Image-reject mixers and single-sideband mixers play a key role in many of todays microwave and RF systems [1, 2, and 3]. IRMs and SSMs reduce system cost and complexity by removing the need for expensive pre-selection, and one or more stages of up- or down-conversion. IRMs simplify down-conversion by employing phase-cancellation techniques to separate the down-converted products resulting from the undesired image and desired RF inputs. Similarly, SSM simplify up-conversion by separating the up-converted lower sideband (LSB) from the up-converted upper sideband (USB). In both IRMs and SSMs, two mixing products are separated and channelized into two different output ports to be further processed or terminated. This article provides a working knowledge of present IRM and SSM technology. It gives an overview of what these devices do, how they operate, and some practical performance considerations. In addition, two appendices are given: one that provides a simplified analysis procedure for evaluating quadrature mixer circuits, and another that correlates image rejection and sideband suppression with circuit parameters.
Get your grade
or your money back
using our Essay Writing Service!
Keywords- IRM, SSM,
The image is an unwanted input signal to the mixer. Its frequency is above or below the local oscillator (LO) frequency by an amount equal to the IF frequency. The image and desired inputs both mix with the LO and down-convert to the same frequency. This poses a problem in conventional DB (double-balanced) mixers because the two down-converted products interfere with each other, since they exit at the IF port together. IRMs avoid this problem by channelizing the two products into separate output ports.
Conventional double-balanced mixers use filters to block the image from entering the mixer, so that no down-converted image is allowed to be generated by the mixer. Since the desired and image signals are always separated in frequency by twice the IF frequency, the IF frequency must be high enough to allow the pre-selector in front of the mixer to block the image, but still allow the desired if signal to enter the mixer. As the IF frequency is reduced, the desired and image signals move closer together in frequency, forcing the selectivity of the pre-selector to increase in order to separate the two adjacent input signals. Pre-selector complexity also increases for tuneable receivers because the pre-selector must track with the LO frequency, to maintain the normally constant IF output frequency. Also since the IF frequency must be relatively high to simplify pre-selection, a number of down-conversion stages are required to down-convert the RF input to the baseband frequency for detection. In comparison to conventional DB mixers, IRMs achieve image-rejection through phase cancellation, not filtering, so the frequency spacing between the image and desired inputs can be negligible. This means that down-conversion can be accomplished without pre-selection, and in fewer stages, saving the cost of extra mixers, amplifiers, local oscillators, and fitters. For similar reasons, up-conversion can also be simplified by using single-sideband mixers.
II. What IRMS and SSMS Do
Figure 2 shows the circuit configuration used for image-reject mixers and single-sideband mixers. The only differences between them are their respective applications and parameters.
Figure 1: IRM Application
IRM: Figure 1 shows how the circuit of Figure 2 is operated as an IRM. The signal at fR1 will down-convert to exit at I1 , and the signal at fR2 will down-convert to exit at I2. If fR1 is the desired signal, then fR2 is its image. Ideally, none of the down-converted image signal exits the desired IF output port. However, since amplitude and phase imbalances exist in practical circuits, some of the down-converted image will be present at the desired IF output port. Image rejection is defined as the ratio of the down-converted image signal power exiting the desired IF port, to that of the desired signal, exiting the same IF output port. For example, if the down-converted image and desired signal levels at I1 are -30 dBm and -10 dBm respectively, then the image rejection is 20 dB. Good image rejection requires close amplitude and phase matching, low mixer VSWR, and directivity.
Figure 2: Block diagram of Image Rejection and Single Side Band Mixer.
Figure 3: SSM Application.
SSM: Figure 3 shows how the circuit of Figure 2 is operated as a single-sideband mixer. The SSM provides a single-sideband suppressed carrier output. A LSB or USB output can be selected by choosing which I port to drive with the IF signal. An IF into I1, results in an LSB output, and an input into I2, results in a USB output. SSMs have two main parameters: sideband suppression and carrier suppression. Sideband suppression is analogous to image rejection, and is defined as the ratio of the undesired sideband signal power to that of the desired sideband signal power at the IF output port. Carrier suppression is a measure of how much of the carrier signal leaks through the SSM to become present at the RF output, and is defined as the ratio of the carrier-power level at the output port to that of the desired.
Always on Time
Marked to Standard
III. How IRMS and SSMS Operates
In any mixer, the phase angles of its RF and LO input signals are conserved through the mixing process, so that the phase of the IF output equals the sum of the IF and LO input phase angles, multiplied by their respective harmonic coefficients, m and n. These coefficients define the inter-modulation products exiting the mixer fIM = mfR + nfL, where m and n are positive or negative integers, for the desired and image down-converted products, m and n equal ±1, referring to Figure 1, if the frequency of the down-converted desired signal is fIF = fL - fR1, then n = 1 and m = -1, and its phase angle will equal (θL - θR1), where θL and θR1 are the phase angles of the LO and RF inputs, respectively. Similarly, the frequency of the down-converted image signal is fIM = fR2 - fL so that m = 1 and n = -1, and its phase angle equals (θR2 - θL).Figure 2 show that both IRMs and SSMs comprise two mixers, two quadrature power dividers, and one in-phase power divider. These are all passive devices, and can act together to significantly enhance system cost effectiveness, performance, and reliability. Mixers M1 and M2 have IF output currents, I1' and I2', respectively. The phase angles of I1' and I2' are 0° and 90°, respectively. For both mixers, (qL is set equal to zero because the LO is applied in-phase to M1 and M2). Also, since the IF inputs to M1 and M2 are in quadrature; i.e., 90° out of phase with respect to each other, (qR for M1 is set equal to zero), and (qR for M2 is set equal to 90°). Hence, I1' = Imn<∠0° and I2' = Imn∠<m90°. Imn is the same for M1 and M2 because the two mixers are assumed to have matching conversion-loss characteristics.
I1' and I2' combine in the output quadrature power divider in such a way as to channelize the (fL - fR1) product into output port I1, and the (fR2 - fL) product into output port I2. When down-converting, one product is taken to be the desired output, and the other is taken to be the image output, which is terminated.
When up-converting, I1 and R1 are interchanged, as are I2 and R2, so that the inputs to the mixer are a low-frequency signal injected at I1 and I2, and a microwave carrier injected at the LO port. The outputs are the LSB (fL - fI1) product that exits at R1, and the USB (fL + fI2) product that exits at R2.
IV. Practical Performance Consideration
The conversion loss of an IRM includes the losses due to the quadrature hybrids and in phase power splitter, in addition to the mixer conversion loss. This additional circuitry increases the conversion loss, but not to unacceptable levels. Typical conversion loss is 8.0 dB from 8 to 18 GHz.
The amount of image rejection obtained with an IRM is determined by the circuit amplitude and phase balance. Since circuitry imbalances are frequency dependent, image rejection is also frequency dependent.
Inter-modulation products are more critical for the SSM, since there are several spurious products close to the desired output . Suppression of the carrier signal, at frequency fc, is also important.
A high-level fc signal provides good inter-modulation suppression, but poor carrier suppression; whereas, a high-level fIF signal provides good carrier suppression at the expense of reduced inter-modulation suppression. The carrier suppression is determined by the mixer L-R isolation.
V. APPENDIX A: SIMPLIFIED ANALYSIS OF QUADRATURE MIXERS
This analysis shows which mixing products will exit the various ports of a quadrature mixer, but without the mathematical simplifications included here. The approach is to determine the Fourier series for the current in each mixer diode, then sum these currents to determine which mixing products exit the various ports. For example, the IRM of Figure 4 is analyzed. The current in each diode is assumed to flow from anode to cathode, and is written as a double Fourier series:
This double series results from multiplying the diode conductance waveform, which is governed by the LO signal, by the waveform for the voltage across the diode, which is governed by the IF signal. The amplitude portion
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
of the Fourier series can be reduced to Knm, or K for short, since we are only concerned with phase.
The phase angle θ corresponds to the difference in phase between the LO input and each of the diode currents. The phase angle ¸ corresponds to the difference in phase between the RF input and each of the diode currents. Four assumptions are made in this analysis:
1. Perfect circuit balance and perfect quadrature couplers
2. Identical diodes
3. Large-signal LO.
4. Small-signal RF.
The current in diode 1 can be written as:
Φ equals À because the if signal is 180° out of phase with the assumed direction of current flow in diode 1 (anode-to-cathode) ¸ equals zero because the LO signal is in phase with the current-flow in diode 1. The current in diode 2 can be written as:
Both Φ and ¸ equal zero because the RF and LO inputs are in phase with the current-flow in diode 2. The current in diode 3 can be written as:
Φ equals (À/2 + À): The À/2 comes from the RF quadrature-hybrid, and the ¸ comes from the current flow in diode 3 being 180° out of phase with the RF signal exiting the hybrid. The current in diode 4 can be written as:
Φ equals À/2 because of the RF quadrature hybrid, and ¸ equals À because the LO signal is 180° out of phase with the current flow in diode 4.
Once the four individual diode currents have been determined, they can be combined to form the IF outputs at I1 and I2. The current exiting I1 can be written as:
Currents i3 and i4 are multiplied by j because of the 90° phase shift in the IF quadrature coupler. Currents i2 and i4 are negative because they are entering (instead of exiting) at the node connecting the diodes to the IF coupler. Similarly, the current exiting I2 can be written as:
Notice first that the R±L products exit at I1, and the L-R product exits at I2 Also, notice that every other odd product exits I1 and I2. Mixing products (±L+R), (±L+5R), (±L+9R), etc. and (L-3R), (L-7R), (L-11R), all exit at I1. And mixing products (L-R), (L-5R), (L-9R), etc., and (L+3R), (L+7R), etc., all exit at I2. Finally, notice that the products exiting at I1 and I2 are always in quadrature with each other. When analyzing IM suppression, the bandwidth of the output port must be the proceeding analysis can be used to quickly analyze mixer/quadrature-hybrid networks to determine which products will exit the mixer ports. The phase angle of each diode current can be written in its final form in terms of jm, jn, (-1)m, (-1)n by inspection, and then summed.
VI. APPENDIX B: IMAGE REJECTION AS AFUNCTION OF AMPLITUDE AND PHASE MATCH
This analysis shows the relationship between image rejection and amplitude phase imbalances . Image rejection is defined as the ratio of the magnitude of the image signal and the desired signal. Therefore, the image rejection at I1 in Figure 2 is:
Figure 4: Schematic diagram of the image-reject mixer
From equation (1) |I1'| = |I2'| = Imn; using this and rewriting equation (1), we obtain:
From equation (4) we obtain the following equations for I1 (m=+1) and I1 (m=-l)
For practical applications, I1' and I2' are not exactly amplitude and phase matched. If an amplitude imbalance factor of A and a phase imbalance factor of θ are included in equations (5A) and (5B), we obtain:
The factor A is equal to the sum of the individual amplitude imbalances in the RF and IF hybrids and the two mixers. The factor θ is the total phase imbalance which is due to the sum of the deviation from quadrature in the RF and IF hybrids, and the phase imbalance of the two mixers.
A and θ also include the effects of hybrid directivity and impedance mismatches between the hybrids and the mixers. Imperfect hybrid directivity causes additional phase errors, and impedance mismatches causes amplitude ripple .
Substituting equations (5A) and (5B) into (3) results in the following equation for image rejection as a function of A and θ:
The effect of A and q on image rejection is illustrated in Figure 5 .
Example: If the if hybrid amplitude imbalance is +0.5 dB, the IF hybrid amplitude imbalance is +0.5 dB, and the mixer amplitude match is -0.5 dB, the total amplitude imbalance is 0.5 + 0.5 - 0.5 = 0.5 dB. If the total phase imbalance is 10 degrees, the image rejection is 20.7 dB. This estimate of the image rejection is optimistic, since it does not include the effects of VSWR and imperfect hybrid directivity
Figure 5: Image rejection vs. amplitude and phase imbalance.
In summary, image-reject and single' sideband mixers provide a valuable means of solving difficult system problems posed by conventional double-balanced mixers. Using phase cancellation instead of filtering for image rejection and sideband suppression, fewer expensive components, such as mixers, VCOs, and amplifiers are required. This means that reliability is increased and cost is reduced. The theory and operation of IRMs and SSMs has been discussed, and key parameters have been defined. The tradeoffs between sideband, inter-modulation and carrier suppression for up-converter applications are outlined and practical design guidelines given. IRMs and SSMs are increasingly solving key system problems.