Fig 2 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Principle and operation of the proposed dual-path noise-canceling LNA. (a) M 1 noise-canceling scheme. (b) M 2 noise-canceling scheme.
Source publication
In this article, a 22.9-38.2-GHz dual-path noise-canceling low noise amplifier (LNA) is proposed, which can achieve a low noise figure (NF) by reducing the noise of both paths. Such LNA consists of one common gate (CG) amplifier with one three-stage transformer, one resistive feedback common-source (CS) amplifier, and two amplitude-adjusting amplif...
Contexts in source publication
Context 1
... This article is organized as follows. The prototype with principle and theoretical analysis of the dual-path noise-canceling LNA is provided in Section II, while Section III presents the LNA circuit implementation. In Section IV, an LNA is fabricated, measured, and compared with the state of the arts. The conclusion is summarized in Section V. Fig. 2 shows the principle and operation of the proposed dual-path noise-canceling LNA. The input signal is divided into two paths (i.e., paths I and II) at the input node of the proposed LNA. Path I consists of a CG stage, a feedback transformer (i.e., L 1 and L 2 ), and an amplifier A 1 . Path II is formed by a resistive feedback CS stage ...
Context 2
... a feedback transformer (i.e., L 1 and L 2 ), and an amplifier A 1 . Path II is formed by a resistive feedback CS stage and an amplifier A 2 . After amplified by paths I and II, the input signal is recombined at the differential output. In the proposed noise-canceling LNA, the noise of both M 1 and M 2 can be reduced simultaneously. As shown in Fig. 2(a), the noise introduced by M 1 of CG stage (i.e., modeled as noise current I n1 ) generates two out-of-phase noise voltages at its drain and source. The noise at the drain of M 1 is amplified by A 1 and then canceled at the differential output ports by its in-phase replica in path II. Meanwhile, Fig. 2(b) shows that in the CS stage, the ...
Context 3
... can be reduced simultaneously. As shown in Fig. 2(a), the noise introduced by M 1 of CG stage (i.e., modeled as noise current I n1 ) generates two out-of-phase noise voltages at its drain and source. The noise at the drain of M 1 is amplified by A 1 and then canceled at the differential output ports by its in-phase replica in path II. Meanwhile, Fig. 2(b) shows that in the CS stage, the noise introduced by M 2 (i.e., modeled as noise current I n2 ) generates two in-phase noise voltages at its drain and gate through the feedback resistor. Those in-phase noise voltages are amplified by A 2 and path I and then canceled at the differential output as commonmode voltages. Good NF can be ...
Context 4
... stable in any source and load VSWR [36]. As shown in Fig. 19, the measured input 1-dB compression point (IP 1 dB ) and input third-order intercept point (IIP3) are −13.2 to −6.6 and −3.6 to 3.2 dBm, respectively. Similar to other multi-path-combined type amplifier, good linearity is achieved in the proposed LNA due to the dual-path topology. Fig. 20 shows that the measured NF is 2.65-4.62 dB within the frequency range of 22.9-38.2 GHz. Compared to simulation, the measured NF response exhibits a slight frequency shifting to higher frequency. Such variation can be calibrated using the phase-tuning lines by the following procedures: 1) figure out the performance variation trends in ...
Context 5
... Such variation can be calibrated using the phase-tuning lines by the following procedures: 1) figure out the performance variation trends in different simulation settings and 2) compared the measured and simulated results and then change the phase setting of phase-tuning line following the trends obtained in step 1. For example, as shown in Fig. 20, similar frequency shifting can be obtained in simulation once the input capacitance of M 3 is reduced. Therefore, by tuning the phase setting to case 4 (increases about 7-fF input capacitance of M 3 ), the measured NF can be matched to the simulation. are introduced for the fair comparison between the proposed LNA and previous works, ...
Context 6
... Fig. 21 shows that good in-band phase balance is achieved in the proposed dual-path noise-canceling LNA. However, the PVT variation and simulation inaccuracies will change the phase of both paths and hence result in performance degeneration. Such phase variation can be calibrated by introducing phase-tuning lines to both paths. As shown in Fig. 22(a), the proposed phase-tuning line consists of a coplanar waveguide (CPW) with grounded shield and four switchable capacitors. The switchable capacitors are controlled by the voltage V c1,2,3,4 , while the grounded shield [37] is utilized to reduce the substrate loss of the phase-tuning lines. The capacitance tuning range of each ...
Context 7
... The switchable capacitors are controlled by the voltage V c1,2,3,4 , while the grounded shield [37] is utilized to reduce the substrate loss of the phase-tuning lines. The capacitance tuning range of each phase-tuning line is about 7 fF, which results in about 4 • -6 • maximum available phase variation within the operation band. As shown in Fig. 22(b) and (c), the NF and gain of the proposed LNA vary with the phase settings. Note that, the proposed phasetuning line is located in the inter-stage matching network of both paths, where the capacitance variation can not only change the phase shifting but also influence the gain and BW of the proposed LNA. Therefore, the further increase ...
Similar publications
Wireless communication continues to lead the world. The term “wireless” significantly influences everything from telephone to satellite communication. Recent years have seen the need for appropriately built Low Noise Amplifiers (LNAs) in receiver front end because of the modem wireless communication in broadcast and microwave radio. In this paper a...
Citations
... A S THE first-stage of the wireless receiver, low-noise amplifier (LNA) plays a crucial role in determining the signal-noise-ratio of a communication link [1]. In the past few years, various techniques have been proposed to improve the LNA's performance [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], e.g., noise reduction [4], [5], [6], [7], G m -boost [8], [9] and bandwidth extension [10], [11], [12], [13], [14]. For high data-rate wireless applications such as satellite communication, gain flatness is also a critical parameter. ...
... A S THE first-stage of the wireless receiver, low-noise amplifier (LNA) plays a crucial role in determining the signal-noise-ratio of a communication link [1]. In the past few years, various techniques have been proposed to improve the LNA's performance [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], e.g., noise reduction [4], [5], [6], [7], G m -boost [8], [9] and bandwidth extension [10], [11], [12], [13], [14]. For high data-rate wireless applications such as satellite communication, gain flatness is also a critical parameter. ...
This letter presents a broadband current-reused low-noise amplifier (LNA) monolithic microwave integrated circuit (MMIC) for wireless communication. Feedback and gain-compensation techniques are applied in the design of the proposed LNA MMIC to obtain the performance of high gain and low in-band gain ripple with ultralow noise figure (NF) in wide operational bandwidth. For verification, the proposed LNA is fabricated in 0.15-
$\mu \text{m}$
gallium arsenide (GaAs) technology with a size of 1.21 mm2, realizing the measured gain of over 27 dB, NF of less than 1.1 dB, output 1-dB compression point (OP
$_{1\,\mathrm {dB}}$
) of greater than 9.2 dBm at 18 GHz and in-band gain ripple of about ±0.17 dB from 10 to 18 GHz. The total power dissipation of the proposed LNA MMIC is less than 125 mW.
... To achieve such LNA designs, CMOS technology is widely adopted to enjoy the flexibility of circuit topology and high integration level [7]- [11]. However, the CMOS process has its natural drawbacks such as low drain voltage, limited transconductance, channel length modulations, etc [12], [13]. ...
... However, its main limits on transconductance, f t /f max , NF, and linearity are also obvious even with the CMOS scaling. For higher working frequencies, wider operating bandwidth, and much better performance, though the system-on-chip (SoC) can be adopted to achieve them and promote integration, designers have to resort to complicated structures on advanced nodes [7], [10], [47], [48]. ...
With the fast development of advanced communication systems, wideband low-noise amplifiers (LNAs) have become more and more critical as they determine the overall performance of the communication systems. A well-designed LNA can keep enough margin for the successive sub-circuits in the receiver chain. At the same time, with the maturity of III-V semiconductor technologies, the performance of LNA, especially the bandwidth has been greatly enhanced. This brief discusses the design process considerations and the limiting factors of LNA bandwidth. Subsequently, a comprehensive review of the recent advances in III-V-based LNA circuits and the techniques for operating bandwidth extension and performance improvement is presented. Finally, we compare the III-V-based LNA architectures and point out a few possible improvements and trends for III-V-based LNAs. Index Terms-Receiver, low-noise amplifiers (LNAs), III-V semiconductor, bandwidth limit, operating bandwidth extension.
... The concept of mutual NC [19][20][21] involves the auxiliary stage canceling the noise of the main stage while utilizing the main stage to eliminate the noise of the auxiliary stage. The key to realizing this idea is to introduce feedback in the auxiliary stage. ...
... However, with the introduction of Rf, M2 also affects the input impedance, which increases the design complexity. Moreover, the NC conditions of M1 and M2 cannot be satisfied simultaneously [19,20] . It can only be that the noise of one is wholly canceled and the other is partially canceled. ...
This paper presents a comprehensive review of noise canceling (NC) techniques in complementary metal-oxide-semiconductor (CMOS) low-noise amplifiers (LNAs) and receivers (RXs). Initially applied to wideband LNAs, the NC technique addresses the trade-off between wideband input impedance matching (IIM) and noise figure (NF). Subsequently, it found application in transimpedance amplifiers (TIAs) and RXs. Two basic NC structures are the common-gate common-source (CG-CS) and resistive shunt feedback common-source (RSF-CS). As research progresses, several advanced structures have emerged, including mutual noise canceling, nested noise canceling, and frequency translational noise canceling (FTNC). This paper elucidates the fundamental principle and definition of the NC technique, and systematically summarizes and classifies the most representative structures over the past two decades. Additionally, each structure's NC principles are analyzed, revealing their intrinsic connections to the two basic structures.
... The complexity of the design further increases if one considers the problems of impedance matching, low power consumption, linearity, and stability. Figure 3 presents the NF versus the gain of current state-of-the-art GaAs [111], [112], GaN [111], [112], SiGe [131], [132], [133], [134], [135], [136], [137], [138], and Si [130], [139], [140], [141], [142], [143], [144], [145] LNAs Figure 2. The PAE versus the P sat of current state-of-the-art GaN [106], [107], [108], [109], [110], [111], [112], GaN [111], [116], Si [117], [118], [119], [120], [121], [122], [123], [124], and SiGe PAs [125], [126], [127], [128], [129] in the Ku/Ka frequency bands. in the Ka/Ku frequency bands. LNAs based on GaAs dominate the low-NF part of the plot, followed by GaN devices. ...
... SiGe LNAs present an NF in the 2-3-dB range in the Ka-/Ku-bands, with gains that span the 15-30-dB range [131], [132], [133], [134], [135], [136], [137], [138]. Sibased LNAs present more modest gain levels in the 12-20-dB range but show NFs close to those of GaAs (i.e., nearly 1 dB) up to 3.5 dB [130], [139], [140], [141], [142], [143], [144], [145]. ...
... Figure 3. The NF versus the gain of current state-of-the-art GaAs [111], [112], GaN [111], [112], SiGe [131], [132], [133], [134], [135], [136], [137], [138], and Si [130], [139], [140], [141], [142], [143], [144], [145] LNAs in the Ku/Ka frequency bands. ...
Low-Earth orbit (LEO) constellations are revolutionizing the world of satellite communication (Satcom), providing new opportunities to manufacturers and operators and enabling innovative and attractive services to users.
... The input impedance Z in and NF of CG are inversely related to the g m of transistors, which means a higher g m improves NF but decreases the Z in , leading to possible impedance mismatch [26]. To alleviate this constraint, g m -boost with noise cancellation (NC) technique is often employed to enhance gain and NF [24], [25]. Nevertheless, the issues of linearity and complicated circuit topology remain. ...
... The dominant noise source and its effects on the overall NF of the LNA should be investigated first. Fig. 8 shows the simulated noise contributions of the proposed LNA at 6.5 GHz without the intrinsic noise from the signal source at the schematic level, which does not affect the calculations for NF [24], [30]. From Fig. 8, the main noise sources are R F 1 , M 1 , R F 2 , and R L1 from the input inductor L 1 . ...
... where ϕ is a fitting parameter [46] for GaAs pHEMT device. From (24), OIP3 can be maintained and even improved through selecting smaller g m , g dn3 or g mn3 since g d is usually not dominant if the channel-length modulation effects are not strong. However, as A v ∝ g m , improving OIP3 via smaller g m is not possible without deteriorating the gain. ...
In this paper, a 2-stage compact wideband low-noise amplifier (LNA) monolithic microwave integrated circuit (MMIC) with multiple resistive feedback (MRFB) is presented. From the DC point of view, the proposed MRFB functions as a self-biasing structure to bias transistors in the optimal condition, improving the noise figure (NF) and linearity. Meanwhile, by employing MRFB with source degeneration and input inductor, the proposed LNA achieves wideband flat gain at the AC side. In comparison with traditional topologies, a wide bandwidth of more than 11.5 GHz with low noise figure of less than 2.5 dB can be achieved. To verify the proposed LNA structure, a chip prototype is fabricated in a 0.15-
$\mu$
m GaAs E-mode pHEMT process with a compact die size of only 0.75 mm
$^2$
including all the testing pads. From the measurement results, the proposed LNA circuit features a 1.0 to 12.5 GHz 3-dB working bandwidth (172% fractional bandwidth), 23.6 peak gain, 1.51 dB minimum NF, 66.7
$\pm$
15 ps group delay, and 24.3/12.6 dBm best OIP3/OP1dB, respectively. The total DC power is around 87.5 mW from a single 2.5-V power supply.
... To overcome the above-mentioned limitations, several wideband LNA architectures have been developed. In the CMOSbased mm-Wave LNA circuits, popular structures include the resistive feedback [10], the coupled inductive peaking [11], [12], and the g m -boost with noise cancellation [13]- [15]. They feature good impedance matching, flat gain response, small die size, and low DC power consumption. ...
... It leads to the fact that the energy coupling from the drain to the gate (feedback) is dominant. Hence, the unidirectional coupling assumption here is valid [13], [30]. ...
This article presents a 3-stage millimeter-wave low-noise amplifier (LNA) monolithic microwave integrated circuit (MMIC) design for broadband applications. The proposed structure consists of cascading common-source (CS) LNA with coupled interstage feedback (CIF) paths at the second and third stages. By tuning the coupling factor and the inductance of the CIF, enhanced overall gain and extended working bandwidth can be achieved simultaneously. The proposed CIF formed by a simple on-chip edge-coupled parallel metal-lines features small size and high flexibility. To facilitate circuit design and optimization, an equivalent circuit for the proposed CIF structure is developed. Besides, to address the trade-off between impedance matching and noise performance, a noise matching strategy focusing more on higher frequencies is presented to achieve good noise performance. With the proposed techniques, an LNA prototype is demonstrated in a commercial 0.15-
$\mu$
m GaAs E-mode pHEMT process. The fabricated LNA has a die size of 1.15 mm
$^2$
and DC power consumption of 109 mW. The measurement results show a peak gain of 25.6 dB with a 3-dB bandwidth of 9
$\sim$
42 GHz, the minimal noise figure (NF) of 1.91 dB, a group delay of 68.7
$\pm$
20.5 ps, and 20.8 dBm best OIP3 under 2.0-V VDD.
... As a result, the low noise amplifier with a low noise figure, high gain, and large bandwidth have become one of the research hotspots. Benefit from the advantages of high integration and relatively low cost of CMOS technology, several mm-wave LNAs in the CMOS process have been reported [1,2,3,4,5,6,7]. However, due to the lossy silicon sub-strate and low carrier mobility, these LNAs have relatively poor noise figure, gain, and linearity performance. ...
This letter presents a 28-38 GHz low-noise-amplifier (LNA) monolithic microwave integrated circuit (MMIC) using 0.15µm GaAs pseudomorphic high electron mobility transistor (pHEMT) process. Inductive degeneration and shunt-series peaking techniques are employed to obtain simultaneous input and noise matching (SINM) and wideband flat gain. A novel wideband multistage noise matching technique based on a high-pass matching network is proposed to achieve low NF and further improve gain flatness over a wide frequency range. A four-stage LNA prototype is designed and fabricated based on the proposed technique. Measurement results show that the proposed LNA has a 1.6-dB minimum NF and an average gain of 23.7dB in the entire operating band. S11 and S22 are better than -10dB from 30 to 37 GHz. The measured output 1-dB compression point (OP1dB) is higher than 14.8 dBm from 28 to 38 GHz. The chip is realized within 2.1*1.5 mm² and consumes 56 mA current from a 5 V supply.
... Meanwhile, Fig. 7 reveals that the linearity of the proposed absorptive IF amplifier shows little degeneration compared to the conventional IF amplifier. Fig. 8 shows the schematic of the proposed dual-path noisecanceling LNA [27]. The wideband input matching is mainly achieved by the common-gate (CG) stage with inductor L gs , while a dc-block capacitor is used in the input of CS stage. ...
... Note that the fully noise-canceling condition could not be satisfied within a wide frequency range in the proposed dual-path noise-canceling LNA since the noise-canceling ratio is varied with the frequency-dependent inductance and capacitance. Therefore, the component values in the proposed LNA are optimized to achieve improved NF, which results in a partial noise-canceling operation in both CG and CS paths [27]. The noise contribution in Fig. 9(b) reveals that the noise introduced by the current of M a1 and M a2 is only about 8.4% and 9.9%, respectively, while the noise is mostly contributed by the passive loss and parasitic resistor of M a1 and M a2 (i.e., gate bias-independent resistance r gbi , source parasitic resistance r s , and drain parasitic resistance r d ). ...
This article presents a 22.6–39.2-GHz reflectionless receiver (RX) capable of reducing the in-band intermodulation and high-order mixing products. Such RX consists of an absorptive IF amplifier, a double-balance passive mixer, and a dual-path noise-canceling low noise amplifier (LNA). The absorptive IF amplifier is utilized to suppress the out-of-band signal reflected back to mixer, which can reduce the intermodulation and high-order mixing products caused by the remixing of the out-of-band signals. Meanwhile, the dual-path noise-canceling LNA is used for achieving low noise figure (NF) in a wide frequency range. To verify the aforementioned principle, a reflectionless RX is implemented and fabricated using a conventional 28-nm CMOS technology. The measurement shows the minimum NF of 3.6 dB and the peak input third-order intercept point (IIP3) of −11.7 dBm. Besides, the proposed RX can support multi-Gb/s, 256-quadratic amplitude modulation (QAM)/1024-QAM modulation signals.
... However, it is challenging to realize a wideband LNA covering several tens of GHz because of those mutually-constraint requirements, i.e., gain, noise figure (NF), and linearity, especially in the mm-Wave domain where high loss, high leakage, and strong mutual couplings come into play. Many novel typologies have been developed to overcome those limits, such as the inductive peaking [11], [12], g m -boosting [13], [14], forward combining [15], Cascode with active load [16], [17], and coupled input matching [18]. However, performance can still be improved. ...
In this letter, a 10-43 GHz low-noise amplifier (LNA) monolithic microwave integrated circuit (MMIC) is designed in a commercial 0.15-µm GaAs E-Mode pHEMT technology. In the proposed LNA circuit, a novel coupled-line-based positive feedback structure is employed with the band-pass characteristic. By carefully tuning its coupling factor and arm length, the center frequency fc and the intensity of the feedback can be controlled, respectively. Subsequently, targeting fc at the higher cutting-edge of the working band leads to compensated gain roll-off and extended bandwidth. Incorporating 3-stage common-source architectures, an LNA prototype is fabricated with a size of 1.05 mm 2 including pads. Under 2-V VDD, good performance is obtained, including 24.6 dB peak gain with 3-dB bandwidth of 33 GHz, 2.4-3.0 dB NF, 54.5±13.8 ps group delay, and 12.3/21.5 dBm best OP1dB/OIP3. The total DC power is 110 mW.
... The models in Section 4.3 are now used to develop a measure of link rate coverage for OtI scenarios with "traditional" or Low-e glass. Table 5 defines typical parameters for the 28 GHz BS and UE representative of recent advances in state-of-the art mmWave hardware [43][44][45][46][47]. We select conservative values for these parameters to reduce the possibility of overestimating data rates, and we include an additional 5 dB of losses in . ...
Outdoor-to-indoor (OtI) signal propagation further challenges the already tight link budgets at millimeter-wave (mmWave). To gain insight into OtI mmWave scenarios at 28 GHz, we conducted an extensive measurement campaign consisting of over 2,200 link measurements. In total, 43 OtI scenarios were measured in West Harlem, New York City, covering seven highly diverse buildings. The measured OtI path gain can vary by up to 40 dB for a given link distance, and the empirical path gain model for all data shows an average of 30 dB excess loss over free space at distances beyond 50 m, with an RMS fitting error of 11.7 dB. The type of glass is found to be the single dominant feature for OtI loss, with 20 dB observed difference between empirical path gain models for scenarios with low-loss and high-loss glass. The presence of scaffolding, tree foliage, or elevated subway tracks, as well as difference in floor height are each found to have an impact between 5-10 dB. We show that for urban buildings with high-loss glass, OtI coverage can support 500 Mbps for 90% of indoor user equipment (UEs) with a base station (BS) antenna placed up to 49 m away. For buildings with low-loss glass, such as our case study covering multiple classrooms of a public school, data rates over 2.5/1.2 Gbps are possible from a BS 68/175 m away from the school building, when a line-of-sight path is available. We expect these results to be useful for the deployment of mmWave networks in dense urban environments as well as the development of relevant scheduling and beam management algorithms.
