Pulse or digital communications – Transmitters
Reexamination Certificate
1999-01-26
2002-12-10
Chin, Stephen (Department: 2634)
Pulse or digital communications
Transmitters
C375S324000
Reexamination Certificate
active
06493398
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to digital communications systems in which the carrier signal is varied in accordance with bursts of binary data being transmitted. More particularly, the present invention pertains to a burst mode data communications technique in which the carrier signal is optimized for bit synchronization and immunity to impairments arising from signal multi-path, in wireless applications, and reflections, due to impedance discontinuities in wire line applications.
2. Description of the Related Art
Digital modulation is the means by which a carrier, inherently, analog in nature, is made to carry digital information in a communications channel. This involves altering the amplitude of the carrier, the angular velocity of the carrier, or both. The objective is to create discrete phase/amplitude states in a manner that leaves little chance for ambiguity between the states. These discrete states, commonly designated as symbol states, correspond, to one or more binary bits of data. By regenerating the binary data at intervals along a transmission path, re-transmitting data that is determined to be corrupt and applying error correction, digital data can be transmitted over great distances with no degradation of quality, even in the presence of a significant level of background noise. This is in stark contrast to analog modulation in which noise effects accumulate through the transmission path, usually without remedy.
In digital communications channels, noise is only one problem which must be overcome. In wire line applications impairments in cables cause impedance discontinuities, whereby resultant distortion of the signal and recovery of the digital data at the receiver becomes much more difficult. In wireless applications, multi-path is a degrading factor, especially with antennas of low, directivity. The transmitted signal traverses the most direct and desirable path to the receiving antenna. However, the signal may traverse an indirect path as well, being reflected from a buildings or other object, and also arrive at the receiving antenna. Multi-path causes a summation of such reflected signals, delayed in time, with the main signal. This also makes recovery of the digital data more difficult.
The prior art uses techniques such as training sequences, which allow equalization to be gradually varied until the equalization settings are optimized. This approach works well with continuous data streams that last for seconds or more, but does not work very well for burst mode transmissions, which may originate from multiple points. The transmitter in this case bursts out data for short periods of time and then becomes quiet to allow other transmitters a chance to send data. For efficient use of the channel, each burst of data normally contains a preamble or header which is used for synchronization and for setting the receiver voltage thresholds. Then the payload data is sent followed by a trailer to gracefully end the transmission.
When channel impairments are present, the preamble is not always lengthy enough to allow an equalizer to train effectively. Each transmitter has its own unique path to a receiver, and associated impairments, which means the equalizer must be prepared to train on every burst or packet sequence of data. If additional header data and training time is allowed for data recovery, the effective data throughput can go down considerably. Some training sequences last for more than a second, which could translate into millions of throw away header bits at the beginning of every data burst. This is clearly undesirable when the goal is to get the maximum utilization of a communications channel.
Carrier signals for digital modulation are typically based on sinusoidal waveforms because such waveforms require the least amount of bandwidth. There exists three classical forms of digital sinusoidal modulation: amplitude shift keying (ASK), frequency shift keying (FSK) , and phase shift keying (PSK). Improvements have been made in digital sinusoidal modulation, however all of the improvements have been based on the three classical techniques previously mentioned. In ASK, the amplitude of the carrier signal is varied or shifted in response to changes in the digital data. In FSK, the frequency of the carrier signal is varied or shifted in response to changes in the digital data. In PSK, the phase of the carrier signal is varied or shifted in response to changes in the digital data.
There are also certain disadvantages associated with the classical modulation techniques. For example, ASK is especially susceptible to atmospheric noise and fading. FSK requires that an associated receiver detect two discrete frequencies before the frequency can be acquired and detected. This presents delays due to the additional time required to receive the several cycles of each frequency. PSK requires complex receiver circuitry in order to detect phase changes. Furthermore, elaborate filtering is necessary to control spurious outputs resulting from the discontinuities associated with the phase changes.
One disadvantage, however, is common to all of the classical modulation techniques and their derivative improvements. This is the use of fixed time slots for varying the characteristics of the carrier signal. When fixed time slots are used, variations in the carrier signal occur at random points along the sinusoidal waveform, thus resulting in spurious frequencies and expanding of the modulation bandwidth. Complex filtering becomes necessary in order to reduce the amplitude of these spurious frequencies. As the bit rate increases, the variations in the carrier signal occur more frequently, thus posing a challenging demodulation task.
It is well known that the amount of spurious output generated by the variation of a sinusoidal waveform is dependant upon the instantaneous value of the slope of the waveform when the changer occurs. Thus, a change which occurs at exactly the midpoint or highest kinetic energy point of the waveform generates the greatest amount of spurious output because the slope value is at its maximum. If the change occurs at exactly the peak of the waveform however, the least amount of spurious output is generated because the slope value is at its minimum, i.e., zero kinetic energy.
Representation of discrete states of digital data is accomplished through various base band encoding techniques.
FIG. 2A
illustrates various known base band encoding techniques. In non-return to zero level (NRZ-Level) encoding
28
, digital code is produced by instantaneously shifting voltage levels at fixed bit time intervals so that two unique binary symbols are represented, a one and a zero. Thus, a one is represented by one level, while a zero is represented by the other level. The NRZ-Mark digital code
30
is produced by instantaneously shifting voltage levels at fixed bit intervals only when a one is transmitted and not changing levels when a zero is transmitted. The RZ digital code
32
is produced by instantaneously shifting voltage levels at half bit-time intervals when a one is transmitted and not changing levels when a zero is transmitted. A BI-PHASE-Level digital code
34
is produced by instantaneously shifting voltage levels at half bit-time intervals so that a one is a high level during the first half of the bit time and a zero is a high level during the second half of the bit time. The NRZ-4Level digital code
36
is produced by shifting voltage levels instantaneously after two, bit-time intervals. A one--one is transmitted by the top level, a one-zero is transmitted by the next lower level, a zero-one is transmitted by the next lower level, and a zero-zero is transmitted by the bottom level. By encoding two bit times into one symbol, NRZ-4Level encoding reduces the effective transitioning rate in half. Similarly, three bit times could be encoded into NRZ-8Level encoding to reduce the transitioning rate by one third. Multi-level encoding suffers from a drawback in that it necessitates a more complex receiver to detect
Chin Stephen
Kim Kevin
LandOfFree
Burst mode digital communications system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Burst mode digital communications system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Burst mode digital communications system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2996418