I. INTRODUCTION
Beamforming is becoming a crucial issue for a number of radio frequency (RF) applications, ranging from wireless access networks to radars. In particular, antennas with non-mechanical beam-steering capabilities are required, in order to finely control their pointing directions without physically moving the antenna. Phased-arrayed antennas (PAAs) are composed by several discrete elements and allow us to obtain the desired beam-forming functionality, by controlling the phase of the RF signal at each antenna element. Active PAAs are attracting a growing interest thanks to their flexibility, since they are composed by a large number of active transmit/receive modules (TRMs), which can locally modify the phase of the transmitted signals. However, in this kind of PAAs, the cost of the large number of TRMs, the system complexity, and the remotization of the antenna still represent non-negligible issues.
PAA problems can be addressed by resorting to photonics, instead of focusing on traditional electronic solutions. Indeed, photonics techniques allow us to effectively control the phase of RF signals independently from their frequency, by means of phase shifters (PSs) [Reference Bui, Mitchell, Ghorbani, Chio, Mansoori and Lopez1] or true-time delay (TTD) [Reference Scotti, Ghelfi, Laghezza, Serafino, Pinna and Bogoni2, Reference Yaron, Rotman, Zach and Tur3]. At the same time, photonics guarantees low weight and dimensions, immunity to electromagnetic interference (EMI), and a relative low cost thanks to its high scalability. Moreover, the inherent broad bandwidth of photonic devices can help in realizing more flexible and frequency-agile systems. Schemes based on high-performance PSs have been demonstrated, but the achieved beam-scanning angle was limited [Reference Bui, Mitchell, Ghorbani, Chio, Mansoori and Lopez1]. On the other hand, TTD-based architectures can be implemented either exploiting chromatic dispersion [Reference Scotti, Ghelfi, Laghezza, Serafino, Pinna and Bogoni2], or realizing a network for adaptively switching the signals on paths with different delays [Reference Yaron, Rotman, Zach and Tur3]. The photonic implementations of the TTD approach have demonstrated to be easier than the electronic ones, but they may require very long spools of optical fiber unless an integrated system is employed. A more flexible implementation employing a programmable bandwidth-variable wavelength-selective switch (BV-WSS) [Reference Schröder, Roelens, Du, Lowery, Frisken and Eggleton4] has been presented in [Reference Yi, Huang and Minasian5], working on a single wide-bandwidth RF signal.
Recently, we have proposed a scheme for a beam-forming network (BFN) [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6], based on a BV-WSS for the simultaneous and independent beam steering of multiple RF signals. The software-programmable WSS have been exploited as a wavelength router to feed multiple antenna elements, and as a tunable PS to independently control the phase of each signal at any antenna element. Moreover, we experimentally validated the scheme by resorting to bulk components. Independent steering of multiple beams has been proposed in [Reference Yaron, Rotman, Zach and Tur3] and [Reference Vidal, Piqueras and Martí7], based on chromatic dispersion in optical fibers, and tunable notch filters and couplers, respectively; these approaches, however, are bulky and suffer from a relatively complex design. In this paper, we provide a brief summary of the experimental results of our previous work [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6]; additionally, we suggest two possible integrated implementations based either on PS or on TTD, depending on the bandwidth of the considered signals.
II. CONCEPT
Figure 1 reports the general scheme of the architecture proposed in [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6]. M independent RF signals RF1…RF M , with central frequencies ƒ RF1,…, ƒ RFM , are up-converted to the optical domain by modulating N-independent lasers at wavelength λ 1,…, λ N in an electro-optical intensity modulator (IM). After the IM, the optical signal is equally split to K × N-outputs BV-WSSs. The employed BV-WSSs are optical devices based on Liquid Crystal on Silicon (LCoS) technology [Reference Schröder, Roelens, Du, Lowery, Frisken and Eggleton4], and are capable of routing the signals to one or more among the N ports, controlling the amplitude and phase of the optical signal at each port, with a pixel resolution of 1 GHz (assumed that the employed optical carriers are in the optical C-band, i.e. about 1.5 µm).
Fig. 1. General scheme for phase-controlled beam steering.
The bulk BFN introduced in [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6] is an all-optical signal processing architecture for transmitting M independent signals, composed by K, BV-WSS-based sub-blocks. Processing signals in the all-optical domain allows finely controlling their phase (or, alternatively, their propagation delay). Eventually, a set of K × N photodiodes (PDs) convert the replicas of each optical sideband (SB) in an RF signal, whose carrier frequency is given by the frequency distance between the kth optical carrier and the SB central frequency. Every PD corresponds to one TRM driving a PAA element and, being the optical phase shifts translated into equivalent RF phase shifts, the overall obtained radiation pattern is composed by several, independently steered lobes.
The core of the presented scheme is the BV-WSS, which enables an arbitrary, reconfigurable phase shift on independent multiple RF carrier signals. The phase shifting accuracy strictly depends on the performance of this device. In a BV-WSS, the input light is spatially dispersed by a diffraction grating on an LCoS pixel grid that processes the incoming optical signal and reflects it back to the grating, where the processed light is focused to the selected output port [Reference Schröder, Roelens, Du, Lowery, Frisken and Eggleton4]. Each pixel acts on a portion of the spectrum, whose span depends on the pixel dimension. It processes both amplitude and phase of the signal, since it is possible to finely tune the liquid crystals transparency and refractive index via an applied software-controlled electric field.
A critical issue to be addressed any time optical/electrical conversions are considered is the power efficiency. In particular, losses are important in the down-conversion stage, since for signal up-conversion, an electric driver at the RF input of the electro-optical modulator should ensure the desired modulation depth, provided the modulator response is linear over the input signal dynamic range. The down-conversion process, on the other hand, suffers from the limited responsivity of the PD and from its saturation power, which limits the maximum input optical power and, in turn, the overall conversion efficiency. As usually done in microwave photonics application, a proper management of the amplification stages in the optical [Reference Singh8] as well as in the electrical section may help to get around the problem. Furthermore, a typically suggested approach for improving the optical signal-to-noise ratio, and ultimately increase the power efficiency of the link, is to employ optical single-side-band (SSB) modulation. Furthermore, the RF signals are employed to modulate several continuous wave (CW) lasers, each one with its own power. Hence, by exploiting a multi-carrier approach higher conversion efficiency for the RF signals can be obtained at the expenses of larger number of employed optical CW sources.
The proposed architecture finds its natural application in any field, where a steerable antenna beam is required, ranging from radar to mobile communications. Moreover, this system is suitable for antennas remoting, since the BV-WSSs’ output is in the form of an optical signal that can propagate for kilometers over an optical fiber with extremely low losses, negligible distortions, and total immunity to EMI.
III. EXPERIMENTAL ACTIVITY: SETUP AND RESULTS
From an experimental point of view, in [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6] we have demonstrated the validity of the proposed architecture by the performance analysis of one sub-block of the overall network. More details of the experimental activity for the validation of the BFN sketched in Fig. 1 can be found in [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6]. Here, we report a summary of the obtained results, which can be helpful to introduce the following sections of this paper, with considerations around an integrated implementation of the presented scheme.
In Fig. 2, we report the power spectral density (PSD) of the employed optical signals together with the programmed amplitude and phase response of the BV-WSS. Following a typical radio-over-fiber approach, four CW lasers with 11 dBm output power and 200 GHz spacing are modulated by two RF tones at ƒ RF1 = 12.5 GHz and ƒ RF2 = 37.5 GHz (black solid line in Fig. 2), where the four optical carriers with double-sideband modulations at ƒ RF1 and ƒ RF2 are clearly visible. Inside the BV-WSS, the incoming wavelengths are separated and spatially deflected through a diffraction grating; thereafter, the light impinges on a two-dimensional matrix of liquid crystal pixels, which can independently influence amplitude and phase of the impinging light, and route each pixel output to any of the output ports. The pixels matrix response can be controlled via software, imposing a phase and amplitude mask to each output port. This way, a fine-grained filter is obtained, which can shape optical signals all over the whole C-band (1530–1565 nm). The results obtained with the unmodulated sinusoidal RF carriers are expected to apply also for signals with bandwidth of few MHz, as those employed in remote-sensing applications, where a constant phase shift can be applied over the signal bandwidth without incurring in squinting phenomenon [Reference Tur, Yaron and Oded9].
Fig. 2. Optical spectra of the 0.16 nm-spaced input optical carriers modulated at 12.5 and 37.5 GHz (black solid curves), the response of the WSS ports in amplitude (gray dotted lines – refer to the vertical axis on the left) as measured with a wideband, flat noise source. The imposed phase mask is traced by the dash-dotted line (refers to the vertical axis on the right).
As depicted in Fig. 2, the phase of every carrier is not modified (i.e., optical phase set at 0°). On the other hand, each upper sideband (USB) undergoes a different phase shift with respect to the related carrier. The phase of the SBs corresponding to RF1 ranges from −90° at Port1, to 180° at Port4 increasing by 90°-wide steps. Likewise, the phase of the SBs related to RF2 decreases by 90° steps from 180° at Port1 to −90° at Port4. Eventually, a 40 GHz-bandwidth PD is used for down-converting the optical signals into the RF domain, thus obtaining four outputs apt to feed four elements of a PAA.
The independent tunable steering of multiple signals has been evaluated by measuring the relative phase shift between the BV-WSS output ports. In the case ƒ RF = 12.5 GHz, a phase shift Δφ = ±90° corresponds to a delay Δτ = ±20 ps. In the case ƒ RF = 37.5 GHz, Δφ = ±90° corresponds to Δτ = ±6.67 ps. The time delays can be precisely measured by comparing the time markers on the sampling oscilloscope (an Agilent 86100C) [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6], or by comparing the oscilloscope traces reported in Fig. 3 The correct behavior of the proposed scheme is demonstrated by the delay imposed between the signals at 12.5 GHz (Fig. 3(a)), and 37.5 GHz (Fig. 3(b)). The measured time shifts exhibited by the signals (rightward for 12.5 GHz, leftward for 37.5 GHz) are ~20 ps for the 12.5 GHz SBs (with an error of ~4%) and ~6.7 ps for the 37.5 GHz SBs (with an error of ~0.45%).
Fig. 3. Oscilloscope traces related to the curves reported in Fig. 4. RF signals at 12.5 GHz (a) and 37.5 GHz (b) from Port1 (P1), Port2 (P2), Port3 (P3), and Port4 (P4), from black to light gray. The legend vertical order follows the order of the offset sinusoidal curves.
Fig. 4. Measurement of the actual front of the step-like imposed phase shift for three values of Δφ: π/2 (gray line), π (black line), 3π/2 (light-gray line).
IV. INTEGRATION OF THE PROPOSED ARCHITECTURE
The presented scheme has been realized resorting to bulk devices, typically employed for communication purposes, in order to demonstrate its working principle. The pixels response of the WSS employed in the reported experimental activity is 10 GHz-wide and it turns to be relatively large, actually representing a limitation for the performance of this LCoS technology-based architecture.
The phase transition fronts of the exploited BV-WSS are reported in Fig. 4. A 40 GHz signal has been employed to modulate an optical carrier. A step-like phase transition with variable Δφ has been imposed between the carrier and its 40 GHz USB. By tuning the carrier wavelength with 0.01 nm steps (1.25 GHz), the USB has been swept across the phase transition, measuring the actual phase shift of the photodetected signal on an oscilloscope, from 0° to the set value of Δφ. The imposed phase steps are actually implemented as ramps. In the case Δφ = 90° (gray curve in Fig. 4, the transition bandwidth is about 8.2 GHz. If Δφ = 180° (black curve), the phase transition is steeper, and it spans over about 3.8 GHz. Finally, in the case Δφ = 270° (light-gray curve), a bandwidth around 6.3 GHz has been measured. If the considered RF signal has a large bandwidth, it might suffer from a non-uniform phase shift over its spectral components, thus leading to the undesired effect of beam squint. This phenomenon is not an issue if considering applications involving signals with relatively narrow bandwidths, such as radar and remote sensing. In any case, the proposed architecture needs to be further optimized by adopting ad-hoc designed devices.
Recently, a small-size and high-resolution BV-WSS has been demonstrated, with 0.8 GHz-bandwidth pixels and 0.2 GHz pixel-to-pixel spacing [Reference Rudnick10]. Such a device can thus be considered for the improvement of the performance of the proposed architecture, though it represents a miniaturized but not integrated system. In order to obtain an extremely compact scheme, and endow it with complete mechanical robustness, the development of a cost-effective and small-sized BFN based on the proposed architecture inevitably implies optical integration.
The proposed BFN is divided in sub-blocks. In its bulk-components version [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6], the phase stability is an issue between the signals summing up from different sub-blocks, since within the same BV-WSS the optical waves propagate together, undergoing the same phase changes. Indeed, no unwanted phase mismatch has been detected and no stability issue arose in the experiment. As regards the amplitude stability issue, a proper equalization is possible thanks to the BV-WSS itself, allowing us to independently control the attenuation at each output port. On the other hand, the inter-sub-blocks phase stability problem is not completely solved by resorting to integrated photonic circuits, since it is not possible to control the circuits dimension with any desired accuracy; however, a feedback-based system can be implemented for an automatic control of the phase shift/time delay imposed by each micro-ring resonator (MRR) [Reference Burla11]. Similarly, the amplitude of the signals propagating through the integrated circuit can be suitably controlled by means of optical attenuator in order to accurately balance the different power losses within each arm.
A) Operation principle of the integrated scheme
The chosen approach for realizing an integrated version of the proposed architecture is based on MRRs in an all-pass filter (APF) configuration [Reference Bogaerts12]. A MRR-based APF consists of a resonant cavity implemented with a circular or race-track waveguide, which is connected to input/output ports through a straight bus waveguide. Such a MRR-based APF produces a frequency selective phase-response, which results in a bell-shaped group delay (GD) response (i.e., the derivative of the phase with respect to angular frequency) centered at a resonant wavelength of the cavity. MRR-APF can be conveniently fabricated for instance with low-loss, CMOS-compatible, silicon over insulator (SOI) or silicon nitrade (SiN x ) technologies. The optical properties of the MRR, i.e. its phase/GD response, can be tuned by varying the coupling strength between the ring and the bus straight waveguide, which can be achieved by means of variable couplers [Reference Chen, Sherwood-Droz and Lipson13]. Further, the resonant wavelength of the cavity can be tuned by introducing a differential optical path change, i.e. a phase shift for the field circulating in the cavity, by acting on the refractive index of the waveguide. A viable solution for the realization of a phase shift in SOI/SiN x -based MRR is to exploit the strong thermo-optic effect in silicon, through metallic heaters placed above the ring cavity.
B) The multicarrier integrated optical BFN
The structure of a multicarrier, multibeam, integrated BFN is sketched in Fig. 5, representing an integrated version of the one reported in Section II and depicted in Fig. 1. As shown in the figure, N different optical carriers, each with M SBs, generated through single-sideband (SSB) optical modulation, are considered. The signals are fed into a wavelength demultiplexer (DEMUX) stage that can be suitably integrated within the circuit resorting to the same technology as for MRRs, and which is responsible for separating the different carriers and respective SBs at its N output ports. After the DEMUX block, each optical carrier, together with its SBs, is routed to a BFN. There, the carrier and SBs comb is equally split over K-separated paths, and each SB propagates in one or more different MRRs, undergoing to a different phase shift, similarly to what described in the previous sections, where the BV-WSS-based BFN is illustrated. At the outputs of the BFNs, a PD drives the radiating elements of one out of K different PAAs, each composed by an array of N elements. The BFN blocks can be implemented employing MRRs, in two different ways: either resorting to PS or to TTD.
Fig. 5. Integrated BFN general scheme. SB, Side Band; DEMUX, Demultiplexer; BFN, Beam-forming Network. PD, Photodiode.
An integrated implementation of the proposed architecture, exploiting PS for beam steering, can be obtained referring to the scheme sketched in Fig. 6. One optical carrier λ j out of the N wavelengths considered in Fig. 5 enters the kth BFN. For the sake of simplicity, only three modulated sidebands (SB ji in the figure) are depicted, and SSB modulation is assumed. This BFN is based on MRRs to implement add/drop (A/D) blocks and the phase shift blocks. As previously mentioned, by slightly adjusting the resonant frequencies of the MRRs, by means for instance of thermal heaters, a different phase shift Δφ jik can be introduced over each SB (where 1 ≤ i ≤ M, 1 ≤ j ≤ N, 1 ≤ k ≤ 4). As depicted in the figure, at the first A/D stage, SB j3 is dropped and propagates toward the MRR imposing the desired phase shift Δφ j3k . Similarly, the remaining SBs are dropped one by one and each of them undergoes to a controlled phase shift Δφ jik thanks to the MRRs. All the optical signals are eventually recombined, making them beat in a PD, converting the SBs back to the RF domain to feed an element of the PAA. Thus, the different-phase radiated RF signals sum up, producing a beam with the desired orientation. The phase relations between the signals and the carrier can be accurately controlled thanks to a feedback network driven by the PD output, as shown in Fig. 6. RF signals phase can be measured much more easily, thus mitigating the problem of undesired phase mismatch.
Fig. 6. Integrated version of the proposed BFN architecture, based on MRRs. The phase response profile of the MRRs, ranging in the 0–2π interval, is sketched. Δφ jik is the phase shift on the ith SB of the jth optical carrier on the kth demux output. SB, Side Band; PD, Photodiode.
With respect to the bulk architecture presented in the previous sections, the MRR-based implementation offers the possibility of providing a TTD instead of PS, for the signals traversing the ring structure, thus making it suitable also for broad-band signals and/or large array size, should the beam squint [Reference Longbrake14] represent an issue. The operation principle for the TTD-based integrated version of the BFN is schematized in Fig. 7, which extends to a multi-channel scenario the binary-tree architecture that has been proposed in [Reference Meijerink15]. Here again, for the sake of clarity and without any loss of generality, we can still consider one optical carrier λ j , which is SSB modulated by two RF signals to generate two SBs (SB j1 and SB j2 in the figure). Similarly to the previous scheme, the optical signal is split over four different paths, in which MRRs are used to provide different values of group delay. The amount of GD can be increased without sacrificing the useful bandwidth by cascading two or more rings with slightly detuned resonant frequencies [Reference Rasras16]. A dedicated ring cascade for each SB is present on each path. As discussed, the couplers and the time delays in the MRRs can be properly tuned thermally, in order to provide the desired GD response at the wavelength corresponding to either SB j1 or SB j2, leaving unaffected the other signal. In particular, in the topmost path (from IN to OUT1), the first ring (deep gray) delays only SB j1 by a certain GD τ 1, whereas the second ring (light gray) delays only SB j2 by a different GD τ 2. Similarly, on the following lower path, a cascade of two couples of properly shifted MRRs is used to produce a GD of 2τ 1 and 2τ 2 for SB j1 and SB j2, respectively. Following this principle, for K-outputs BFN, the signal SB ji coming out from OUT-K, is delayed by Kτ i ; thus, an increasing GD with constant independently tunable steps going from OUT1 to OUT4 can be achieved for both SBs [Reference Meijerink15]. After being delayed in the optical domain, the signals eventually beat in the PDs, generating an RF copy of SB j1 and SB j2, by means of an optical down-conversion process, with a central frequency equal to the difference between the carrier and the SBs optical frequency. The output of each PD can thus be used to feed a PAA element and, by the superposition of the RF signals radiated from each element, the steering of every transmitted signal is effectively achieved. By changing the delay τ i of each path and by acting on the MRR phase and coupling strength, the RF signal can thus be steered at any direction, with a typical sub-ps resolution for the induced delays.
Fig. 7. Working principle of the integrated BFN based micro-ring resonators. The dashed and dotted curves represent the cascaded MRRs overall response, in terms of GD, for two different wavelengths. SB, Side Band; PD, Photodiode.
V. CONCLUSIONS
In this paper, we have proposed a novel scheme exploiting signal processing at a photonic level for realizing the independent and simultaneous beam steering of multiple RF signals. Starting from the scheme proposed in [Reference Serafino, Malacarne, Ghelfi, Porzi, Presi and Bogoni6], which is based on the use of BV-WSSs, here we propose integrated implementations of that architecture. BV-WSS allows controlling via software the spectral shaping of the optical amplitude and phase at each output port; the integrated architecture can perform either PS or TTD, depending on the bandwidth of the signals the BFN is designed for.
The capability of the proposed system for simultaneously obtaining an independent, tunable steering of different signals has been exhaustively demonstrated, also assessing the degree of precision of the phase control. However, the exploited BV-WSS is a device purposely designed for optical wavelength division multiplexing communication systems, and we believe that a specific design for beam-forming applications could achieve better performance. The scheme therefore appears as a promising solution for feeding the numerous TRMs in a phased-array antenna, digitally controlling the direction of several RF signals simultaneously and independently.
One of the integrated proposed versions of the multi-beam BFN exploits MRRs to implement A/D blocks as well as tunable PS elements, orienting the beam-forming strategy on PS. Alternatively another integrated design has been considered, exploiting TTD and integrated optical splitters, ideal for dealing with broadband signals.
Giovanni Serafino received the M.S degree in Telecommunication Engineering in 2009 from Università di Pisa, Italy, and the Ph.D. degree in Emerging Digital Technologies from the Scuola Superiore Sant'Anna of Pisa, Italy, in 2013. From 2009 he's with the National Laboratory of Photonic Networks (LNRF) of Scuola Sant'Anna in Pisa, Italy. He's currently post-doc researcher in LNRF for Scuola Sant'Anna. His early research interests have been in the areas of all-optical signal processing, microwave photonics, fiber-optic transmission systems, and reconfigurable optical networks. From 2009 he has been studying applications of microwave photonics techniques to radar systems and wireless communications.
Antonio Malacarne was born in Livorno, Italy, in 1978. He received the Ph.D. degree in ICT (Telecommunication Area) at Scuola Superiore Sant'Anna of Pisa (SSSA), Italy, in 2009. He currently is a research fellow at SSSA in both the areas “High-capacity optical communications” and “Digital and microwave technologies”. His career includes a two-year long post-doctoral position at the National Institute of Scientific Research (INRS) in Montreal, Canada, with Prof. J. Azaña. He's currently involved in numerous European/International/National projects. He is the author or coauthor of 30 manuscripts published in international journals, more than 60 contributions for international conferences and 5 international patents. His current research interests include integrated optical transceivers, optical interconnects, photonic processing and microwave photonics.
Dr. Claudio Porzi received MS degree in Electronics Engineering from “La Sapienza” University of Rome, Italy, on May 2000, and Ph.D. in Telecommunications from “Scuola Superiore Sant'Anna” University, Pisa, Italy, on January 2006. Since February 2014 he is with CNIT (Interuniversity National Consortium of Telecommunications). Between January 2008 and February 2014 he has been Assistant Professor at Scuola Superiore Sant'Anna University, Pisa, Italy. His main research interests include nonlinear optical signal processing, photonics integrated circuits and semiconductor optoelectronic devices, microwave photonics, optical packet-switched networks, and access networks. He co-authored more than 80 papers published on international journals or presented at major international conferences, and 5 international patents. In 2010 he was granted of the “Best Paper Award” at the international communications Conference ICC 2010- Optical Networks and Systems Symposium.
Paolo Ghelfi received the M.S. degree in electronics engineering from the University of Parma, Italy, in 2000. From 2001 he is with the National Laboratory of Photonic Networks of CNIT in Pisa, Italy, where he is now Head of Research. In 2008 he has been visiting scientist at KIST in Seoul, South Korea.
His early research interests have been in the areas of fiber-optic transmission systems, all-optical signal processing, and reconfigurable optical networks. From 2009 he has been studying applications of microwave photonics techniques to radar systems and wireless communications.
He has authored or co-authored more than 30 papers on International Journals, more than 90 papers on International Conferences, and 20 patents.
Marco Presi received the Laurea degree in physics from the University of Rome “La Sapienza”, Rome, Italy, in 2001, and the Ph.D. degree in applied physics from the University of Pisa, Pisa, Italy, in 2006. Since 2007, he has been a Research Associate with the Scuola Superiore Sant'Anna, Pisa, Italy, where he is currently engaged in research on wavelength-division-multiplexing optical systems. He coauthored more than 50 papers published in international peer-reviewed journals and in international conferences.
Antonio D'Errico, born in San Severo, Italy, in 1974, received the Ph.D. degree cum laude in telecommunication engineering from the Scuola Superiore Sant'Anna, Pisa, Italy, in 2008. He is currently with Ericsson Research as Experienced Researcher. His research interests include advanced technological solutions for optical networks. He is author of more than 75 papers published in international journals, conference digests, and patents. He is on the board of reviewers for a number of international journals fields of optics and photonics.
Marzio Puleri (M′76-SM′81-F′87) received the M.S. degree in electronic engineering from the University of Rome “La Sapienza” in 1992. Since 1993 he has been working at Ericsson Telecomunicazioni S.p.A. From 1993 to 1998 he worked on integrated circuit design and related methodology. From 1999 to 2007 he worked as System Manager on microwave point-multipoint radio transport networks. In 2008 he joined the Ericsson research group in Pisa, working on optical packet networks and 4G/5G mobile networks. His research interests covers several fields: microelectronics, automated SW and microelectronics design methodologies, packet switching, IP-based networks, radio networks, optical packet networks, Artificial Intelligence, network diagnostics, robotics.
Antonella Bogoni, head of research area of CNIT (National Inter-University Consortium for Telecommunications), dedicated her research activity to photonics technologies for ultra-fast optical communication systems and microwave photonics for communication and surveillance systems. In 2009 she obtained an ERC starting grant for developing a photonic-based fully digital radar systems, and in 2012 and 2015 she got two additional ERC grants within the “proof of concept” program in order to convert her research results into a pre-industrial product for airport security and environment monitoring. In2008 and from 2009 to 2010 she collaborated with the University of Southern California (USC) in Los Angeles as winner of a “Fulbright” scholarship. She is co/author of 52 patents, 8 books and chapters and more than 130 papers on the main scientific international journals. She is chair of international conferences and workshops, including general chair of Photonics in Switching 2014 and Sub-committee Chair of ECOC 2015.
