The mixer is a critical component in wireless transceivers and plays a significant role in determining the overall system performance. Two key parameters that define the quality of a mixer are linearity and conversion gain. In receiver systems, the mixer provides a certain level of conversion gain, which simplifies the design of subsequent modules and enhances the system's noise performance and sensitivity. Linearity, on the other hand, determines the maximum signal strength the mixer can handle without distortion.
As communication systems evolve to meet higher performance demands, both downconverters in receivers and upconverters in transmitters require improved linearity. This has made the design of mixers with high gain and high linearity a major focus in the industry. In CMOS circuit design, techniques such as current multiplexing and current injection are commonly used to enhance linearity and conversion gain. However, current injection is not ideal for low-power applications since it requires a large injected current to achieve better performance, which increases power consumption.
In this paper, we present a high-gain and high-linearity mixer that leverages current multiplexing and the even harmonics of the local oscillator (LO) signal. This approach allows for better isolation and improved performance without excessive power consumption.
1. Circuit Design and Analysis
1.1 Circuit Structure
In most receiver topologies, LO signals can leak into the RF path due to parasitic capacitance or substrate coupling. This leakage can occur through two main paths: one where the LO signal appears at the intermediate frequency (IF) output, and another where it enters the low-noise amplifier (LNA) input and gets amplified before mixing. This results in DC offset and reduced dynamic range, especially in zero-IF systems.
To address this issue, the mixer described in this paper uses an even harmonic topology, as shown in Figure 2. This structure incorporates a local oscillator frequency multiplier and a current multiplexing circuit to improve port isolation, conversion gain, and linearity. The differential LO input ensures that the AC signal is shorted at node A, improving isolation. When using short-channel transistors, the differential pair generates the second harmonic of the LO signal, which is then mixed with the RF signal. The resulting IF frequency is given by fIF = |fRF - 2fLO|.
Using even harmonics of the LO eliminates signal leakage and reduces the required LO frequency, making the design simpler. An inductor LE is included to boost the amplitude of the second harmonic entering the mixer, enhancing linearity and noise performance. It also helps expand the dynamic range of the current multiplexing circuit. The IF output is connected to a source follower as an output buffer.
1.2 Current Multiplexing Circuit Analysis
The current multiplexing structure at the RF input consists of MRFP1, MRFN1, MRFP2, and MRFN2, as shown in Figure 2. The symmetric configuration allows the total transconductance to be the sum of the individual transconductances. This increases the gain of the mixer.
According to the channel length effect, the cross-coupled current expression is:
$$ I_{out} = n \cdot (V_{in} - V_t)^2 \cdot (1 + \lambda V_{DS}) $$
Here, $ n $ is the transconductance parameter, $ V_{in} $ is the input signal, $ V_{ov} = V_{GS} - V_t $ is the overdrive voltage, $ \lambda $ is the channel length modulation factor, and $ V_t $ is the threshold voltage. From this equation, the transconductance of the current multiplexing structure is the sum of the transconductances of the two transistors.
When the input signal is positive, the MRFN transistors operate in saturation while the MRFP transistors are in cutoff, effectively acting as resistors. This creates an n-channel common-source amplifier. Conversely, when the input is negative, the structure behaves like a p-channel common-source amplifier. This push-pull configuration improves the dynamic range and linearity of the circuit.
1.3 Frequency Multiplier Circuit
To further understand the principle of LO signal multiplication, we analyze the inductive frequency multiplier circuit shown in Figure 3. The leakage currents from MLON1 and MLON2 can be expressed as:
$$ I_{LON+} = n \cdot (V_{LO} - V_{TN})^2 \cdot (1 + \lambda V_{DS}) $$
$$ I_{LON-} = n \cdot (-V_{LO} - V_{TN})^2 \cdot (1 + \lambda V_{DS}) $$
The total current through the multiplexing and frequency multiplying circuits is the sum of these two components, resulting in a second harmonic signal at node VCOM. The presence of an inductor LE increases the voltage at the mixing point, enhancing the amplitude of the second harmonic signal and improving the mixer’s linearity and noise performance.
Frequency multiplier circuit
Figure 3: Frequency multiplier circuit
Insert Type Rotor,Insert Type Dc Rotor,Insertion Type Variable Frequency Rotor,Variable Frequency Dc Rotor
Suzhou Mountain Industrial Control Equipment Co., Ltd , https://www.szmountain.com