This paper describes a 2-20 GHz 6-bit True-Time Delay. A total equivalent electrical length in air of 43.5 mm (145 ps) is achieved over a 2-20 GHz bandwidth. Digital drivers and a serial-to-parallel converter are integrated on the same MMIC. The ED02AH 0.2 µm PHEMT process from OMMIC is used. The time delay elements are real-ised using constant-R networks. The three smallest bits make use of a self-switched version of the constant -R networks while the 3 largest bits use a topol-ogy with single-pole double-throw (SPDT) switches and constant-R networks in the delay path. Measurement results for a typical chip are presented.
Wireless sensor nodes present a difficult compromise at the system level between low power consumption, interference resilience and ease/size of the application. Recent low-power designs employ receiver architectures in which the LO runs at or below the received frequency (in TX at the required RF frequency). These architectures facilitate LO generation and distribution at a favorably low frequency, but cause a systematic, hard-to-circumvent susceptibility to out-of-band blockers in RX [1] and pulling in TX [1,2]. The presented Bluetooth Low-Energy (BLE) transceiver, as shown in Fig. 13.3.1, was purposely designed with a classic double-frequency VCO architecture to avoid these inherent weaknesses, but still significantly reduces the power consumption of designs of the same architecture [3,4]. In order to ease application, an on-chip DC-DC converter supports battery operation between 0.9V and 3.3V. An on-chip T/R switch, RX balun and impedance matching allow 50Ω chip antennae to be used directly on the single-ended RFIO pin, without the need for external RF filtering. This enables minimum-sized modules and a fast time-to-market.The VCO tank impedance is maximized by minimizing the number of cells in the capacitor array, and thus its parasitic capacitance. Eight identical unit cells are biased in three possible conditions, comprising 36 possible combinations. The sixteen settings that give the best linearity are selected through an LUT. The unit cells are also used in the drain tank circuit of the LNA, as shown in Fig. 13.3.3. The LO buffers are based on a stacked architecture, as shown in Fig.13.3.2. The reduced output swing minimizes the dynamic power consumption and the storage element (C S >> C L ) allows charge reuse without requiring a specific phase relation between the signals. The stacked architecture can be extended to more than two elements at the expense of symmetry. The four stacked buffers drive the LO lines towards the mixer. When the losses incurred from the AC coupling capacitors are ignored, the power is potentially reduced by 75% when compared to the buffering of all signals with rail-to-rail inverters. The quadrature mixer is a switched transconductor design [5], loaded with a transimpedance amplifier with a bandwidth of 2MHz.A transformer-coupled differential-mode LNA is applied for optimum power-supply noise rejection. The LNA is co-designed with the single-pin RFIO interface for simultaneous noise and impedance matching, with their cascade providing 5.5dB noise figure and 27dB voltage gain from the antenna input to the LNA output, consuming 1mA of current. This gain is made possible by biasing the LNA core transistors in weak inversion (g m /I d =20V -1 ), with cascode devices, employing an output tank load with a loaded Q=10 and designing the input transformer to provide g m -boosting. The cascode stage optionally provides 6dB gain reduction. The output LC tank employs a calibration bus that is multiplexed with two calibration registers: one for tuning the tank at the lower half of the BLE b...
This paper discussed the impact of frequency compensation on the linearity of 2 nd and 3 rd order negative feedback amplifiers from a synthesis point of view. For each compensation method, simple expressions are derived that relate the amount of bandwidth required to get a desired linearity performance up to the end of the information band. The phantom zero technique is concluded to be by far the best compensation method what linearity is concerned.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.