. Right. Yes, but you cannot enter all the time so there are always certain areas, you can enter H O V lane, and then you continue and then again certain areas you can leave. So that will be interesting, but so still not the totally deployed and what I understand in China because I thought that you folks are very ahead of us but so it's still going on. So it's interesting. Okay, sorry for interrupting you. There's one question in the Q&A box if you want to check. Okay, good thanks to for reminding me. The question is, what will be your moon shot for 6G. I am a little bit too old to have a wild imagination for moonshot. Okay, because metaverse definitely is a moonshot of a moonshot, I think. But I think a reasonable degree of a digital twin network for 6G would be my moonshot. I think that would definitely be that would require a very efficient integration of all of the technology, IT, DT and CT. And I think I would be happy if the CG network really were designed with actually embedded intelligence and has a clever way of ensuring the availability and the quality of the data that's needed. And providing sufficient level of autonomous operation, both in management and the control and the scheduling. I don't mean, I think, sorry that it doesn't sound very, I guess, the old words are sexy or over exciting, but I'm too old. You are not. But I'd be happy if 6G would accomplish what I just said. Maybe I can add my opinion about this after I found out what moonshot means. Well, again, if that's that's the meaning of the question. So the moonshot for 6G or 7G or 8G will be the propagation delay problem. Okay, as long as we are not in the quantum communication domain, then we'll have this propagation delay effects. And since we have all these latency problems for the applications also mentioned by Dr. E, that is, like all these metaverse and holograms and some other applications here, the latency is the biggest problem. And whatever we'll do intelligence edge computing with all this AI machine learning. The problem is we cannot beat the nature about propagation delay, unless we move to quantum communication. And although the quantum communication is really since three decades that I'm alive or maybe longer up and coming up and coming everybody is doing quantum communication. I'm not far away from that really. So, believe me, it's like, you know, we cannot deal with these gigantic data centers running around with data centers in our hands, quantum. There is no quantum cell phone even forget other things right. So it's very cumbersome, but long time maybe not in our lifetime but I think the propagation delay must be somehow combat it over the center. And that's the only way is the quantum communication my opinion. So I hope, you know, I could answer your question to in terms of the propagation delay, which is actually good that we are. Another question, which is good that we make, you know, living out of this right we do research and trying to combat this problems. Anyhow, there is another, his name is Ashraf. Like, I'm in India right. Could you please show the slide of your G book once again. Just, can you see it. Yes, we can see it. Okay. So you are the author of this chapter right. Okay. It's edited. I mean, the editors are on the cover, I can see. Yeah. So I hope you're fine. Okay. Okay. Any questions. I think, again, I would like to express my sincere thanks to you. And we close the Q&A session. And unless you will start the whisper corner or what a wisdom corner right. Thank you very much. Thank you for moderating this session and thank you to our speaker for this very interesting and informative presentation. So yeah, let's move to the wisdom corner live life lessons, which is based upon the idea to give a unique and special angle to these series of webinars, adding a personal touch. So successful researcher like Dr. Chilly today will guide students and young scholars in the field of current ICT research, and we'll also share some impactful life lessons. So let's start with my first question. So Dr. Chilly, which is your hard earned life lessons or failure that you would like to share with us today that might perhaps help someone attending the webinar. Okay. So it's hard to to identify a single event of hard earned lesson. I would think that as I, but this could be personal because maybe my view of the world used to be a simplified very much simplified so the right way there is the wrong way, right? But actually, as I get older, I think I've learned that there is always a variety of options because if you, if you have just like wise communication is diversity, right? If you have a diverse background, putting together, you actually could find the best way. And so in when I was younger, I kind of tend to exist. It's my way is the right way. But actually, it's only after I joined the old 18 develop research. It's from my esteemed colleagues. I learned that really, there is not only one right way in solving a problem. Okay. So just be open minded. I think that's very important. So the second question is the rich strengths and capabilities. Do you think that students and young scholars should be most focused on developing and how they should accomplish this? I think, to me, I think what I benefit to the most throughout my life is curiosity. I think life is a very precious journey and for your, I mean, the opportunity to learn continuous learning and be curious and also do not take the answer from the books as the only answer and final answer. There to challenge. There to ask questions. There to try things that there is no set paths. There's no one else has done it before. I think that is a capability that would bring you a very exciting career. And in which field in particular and which topics would you recommend students to study? You know that there's a lesson that I learned actually at the high school middle school graduation ceremony of my youngest boy. The lesson was given to all of the parents in that ceremony by the middle school principal. The principal reminded all of the parents in that event that please notice your kids, meaning the middle school graduate in those days. By the time your kids is ready to become part of a, become adult, enter the society, the world would be so different from that day to day in that day. So do not assume you'll always know what's best or what's the only thing that your, your son or daughter should pursue because they will know better than you because the world is constantly changing. Okay, but having said that, I would, I wouldn't want to say, in addition to as long as you are always interested in learning and you are curious, and you will benefit from it. In addition to that, I think any field you are interested in, as long as you love it, as long as you have passion, you will do well. There is one thing I think going forward very important that is for most of the people. You would, you would do well by having cross domain knowledge, cross field knowledge and experience. Okay, so you need to in a way that for the thing you really, really love, you'll need to have a depth. Okay. But in addition to that, it's important, you also have a knowledge of, you know, in addition to this vertical depth, you'll have a horizontal layer of cross domain knowledge. That is very important. Okay, and for example, we talked about intelligence, AI machine learning a lot. I think going forward doesn't matter what field you are. Okay, you will probably always need a little bit of ability, utilizing AI and machine learning as a tool to enhance what you're doing. But you also need to keep in mind that after all machine learning AI, pure data driven solution may not be the most efficient approach because a brute force throwing in, you know, computers and the computing power may not be the right way to do. First of all, it's not very green, right? So you should also remember there's a lot of experience and knowledge that has been built up prior to this day. So the best way going forward would be AI plus HI. Okay. At the end of the day is hybrid with human intelligence. I think going forward that would be some principle that doesn't matter what field you are in, you need to keep in mind. Wow. Thank you so much. One more thing is I would like to call to action and attention for all the young scientists and technologists to remember why you are doing it. And what is the real meaningful and valuable goal of what you do? I think, for example, when we talk about 6G, very early on, many researchers already thought very well to incorporate UN SDG, the sustainable development goals into consideration. So even if we are mostly majoring STEM, I think reasonable consideration of the societal impact and what's really valuable, what's really sustainable, I think it's very important to have that mindset. Because the future, for better or for worse, will be enjoyed, hopefully not suffered by the young technologists that we are sharing this with. Thank you so much. Very interesting. Wow. And tell us one of the most tangible contributions that you have made in your career that have an impact on your life, professional or personal life and that you're most proud of. Technically, I have done a lot of interesting thing. I always consider myself very lucky that I work in this field. Actually, my PhD dissertation in Stanford University was on satellite communications. But ever since I started working, I stayed on the terrestrial system mainly, even though now for 6G some people are start looking at satellite again. But I think I was very lucky to be part of this very exciting industry in its flying and growing phase for many decades. I did something I think was fundamental in the CDMA system. But to me, I guess the latest is something that you remember the clearest, right. I think my effort going from green and soft to open and smart that carry forward our effort from CRAN to wireless big data to to O-RAN is something personally I feel the strongest about of course this is the past 10 years and so maybe I remember it to the most, but I kind of feel that the impact, particularly of the Alliance is very, very visible, even though it's not without the controversial controversy because of some regional issues and industry incumbent players versus you know the emerging new players. But I feel that clearly focus on this ITCT and DT convergence and recognizing that this is really a necessary and essential way for for the future. Because our ecosystem, our industry, the wireless communication industry has enjoyed great growth in the first three, four decades is still growing, but we have seen, since more than 10 years ago, we have seen that the industry is really kind of stagnating. And there is obvious upcoming transformation that needs to take place to change the ecosystem so that we can embrace and welcome a lot more new players joining this ecosystem and making it an open embracing and the agile future. I think so far I feel that the Orion Alliance is doing that, even though there's still a long way to go to the time that we can say mission accomplished, but this is something that I feel very close to heart and very happy about. And I think I look forward to a lot more good works and the great works together with everyone globally on this. Thank you. And my last question, something more personal maybe is there is there a motor and a for is my book, a movie, a piece of art and music that you would like that describes you best and you would like to share with us before closing. I think it's nothing fancy, but it's just something that I actually wrote in the, in the graduation book, when I was graduating from college, actually I was still in Taiwan at the time. And, you know, there's, we have a tradition of writing something. Okay. Each of the graduates would write something on the graduation book. I wrote something very simple plan, which is do what you love and love what you do. Okay. I think both are very important. And I feel that somehow just from the bottom of my heart, that's something that I have been living with do what I love and love what I do. And I feel very blessed with that. Thank you so much really for being so generous and for sharing with us these really inspiring life lessons and pills of wisdom. Thank you. Thank you very much. So I'd like to ask Ian to join us before we close. And yeah, thank you so much. Yeah, now we can see you. Yes. I would like to thank you both. Ian, the floor is yours, please. Thank you, Alessia. Thanks for leading the last session. Again, really thank you for spending time with us. Hope we'll see you in person someday again. And enjoy the rest of the week. And I think it's evening right in Beijing. Are you in Beijing or in the US. I am currently in Beijing. Okay, so it must be evening so enjoy your evening. Yeah, so stay safe and healthy and good to see you really. Thank you very much. And we look forward to seeing you all also for the next webinar at 27th of September. Thank you again. Thank you. Bye bye. Thank you. Thank you. Thank you. Thank you.