Telecommunications – Receiver or analog modulated signal frequency converter – Frequency modifying or conversion
Reexamination Certificate
2002-09-30
2004-10-26
Urban, Edward F. (Department: 2685)
Telecommunications
Receiver or analog modulated signal frequency converter
Frequency modifying or conversion
C455S333000, C455S323000, C327S355000
Reexamination Certificate
active
06810242
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to frequency mixers and, more particularly, to subharmonic mixers.
2. Related Art
A key principle of a frequency mixer is that, in mixing multiple electrical signals together, it adds and subtracts frequencies to produce new frequencies. In the field of signal processing, the process of multiplication in the time domain is recognized as equivalent to the process of convolution in the frequency domain. Mixers produce distortion or multiplication products that reduce or diminish the quality of the output signal. Much of the art and science of making good use of multiplication in mixing goes into minimizing these unwanted multiplication products (or their effects) and making multipliers provide their frequency translations as efficiently as possible.
Mixers can also create nonlinear distortion. Nonlinear distortion may take the form of harmonic distortion, in which integer multiples of input frequencies occur, or intermodulation distortion (IMD), in which different components multiply to form new components. Any departure from absolute linearity results in some form of nonlinear distortion.
Standard mixer design involves significantly nonlinear multiplication. Typically, the switching operation of a mixer causes the local oscillator signal (“LO”) to act effectively as a square wave. There are several advantages to such switching action, including reduced noise, improved gain, insensitivity to device mismatch and variation, insensitivity to exact LO strength, and simplified design. A disadvantage, however, is that odd-order mixing products (“OMPs”) are generated. An OMP is generally defined as the product of one input and an odd harmonic of another input.
The development of advanced receiver architectures, especially a direct conversion receiver (DCR), are greatly aided by use of a subharmonic mixer (SHM). While generally performing better than standard mixers in certain parameters important to DCR, a SHM generally under-performs in at least some of the more standard figures of merit (i.e. noise-figure (NF)) as compared to standard mixers.
The reason for the interest in a SHM for a DCR is due to the finite isolation between the radio frequency (RF) and LO paths. The problems of LO self-mixing and blocking signal coupling exist with a SHM. The undesired coupling of large blocking signals present in the RF path to the LO path cause direct current (DC) offsets at the output of the baseband signal. LO self-mixing and unmodulated blocker coupling cause DC offsets, which corrupt data in some modulation schemes, such as the modulation scheme used in Global System for Mobile communications (GSM).
Since an SHM uses a frequency that is one-half (½) of the frequency of the desired RF signal, the problem of parasitic coupling in a receiver using an SHM is greatly reduced. Ignoring second order effects, coupling of large blocking signals to the LO path no longer effects mixer operation. This is the main reason for the large interest in an SHM, even though they tend to be noiser than traditional mixers.
Two known techniques for building subharmonic mixers and especially subharmonic mixers that operate at one-half (½) the LO have several weaknesses. The best-known technique is that described in the paper “A Wide-Bandwidth Si/SiGe HBT Direct Conversion Sub-Harmonic Mixer/Downconverter,” L. Sheng et al, IEEE Journal of Solid-State Circuits, Vol. 35, No. 9, September 2000. This technique involves stacking two standard, double-balanced mixer cores driven by ½ LO signals split 90 degrees. Two sets of hard-switched bipolar transistor pairs driven by ½ LO signals split by 90 degrees provide a functional equivalent of a downconverter that can be represented by the equation: BB=RF*sin(f
lo
*t)*cos(f
lo
*t)=RF*sin(2*f
lo
*t), where BB is a baseband signal, RF
in
is the radio frequency input signal, and f
lo
is the local oscillator frequency. This method has several problems as compared to standard (Gilbert-style) mixers. One problem is that the use of two stacked cores requires significantly more headroom than a single core. Headroom is the amount of additional signal above the nominal input level that can be sent into or out of an electronic device before clipping distortion occurs. This not only means that the core requires extra headroom, but that the LO driving it is limited to avoid driving devices into saturation. Another problem is that the bottom stack's common-emitter node oscillates at 2*f
lo
, the receive frequency. This oscillating results in undesired mixing of large blocking signals and may desensitize the radio as a result. In addition, during the double mix-down that occurs, the receive frequency is mixed to a frequency near the LO, raising the possibility of IIP2 degradation. IIP2 is the theoretical input level at which the second-order two-tone distortion products are equal in power to the desired signals.
Another technique for building subharmonic mixers is described in U.S. Pat. No. 6,370,372 to Molnar et al., which is assigned to the assignee of the present invention and herein incorporated by reference. The technique described in this patent overcomes the above mentioned deficiencies by the use of a single four-way comparison SHM driven by carefully constructed stair waves which provide hard switching. This structure requires only one layer of switching core and its performance hinges on the generation of the stair waves. Four transistors driven by four 90-degree split stair waves guarantee only one device is on at a time, each for ¼ of an LO cycle. By summing the 0, 180 degree outputs and the 90, 270 degree outputs, one gets effective mixing at two times the LO frequency. One problem is that the mixer loads each stair wave asymmetrically (only the top ¼ of the signal actually drives a transistor that is on). This means that the stair waves tend to generate some 2
nd
harmonic distortion. Another problem is that the stair wave requires twice as much headroom as is needed to switch the mixer core. As a result, headroom is limited by the drivers, making it harder to get very large swing and fast switching in the mixer core. A third problem is that the LO requires harmonic content (3
rd
order harmonic especially) to operate as a good stair wave, so inductance/capacitance (LC) tuning is not an option. Finally, to ensure hard switching without excessive swing, the drive impedance should preferrably be relatively low. All this contributes to a less than optimal NF and high power consumption in the LO drivers.
Therefore, there is a need for an SHM that includes the benefits of this type of mixer, while overcoming its problems.
SUMMARY
A subharmonic mixer and a method of downconverting a received radio frequency signal is described. The subharmonic mixer of the present invention uses two stacks of switching cores with high order symmetry to reduce unwanted harmonic generation and uses complementary metal-oxide semiconductor (CMOS) transistors to improve headroom.
In one embodiment, the subharmonic mixer includes a local oscillator interface, a first switching stage, a second switching stage, and a baseband output. The local oscillator interface receives a local oscillator signal. The local oscillator signal includes a waveform with four equally spaced phase components. The first switching stage receives an input current and the local oscillator signal to supply four intermediate currents. Each intermediate current is responsive to one of the phase components and the input current. The second switching stage receives the intermediate currents and the local oscillator signal to supply eight baseband currents. Each baseband current is responsive to one of the two phase components adjacent to the phase component used to generate the corresponding intermediate current. The baseband output is generated by summing the baseband currents.
In another embodiment, a method includes receiving an RF signal, providing four equally phase shifted local oscillator signa
Hatcher Geoffrey
Magoon Rahul
Molnar Alyosha C.
Jackson Blane J.
Skyworks Solutions Inc.
Thomas Kayden Horstemeyer & Risley LLP
Urban Edward F.
LandOfFree
Subharmonic mixer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Subharmonic mixer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Subharmonic mixer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3326807