To embed our video on your website copy and paste the code below:
<iframe src="https://www.youtube.com/embed/9dTNkzC9I28?modestbranding=1&rel=0" width="970" height="546" frameborder="0" scrolling="auto" allowfullscreen></iframe>
Guy Daniels, TelecomTV (00:24):
Hello, you are watching the Open Run Summit part of our year-Round DSP Leaders Coverage. I'm Guy Daniels now as open run deployments start to scale and feature not only in greenfield sites but also existing Brownfield networks. The wider industry is closely following the actions of the major telcos and one of these leading CSPs is Orange and I'm delighted to say that joining me now is AUSA hat who is director of innovation in Radio and Environment at Orange. Hello Teresa. Good to have you back on the program. It's been a couple of years since we last spoke. Now let me first of all start by asking you about the position of Orange on Open Run. What's its assessment of the technology and its viability to orange?
Atoosa Hatefi, Orange (01:17):
Yeah, actually we're confident that open RAN will happen, so the question is not if, but it's when. So we are interested in open RAN and our interest is relying on three main pillars, so openness, cloudification and intelligence and these different pillars are not all at the same level of maturity. So if you start by the openness, so it's a key requirement for us in order to increase under diversity to SWAs. In short term we are actually considering open frontal ready equipment as a main requirement and it's key for us. But in longer term we would like also to extend this openness to the cloud infrastructure part and we would like to actually see a smooth, let's say portability of vRAN software across diverse cloud infrastructure. And this is today less major compared to the open front top.
(02:18):
The second pillar cloudification, so actually we have a strong interest towards unified cloud infrastructure and orchestration with native CICD in order to reduce the opex. Cloudification is today more or less major for rural use cases and this is already actually the case. I mean we are already actually trialing it in Romania in a pilot and it is successful for V two G and V 4G and soon we'll test V 5G. But for more complex scenarios, particularly in urban with high capacity, let's say radio configuration with massive MIO, there are still some improvements to be done on the cloud infrastructure part, particularly in terms of dimensioning and energy consumption. But we are confident that this also will happen in the near future, particularly with the next generation of chip sets. And the last pillar intelligence, so higher automation and achieving this level four is one of our main objectives with higher customization, higher reactivity in terms of network operation, network management and network optimization. We are targeting actually a multi-vendor, multi technology and multi-domain SML including non realtime R which should be able to work not only with open cloud ran but also with traditional ran. And in that spirit we are considering the migration of our CS L towards non real-time R including with our traditional RAN network.
Guy Daniels, TelecomTV (04:00):
Fantastic positive message there and thanks so much for going through the various elements that you keep your focus on. You mentioned some progress that you're making there and some trials and what you, are you able to provide us with a summary update on what your deployment plans may be?
Atoosa Hatefi, Orange (04:20):
Yeah, actually deployment of open RAN and its introduction in our networks will be gradual as the Brownfield operator we are very demanding requirements and also introduction of open RAN comes with some impacts in terms of operation and which requires also lots of upskilling on our side. That's why we are considering this migration from traditional ran to open RAN as a journey. We are today in a learning process we are progressing very well thanks to different initiatives that we have. So pko in the past different lab testing including or plot fest in our premises and also the ongoing pilots in Romania. So in terms of introduction, we are targeting first Europe on a case by case basis, so as I said, rural is the first case that we're considering, so we have already a pilot ongoing and urban would come next when it is major, particularly for high let's say capacity radio configurations in urban and once the improvements in terms of dimension and energy consumptions are already there. The deployment scenarios that I mentioned was rural, urban, once major, but at the same time we are also actually studying and evaluating some other opportunities for deployment, particularly in the context of private networks and this could also be beneficial in order to be able to have a kind of unified cloud infrastructure for different type of workloads for enterprise use cases.
Guy Daniels, TelecomTV (05:56):
Great, thanks very much. That update toa, now open run has the promise to extend the range of potential suppliers in theory lead to more competition, but can smaller alternative vendors really compete and have direct relationships with telcos or do they need to be preintegrated and bundled through more established vendor, perhaps one who already has that existing commercial relationship with a telco?
Atoosa Hatefi, Orange (06:23):
Yeah, actually it's true that open rank comes with this promise of mix and match between different vendors increasing let's say vendor diversity, increasing flexibility of deployment including with smaller vendors. But we should not forget that it's a new model particularly for brownfield operators who are used to single ran deployments. So we need to consider again this introduction step-by-step. I would say that in the initial phase, pre-integrated solutions bundled with well-established vendors could actually speed up the readiness of deployable solutions. But in the longer run we can actually work and we are actually working actively in order to have a kind of coordinated testing and a global certification framework to manage in a more efficient way this intergroup testing between different vendors and reduce this, I said test burden on operators and avoid all of these redundant type of testing and this should at the end help this smaller vendors to, and this should give them a wider access actually for integration and for interoperability type of testing.
Guy Daniels, TelecomTV (07:38):
Yeah, we'll be certainly welcomed. There's growing interest in incorporating AI within the ran including its ability to optimize operations. There's growing interest to incorporate AI everywhere. What is the likelihood of an AI native open ran? Is this something orange would be keen to develop or keen to see being developed?
Atoosa Hatefi, Orange (08:00):
Yes, indeed. Actually AI is taking a growing role in the RAN at different levels. We have AI for operations which started to become actually available in commercial networks for multiple use cases such as predictive maintenance such as anomaly detection or root cause analysis and so on. We have also AI for network optimization. So here initially we're considering AI for radio high layers for example to optimize mobility, to optimize load balancing energy efficiency. But then potentially later we can also consider AI for lower radial layers including radio physical layer. But of course it'll depend on the gain versus pain. So I would say that as such, open ran is indeed aimed to be AI native and we need to consider several say architecture principles to introduce it. As I said before, a multi-domain and multi-vendor SMO is key with open interfaces such as O one, O2, A one that could bring intelligence to our network and to the whole automation processes. And then we also need to consider support of our week applications, particularly to a non-real time r, again with supportive open interfaces and in a multi-vendor environment. And this also needs to be able to work with our traditional right
Guy Daniels, TelecomTV (09:32):
Now you mentioned Rick there. How closely are you following developments with the R? What role do you see for xap and R apps in your radio network?
Atoosa Hatefi, Orange (09:43):
Actually, we're actively following the development of Rick and we have several initiatives, test initiatives inhouse. I would say for us this RIG framework comes with several benefits. It brings a kind of standardized framework for multi-vendor operation and interworking, which actually help us to have a better, let's say interworking compared to what we have today with csun. It also brings more, let's say differentiation for operator with customized apps and it bring also kind of more homogeneous experience across different RAN vendors with third party applications. So these are the main benefits that we see with standardized RIG framework. And in that sense our main objective, our first main objective is to actually ensure the migration of our CSUN towards the non realtime. Rick, as I said, not only for open ran cloud ran, but also with traditional ran the role of SAP and near realtime. Rick is not that clear today for us compared to the non real time R, which is more major today. We still need to evaluate and we are evaluating actually the real gains and use cases with near real time ECAP that could really bring value for us.
Guy Daniels, TelecomTV (11:13):
That's great, thank you very much. And a final question for you. We've been discovering during this summit that good progress has been made, but there are still challenges ahead. So what do you see as the challenges that Open Ran still needs to overcome in order to fulfill its potential?
Atoosa Hatefi, Orange (11:32):
Indeed, there are few challenges that needs to be overcome to allow wide scale deployment of open ran, but we don't see any showstopper actually that prevent open RAN to be widely adopted in the near future. So among the challenges that I would like to mention, so I would say first software parity feature parity compared to traditional ran, but we see that today it's improving very well in the ecosystem depending on the vendor of course we have already solutions that are available in the market with I would say near feature parity compared to the traditional ran. Then we have energy efficiency to be improved, particularly for high capacity radio configuration. And this is more related to the energy efficiency of the cloud infrastructure as I said before. So some improvements in terms of let's say dimensioning and energy efficiency still to be done, but we're confident that this will happen with the next generation of chipset and also with the target of one server per site that is now considered in the near future. Then another aspect I would like to mention is interoperability. That still needs to be improved both on the open frontal and also on the portability of the vRAN software over different say type of cloud infrastructure. We have the SMO maturity that needs to be improved in order to truly allow multi-vendor and multi doin management of both cloud ran, open, RAN and traditional ran with a real open rec marketplace.
(13:21):
Then there is this, let's say challenge related to integration, which is still remains actually. We think that the global integration framework in the industry could help to reduce the test burden, the INTERABILITY test burden. So on our side in the short term, we are relying on vendors to take the lead for this integration part. However, in the long run we don't close the door for us to take a more active role actually in the whole process, which would also bring us actually more autonomy in the choice of our partners. Another point that I would like to actually add is related actually to operators side and the fact that we need also to have a change of mindset with introduction of open ran and to adapt our processes and particularly for example, related to vendor management and also all of the reskilling and upskilling that would be required for automation and ization. This is also an important point to keep in mind and to finish, I would also like to highlight a more, let's say generic question for the whole ecosystem longer term, and this is related actually the question whether two lines of products will be maintained over time or if eventually it'll converse to single line of products, which will be based on open ran cloud trend.
Guy Daniels, TelecomTV (14:58):
Very interesting and a great question there at the end. We'll put that to our panels later in the summit and see what they think as well. We must leave it there. Thank you so much for taking part in our show today. Great insights. Now don't forget the full scheduler programs and speakers can be found on the telecom TV website and please do take part in our poll. Goodbye for now.
Hello, you are watching the Open Run Summit part of our year-Round DSP Leaders Coverage. I'm Guy Daniels now as open run deployments start to scale and feature not only in greenfield sites but also existing Brownfield networks. The wider industry is closely following the actions of the major telcos and one of these leading CSPs is Orange and I'm delighted to say that joining me now is AUSA hat who is director of innovation in Radio and Environment at Orange. Hello Teresa. Good to have you back on the program. It's been a couple of years since we last spoke. Now let me first of all start by asking you about the position of Orange on Open Run. What's its assessment of the technology and its viability to orange?
Atoosa Hatefi, Orange (01:17):
Yeah, actually we're confident that open RAN will happen, so the question is not if, but it's when. So we are interested in open RAN and our interest is relying on three main pillars, so openness, cloudification and intelligence and these different pillars are not all at the same level of maturity. So if you start by the openness, so it's a key requirement for us in order to increase under diversity to SWAs. In short term we are actually considering open frontal ready equipment as a main requirement and it's key for us. But in longer term we would like also to extend this openness to the cloud infrastructure part and we would like to actually see a smooth, let's say portability of vRAN software across diverse cloud infrastructure. And this is today less major compared to the open front top.
(02:18):
The second pillar cloudification, so actually we have a strong interest towards unified cloud infrastructure and orchestration with native CICD in order to reduce the opex. Cloudification is today more or less major for rural use cases and this is already actually the case. I mean we are already actually trialing it in Romania in a pilot and it is successful for V two G and V 4G and soon we'll test V 5G. But for more complex scenarios, particularly in urban with high capacity, let's say radio configuration with massive MIO, there are still some improvements to be done on the cloud infrastructure part, particularly in terms of dimensioning and energy consumption. But we are confident that this also will happen in the near future, particularly with the next generation of chip sets. And the last pillar intelligence, so higher automation and achieving this level four is one of our main objectives with higher customization, higher reactivity in terms of network operation, network management and network optimization. We are targeting actually a multi-vendor, multi technology and multi-domain SML including non realtime R which should be able to work not only with open cloud ran but also with traditional ran. And in that spirit we are considering the migration of our CS L towards non real-time R including with our traditional RAN network.
Guy Daniels, TelecomTV (04:00):
Fantastic positive message there and thanks so much for going through the various elements that you keep your focus on. You mentioned some progress that you're making there and some trials and what you, are you able to provide us with a summary update on what your deployment plans may be?
Atoosa Hatefi, Orange (04:20):
Yeah, actually deployment of open RAN and its introduction in our networks will be gradual as the Brownfield operator we are very demanding requirements and also introduction of open RAN comes with some impacts in terms of operation and which requires also lots of upskilling on our side. That's why we are considering this migration from traditional ran to open RAN as a journey. We are today in a learning process we are progressing very well thanks to different initiatives that we have. So pko in the past different lab testing including or plot fest in our premises and also the ongoing pilots in Romania. So in terms of introduction, we are targeting first Europe on a case by case basis, so as I said, rural is the first case that we're considering, so we have already a pilot ongoing and urban would come next when it is major, particularly for high let's say capacity radio configurations in urban and once the improvements in terms of dimension and energy consumptions are already there. The deployment scenarios that I mentioned was rural, urban, once major, but at the same time we are also actually studying and evaluating some other opportunities for deployment, particularly in the context of private networks and this could also be beneficial in order to be able to have a kind of unified cloud infrastructure for different type of workloads for enterprise use cases.
Guy Daniels, TelecomTV (05:56):
Great, thanks very much. That update toa, now open run has the promise to extend the range of potential suppliers in theory lead to more competition, but can smaller alternative vendors really compete and have direct relationships with telcos or do they need to be preintegrated and bundled through more established vendor, perhaps one who already has that existing commercial relationship with a telco?
Atoosa Hatefi, Orange (06:23):
Yeah, actually it's true that open rank comes with this promise of mix and match between different vendors increasing let's say vendor diversity, increasing flexibility of deployment including with smaller vendors. But we should not forget that it's a new model particularly for brownfield operators who are used to single ran deployments. So we need to consider again this introduction step-by-step. I would say that in the initial phase, pre-integrated solutions bundled with well-established vendors could actually speed up the readiness of deployable solutions. But in the longer run we can actually work and we are actually working actively in order to have a kind of coordinated testing and a global certification framework to manage in a more efficient way this intergroup testing between different vendors and reduce this, I said test burden on operators and avoid all of these redundant type of testing and this should at the end help this smaller vendors to, and this should give them a wider access actually for integration and for interoperability type of testing.
Guy Daniels, TelecomTV (07:38):
Yeah, we'll be certainly welcomed. There's growing interest in incorporating AI within the ran including its ability to optimize operations. There's growing interest to incorporate AI everywhere. What is the likelihood of an AI native open ran? Is this something orange would be keen to develop or keen to see being developed?
Atoosa Hatefi, Orange (08:00):
Yes, indeed. Actually AI is taking a growing role in the RAN at different levels. We have AI for operations which started to become actually available in commercial networks for multiple use cases such as predictive maintenance such as anomaly detection or root cause analysis and so on. We have also AI for network optimization. So here initially we're considering AI for radio high layers for example to optimize mobility, to optimize load balancing energy efficiency. But then potentially later we can also consider AI for lower radial layers including radio physical layer. But of course it'll depend on the gain versus pain. So I would say that as such, open ran is indeed aimed to be AI native and we need to consider several say architecture principles to introduce it. As I said before, a multi-domain and multi-vendor SMO is key with open interfaces such as O one, O2, A one that could bring intelligence to our network and to the whole automation processes. And then we also need to consider support of our week applications, particularly to a non-real time r, again with supportive open interfaces and in a multi-vendor environment. And this also needs to be able to work with our traditional right
Guy Daniels, TelecomTV (09:32):
Now you mentioned Rick there. How closely are you following developments with the R? What role do you see for xap and R apps in your radio network?
Atoosa Hatefi, Orange (09:43):
Actually, we're actively following the development of Rick and we have several initiatives, test initiatives inhouse. I would say for us this RIG framework comes with several benefits. It brings a kind of standardized framework for multi-vendor operation and interworking, which actually help us to have a better, let's say interworking compared to what we have today with csun. It also brings more, let's say differentiation for operator with customized apps and it bring also kind of more homogeneous experience across different RAN vendors with third party applications. So these are the main benefits that we see with standardized RIG framework. And in that sense our main objective, our first main objective is to actually ensure the migration of our CSUN towards the non realtime. Rick, as I said, not only for open ran cloud ran, but also with traditional ran the role of SAP and near realtime. Rick is not that clear today for us compared to the non real time R, which is more major today. We still need to evaluate and we are evaluating actually the real gains and use cases with near real time ECAP that could really bring value for us.
Guy Daniels, TelecomTV (11:13):
That's great, thank you very much. And a final question for you. We've been discovering during this summit that good progress has been made, but there are still challenges ahead. So what do you see as the challenges that Open Ran still needs to overcome in order to fulfill its potential?
Atoosa Hatefi, Orange (11:32):
Indeed, there are few challenges that needs to be overcome to allow wide scale deployment of open ran, but we don't see any showstopper actually that prevent open RAN to be widely adopted in the near future. So among the challenges that I would like to mention, so I would say first software parity feature parity compared to traditional ran, but we see that today it's improving very well in the ecosystem depending on the vendor of course we have already solutions that are available in the market with I would say near feature parity compared to the traditional ran. Then we have energy efficiency to be improved, particularly for high capacity radio configuration. And this is more related to the energy efficiency of the cloud infrastructure as I said before. So some improvements in terms of let's say dimensioning and energy efficiency still to be done, but we're confident that this will happen with the next generation of chipset and also with the target of one server per site that is now considered in the near future. Then another aspect I would like to mention is interoperability. That still needs to be improved both on the open frontal and also on the portability of the vRAN software over different say type of cloud infrastructure. We have the SMO maturity that needs to be improved in order to truly allow multi-vendor and multi doin management of both cloud ran, open, RAN and traditional ran with a real open rec marketplace.
(13:21):
Then there is this, let's say challenge related to integration, which is still remains actually. We think that the global integration framework in the industry could help to reduce the test burden, the INTERABILITY test burden. So on our side in the short term, we are relying on vendors to take the lead for this integration part. However, in the long run we don't close the door for us to take a more active role actually in the whole process, which would also bring us actually more autonomy in the choice of our partners. Another point that I would like to actually add is related actually to operators side and the fact that we need also to have a change of mindset with introduction of open ran and to adapt our processes and particularly for example, related to vendor management and also all of the reskilling and upskilling that would be required for automation and ization. This is also an important point to keep in mind and to finish, I would also like to highlight a more, let's say generic question for the whole ecosystem longer term, and this is related actually the question whether two lines of products will be maintained over time or if eventually it'll converse to single line of products, which will be based on open ran cloud trend.
Guy Daniels, TelecomTV (14:58):
Very interesting and a great question there at the end. We'll put that to our panels later in the summit and see what they think as well. We must leave it there. Thank you so much for taking part in our show today. Great insights. Now don't forget the full scheduler programs and speakers can be found on the telecom TV website and please do take part in our poll. Goodbye for now.
Please note that video transcripts are provided for reference only – content may vary from the published video or contain inaccuracies.
Atoosa Hatefi, Director of Innovation in Radio & Environment, Orange
Atoosa Hatefi outlines Orange’s position on Open RAN and how viable the technology is for the telco. She also comments on the expanded supplier ecosystem inherent within the Open RAN market and how that works in practice. There is also growing interest in incorporating AI in the radio access network, especially for its ability to optimise operations, which could lead to the development of an AI-native Open RAN.
Recorded May 2024
