Corning (Israel)

Next generation data centers are challenged with the task of delivering increased bandwidth and data rates at manageable power consumption. Co-packaged optics offer a path forward by enabling significant power reduction when electronic and photonic chiplets are locally mounted on a packaging substrate. In this paper, we discuss a glass substrate solution where both optical and electrical connectivity can be fabricated using ion-exchange (IOX) waveguides embedded into the glass and low-loss electrical routing. Design aspects are discussed to support 102.4 Tb/s data center switch applications. Signal transmission between the ASIC and PIC is completed with a total of 5 metal layers utilizing fine-line signal routing separated by ground planes. Through glass vias (TGVs) provide both power delivery for the mounted components and data management connectivity. Low-loss evanescent coupling occurs between the waveguides embedded in the glass substrate and photonic chips with high-density silicon nitride optical IOs. This design enables low-cost assembly methods involving the pick and place alignment of optical components and offers connectivity with low-profile mechanical transfer (MT) ferrule based fiber connectors.

We review the role of optical fiber as a transport medium in the three generations of in-building wireless networks. We show that fiber becomes a necessity for 5G radio access networks (RAN) operating in high-bandwidth regions of the spectrum at millimeter wave (mmWave) frequencies.

This article introduces an Initial Access Fairness Beamforming (IAFB) approach for wireless networks using millimeter wave and terahertz frequencies and proposes novel synthesis method. The fairness is establishing independence between user location and the initial access signals received power level. We propose codeword synthesis with phase-only control which is beneficial for passive implementation of the receiver (Rx) and improves power utilization at the transmitter (Tx), which reduces mutual coupling impact for both Rx and Tx directions. In contrast to classical shaped beam synthesis methods where desired radiation patterns are targeted, we target the fairness of the received power distribution at the desired coverage area by solving a multi-objective nonconvex optimization problem. We address the known nonconvex optimization issues of initial state sensitivity and convergence rate by splitting the problem into a sequence of concatenated sub-problems, where the solution of one sub-problem defines the initial state for the next sub-problem.

In this paper we propose a new method of occupied spectrum analysis for channel detection in a shared spectrum environment. Our approach is based on iterative multi-stage multi-resolution scanning using configurable Sliding Discrete Fourier Transform (SDFT) aided by an Unsupervised Machine Learning (UML) clustering method. The proposed low-complexity high-accuracy real-time spectrum scanning and channel detection is simulated for multiple Radio Access Networks (RAN) of Long-Term Evolution (LTE) & Fifth Generation (5G) channels in a shared frequency band. The results of the simulation show possible successful utilization of the proposed method as a sensing tool for spectrum sharing management and other applications where accurate channel detection occupancy is required.