Deep Q-network-based auto scaling for service in a multi-access edge computing environment

In 5G networks, it is necessary to provide services while meeting various service requirements, such as high data rates and low latency, in response to dynamic network conditions. Multi-access edge computing (MEC) is a promising concept to meet these requirements. The MEC environment enables service providers to deploy their low latency services that are composed of multiple components. However, operating a service manually and attempting to satisfy the quality of service (QoS) requirements is difficult because many factors need to be considered in an MEC scenario. In this paper, we propose an auto-scaling method using deep Q-networks (DQN), which is a reinforcement learning algorithm, to resize the number of instances assigned to service. In our evaluation, compared to other baseline methods, the proposed approach maintains the appropriate number of instances effectively in response to dynamic traffic change while satisfying QoS and minimizing the cost of operating the service in the MEC environment. The proposed method was implemented as a module running in OpenStack and published as open-source software.     

Empirical millimeter-wave wideband propagation characteristics of high-speed train environments

Owing to the difficulties associated with conducting millimeter-wave (mmWave) field measurements, especially in high-speed train (HST) environments, most propagation channels for mmWave HST have been studied using methods based on simulation rather than measurement. In this study, considering a linear cell layout in which base stations are installed along a railway, measurements were performed at 28 GHz with a speed up to 170 km/h in two prevalent HST scenarios: viaduct and tunnel scenarios. By observing the channel impulse responses, we could identify single- and double-bounced multipath components (MPCs) caused by railway static structures such as overhead line equipment. These MPCs affect the delay spread and Doppler characteristics significantly. Moreover, we observed distinct path loss behaviors for the two scenarios, although both are considered line-of-sight (LoS) scenarios. In the tunnel scenario, the path loss exponent (PLE) is 1.3 owing to the waveguide effect, which indicates that the path loss is almost constant with respect to distance. However, the LoS PLE in the viaduct scenario is 2.46, which is slightly higher than the free-space loss.     

Passivation of Surface and Interface States in AlGaN/GaN HEMT Structures by Annealing

The Ni/AlGaN interfaces in AlGaN/GaN Schottky diodes were investigated to explore the physical origin of post-annealing effects using electron beam induced current (EBIC), current–voltage (I–V) characteristics, and X-ray photoelectron spectroscopy (XPS). The EBIC images of the annealed diodes showed that the post-annealing process reduces electrically active states at the Schottky metal/AlGaN interfaces, leading to improvement of diode performance, for example a decrease in reverse leakage current and an increase in Schottky barrier heights. Pulsed I–V characteristics indicate the Fermi level is up-shifted after annealing, resulting in a larger sheet carrier density at the AlGaN/GaN interface. Unintentional oxidation of the free AlGaN surface during the post-annealing process, revealed by XPS analysis, may prevent electron trapping near the drain-side of the gate edges. We suggest that the post-annealing process under an optimized conditions can be an effective way of passivating AlGaN/GaN heterojunction field-effect transistors.     

Performance evaluation of a new BPSK-OFDM timing estimation algorithm

Previously, the timing estimation in OFDM system employs cyclic prefix has been presented with the assumption of independent identified distributed OFDM symbol. The information used to find the symbol synchronization depends on the length of cyclic prefix. Actually, the data in the BPSK-OFDM modulated symbol, transferred by inverse fast Fourier transform, is with a complex symmetry character. With this character, more information from the whole OFDM symbol could be provided for the symbol timing. In this paper, the proposed algorithm uses the autocorrelation of the received signal to obtain the symbol timing. The simulation of Beek’s algorithm and the proposed scheme are presented. The performances of these algorithms are evaluated based on the missing symbol probability and the estimator mean square error. The simulation results show that the proposed algorithm is suitable to achieve a better symbol synchronization.     

Fabrication of Dicing-Free Vertical-Structured High-Power GaN-Based Light-Emitting Diodes With Selective Nickel Electroplating and Patterned Laser Liftoff Techniques

Through the use of selective nickel (Ni) electroplating, patterned laser liftoff technique, and surface roughing of the top n-GaN epilayer, a novel process for the fabrication of vertical-structured metal-substrate GaN-based light-emitting diodes (VM-LEDs) to avoid difficulties in Ni substrate dicing and improve device yield was proposed and demonstrated. In conjunction with a sidewall passivation with SiO₂ and keeping the size of epilayer smaller than that of Ni island, a considerable improvement in yield and device performance were shown. As compared to conventional lateral-structured GaN-based LEDs, VM-LEDs show an increase in light output power about 174% at 350 mA with a significant decrease in forward voltage from 3.5 to 3.17 V     

A 2.4-GHz Low-Power Low-IF Receiver and Direct-Conversion Transmitter in 0.18-μm CMOS for IEEE 802.15.4 WPAN Applications

In this paper, a low-power low-IF receiver and a direct-conversion transmitter (DCT) suitable for the IEEE standard 802.15.4 radio system at the 2.4-GHz band are presented in 0.18-mum deep n-well CMOS technology. By using vertical NPN (V-NPN) bipolar junction transistors in the baseband analog circuits of the low-IF receiver, the image rejection performance is improved and the power consumption is reduced. In addition, by applying the V-NPN current mirrored technique in a DCT, the carrier leakage is reduced and the linearity performance is improved. The receiver has 10 dB of noise figure, -15 dBm of third-order input intercept point, and 35 dBc of image rejection. The transmitter has more than -2 dBm of transmit output power, -35 dBc of local oscillator leakage, and -46 dBc of the transmit third harmonic component. The receiver and transmitter dissipate 6 and 9 mA from a 1.8-V supply, respectively     

35-nm Zigzag T-Gate $hbox{In}_{0.52}hbox{Al}_{0.48} hbox{As/In}_{0.53}hbox{Ga}_{0.47}hbox{As}$ Metamorphic GaAs HEMTs With an Ultrahigh $f_{max}$ of 520 GHz

Metamorphic GaAs high electron mobility transistors (mHEMTs) with the highest-f _max reported to date are presented here. The 35-nm zigzag T-gate In_0.52Al_0.48As/In_0.53Ga_0.47As metamorphic GaAs HEMTs show f _maxof 520 GHz, f _T of 440 GHz, and maximum transconductance (g _m) of 1100 mS/mm at a drain current of 333 mA/mm. The combinations of f _max and f _T are the highest data yet reported for mHEMTs. These devices are promising candidates for aggressively scaled sub-35-nm T-gate mHEMTs.     

Acoustooptic Tunable Gap-Type Bandpass Filter With a Broad Stopband

We demonstrate a novel all-fiber gap-type tunable bandpass filter configuration that consists of a broadband hollow optical fiber (HOF) acoustooptic tunable filter (AOTF) concatenated with a narrowband single-mode fiber (SMF)-AOTF. Owing to the unique mode coupling properties of HOF-AOTF and SMF-AOTF, a narrow passband channel of a full-width at half-maximum (FWHM) of ~4 nm could be formed in a broad stopband platform of a 90-nm FWHM successfully. In this scheme, the center wavelength and rejection efficiency of both passband and stopband were found to be flexibly tunable by adjusting the frequency and the voltage of radio-frequency signals applied to individual acoustic transducers.     

Low-Loss Polymer-Based Long-Range Surface Plasmon-Polariton Waveguide

To improve the propagation loss of polymer-based long-range surface-plasmon-polariton (LR-SPP) waveguide devices at the telecom wavelength range, low-loss LR-SPP waveguides were fabricated in an ultraviolet-curable acrylate polymer with a low refractive index and absorption loss. A propagation loss of 1.72 dB/cm at a wavelength of 1.55 mum was achieved with a 14-nm-thick and 3-mum-wide metal stripe.     

A WDM-PON Employing Fabry–PÉrot Laser Diodes Without a Seed Light Injection

We propose and demonstrate a wavelength-division-multiplexed passive optical network by employing double-contact Fabry-Perot laser diodes (F-P LDs) without a seed light injection. To avoid the high mode partition noise at low frequency, we use a binary phase-shift keying as a modulation format at a low relative intensity noise window. An error-free transmission is achieved by compensating a lasing envelope shift due to temperature variation with the double-contact F-P LD.