Telecommunications – Receiver or analog modulated signal frequency converter – Noise or interference elimination
Reexamination Certificate
1998-09-15
2001-05-01
Urban, Edward F. (Department: 2683)
Telecommunications
Receiver or analog modulated signal frequency converter
Noise or interference elimination
C455S326000, C455S333000, C455S311000, C455S307000
Reexamination Certificate
active
06226509
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates, in general, to a circuit and operating method for image rejection and also to the configuration of an image reject mixer. More particularly, but not exclusively, the invention is applicable to integrated circuit technology for a radio communication device, such as a Digital Enhanced Cordless Telecommunication (DECT) chip-set.
SUMMARY OF THE PRIOR ART
With an increasing demand by the general public for smaller and more ergonomic designs, telecommunication equipment (and particularly subscriber/handset) manufacturers have sought higher levels of functional integration within their respective integrated circuit (IC) designs. Indeed, subscriber terminals, for example, can now be realized through the use of as little as two ICs (which are sometimes referred to as “chips”). However, in order to achieve this level of integration, manufacturers' have had to develop improved circuitry that consumes relatively low amounts of power (typically driven by a voltage at or below approximately 3.3 volts) and which circuitry is optimized in terms of both configuration and power dissipation.
As will be understood, the term “subscriber terminal” is generic and encompasses transceiver, transmitter and receiver equipment, while such equipment can operate in wireline, radio frequency (RF) or optical environments. Indeed, the various architectures referred to above can be realized using either analog or digital techniques.
As will be readily understood in relation to general communication theory, a carrier signal is in some way modulated by an information-bearing signal in order to support information transfer. In a radio frequency environment, for example, many different forms of modulation exist, such as Gaussian minimum shift keying (GMSK) that utilizes phase modulation (through in-phase and quadrature components) or orthogonal frequency division multiplexing (OFDM) in which a composite signal envelope is created from a multiplicity of sub-channel carriers using, for example, frequency modulation techniques. Modulation of the information-bearing signal onto the carrier signal has the effect of producing a channel having a finite bandwidth that is related to the rate of transmission of information according to Nyquist's theorem. In other words, a modulated radio signal contains its information in a range of frequencies distributed about a central carrier (or channel) frequency.
In all cases, the carrier signal is within a stipulated frequency range for a particular form of transmission, e.g. downlink and uplink transmissions in the Global System for Mobile (GSM) communication each have an allocated band within the range of 935 MHz to 960 MHz while the uplink and downlink bands support respective duplex channel pairs that are separated by a 45 MHz spacing.
With respect to the reception and subsequent demodulation of an RF signal, for example, there are presently two lines of thought. First, the RF signal can be directly mixed down to a zero intermediate frequency (IF) where the carrier signal is located at dc. However, any dc offsets (and, in particular, time varying dc offsets) that occur at the output of a mixer appear as part of the signal, such dc offsets therefore have the unwanted effect of corrupting data integrity. Unfortunately, these dc offsets cannot be filtered out without removing wanted information in the RF signal. Consequently, the sensitivity of a receiver is limited by the level of the dc offsets.
Mixing straight down to dc, however, has many advantages since both filtering with low pass filters at dc and digitizing the signal is much easier than at higher frequencies.
The accepted but alternative line of thought with respect to information recovery from a modulated carrier involves the generation and use of an intermediate frequency (IF) signal from the RF signal; the IF signal, whilst being at reduced frequency relative to the carrier, still has a relatively large frequency displacement with respect to baseband (dc). The problem with the use of such an intermediate frequency is that the wanted signal at the relatively low IF can be very easily confused with its image signal. More specifically, prior to further down mixing to baseband, the wanted signal sits above the local oscillator by the relatively low intermediate frequency, whilst an image signal sits below the local oscillator by the same amount. On down mixing, in general, the image frequency is translated to a negative frequency, which negative frequency can only be distinguished from a positive frequency using a quadrature mixer in which the degree of resolution is limited by the accuracy of the quadrature mixer. Consequently, the use of an intermediate frequency can result in the incorrect interpretation of data, while elimination of any associated uncertainty is achievable at the cost of the provision of a more accurate quadrature mixer.
For complete understanding, it should be appreciated that the term “negative frequency” relates to the direction of rotation of a phasor (vector) used to generate a time domain data from frequency domain data, while it will be understood that the term “frequency” is the rate of change of phase. More particularly, once the initial phase and amplitude of a carrier are determined, the sign of the frequency determines whether the phasor rotates in a clockwise or counter-clockwise rotation, and hence whether the frequency is positive or negative.
In a communication system, the amplitude of the wanted frequency that is recovered is often relatively small (as a consequence of normal propagation losses), while a received interferer at the image frequency has significantly larger amplitude. In this respect and as will be appreciated, the choice of IF is critical in coping with the presence of such an interferer.
Specifically, if a zero IF is used then the wanted signal is its own image. Since the image and the wanted are therefore at the same power level it is only necessary to reject the image by enough to ensure the wanted signal can be cleanly demodulated. Quadrature mixers offering thirty decibels (30 dB) of image rejection are quite sufficient for this.
If a higher IF is used, then the wanted and image signals may be separated by sufficient frequency (before mixing) to allow the image signal to be attenuated by filtering before mixing. In other words, to allow effective filtering of the image signal, present systems select the IF signal to ensure that the difference frequency (or 2*IF) facilitates filtering, i.e. a distinguishable frequency separation exists between the image signal and the frequency envelope of the wanted signal. In fact, to ensure the filtering can be achieved, the IF needs to be a reasonable fraction of a frequency of the local oscillator (LO). In practice, this means that for digital radios with RF frequencies around 1 GHz (gigaHertz), the IF is currently of such a high frequency that it cannot be simply digitized or filtered. Clearly, the sum component of the IF and the RF signal is easily filterable since it is above the RF band.
If the IF is too low, then it is impossible to filter the image whilst the signals are at RF. Thus, the image signal must be restricted to be no greater than the level which an image reject mixer can sufficiently suppress. Clearly, however, the level of suppression is somewhat dependent upon the configuration and type of components (and the manufacturing techniques) employed to construct the image rejection mixer and/or filter and so improved performance can usually be attained through increased manufacturing costs. However, such increases in cost may render a product commercially non-viable and so performance of the product is therefore often compromised.
In digital radio systems, it is only those high amplitude signals (i.e. high level interferers) that are close to the wanted signal that cannot be rejected. Moreover, even in such digital systems, imperfections in conventional filtering and phase noise in local oscillators prevent effective rejection of high level interferers.
Babla Chettan
Mole Peter John
Van Vliet Henric
Lee Mann Smith McWilliams Sweeney & Ohlson
Nortel Networks Limited
Perez-Gutierrez Rafael
Urban Edward F.
LandOfFree
Image reject mixer, circuit, and method for image rejection does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image reject mixer, circuit, and method for image rejection, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image reject mixer, circuit, and method for image rejection will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2510218